Showing posts with label Boceprevir FDA Transcript. Show all posts
Showing posts with label Boceprevir FDA Transcript. Show all posts

FDA Transcripts Telaprevir/Boceprevir April Advisory Committee Meeting

Telaprevir


  • Transcript for the April 28, 2011 Meeting of the Antiviral Drugs Advisory Committee7  (PDF - 625KB)



  • FOOD AND DRUG ADMINISTRATION
    CENTER FOR DRUG EVALUATION AND RESEARCH (CDER)
    ANTIVIRAL DRUGS ADVISORY COMMITTEE (ADVAC)

    Thursday, April 28, 2011
    Today's agenda involves the discussions of new drug application, NDA 201-917, telaprevir, manufactured by Vertex, with a proposed indication for the treatment of chronic hepatitis C genotype 1 infection in combination with peginterferon and ribavirin in adult patients with compensated liver disease who are previously untreated or who have failed previous therapy.

    This is a particular matters meeting during which specific matters related to Vertex's telaprevir will be discussed. Based on the agenda for today's meeting and all financial interests reported by the committee members and temporary voting members, no conflict of interest waivers have been issued in connection with this meeting.

    We will now proceed with the FDA opening remarks from Dr. Debra Birnkrant. I would like to remind public observers at this meeting that while this meeting is open for public observation, public attendees may not participate except at the specific request of the panel.
    Introduction/Background

    DR. BIRNKRANT: Well, good morning, everyone. This is not exactly the right slide set for today, but it's the same theme as yesterday's. There should be a second set of slides. If not, I can just speak from notes.
    MR. TRAN: I apologize.

    DR. BIRNKRANT: Okay. No problem.
    So today we'll be discussing the telaprevir NDA that was submitted under 201-917 by Vertex Pharmaceuticals. Telaprevir is an NS-34A protease inhibitor that was studied in combination with pegylated interferon and ribavirin in multiple populations. The trials assessed response-guided therapy, or RGT. When added to standard of care, higher SVR rates were seen.
    In addition, there were higher SVR rates seen in difficult-to-treat populations. However, their representation was limited, as you will see. There were additional toxicities when telaprevir was added to standard of care, that is, pegylated interferon and ribavirin, but these toxicities can be managed and monitored.
    The FDA presentation will highlight the following. The clinical program will be summarized, including treatment duration in various populations. There will also be a limited resistance assessment, and there will be a limited discussion on IL28B. The second half of FDA's presentation will focus on safety.

    Dr. Jadhav from our pharmacometrics group will discuss the premise that previous exposure to pegylated interferon and ribavirin appears to be less important than early virologic response in relation to RGT for prior relapsers.
    Well, the positive thing about not having my slides here is that you don't have to look at the pipeline again.
    [Laughter.]

    DR. BIRNKRANT: But I will say it is still full. And it's exciting today that we'll be talking about one of the direct-acting antivirals that is in the pipeline.
    Our questions to the committee this afternoon relate to the safety profile of telaprevir, focusing on rash and anemia. The second question to the committee will be related to the risk-benefit of telaprevir, and that will be when the committee votes. Then there will be discussions related to efficacy and treatment duration in certain populations. And as before, we will ask about additional studies that need to be conducted.
    The agenda is as follows. Following my remarks, Vertex Pharmaceuticals will present their data. This will be followed by a questioning period. After the break, FDA will make their presentation. This will also be followed by a period of questions for clarification purposes.
    Lunch will take place around noon. This will be followed by an open public hearing at 1:00. Following the open public hearing, we'll have the
    charge to the committee where we'll review the questions again. And then the committee can question or raise questions of the sponsor and the FDA. And then we'll go through the questions, take a vote, and conclude the meeting.
    Thank you very much.

    sponsor presentations

    DR. CARGILL: Thank you.
    We will now proceed with the sponsor presentations. I would like to remind public observers at this meeting that while this meeting is open for public observation, public attendees may not participate except at the specific request of the panel.
    Both the Food and Drug Administration and the public believe in a transparent process for information-gathering and decision-making. To ensure such transparency at the Advisory Committee meeting, FDA believes that it is important to understand the context of an individual's presentation.
    For this reason, FDA encourages all participants, including the sponsor's non-employee presenters, to advise the committee of any financial relationships that they may have with the firm at issue, such as consulting fees, travel expenses, honoraria, and interest in the sponsor, including equity interests and those based upon the outcome of the meeting.
    Likewise, FDA encourages you at the beginning of your presentation to advise the committee if you do not have any such financial relationships. If you choose not to address this issue of financial relationships at the beginning of your presentation, it will not preclude you from speaking.
    Thank you.

    Sponsor Presentation – Robert Kauffman
    DR. KAUFFMAN: Thank you, Dr. Cargill.
    Dr. Cox, Dr. Birnkrant, committee members, FDA staff, and guests, good morning. I'm Bob Kauffman, chief medical officer at Vertex Pharmaceuticals. On behalf of Vertex, it's a privilege to prevent the NDA for telaprevir for your consideration.

    As you know, telaprevir is an inhibitor of the HCV protease. The indication under consideration this morning is as follows. Telaprevir is indicated, in combination with pegylated interferon and ribavirin, for the treatment of genotype 1 chronic hepatitis C in adult patients with compensated liver disease, including cirrhosis.
    The indication includes patients who are treatment-naiςve and those who have been previously treated. These include prior null responders, partial responders, and relapsers.
    Telaprevir has been in clinical development since 2004. Its extensive development program comprised more than 40 studies that are included in the NDA. Phase 3 development started in 2008, following a clinical advice meeting with FDA. We completed the rolling NDA submission in November of 2010. Notably, the safety database for telaprevir includes nearly 4,000 patients.
    As you will hear today, hepatitis C is a significant public health problem. There is an unmet need for treatment options that can increase sustained virologic response rates, or SVR, and reduce the treatment burden. Today we will present results of our comprehensive clinical development program in hepatitis C. Our data show that adding telaprevir to pegylated interferon and ribavirin produces substantial clinical benefit compared to the current treatment. This is exemplified by an SVR rate of 79 percent in treatment-naive patients.
    Telaprevir's efficacy advantage is evident across subpopulations, including patients who have not achieved an SVR with prior therapy. Telaprevir also reduces treatment burden by reducing the treatment duration for up to two-thirds of treatment-naive patients, and our data show that patients with prior relapse can also benefit from response-guided therapy. The safety profile of telaprevir has been well characterized, and in our trials we implemented practical management strategies to protect patients.
    Following this introduction, Dr. Ira Jacobson will present a brief background on the burden of disease and the current treatment landscape of HCV. Then I'll provide an overview of the telaprevir clinical development program, focusing on early development.
    Dr. Shelley George will discuss efficacy data from our Phase 3 program, followed by Dr. Priya Singhal, who will present the safety data. Then finally, I'll come back to the podium to summarize the benefit-risk assessment for telaprevir and offer some concluding remarks.
    Following the presentation, we will be available to answer the committee's questions. In addition to Dr. Jacobson, we also have Dr. Robert Stern from Harvard with us to address questions as well.
    Now I'd like to turn the podium over to Dr. Jacobson to discuss the current treatment landscape of HCV. Dr. Jacobson?
    Sponsor Presentation – Ira Jacobson

    Dr. Jacobson
    DR. JACOBSON: Thank you, Dr. Kauffman.
    Good morning. It's a privilege to appear before you today. I wish to disclose that I'm a consultant and investigator for Vertex and have a similar relationship with Merck.
    Following yesterday's excellent review by Drs. Birnkrant and Murray, I'd like to take just a few moments to provide a clinician's perspective on hepatitis C, a major public health problem for which there is a pressing need for improved treatment.
    One hundred seventy million people globally are estimated to be infected with hepatitis C virus, or HCV. Three to four million people become infected annually, including tens of thousands in the United States. Genotype 1 HCV, the most difficult to treat, accounts for nearly three-quarters of all infections in the United States.
    The magnitude of the public health problem posed by HCV is underscored by how much more common it is than HIV and hepatitis B virus in the United States. According to the recent report from the Institute of Medicine of the National Academy of Sciences, up to 75 percent of the 3 to 4 million Americans with hepatitis C have not even been diagnosed yet. This presents a major public health challenge that needs to be addressed in parallel with the introduction of increasingly effective therapies.
    Among the 70 to 75 percent of people who develop chronicity after acute infection, 10 to 20 percent develop cirrhosis after 20 years, and the incidence of cirrhosis continues to increase over the ensuing decades. Hepatitis C patients with advanced scarring, especially cirrhosis, have to live with the ever-present risk of liver cancer which may occur even in other well-compensated liver disease, along with the possibility of future decompensation and life-threatening complications.
    Beyond the emotional burden imposed by the knowledge that one has hepatitis C, I'd like to share with you the problems faced by patients with cirrhosis. They frequently suffer from symptoms such as fatigue, weight loss, depression, muscle wasting, and impaired cognition. These symptoms decrease quality of life and may be disabling.
    The most catastrophic complications of
    advanced liver diseases include gastrointestinal bleeding, ascites, bacterial infections, encephalopathy, and liver cancer. Hepatitis C is the most frequent indication for liver transplantation in the United States. Although transplantation can be life-saving, recurrent hepatitis C in a transplanted liver may be associated with a rapidly progressive course.

    Finally, as shown on the lower right of the slide, the mortality attributable to HCV in the United States is already substantial and expected to grow. According to a recent study, the prevalence of decompensated cirrhosis associated with hepatitis C in the United States began to increase after 1990, as shown on the left. In the absence of more effective therapy, the number of cases of decompensated cirrhosis is expected to increase to about 145,000 cases annually by 2020.
    Similarly, the incidence of liver cancer in patients with hepatitis C and advanced fibrosis also began to increase after 1990. Should the risk of liver cancer in people with fibrosis remain unchanged, the incidence of hepatitis C-related liver cancer is projected to peak in 2019 at about 14,000 cases per year.

    A study published just this month from the VA HCV database affirms the increasing prevalence of cirrhosis and its complications among hepatitis C patients. In this database, perhaps the largest HCV database in the world, the prevalence of cirrhosis and decompensated cirrhosis more than doubled over a 10-year period, and the prevalence of liver cancer increased 19-fold. The rise in these prevalence rates remained highly significant after adjustment for gender and increasing age of the HCV cohort.
    Of all cancers, liver cancer has the fastest-growing death rate in the United States, and the preponderance of this has been attributed to hepatitis C. From a personal perspective, hardly a week goes by in my practice without the detection of at least one new case of liver cancer in hepatitis C patients with advanced liver scarring who undergo regular screening.

    At our weekly tumor boards, an alarming number of patients with liver cancer, the majority from hepatitis C, are presented to the team of specialists needed to provide optimal care for this often lethal condition.
    Hepatitis C imposes a significant extrahepatic burden in addition to its better-known effects on the liver. For example, hepatitis C is associated with an increased risk of adult onset diabetes. This may be related to the inflammatory response induced by HCV, featuring up-regulation of cytokines which promote insulin resistance. A recent study showed a lower risk of subsequent diabetes in patients with sustained virologic response, or SVR, than in non-responders to a course of therapy.

    Serious B cell proliferative disorders, including cryoglobulinemia and non-Hodgkin's lymphoma, arise more frequently in patients with hepatitis C. These B cell disorders may remit after HCV is eradicated. Patients with hepatitis C may suffer from diminished quality of life related
    to such factors as depression and cognitive impairment, along with inflammatory disorders, including arthritis and Sjogren's syndrome.
    These observations underscore that hepatitis C should be regarded as a systemic disorder. Successful treatment may reduce the impact of the extrahepatic manifestations of hepatitis C virus infection in addition to the burden of liver disease.
    The ultimate successful treatment of a viral infection is viral eradication or cure. The biologic features of HCV, unlike those of HIV and hepatitis B, do indeed make it potentially curable. As shown on this slide, both HIV and HBV are able to archive or embed their genomes in forms that are not directly susceptible to currently available therapeutic agents, which effectively suppress viral replication but do not directly target archived viral genomes.

    In contrast, there is no known archival form of the HCV genome. The implication of this is the ability to eradicate HCV with a finite course of
    therapy. Consistent with this, over 99 percent of patients who have a sustained virologic response six months after completion of therapy remain free of detectible virus in the long term.
    With the current standard of peginterferon and ribavirin, roughly 40 to 50 percent of patients with genotype 1 HCV have SVR, with the figures at the lower end predominating in the United States. Most genotype 1 patients require a full 48 weeks of therapy to optimize the chance of SVR.
    Options in genotype 1 patients with prior treatment failure are very limited, with only about 10 percent of nonresponders, and, at most 25 percent of relapsers attaining SVR with retreatment. Many of our patients have been subjected to repeated courses of therapy to no avail.
    Peginterferon and ribavirin therapy has a number of adverse effects. This leads to a shared hope among patients and physicians that with new agents, we can reduce the duration of therapy required to optimize the chance of cure.

    Factors that adversely affect the chance of response to interferon include high viral load in genotype 1 patients, advanced liver scarring, older age, heavy body weight, and insulin resistance.

    African Americans have lower rates of SVR than other ethnic groups. This has recently been explained, at least in part, by the discovery of polymorphisms in the region of the IL28B or lambda interferon 3 gene. These polymorphisms have a major impact on response to interferon.
    For unknown reasons, persons of African heritage have a high prevalence of the T allele at the best-characterized of these loci, which is associated with nonresponse, whatever the ethnicity of the patient. In addition, about a third of HIV-infected persons have hepatitis C. Coinfection is associated with more rapidly progressive fibrosis and lower rates of response to treatment than monoinfection.

    The benefits of SVR include improved histology, along with a reduction in several critical outcome measures, including
    Adecompensation, de novo formation of varices, hepatocellular carcinoma, and actual mortality. Several studies support the contention that SVR improves outcomes in patients with hepatitis C.

    The HALT-C trial was designed to study maintenance interferon therapy in patients with advanced hepatitis C who were nonresponders to previous treatment. They initially received full dose combination therapy with peginterferon and ribavirin, which was continued to 48 weeks if response occurred on treatment. In patients achieving SVR on this regimen, rates of compensation were significantly lower than nonresponders with respect to decompensated liver disease, transplantation, liver cancer, and liver-related death. The HALT-C trial is one of several studies indicating a sharp reduction in the incidence of liver cancer after SVR has been attained.
    An important study from the VA system showed that SVR improves long-term survival. In nearly 17,000 patients who completed treatment no later
    than mid-2008, SVR occurred in 35 percent of genotype 1 patients and in higher numbers of those with genotypes 2 or 3. Multivariate modeling accounting for a wide variety of comorbidities common in this population demonstrated an association between SVR and a highly significant reduction in all-cause mortality.

    With new antiviral agents, we can look forward to a significant improvement in the rate of SVR in genotype 1 patients. By tailoring the duration of therapy according to the kinetics of response, a concept known as response-guided therapy, we expect to shorten the duration of therapy in a much larger proportion of patients than is currently possible.
    Cure of infection in more patients should prevent many life-threatening complications of hepatitis C and reduce the need for transplantation. We need to reverse the rising tide of morbidity and mortality and reduce the human toll taken by this disease.

    Now I'd like to invite Dr. Kauffman back to
    the podium to provide an overview of the sponsor's clinical development program. Thank you.

    Sponsor Presentation – Robert Kauffman
    DR. KAUFFMAN: Thanks, Dr. Jacobson. And now let's turn our attention to the telaprevir development program. I'll touch on a few important aspects of clinical pharmacology and virology and then discuss the key learnings from our Phase 1 and 2 studies that guided Phase 3 development.
    As you've heard, telaprevir is an orally bioavailable inhibitor of the hepatitis C virus protease. In in vitro studies, it has demonstrated reversible and tight binding to the HCV protease active site. It forms a stable enzyme inhibitor complex with a long half-life of about one hour. The inhibition constant of this stable complex is 7 nanomolar. Telaprevir has an IC50 in the replicon system of 350 nanomolar, concentrations that are readily exceeded in vivo.
    The goals of our clinical development program were to increase SVR rates for both treatment-naive and treatment-experienced patients
    and to shorten treatment duration where appropriate. To achieve these goals, Vertex has conducted a broad clinical program comprising five Phase 2 studies and three Phase 3 studies.

    This program was supplemented by a clinical pharmacology and virology program to characterize the pharmacokinetics and pharmacodynamics of telaprevir to evaluate for drug-drug interactions and to understand the potential for viral resistance. The integrated data from these studies provided key information on the dose and duration of each of the components of the treatment regimen.

    Here I'm providing the highlights of the clinical pharmacology program for telaprevir. Telaprevir has a plasma half-life of 9 to 11 hours at steady state. Food substantially increases bioavailability, so telaprevir must be administered with food.
    Telaprevir is metabolized by both CYP3A4 and non-CYP mechanisms and is excreted primarily through the GI tract. It's notable that telaprevir plasma concentrations are not substantially
    increased by metabolic inhibition at steady state, providing a safety factor during co-administration with other drugs.

    We have conducted a comprehensive clinical pharmacology program that has thoroughly characterized the potential for drug-drug interactions with telaprevir. Telaprevir is a substrate and potent inhibitor of CYP3A and a substrate of PGP. Telaprevir may saturate or inhibit PGP in the GI tract. In these respects, it's similar in its drug interaction potential to some HIV protease inhibitors you may be familiar with.
    We evaluated the drug-drug interactions of telaprevir with many drugs commonly used in patients treated for hepatitis C, and these are listed here. There is an interaction with the estrogen component of oral contraceptives that may reduce their effectiveness.

    We also evaluated antiviral drugs used in HIV as well as immunosuppressive agents and drugs used by patients with opioid dependency. Two
    additional studies are ongoing and their clinical phase is complete, and these are studies of interactions with raltegravir and buprenorphine. These data will provide important information for prescribing physicians.
    We have also studied the clinical pharmacology of telaprevir in special populations. Mild hepatic impairment has minimal impact on telaprevir pharmacokinetics. Notably, in patients with moderate hepatic impairment, plasma concentrations of telaprevir are reduced by 46 percent. In a single-dose study, severe renal impairment did not have a substantial effect on telaprevir pharmacokinetics.

    To further characterize the PK in special populations, we are collecting PK data in our ongoing coinfection study, and we are planning a study in people with recurrent hepatitis C after liver transplantation.
    To identify the appropriate dose of telaprevir, we evaluated viral dynamics, both in monotherapy and in combination therapy. The first
    viral dynamic study evaluated three dose levels of telaprevir alone over 14 days. I draw your attention to the very rapid initial decline in HCV RNA for all dose groups. Note also the more consistent decline at the 750-milligram dose level shown in orange. These data, along with the results of concentration effect modeling, led us to select 750 milligrams are the clinical dose.

    The figure on the right shows the viral kinetics for peginterferon alone, and for telaprevir with and without coadministration with peginterferon. Telaprevir has about a 1,000-fold greater antiviral compared to peginterferon. You'll note the pattern of viral decline for telaprevir monotherapy in the second study was the same as seen in the first study, and these are the lines in orange. But with co-administration, an even greater rate of decline is shown, as seen in the pink line, because co-administration also inhibits telaprevir-resistant variants.
    HCV exists in infected individuals as a genetically diverse viral population due to its rapid, error-prone replication. This means that drug-resistant variants preexist and can be selected by the antiviral agent. Across our program, we have conducted an analysis of viral variants, representing sequences from over 3400 patients at baseline, and in 740 patients who did not achieve an SVR.
    The resistance profile has been found to be consistent across patient populations. Overall, 12 to 22 percent of all patients starting treatment have had detectible resistant variants after treatment failure. Follow-up studies have shown that resistant variants tend to diminish over time in the absence of treatment, and Dr. George will provide more information on this topic in her talk that follows mine.

    To define telaprevir's resistance profile, we performed sequence analyses in patients who did not have an SVR. We consistently detected substitutions at four positions in the NS3 protease region, and those are positions 36, 54, 155, and 156. The double variant, V36M plus R155K, is also commonly found.

    Variants are shown here in increasing degree of resistance from left to right. It's important to note that all variants were less fit than wild type, providing a mechanism by which they diminish in frequency over time in the absence of drug-selected pressure. There is cross-resistance among different protease inhibitors. However, all variants are sensitive in vitro to other classes of antiviral agents and to peginterferon and ribavirin.

    response-guided therapy
    One of the topics we'll discuss today is the use of response-guided therapy. With response-guided therapy, treatment duration is determined by the rate of viral decline. Let me show you how the combination of viral dynamic modeling and empirical data were used to help estimate the treatment duration for achieving SVR in our trials.
    To achieve an SVR, the total body burden of viable virus needs to be below one copy, which is represented here by the black dashed line. The blue horizontal line shows the limit of quantification of the RNA assay. This means the virus is undetectable once it falls below this line; but of course it may still be present.

    The typical response to peg and ribavirin is shown here in red. It indicates the current 48-week duration of treatment to achieve an SVR. Only a single line is indicated because there is no known virologic resistance to peg and ribavirin.

    With telaprevir-based therapy, total body eradication requires elimination of both wild type and resistant variants. They are represented here by the purple and green lines. This rate is much faster, on average, for a telaprevir regimen compared to peg and ribavirin.
    Through an evaluation of the time to eradication of wild type virus and viral variants and empirical data from the Phase 2 program, we estimated durations of treatment necessary for complete viral eradication. These hypotheses were then tested in subsequent clinical trials.
    In Phase 2, we sought to answer several basic questions necessary to further advance the program. The primary results of those studies are summarized here. With respect to patient populations, we demonstrated efficacy in both treatment-naive and treatment-experienced patients. With respect to safety, we encountered severe rash for the first time and implemented successful management strategies for rash and for anemia.
    The dose was based on Phase 1 data, as I've shown you a few moments ago, and PK/PD modeling, and was subsequently reconfirmed with PK/PD analyses of safety and efficacy. The 12-week duration of telaprevir was based on Phase 2 data and was supported by viral modeling. There was no demonstrable advantage to more than 12 weeks of telaprevir.
    With respect to the treatment regimen, in Phase 2 we demonstrated that the absence of ribavirin in the regimen substantially increased breakthrough and relapse rates, and, therefore, ribavirin has been shown to be a key component of the regimen.

    The potential of response-guided therapy was demonstrated in treatment-naive patients with 61 to 69 percent SVR rates and low rates of relapse with 24 weeks of treatment.
    I'll now turn the podium over to Dr. Shelley George, who will show you how these concepts were incorporated into the Phase 3 program and the outcome of those studies.


    Sponsor Presentation – Shelley George
    DR. GEORGE: Thank you, Dr. Kauffman.
    Good morning. I am Shelley George, vice president, HCV Therapeutic Area Lead in Medical Affairs at Vertex. It is a privilege to present the telaprevir Phase 3 efficacy data today.
    As you have heard from Dr. Jacobson, sustained viral response, or SVR, is considered to be a virologic cure in HCV-infected patients. However, the standard treatment of pegylated interferon and ribavirin, or PR, for 48 weeks, leaves 50 to 60 percent of genotype 1 treatment-naive patients without a cure, and there are limited options for successful retreatment. Given these limitations, increasing SVR rates and
    shortening treatment duration were important goals for the telaprevir Phase 3 development program.
    I will start my talk by reviewing critical design features shared by all of the Phase 3 studies. These studies were designed to assess the efficacy and safety of 12 weeks' treatment with telaprevir, 750 milligrams thrice daily, in combination with either 24 or 48 weeks of pegylated interferon and ribavirin, according to the package insert, based on treatment response.

    The Phase 3 studies included both treatment-naive and treatment-experienced patients with genotype 1 chronic hepatitis C. Enrolled patients had varying degrees of liver disease, including compensated cirrhosis. They were from different geographic regions and had diverse racial and ethnic backgrounds.
    The primary endpoint in all three of the studies was the proportion of patients achieving SVR, defined as having undetectable HCV RNA levels 24 weeks after the last planned dose of study drug. This endpoint is more conservative than the last actual dose because it requires a longer follow-up period for patients who prematurely discontinue treatment.
    Now, each of the Phase 3 studies also had unique features, such as response-guided therapy and lead-in dosing. These features were designed to address issues vital to the recommended treatment protocol. Study 108, known as ADVANCE, was a pivotal Phase 3 study designed to establish the superiority of telaprevir in combination with peg/ribavirin compared with the standard regimen in treatment-naive patients.
    The duration of PR treatment was determined using a response-guided approach with the following criteria. Patients with undetectable HCV RNA at weeks 4 and 12 of treatment, known as an extended rapid viral response, or eRVR, had a planned treatment duration of 24 weeks of pegylated interferon and ribavirin. Those who did not achieve an eRVR had a planned treatment duration of 48 weeks of pegylated interferon and ribavirin. In this study only, both 8-week and 12-week telaprevir
    treatment durations were evaluated.

    Moving on to study 111, or ILLUMINATE, this was a supportive phase 3 study designed to confirm response-guided therapy in treatment-naive patients. The third study, C216, known as REALIZE, was a pivotal phase 3 study designed to establish the superiority of a telaprevir-based regimen among treatment-experienced patients, including prospectively defined prior relapses, prior partial responders, and prior null responders. A unique feature of this study was the inclusion of an arm with a 4-week lead-in with PR before commencing telaprevir treatment.
    Now I will review the highlights of pivotal study 108, ADVANCE. This was a randomized, double-blind, placebo-controlled, multicenter study. Patients were randomized to one of three treatment groups in a 1 to 1 to 1 ratio, stratified by HCV genotype and baseline viral levels.
    Telaprevir was given in combination with PR for either the first 8 weeks, noted here as T8/PR, or for the first 12 weeks, shown as T12/PR.
    Patients in the 8-week telaprevir group received telaprevir-matched placebo during weeks 9 through 12. Patients who achieved an extended rapid viral response, eRVR, had a planned pegylated interferon and ribavirin treatment duration of 24 weeks. Those who did not achieve eRVR had a planned PR treatment duration of 48 weeks.

    For the control group, the total treatment duration was 48 weeks, with telaprevir-matching placebo given for the first 12 weeks and peg/ribavirin dose for the entire 48-week period. All patients were followed to week 72 for a common assessment time point.
    The patient demographics and baseline characteristics in the study were well-balanced across the treatment groups. Notably, 60 percent of sites were located in North America, and 40 percent of sites were located in Europe and other countries. Factors associated with a poor response to peg/ribavirin therapy include black or African American or Hispanic or Latino race or ethnicity, high baseline viral levels, and more
    advanced liver disease.

    Of the 1,088 patients who received at least one dose of study drug, 9 percent of patients enrolled in this study were black or African American and 11 percent were Hispanic or Latino. The majority of patients in each of the three treatment groups had high baseline viral levels, 15 percent of patients had bridging fibrosis, and 6 percent had cirrhosis.
    Sustained viral response rates were markedly and significantly higher in the telaprevir-treated groups as compared with the standard treatment group. In the 12-week telaprevir group, the SVR rate was 79 percent as compared with 46 percent in the standard treatment arm, an absolute significant difference of 33 percent. In the 8-week telaprevir group, the SVR rate was 72 percent, an absolute significant difference of 26 percent as compared with standard treatment. The SVR rate of 46 percent in the control arm was similar to that reported in the literature.
    The proportion of patients who achieved SVR
    was higher in the 12-week telaprevir group than the 8-week telaprevir-treated group. This study was not designed to compare response rates across the two telaprevir regimens. However, a numerical difference of 7 percent was observed in favor of the 12-week telaprevir-treated group, with a confidence interval which lies entirely to the left of zero. These efficacy results favor the T12 regimen.
    Fifty-eight percent of patients in the study who received a telaprevir-based regimen achieved an extended rapid viral response, or eRVR. These patients were therefore eligible for a planned total treatment duration of 24 weeks. In contrast, only 8 percent of patients achieved eRVR in the control group, and these patients were assigned to receive 48 weeks of treatment per protocol.

    eRVR status is a critical factor in determining treatment outcome. SVR rates were high, ranging from 87 percent to 92 percent, in eRVR-positive patients who received 24 weeks of treatment with a telaprevir-based regimen. SVR was also high among eRVR-positive patients in the control group, who received 48 weeks of treatment per protocol. However, as we've seen in the previous slide, only 8 percent of patients in the control group achieved eRVR.
    By contrast, SVR rates ranged from 52 percent to 60 percent among those patients who did not achieve eRVR and received 48 weeks of treatment with a telaprevir-based regimen. These SVR rates were higher than those achieved by patients who did not achieve an eRVR in the control group, indicating that telaprevir confers benefit for a proportion of these patients as well.

    Overall, relapse rates among telaprevir-treated patients were low, notably, three- to fourfold lower than with standard treatment. Low relapse rates were observed among those patients who achieved eRVR. High SVR rates and low relapse rates observed in patients who achieve eRVR indicate that shortening the duration of therapy did not compromise treatment outcomes in these patients.
    SVR rates were higher in the telaprevir-treated groups than in the control group regardless of baseline demographics and disease characteristics. This forest plot shows the absolute differences and the 95 percent confidence interval for the differences in SVR rates by baseline characteristics between the 12-week telaprevir-treated group and the control group.

    Substantial clinical benefit was achieved across a broad range of patient subgroups, including some groups that traditionally have a worse outcome with standard treatment. These include black, African American, and Hispanic or Latino patients, and those with cirrhosis or high baseline HCV RNA levels. In some subgroups, such as age greater than 65 years, HCV genotype 1, subtype unknown, the low number of patients resulted in wide confidence intervals.

    Telaprevir-based regimens provided a substantial benefit regardless of the stage of liver disease. Patients with less advanced liver disease had higher response rates than patients with bridging fibrosis or cirrhosis.

    However, the substantial treatment difference observed between the telaprevir-treated groups and the control group was maintained regardless of the stage of liver disease.

    SVR rates were substantially higher in the telaprevir-treated groups than in the control group across the races and ethnicities shown here. Hispanic or Latino patients treated with a 12-week telaprevir regimen achieved an SVR rate of 77 percent, similar to the 79 percent SVR rate achieved in Caucasian patients. Black or African American patients treated with a 12-week telaprevir-based regimen achieved an SVR rate of 62 percent, as compared with 25 percent for the standard treatment group. Importantly, the substantial treatment difference between the telaprevir-treated groups and the control group was maintained across all subgroups shown here.
    In conclusion, this study in the treatment-naive patient population demonstrated the clinical benefit of telaprevir, with an absolute significant difference of 33 percent in SVR rates for the 12-week telaprevir-based regimen compared with standard treatment.

    Clinical benefit was achieved across a broad range of patient subgroups, including those that are difficult to cure with standard treatment such as blacks and African Americans, Hispanics or Latinos, and patients with cirrhosis. High SVR rates and low relapse rates support a 24-week treatment duration in those patients who achieve an eRVR.
    Now I will move on to study 111, known as ILLUMINATE. This noninferiority study was designed to support pivotal study 108, with a goal of confirming response-guided therapy among treatment-naive patients. As you have seen in study 108, high SVR rates were achieved among patients who achieved an eRVR and those received a treatment duration of 24 weeks. Study 111 is designed to answer the question of whether or not there is clinical benefit in extending peg/ribavirin treatment duration from 24 weeks to 48 weeks among those patients who achieve an eRVR.
    This was a randomized, open label, multicenter study. Telaprevir was administered in combination with PR for 12 weeks. The total duration of PR treatment was either 24 or 48 weeks. Those patients who achieved eRVR and completed the week 20 visit were randomized in a 1 to 1 ratio to either stop all study treatment at week 24 or to continue treatment with PR to week 48. Those patients who did not achieve eRVR were assigned to receive treatment with PR for 48 weeks.

    Patients who received at least one dose of study drug but who prematurely discontinued treatment before week 20 were not randomized or assigned to a treatment regimen. All patients were followed to week 72.
    Study 111 was designed to determine if extending the PR treatment duration from 24 to 48 weeks in those patients with eRVR would result in additional benefit. Therefore, the primary focus of this presentation will be on the randomized eRVR-positive treatment arms.

    The primary comparison was based on confidence interval estimates to rule out the inferiority of the 24-week regimen to the 48-week regimen among those patients who achieved eRVR. The predefined noninferiority margin was 10.5 percent. Of note, although this was an open label study, HCV RNA levels from tests prior to week 24 and eRVR status were not revealed to the investigator or patient until the end of the study. This was intended to reduce the potential for bias among those patients who received 48 weeks of therapy in the study.
    Patient demographics and baseline characteristics were similar among the randomized treatment groups and generally representative of the overall study population. Notably, 94 percent of sites were located in North America. In the overall study population, 14 percent of patients enrolled were black of African American and 10 percent were Hispanic or Latino. The majority of patients had high baseline viral levels, and of the 540 patients enrolled, 16 percent had bridging fibrosis and 11 percent had cirrhosis.

    SVR rates in the randomized eRVR-positive arms were 92 percent for the 24-week treatment group and 90 percent for the 48-week group. The lower limit of the two-sided 95 percent confidence interval on the difference in SVR rates between the two groups was minus 4.3 percent, to the right of the predefined noninferiority margin of minus 10.5 percent, confirming that 24 weeks of treatment is not inferior to 48 weeks of treatment.
    Overall, 65 percent of patients in the study achieved an extended rapid viral response. The SVR rate in the study overall was 74 percent, similar to the response rate observed in study 108.

    In conclusion, study 111 confirms the use of response-guided therapy in treatment-naive patients treated with telaprevir. There was no evident advantage in extending treatment for 48 weeks as compared with 24 weeks in those patients with eRVR. The treatment outcome in the study is very similar to that of pivotal study 108.
    Next I will review the results of pivotal study C216, known as REALIZE. This study was designed to confirm the superiority of telaprevir in combination with PR, as compared with PR alone, among those patients who did not achieve a sustained viral response after a prior course of PR therapy. This study also included an arm containing a 4-week lead-in with peg/ribavirin prior to starting telaprevir to assess the potential impact of a lead-in on treatment outcomes.

    Patients in this study were prospectively studied with predefined, well-documented, prior nonresponse criteria. The definitions of populations in this study were in accordance with the FDA draft HCV guidance. Patients had either prior relapse, defined as HCV RNA undetectable at the end of a prior course of PR therapy but not achieving SVR, or prior nonresponse, defined as never having had an undetectable HCV RNA level during or at the end of a prior course of PR therapy.

    Prior nonresponse was further categorized as either partial response, greater than a week or 2 log decrease in HCV RNA at week 12, or null response, less than a 2 log decrease in HCV RNA at week 12. Patients with prior null response are the most difficult to cure, with SVR rates ranging from naught to 10 percent after retreatment with pegylated interferon and ribavirin.

    REALIZE was a randomized, double-blind, placebo-controlled study. Patients were randomized in 2 to 2 to 1 ratio to one of three treatment groups, all with a planned treatment duration of 48 weeks. In the first treatment group shown above, patients received telaprevir in combination with PR for 12 weeks, followed by placebo plus PR for 4 weeks, then followed by PR for an additional 32 weeks.

    In the next group, the lead-in group, patients received placebo in combination with PR for 4 weeks, followed by telaprevir plus PR for 12 weeks, followed by PR for an additional 32 weeks once again.
    In the standard treatment group, patients
    received placebo plus PR for 16 weeks, followed by PR for 32 weeks. At the end of treatment, all patients were followed until 24 weeks after the last planned dose of study drug.

    Demographic and baseline disease characteristics were comparable in this study across the treatment groups. Importantly, 60 percent of sites in this study were located in Europe and 40 percent were located in North America and other countries. Five percent of enrolled patients in the study were black or African American, and 11 percent were Hispanic or Latino.
    Of the 662 patients treated, there was a high proportion of patients with poor prognostic factors. The majority of patients had high baseline viral levels across the treatment groups; 18 percent of patients were prior partial responders, and 27 percent were prior null responders, the hardest patients to cure; 22 percent had bridging fibrosis, and 26 percent of patients had cirrhosis. Not shown here, but of note, the proportion of patients with cirrhosis was
    higher in the prior nonresponder population than in the prior relapser population.

    Among prior relapsers, SVR rates in the telaprevir treatment groups were significantly and markedly higher than the control group. SVR rates were 84 percent and 88 percent in the immediate start and lead-in telaprevir groups respectively, as compared with 22 percent in the control group. SVR rates were similar in the two telaprevir-treated groups, with and without a lead-in.

    For prior null responders and prior partial responders, the SVR rates were also significantly higher in each of the telaprevir treatment groups than in the control group. Among prior partial responders, SVR rates were 61 percent and 56 percent in the immediate start and lead-in telaprevir-treated groups, as compared with 15 percent in the control group. Among prior null responders, SVR rates were 31 percent and 33 percent in the immediate start and lead-in telaprevir-treated groups, as compared with 5 percent in the control group. Once again, SVR
    rates were similar in the two telaprevir-treated groups with and without a lead-in.

    Treatment outcomes were similar across the subgroups in the arms with and without a lead-in. Among prior relapses, virologic failure, shown here in purple, occurred rarely and relapse rates, shown here in aqua, were low. Among the prior nonresponders, virologic failure and relapse occurred more frequently and were highest among the prior null responders.
    On-treatment virologic failure and relapse rates were similar between the two telaprevir-treated therapies with and without a lead-in. There was also no difference in the types of emerging viral variants between the two groups.

    This forest plot shows the absolute differences in SVR rates between the 12-week telaprevir-treated groups and the control group, and the 95 percent confidence interval for the difference by subpopulations. SVR rates were higher in each telaprevir group than in the control group in Caucasian patients. However, the number
    of patients in other subgroups by race was too low to allow a meaningful comparison across the subgroups.

    Subpopulations with wide 95 percent confidence intervals reflect small sample sizes, seen in patients over 65 years old, those of black race and Asian race. Among prior relapsers and prior nonresponders, SVR rates were higher in the telaprevir groups than in the control group by baseline disease characteristics such as HCV subtype, HCV RNA levels, and liver disease status.

    The SVR rates for the pooled telaprevir groups in the study were significantly higher than for retreatment with PR alone in all subpopulations; for prior relapsers, 86 percent versus 22 percent in the control group; prior partial responders, 59 percent versus 15 percent in the control group; and prior null responders, 32 percent versus 5 percent in the standard treatment group. SVR rates, virologic failure, and relapse rates were similar between the two telaprevir-treated groups with and without lead-in.

    Response-guided therapy was not prospectively evaluated in the treatment-experienced population in the study. However, data from Phase 2 and 3 studies indicates that the treatment-experienced patient population is not uniform. In fact, patients with prior relapse appear to be more similar to the treatment-naive patient population than to the prior nonresponse population based on both rapidity of the antiviral response on treatment and the SVR rates.

    Data analyses indicate that patients with prior relapse are likely to benefit from response-guided therapy. This is based on the same criteria as for treatment-naive patients, undetectable HCV RNA at weeks 4 and 12 or eRVR. These data analyses are further supported by viral dynamic modeling.
    I will now briefly review the clinical evidence supporting response-guided therapy in prior relapses from Phase 2 and Phase 3 studies. The majority, 58 to 96 percent, of patients in both the prior relapse and treatment-naive patient populations achieved eRVR, shown here in orange.
    Let's now look at the blue bars, which show SVR rates among eRVR-positive patients, starting on the left-hand side with the treatment-naive subpopulation. SVR rates of 92 percent were achieved by treatment-naive patients with eRVR who were treated for 24 weeks in study 108.
    Similar SVR rates, ranging from 89 percent to 100 percent, were achieved by eRVR-positive patients with prior relapse who were treated for 24 weeks in the Phase 2 studies 106 and 107, shown in the center. These high response rates were comparable to the SVR rate of 96 percent achieved by eRVR-positive prior relapses who received 48 weeks of treatment per protocol in study C216.

    In summary, both prior relapsers and treatment-naive patients represent populations with high interferon responsiveness, and in both these patient populations, patients with eRVR achieve high SVR rates. Of note, SVR rates were high in prior relapses regardless of the duration of pegylated interferon and ribavirin. This suggests that achieving eRVR was the critical factor in determining outcome. These results were further confirmed by viral dynamic modeling, and taken together, support response-guided therapy among prior relapsers.

    Now we'll move on to the durability of response achieved with a telaprevir-based regimen. This was investigated in two settings, firstly, in patients who had achieved an SVR during the one-year follow-up period in the Phase 2 clinical studies, and secondly, in an ongoing three-year observational follow-up study known as study 112, or EXTEND.

    Relapse after achieving an SVR with standard PR therapy occurs in less than 1 percent of treated patients. Durability of response for telaprevir was assessed in 361 patients who had achieved an SVR after treatment with a telaprevir-based regimen in a phase 2 study. Patients were also required to have had at least one post-SVR follow-up assessment.
    Two patients had late relapse within six months after achieving SVR during the study follow-up period. All other patients with SVR, who have been followed up for three years after end of treatment, continue to have undetectable HCV RNA.

    Study 112 is an ongoing observational three-year follow-up study to evaluate the durability of response in a cohort of patients with SVR following telaprevir-based treatment in a prior trial. Patients were given the option to enroll in this follow-up study, which was initiated during the course of the Phase 3 program. The current interim analysis included 123 patients with SVR in a phase 2 study. Patients in the Phase 3 studies had not been enrolled at the time of this analysis.
    Durability of response was demonstrated in 122 of 123 patients, who were followed for 5 to 35 months after SVR, with a median follow-up time of 22 months. One patient had late relapse before he enrolled in study 112. This is the same patient that was previously reported during the follow-up period in Phase 2. No late relapses have occurred during the observational period of study 112, which is ongoing.

    Now I'd like to move on to the evolution of resistant variants after treatment in patients who did not achieve an SVR in a prior trial. This was studied during the Phase 3 program, with a median follow-up period of about one year after treatment failure. In addition, an interim analysis of study 112 has been conducted with a median follow-up period of about two years. I will start with an analysis of patients from the Phase 3 trials.

    Of those patients in Phase 3 trials who did not achieve an SVR, 84 percent of those patients with genotype 1A and 54 percent of those patients with genotype 1B had detectable resistance after treatment failure. The rates and median time to loss of detectable resistant variants by population sequencing was estimated with a Kaplan-Meier curve. There was a difference between subtypes, with a median time of 10 months for genotype 1A patients and 3 weeks for genotype 1B patients.

    These results suggest that resistant variants were no longer detected by population sequencing over time in both treatment-naive and treatment-experienced patients. Notably, about one-quarter of these patients were prior null responders.
    Study 112 was also designed to evaluate changes in HCV variants over time in a separate cohort of patients without SVR following telaprevir-based treatment in a prior trial. For the purpose of this interim analysis, patients were required to have both detectable telaprevir-resistant variants at the post-nadir time point in the previous trial and viral sequencing data available from study 112.

    Of the 56 treatment-naive and treatment-experienced patients from the Phase 2 program, almost 90 percent of patients, 50 of 56, no longer had detectable resistant variants by population sequencing, with a median follow-up time of about two years. We will continue to follow all patients after three years.

    To further confirm the loss of resistant variants, a more sensitive clonal sequence analysis was performed on a subset of patients from
    study 112. These results show that viral populations at follow-up were not enriched in resistant variants as compared with baseline. As there is no data yet on retreatment of these patients, the clinical implications of this finding are not known. However, these emerging data are encouraging.
    In conclusion, interim data from study 112 have demonstrated that sustained viral response with telaprevir is durable. Late relapse is rare and consistent with historical late relapse rates with PR of about 1 percent. In those patients who did not achieve an SVR after telaprevir treatment, emerging data suggest that the frequency of resistant variants declines over time.

    Now I will close my talk with the overall efficacy conclusions from the telaprevir Phase 3 program. Telaprevir-based regimens resulted in significantly higher SVR rates in both treatment-naive and treatment-experienced patients as compared with standard treatment. This clinical benefit was observed across a broad range of patients, including subgroups associated with a poor response to peg/ribavirin. Similar treatment outcomes were achieved in treatment-experienced patients with and without a 4-week PR lead-in.

    Response-guided therapy enabled the majority of treatment-naive patients to achieve high SVR rates with 24 weeks total therapy. Phase 2 and 3 clinical data and viral dynamic modeling support response-guided therapy in those patients with prior relapse to PR treatment.
    Thank you for your attention. I would now likewise to invite Dr. Singhal to the podium to address the safety profile of telaprevir.


    Sponsor Presentation – Priya Singhal
    DR. SINGHAL: Thank you, Dr. George.
    Good morning. I'm Priya Singhal, senior director, Disease Area Safety Lead, Vertex Global Patient Safety. I will present telaprevir safety.
    I'll start with a review of general safety. Following that, I'll discuss analyses performed for specific adverse events, along with their respective management during the Phase 3 program.

    Then I'll close with overall conclusions.
    Before we begin our review, I'd like to draw your attention to the following important considerations. Telaprevir is added for the first 12 weeks of a total 24- or 48-week regimen of peginterferon and ribavirin, also referred to as PR. Peginterferon and ribavirin have a well-characterized safety profile. Adverse events such as anemia, rash, fatigue, and fever are common.
    These PR-associated adverse events also occur in the first 12 weeks when telaprevir is being dosed. However, these adverse events continue to have a significant prevalence until the peginterferon and ribavirin dosing is completed.

    Telaprevir's safety profile has been similar across the treatment-naive and the prior treatment-experienced patients. The safety profile was also similar across the Phase 2 and 3 studies. Hence, safety data were pooled across the Phase 2 and 3 studies. Details of these analyses are in your briefing document. Today's presentation will focus on the pooled safety data from the Phase 3 studies alone.

    general safety
    Let's start with general safety. More than 3,800 patients across the entire clinical development program have been exposed to telaprevir. As Dr. George just discussed, there were a total of three Phase 3 studies. Two of these studies included genotype 1 HCV treatment-naive patients, and one included prior treatment-experienced patients, all with compensated liver disease.

    A total of 1797 patients were randomized to any telaprevir/peginterferon/ribavirin combination regimen, and 493 to placebo, peginterferon, and ribavirin combination regimen. Unless mentioned otherwise, I will present data from the telaprevir/placebo phase. In the placebo-controlled studies, this was the period during which patients received either telaprevir or placebo in combination with peginterferon and ribavirin.
    This here is an overview of patient disposition and key reasons for discontinuation.

    Across Phase 3, 71 percent of patients randomized to telaprevir combination treatment groups completed, compared with 51 percent of patients in peginterferon and ribavirin groups alone.
    Now, as you look at reasons for discontinuation, you can see an additional 4.7 percent of telaprevir-treated patients did not complete due to an adverse event. No deaths occurred during the telaprevir combination treatment, and one occurred during the peginterferon and ribavirin treatment.

    Upon reviewing adverse events, we observed that almost all patients in both groups, 96 percent or more, experienced an adverse event regardless of whether they received telaprevir or placebo. This is reflective of the adverse events associated with peginterferon and ribavirin.
    Reviewing other indicators, you can see in the left column that when telaprevir is added to peginterferon and ribavirin, there is a higher incidence of serious adverse events grade 3 or severe adverse events and adverse events leading to treatment discontinuation.
    Next we will review adverse events in patients receiving telaprevir combination treatment as compared with placebo plus peginterferon and ribavirin. We'll start by looking at the most common AEs. These here are the adverse events that occurred in at least 20 percent of patients in either group. Now, fatigue, pruritus, nausea, rash, anemia, and diarrhea occurred at a 5 percent greater incidence in the telaprevir combination group.

    adverse event severity

    Now let's review adverse event severity. These are the AEs that were reported as grade 3 or severe in 1 percent or more patients in either treatment group. Of these, anemia, fatigue, and rash were reported at a 1.5 percent greater incidence in the telaprevir combination group. Anemia and rash were also the events that led to a higher discontinuation rate in the telaprevir combination group as compared with placebo/peg/ribavirin.
    Next we will review specific analyses for
    rash and anemia and the management of these events in the Phase 3 program. Let's begin with a review of rash.

    Severe rash was first reported in the Phase 2 program. In response, a rash assessment and management plan, including severity grading and general guidance was implemented to ensure patient safety. In Phase 2, a severe rash required all study drugs to be discontinued. Based on the Phase 2 experience, management of severe rash was modified in Phase 3 to require that telaprevir alone be discontinued. Importantly peginterferon and ribavirin could be continued per clinical judgment. In addition, an external rash adjudication panel was appointed to characterize severe rash. This panel was chaired by Dr. Robert Stern from Harvard Medical School, who's here with us today.

    In parallel with these activities, several investigations were undertaken to try and understand the rash etiology. These were metabolite characterization, HLA and PGB analyses,and exposure-response relationships. At this time no mechanism has been identified.
    A special search category, or SSC, was established to ensure that adverse events represented the medical concept comprehensively, and that patients were counted once in the category. As you can see from the first row, there was a rash incidence of 34 percent in the peginterferon and ribavirin group. An additional 22 percent of patients experienced rash when telaprevir was added. The incidence of severe rash among Phase 3 telaprevir-treated patients was 3.7 percent. In Phase 3, the rate of discontinuation of all study drugs due to rash was low, at .8 percent.
    This Kaplan-Meier here represents the pooled data from the placebo-controlled Phase 2 and 3 studies during the overall treatment phase of the studies. Time in weeks is on the X axis and proportion of patients on the Y axis. The orange line is telaprevir combination group, and the grey line is placebo with peginterferon and ribavirin.


    You can observe that there was an earlier onset and higher incidence of rash when telaprevir was added.
    Ninety-three percent of all rash events were mild or moderate, and the majority, more than 90 percent of rash events, did not progress to rash of greater severity. Rash resulted upon treatment completion or discontinuation. The majority of patients with rash received topical corticosteroids and/or systemic antihistamines. No data are available to assess the effectiveness of these treatments.

    As you may be aware, in the literature rash is associated with PR and has often been characterized as an eczematous rash. As we saw before, in Phase 3 severe rash was reported in 3.7 percent of telaprevir-treated patients. The external adjudication panel reviewed all of these cases in addition to several others.
    Severe rash was characterized as follows: primarily an eczematous, pruritic, spongiotic dermatitis that involved less than 30 percent of the body surface area. There was no evidence of vasculitis or type 1 hypersensitivity reactions. Rare cases had features suggestive of severe skin reactions.
    We will now move on to review the rare severe skin reactions observed across the development program.

    To ensure a systemic adjudication, the scoring system from the European Registry of Severe Cutaneous Adverse Reactions was adapted. Importantly, in this scoring system, cases adjudicated as possible were not considered likely to be true cases. The external panel suspected Stevens-Johnson syndrome in three cases. All these cases resolved.
    One of the three cases was assessed as a definite case. It occurred 11 weeks after the last dose of telaprevir, while the patient was still on peginterferon and ribavirin, as well as buproprion and naproxen. It was considered unrelated to telaprevir. Of the other two cases, one was assessed as probable and one as possible.

    The panel also suspected drug reaction with eosinophilia and systemic symptoms, also called DRESS, in 11 cases. Ten resolved. One was reported as resolving but was later lost to follow-up. One of the 11 cases was considered as a definite case. Of the remaining 10, 2 were considered as probable and the remaining 8 denoted as possible.
    The predominate features across these cases were fever, rash, and eosinophilia. As you may appreciate, fever is also a common reaction to peginterferon. Systemic organ involvement, which is a hallmark of DRESS, was absent in 9 cases, and unconfirmed in 2.

    In the Phase 3 program, rash management was included per protocol. Investigators were asked to follow general management principles, including monitoring for all rash events and symptomatic treatment as required. Guidance to distinguish three categories of rash was also included. These categories were: mild or moderate rash, severe rash, and among the severe rash, the category of rare severe skin reactions.

    For mild and/or moderate rash, study drug discontinuations were not required. For severe rash, discontinuation of telaprevir was required. Telaprevir could not be reduced or restarted. Importantly, peg/ribavirin could be continued per clinical judgment. For rare severe skin reactions, which are also described with PR, permanent discontinuation of all study drugs and all medications was required.

    As you will recall, in the Phase 3 program the rate of discontinuation of all study drugs due to rash was .8 percent. This was an improvement over the 5.2 percent that we had observed in the Phase 2 program, suggesting that the Phase 3 rash management plan here, shown here, was effective.

    anemia

    Next we will move on to a review of anemia. This here is a graphical representation of the pooled data from the placebo-controlled Phase 2 and 3 studies. Time in weeks is on the X axis, and the mean hemoglobin in grams per deciliter on the Y axis over the course of the studies.
    The grey line represents placebo plus peginterferon and ribavirin, which is associated with a decline in hemoglobin, as you can see. Now, the orange curve shows that when telaprevir is added, there was about an additional 1 to 1.5 gram per deciliter decrease in hemoglobin.
    The vertical lines here are the relevant time points. The nadir in both groups was between weeks 12 to 14. In the telaprevir combination group, you can see that the mean hemoglobin returned to peginterferon and ribavirin levels after telaprevir was completed by about week 24. At the end of the study, the last vertical line, you can observe that there was no difference between the mean hemoglobins of the two groups.
    This table compares the post-baseline hemoglobin nadirs between groups. 36.8 percent of patients in the telaprevir combination group had hemoglobin values less than 10 grams per deciliter compared with 14.8 percent of patients in the placebo plus peginterferon and ribavirin group.
    In the telaprevir combination group, 27.3 percent of patients stayed 8.5 grams per deciliter, and 9.5 percent were below 8.5 grams per deciliter. The comparative values for placebo plus peginterferon and ribavirin group were lower.

    The potential mechanism of anemia has been evaluated. From a nonclinical perspective, repeat dose toxicity studies in rats and dogs suggested extravascular hemolysis as the mechanism. This was accompanied by a regenerative response. In vitro red blood cell mechanistic studies did not demonstrate a direct effect.

    From a clinical perspective, the exposure-response analyses suggested a relationship between grade 2 or greater hemoglobin decline with telaprevir, peginterferon, and ribavirin exposures. The reticulocyte production index, or RPI, was obtained from the clinical studies. This analysis suggested that peripheral destruction of red blood cells as well as a lower red blood cell production both contributed to the observed anemia in telaprevir combination treatment.

    This here is a summary table of the anemia
    special search category adverse events. From the first row, we observe that 35.7 percent of patients in the telaprevir combination group reported an anemia adverse event compared with 16.6 percent in the placebo and peginterferon/ribavirin group.
    Reviewing the last row, we see that rate of discontinuation of all study drugs for anemia was low at 1.3 percent in telaprevir-treated patients, compared with .4 percent in the placebo combination group. These are the comparative rates for ribavirin dose reduction in the two groups. The rates were higher in the telaprevir combination group.
    In the Phase 3 program, management of anemia was included per protocol as follows: hemoglobin monitoring in accordance with the peginterferon and ribavirin labels, as well as clinical judgment; ribavirin dose modification per the ribavirin label; discontinuation of telaprevir, per clinical judgment, and telaprevir could not be reduced or restarted; and discontinuation of telaprevir was required any time that ribavirin was discontinued.

    Blood transfusions were reported in 6 percent of telaprevir-treated patients and 1 percent of placebo/ peginterferon/ribavirin-treated patients.
    In line with FDA guidance during the Phase 3 program, erythropoietin-simulating agents, or ESAs, were not permitted per protocol. They were used in 1 percent or less of patients across treatment groups. This has allowed for an unbiased assessment of anemia incidence as well as severity.
    Next we will collectively review anorectal adverse events such as hemorrhoids, anorectal discomfort, and other similar events.
    A special search category was established for a comprehensive evaluation. As you can observe from the summary table and the first row, these adverse events were frequent and reported at a higher incidence in the telaprevir combination group. However, these adverse events were rarely severe, with less than .5 percent of patients discontinuing for these events.
    This brings us to the conclusions on telaprevir's safety profile.

    Telaprevir's safety profile is based on exposure in more than 3,800 patients. Addition of telaprevir to peginterferon and ribavirin increased the incidence of specific adverse events. The majority occurred in the first 12 weeks, were mild to moderate, and did not lead to treatment discontinuation.

    Rash and anemia were identified as key telaprevir-associated adverse events. Addition of telaprevir resulted in an increased incidence and severity of these events, as well as treatment discontinuations. Rash and anemia were well characterized during the development program, and both were reversible and manageable. Early recognition of these key events helped us characterize them and develop management strategies that were tested in Phase 3.
    Thank you for your attention. I would now like to turn the podium back to Dr. Kauffman.


    Sponsor Presentation – Robert Kauffman
    DR. KAUFFMAN: Thank you, Dr. Singhal.


    I'd now like to discuss the overall benefit-risk assessment of telaprevir.


    Telaprevir represents a true paradigm shift in the treatment of HCV. Treatment with a telaprevir-based regimen resulted in significantly and markedly higher SVR rates compared to current treatment. This greater response was observed across genotype 1 treatment-naive and treatment-experienced populations, including relapsers, partial and null responders, and other groups that are traditionally considered poor responders to current treatment.
    Null responders, those with the poorest response to peg and ribavirin, have only a 5 percent chance of SVR with current treatment; but with telaprevir, null responders have a substantially improved outcome. Although SVR rates were lower in absolute terms compared to other populations, substantial numbers of null responders can be cured with telaprevir.

    In up to two-thirds of treatment-naive patients, these higher SVR rates can be achieved with 24 weeks of treatment, half the current treatment duration. Results from the clinical development program also support response-guided therapy for patients with prior relapse. Overall, these significant benefits outweigh the risks of adding telaprevir to the current peginterferon/ribavirin treatment.
    Illustrated here are the proposed treatment regimens. Response-guided therapy for treatment-naive and prior relapse patients will rely on HCV RNA evaluations at weeks 4 and 12, evaluations that are already in common use today. The regimen for null and partial responders is straightforward, at 48 weeks for all patients.
    Most patients who do not achieve an SVR had telaprevir-resistant variants after treatment failure. These represent 12 to 22 percent of all patients starting treatment. Telaprevir-resistant variants show cross-resistance to other protease inhibitors but do not confer cross-resistance to direct-acting antivirals with different mechanisms of action; for example, polymerase or NS5A inhibitors.

    Follow-up data indicates that resistant
    variants decrease in frequency over time, including in null responders, who have the highest rate of virologic failure. We recognize the clinical significance of these findings have not been evaluated, but it's encouraging that in most patients, resistant variants diminish and in many cases are no longer detectable over time.
    The data presented earlier by Dr. Singhal showed that the safety profile of telaprevir is well characterized and that adverse events are manageable. Rash and anemia are the most clinically significant adverse events associated with telaprevir.

    Most cases of rash are mild to moderate and are primarily described as eczematous. Rash is reversible after treatment completion. Severe rash is manageable with early recognition and sequential drug discontinuation. Few patients stop all treatment due to rash. Instances of severe skin reactions were rare. They resolved with treatment discontinuation.
    As you saw earlier, telaprevir produces an
    incremental, approximately 1 gram per deciliter, decrease in hemoglobin that responds to ribavirin dose reductions and to ribavirin and telaprevir discontinuation, if necessary. Only a small proportion of patients discontinued treatment due to anemia. SVR is not compromised by ribavirin dose reduction. Adverse event management for rash and anemia have been implemented in Phase 3 and can be applied in practice with appropriate education.


    Vertex is committed to providing information to treating physicians and patients on the risks associated with telaprevir and strategies for minimizing these risks. As we have described, the identified risks of telaprevir are rash and anemia. In addition, we plan to educate on other important topics, including the contraindications, warnings, and precautions outlined in the label; the importance of taking telaprevir with food, and adherence with the dosing regimen to minimize virologic failure; the need for precautions against pregnancy related to the teratogenic potential of ribavirin and potential reduction in the
    effectiveness of oral contraceptives through a telaprevir-drug interaction; and the potential for other drug-drug interactions that may result in the need for alternative medications or dose changes during the 12-week period of telaprevir administration.
    As we seek approval today for the proposed indication, we are also continuing to study the efficacy and safety of telaprevir in specific areas. Although we studied several hundred African American patients in our program, we plan to initiate a study later this year to expand our experience, particularly in those who have not responded to prior treatment and in patients with more advanced fibrosis.
    We have an ongoing Phase 2 study in HIV/HCV coinfection. A phase 3 study will follow once the data are evaluated. Several HIV regimens are being evaluated for use with telaprevir in the current study.
    We also plan to start a pilot study in post-transplant patients later this year.

    Hereto,drug-drug interaction data with the commonly used immunosuppressive agents have been obtained. Studies in pre-transplant patients and patients with hepatic decompensation will be quite challenging. We are in discussions with experts in the field to evaluate the feasibility of an initial study in this population.
    A pediatric dosage form is in development, and we plan a pediatric efficacy study for registration once this is available. And finally, we're also evaluating a twice-a-day telaprevir dosing regimen to provide better dosing convenience.
    In conclusion, as you heard today, hepatitis C is a significant public health problem. There's an unmet need for treatment options that can increase SVR and reduce treatment burden. Telaprevir-based treatment results in significantly higher SVR rates compared to current treatment in patients with genotype 1 hepatitis C.
    Telaprevir also reduces treatment duration for up to two-thirds of treatment-naive patients,
    and response-guided therapy has also been shown to be appropriate for prior relapsers. The safety profile has been well characterized, and we have implemented practical management strategies for rash and anemia to protect patients.
    The data presented today for telaprevir demonstrate that the benefits of telaprevir greatly outweigh the risks and represent a true advance in patient care. Thank you.


    Clarifying Questions from Committee to Sponsor

    DR. CARGILL: Thank you.
    Clarifying questions from the committee for the sponsor? Dr. Clay?
    DR. CLAY: Good morning. I want to commend Vertex on the array of drug-drug interaction studies they've already done as well as the ones they've planned. And in your ongoing and planned future studies, I saw on clintrials.gov that you also are doing one in people who failed telaprevir and continue in that study. So I want to commend you that that's already up there and listed.

    But I want to focus in on some of your safety issues.

    In the material provided by the Food and Drug Administration, they talk about your metabolite PZA, and they mention PZA is associated or a metabolite of niacin. But I'm more familiar with PZA as a metabolite of pyrazinamide, used in the treatment of tuberculosis.
    You're describing a side effect profile that we expect to see from pyrazinamide. You've got increased uric acid, including incidences of gout. You have fever. You have anemia. You have thrombocytopenia. You have a rash. I'm not that familiar with the presentation of the rash in tuberculosis treatment.

    But I guess this gets to the fact that you're giving single therapy to people, and I was curious if in your clinical trials, did you perform a PPD at baseline to see if someone was PPD-positive? And if so, had they already taken medicines for the treatment and prevention of tuberculosis?
    DR. KAUFFMAN: We did not provide that information. We did not necessarily test patients.
    It was really up to the clinicians who enrolled patients to decide if they were eligible for the trial. There was no specific inclusion or exclusion criteria related to prior tuberculosis status.
    I'll point out, though, that the concentrations of pyrazinamide, although it is -- pyrazinoic acid, although it is a metabolite of telaprevir, are much lower than the concentrations that are present when pyrazinamide is used as a therapeutic agent.

    DR. CLAY: I don't doubt that. But the side effect profile really looks like you're giving full-dose pyrazinamide. And I'm just curious how much information is provided to the investigators so that they could say, well, we need to test for TB in people before we give them this medicine.
    I guess maybe it gets to -- I didn't have a great deal of information about how much of your drug ends up being pyrazinoic acid, and so maybe that would be helpful to me to know that.
    DR. KAUFFMAN: Yes. I'll ask Dr. Garg to
    come up and discuss that.

    DR. GARG: Varun Garg, senior director of Clinical Pharmacology.
    If I understand your question correctly, you're asking about how much of a dose difference or exposure difference is there between pyrazinamide and telaprevir. Is that correct?

    DR. CLAY: No, not really. I don't really need to know how much of your drug ends up being pyrazinamide or pyrazinoic acid because it looks like there's a lot of pyrazinoic acid on board here.
    I guess maybe if you could clarify for me -- you're broken down into three different metabolites, and so your major metabolite is this. So maybe go ahead and tell me, in relation to normally what we would expect with administering pyrazinamide, how much pyrazinoic acid made it to the patient's plasma.

    DR. GARG: Right. So in the literature, the pyrazinoic acid concentrations that are described are about 14- to 150-fold higher when you
    administer a single dose of pyrazinamide to healthy volunteers. There is very little data on pyrazinoic acid PK, but from the literature that we could obtain, the levels are anywhere from 14-fold higher to 150-fold higher, or more, with a single dose of pyrazinamide.
    DR. CLAY: In your 24-hour urine collections, which I'm assuming you did, did you sample for 5-hydroxy?
    DR. GARG: No. We actually did not measure pyrazinoic acid in the urine. We only did that in the plasma.
    DR. CLAY: Thank you.

    DR. CARGILL: Thank you. I have a question that relates to slide -- two questions. The first relates to slide CS-6, in which you indicate there were 64 patients in the telaprevir/peg/ribavirin group that were noncompliant compared to 9 in the placebo group. And I guess my questions are, of those who were nonadherent, what were their characteristics?
    My second is, you also indicate in the
    presentation decreased evidence for reduced oral contraceptive effectiveness. And so I would like to know if female patients taking these medications were counseled to use an alternate form of birth control.
    DR. KAUFFMAN: So with respect to your second question, yes, they were. We recommended an additional barrier method of contraception during the period of telaprevir administration and for two months after. These patients did not necessarily have to stop their oral contraceptive, but obviously we could not rely on it for effectiveness, so additional barrier methods were recommended in the protocols.
    DR. CARGILL: And with respect to the first question about the 64 patients who were noncompliant?
    DR. KAUFFMAN: Yes. I don't have information on the characteristics of those patients. We measured adherence through various methods, and, yes, there was some during the telaprevir dosing period.

    DR. CARGILL: Ms. Dee?
    MS. DEE: You know, I noticed in the briefing document that you supplied to us that 50 percent -- I mean -- yes, the investigators thought that there was a rash over 50 percent of the body surface area, and once the DEP was in place, that they characterized it as 30 percent.
    It would seem to me that it would be easy to look and decide between whether it's 30 or 50 percent of a rash on a body surface.
    Why did it go down 20 percent once the DEP reviewed it?
    DR. KAUFFMAN: Yes. When the DEP reviewed these cases, they consider the intervening skin around areas of rash to be not included in the body surface area; whereas when the investigators looked at patients, they kind of estimated the body surface area by what they thought was the amount of rash that covered the skin, and they estimated it higher. The DEP, when they looked at it, they re-estimated the amount of body surface area and downgraded it for that reason. You know, if you
    have a patch of rash in a certain area, the investigators likely counted it as encompassing that whole area, whereas the DEP looked at the amount of intervening skin and made an estimate of the actual coverage of the skin surface area. So there was a difference.
    DR. CARGILL: Thank you.

    Dr. Ghany?
    DR. GHANY: Yes. In your Phase 3 trial where you compared T8 to T12 dosing, it was primarily to limit exposure to telaprevir. Could you tell us if there was a difference in the incidence of rash between those two arms? And if so, then I'd like you to answer a second question for me. Perhaps if you could explain what accounts for the difference in SVR rates between the T12 and T8 groups. I think the delta was 7 percent, but they had similar rates of early rapid virological response as well as similar relapse rates.
    So was it due to higher on-treatment response rates in the T12 group, or was it due to a higher rate of resistance?

    DR. KAUFFMAN: Yes. So let me just give a little background.
    The hypothesis behind the 8-week telaprevir duration was based on observations in Phase 2, where it appeared that there was a higher incidence of severe rash occurring in the third month of therapy, from weeks 8 to 12. Therefore, we decided to evaluate a shorter duration of telaprevir.
    We chose 8 weeks because all of our modeling and Phase 2 data suggested that we would not likely give up very much antiviral efficacy with 8 weeks of telaprevir. And that, of course, was the hypothesis that was tested.
    In the outcome of the study, it's true there was a small difference in the rate of severe rash, with about an absolute 2.5 percent lower rate of rash in the 8-week arm compared to the 12-week arm. But as you point out, there's a difference in the SVR rates, and that's largely accounted for by a somewhat higher rate of our breakthrough or virologic failure in the 8-week arm.
    So the 8-week arm has a little less
    Avirologic activity, if you will, and in a sense, that's the tradeoff for a very small difference in the occurrence of severe rash.
    DR. CARGILL: Thank you.

    Dr. Connick?
    DR. CONNICK: I wanted to ask a few questions about CE-14 and CE-31 relating to evidence supporting efficacy in subpopulations. I guess I want to say it's disappointing that your recruitment of blacks was so low, given the significant amount of disease in that population in this country.
    Let's see. On CE-31, I believe, the data are not -- the confidence interval crosses 1 for blacks. So I wanted to ask, do you have data from other studies that you can combine to convince us that in the treatment-experienced African Americans that there is efficacy?
    DR. KAUFFMAN: So as you point out, in large part because of the geographic distribution of the patients that were enrolled in study C216, with relatively fewer coming from North America, the
    inclusion of black and African American subjects was reduced in that population.
    We have enrolled African Americans in other Phase 2 studies in treatment-experienced patients, and their outcomes are really quite good, although, as has been true throughout the program, overall SVR rates in African American patients are lower than for Caucasians. Even though there's still a great incremental gain with telaprevir, nevertheless the absolute levels are still lower.
    But we have, yes, only a limited experience in the treatment-experienced population. But that was a reason for us considering an additional study that we are going to start later this year to try to expand our experience in that population, particularly in the treatment failure population among African Americans.

    DR. CONNICK: So I noted also on both of those slides, CE-14 and 31, that people over 65 again didn't -- the confidence interval crossed zero. Do you feel that there may not be benefit in treating older individuals, or was this because that was such a small number?

    DR. KAUFFMAN: Our view, obviously, is that the overall efficacy is very high with telaprevir. There's certainly no reason to think, a priori, that there would be any issue in those who are over age 65; that, as you point out, the confidence interval is quite broad because there were only small numbers of patients in that age range that were included.
    DR. CONNICK: Thank you.
    DR. CARGILL: Thank you.

    Ms. Valbh?
    MS. VALBH: Hi. I have a couple of questions. My first question is a follow up on Dr. Cargill's question about compliance. How did you assess compliance in these patients? What tools were used to assess compliance?
    Then my other question is, even though a small percentage of patients used EPO in your trial -- I think it was 14 patients -- at what point was the EPO initiated? So what was the hemoglobin level? And then also, if you can
    comment on, in the anemic patients, after telaprevir was discontinued, comment on the hemoglobin levels and how quickly they came back.
    DR. KAUFFMAN: Yes. There are several questions there. Maybe I can start with the first one on adherence.

    We estimated adherence based on pill counts, on drug accountability records, and on the start and the stop dates in the CRF. And in addition, patients filled out diaries about when they took their medications. Adherence overall was really very good. More than 95 percent of patients took more than 95 percent of their drug.

    Remember, the telaprevir dosing period is only 12 weeks, and therefore patients can be coached through that and can be coached on the importance of taking their medication as prescribed.
    Sorry. Can you remind me now of the other questions?
    MS. VALBH: Sure. On the 14 patients that were placed on EPO, what was the hemoglobin value when EPO was initiated?
    DR. KAUFFMAN: I'll ask Dr. Singhal to come up and address that question.
    DR. SINGHAL: So EPO was used in a small number of patients, and I'd like to just go back to the initial point that it wasn't permitted. Now, there was a special permission given for the C216 study, which was the REALIZE study, in France at the request of the study -- of the AFSSAPS. And so they were allowed to use EPO, and that was really where most of the EPO was used.
    So it wasn't really used in North America for the Phase 3 program. I don't have the exact levels, but it seemed to be based more on a practice rather than on a hemoglobin level.
    MS. VALBH: On a practice.

    DR. SINGHAL: And with regards to your question about hemoglobin recovery, I'd like to point out also on the curves that, as we demonstrated earlier, the hemoglobin rises up as soon as telaprevir is discontinued to peginterferon and ribavirin levels, and then gradually rises up
    to peginterferon/ribavirin levels over the following 12 weeks after telaprevir is completed, so by 24 weeks. And the most significant rise is after peginterferon and ribavirin are pulled off and completed.
    MS. VALBH: Okay. And, I'm sorry, I have one more question. On one of the slides there was a mention that there was guidance on symptomatic management of rash. Can you tell me a little bit about what that guidance was? Did you have an algorithm that, if a rash presented itself, start with a topic corticosteroid, or you start with an antihistamine? What was that guidance?

    DR. KAUFFMAN: Yes. I'll ask Dr. Singhal to come up and tell you about that.
    DR. SINGHAL: A very specific guidance was included in all the Phase 3 protocols, and let me begin by the general principles that were included.
    So the general principles were that physicians needed to follow the general management principles and use systemic antihistamines as required for pruritus or rash, as well as topical
    corticosteroids. There was a specific mention that if they needed to use topical corticosteroids for more than two weeks, they needed to connect with a medical monitor to allow whether it was going to be systemically absorbed.
    Very specifically, systemic corticosteroids were to be used only as indicated and if required when other modalities had failed because systemic corticosteroids, once they were used, we required telaprevir and other drugs to be discontinued. So that was the specific management.
    Now, this was a general principle management across all grades of rash. So I just want to go back to the management that I provided. This was the general management as well as the symptomatic treatment. We do have the details of how many patients used these treatments, and if you like, I can share that with you.
    With regards to study action, it was really very clear. Mild and moderate rash did not require any discontinuation, and that was more than 93 percent, as we saw. And then the severe rash
    required discontinuation of telaprevir alone.
    MS. VALBH: Thank you.
    DR. CARGILL: Thank you.

    Dr. Bigby?
    DR. BIGBY: I think that there's a great deal of unclarity about the definitions of the words that you're using, so I have several questions in that regard. I think probably Dr. Singhal would be the one to address these.


    You used the term "severe rash," and then in that slide at CS-16 at the bottom, you have "severe skin reaction." So could you just tell us what is your definition of severe rash?

    DR. KAUFFMAN: Yes. I'll ask Dr. Singhal to come up.
    DR. SINGHAL: For the Phase 3 program and the Phase 2 program, this definition was very specifically included in the protocol. Severe rash was defined as any rash which, per the investigator -- and maybe I can have SC-10, please; thank you -- which would -- the grade 1 rash was a localized eruption. A grade 2 rash was a diffuse
    eruption which maybe occupied less than 50 percent of the body surface area.

    Then as you see here, the grade 3 rash really encompassed two concepts. The first one was a generalized rash that occupied more than 50 percent of the body surface area. But it also included a rash that may indicate a severe skin reaction. So I'll stop here for a moment.
    The severe skin reactions, which is the second sub-bullet under the last bullet here, would be constituting a potentially life-threatening skin reaction which would be also possibly attributed to drug, such as Stevens-Johnson syndrome. And we wanted to make sure that the management for these two different types of rash was different.
    The first, generalized severe rash that did not have any of the features listed here, required only telaprevir to be discontinued. But if there was any suggestion of bullae, vesicles, mucous membrane involvement, target lesions, or others, it required all study drugs to be discontinued. And that was the distinguishing feature between the severe skin reaction and the severe rash.
    DR. BIGBY: That's clear. Okay. And the second part of the question is, did you have any cases of TEN?
    DR. KAUFFMAN: I'm sorry. I didn't hear that last part.
    DR. BIGBY: Did you have any cases of toxic epidermal necrolysis?
    DR. KAUFFMAN: No, we did not.
    DR. CARGILL: Thank you.

    Dr. McGovern?
    DR. MCGOVERN: I just have a couple of quick questions. The first one is just administration. You might have even mentioned this, but I might have missed it.
    I think your drug is two tablets per dosing interval. I just would like clarification of that. Also, if you can give us some guidance on what you described as food with the pills.
    A third one is related to resistance. You showed us a slide that shows the disappearance of RAVs much faster with genotype 1B patients.
    Do you have any information about the various RAVs between genotype 1Bs and 1A patients, whether the disappearance is related to viral fitness. And also, do you have any information on -- in the lead-in phase of C216, do you have information on whether there was any benefit in terms of the lead-in in terms of the formation of RAVs?
    DR. KAUFFMAN: Yes. I'll take the first part, and then I'll ask Dr. Kieffer to come up and discuss the resistance issue.
    You were right. The dosing was with two tablets taken three times a day. The recommendation for food was that telaprevir be taken with a normal meal or with a snack, and it was asked that the snack contain some fat. It didn't have to be a lot of fat, but some fat. But otherwise, a normal meal is what was recommended.
    Now Dr. Kieffer to answer the remaining questions.

    DR. KIEFFER: So, as you mentioned, we did observe a difference in the evolution of resistant
    variants during follow-up between subtype 1A and 1B.
    If I could have VR-52, please? And so the reason for that is that we do observe different variants between subtypes. So what this slide shows you is the type of variants we observed between subtype 1A and 1B. So the majority of resistant variants observed in subtype 1A include V36M, R155K, and the double variant V36M and R155K.
    In subtype 1B, those patients mostly have V36A, T54A, and A156S or T. And we believe that there's likely a fitness difference in these variants, and that is what causing the observed difference in the decline of the resistance over time in the absence of treatment.
    To your last question about differences in resistant mutations in the lead-in versus no lead-in, if I could show VR-56. This slide will show you the different types of treatment failure and whether or not patients had detectable resistance. The wild type is in green; these are patients that did not have any detectable resistant variants by
    population sequencing. And then orange indicates patients where telaprevir-resistant variants were detectable. And you can see that there's really no difference between the arm with a lead-in or without a lead-in for all categories of response.
    DR. CARGILL: Thank you.

    Dr. Murata?
    DR. MURATA: Thank you. I have questions on anemia management, if there was a brief -- regarding slide 26 of the safety section -- if the guidance on hemoglobin monitoring can be briefly reviewed.
    With respect to the blood transfusions, in addition to the percentages that you've shown, if you can provide additional details; for example, number of red blood cell units transfused per patient or how many transfusions were permitted prior to drug discontinuation.
    DR. KAUFFMAN: Yes. I'll ask Dr. Singhal to come up to address that.
    DR. SINGHAL: Please, may I ask you to repeat your question? I'm not sure I followed it.
    DR. MURATA: So the question is in two parts. One, if you can briefly provide an overview, as you did for the rash, but instead for the hemoglobin monitoring. Second, if you can provide additional information or clarification of the blood transfusions that were permitted and occurred during the study; specifically, the number of transfusions that were permitted prior to drug discontinuation, and on average, for those who required transfusions, how many transfusion events did occur.

    DR. SINGHAL: So with regard to the transfusion, there wasn't any specific guidance within the protocols that linked transfusion to discontinuation. However, there was guidance to follow the ribavirin label.
    So, in general, investigators did follow the guidance, and the guidance was that if it was less than 10 grams per deciliter, then investigators had to reduce ribavirin -- had to follow ribavirin dose modification; and if it was less than 8.5, they needed to discontinue. And whenever they
    discontinued ribavirin, we did have a rule in place that they needed to discontinue telaprevir.
    So, in general, that was the guidance that we provided. Transfusions were used in 6 percent of the patients across the Phase 3 program, and I don't have specific values. These were also -- some of these were based on local practice. And so there wasn't necessarily a trend across the level at which they were used.
    DR. CARGILL: Thank you.

    Dr. Strader?
    DR. STRADER: I'd just like to follow up on that point you just made. Were there any patients who had anemia occur after the 12 weeks of telaprevir was given? And if so -- obviously, if you're discontinuing the ribavirin, the telaprevir dose is already finished -- do you treat those patients with peginterferon alone or is the dose discontinued, both medications discontinued at that point?

    DR. KAUFFMAN: Dr. Singhal?
    DR. SINGHAL: So if I understand your
    question correctly, did anemia occur after telaprevir was discontinued? Yes, there were some cases of anemia that were reported after telaprevir was discontinued. I just want to remind you that the way that the calculation and the displays are done, we pick up the worst and the most severe event of anemia, and these tended to occur in the first 12 weeks.
    So although patients may have experienced an event later on, you wouldn't necessarily see it on that display. And after that, if they had discontinued telaprevir and ribavirin, they could continue through the study on peginterferon, and of course if they had to stop that because the anemia continued. But those were fewer instances, very few.
    DR. STRADER: Two more quick questions. One of them, there's a slide CS-17 where you're showing the rash with telaprevir and with placebo. And it appears that you indicate that the rash continues even though the telaprevir dose is discontinued, and for a long period of time.
    But I was under the impression that once the telaprevir dose was discontinued, the rash resolved. So can you just clarify that slide for me, please?

    DR. KAUFFMAN: Yes. Dr. Singhal?

    DR. SINGHAL: Yes. I can clarify that. The point here was that, in general, we have noted that there is a background rate of rash, mild and moderate rash, with peginterferon and ribavirin, to the rate of about 34 percent. When we added telaprevir to that regimen, there was an increase in the rate of mild and moderate rash. And then there was a severe rash that wasn't really noted at all in the placebo/peginterferon/ribavirin group. It was only noted in the telaprevir groups, as we demonstrated.
    So when telaprevir was discontinued by the rules that we had established, remember, the patients were still on peginterferon and ribavirin, so they had the underlying peginterferon and ribavirin. We believe that that may be contributing to an increased time to resolution.


    In general, the severe rash could take about 4 to 6 weeks to resolve completely. That was the median time to resolution, which is why I wanted to make the point that it doesn't go away immediately.
    DR. STRADER: Okay. One last question. The treatment of treatment-experienced patients, you chose for some reason not to use response-guided therapy in those groups. Is it because you had done some Phase 2 studies that you don't show us here that show that there was a response in response-guided treatment? Why was it that you decided not to do response-guided therapy for treatment-experienced patients?

    DR. KAUFFMAN: When we designed the C216 study, some of those data were not available, and we chose the conservative route of treating patients for the full duration, given that they had already failed treatment. Also, there's only so many questions one can usually ask in a clinical trial, and therefore we decided to keep it simple and just used 48-week treatment for all patients.

    DR. VAN DYKE: Yes. I had one question about viral resistance. You mentioned that there was cross-resistance with other protease inhibitors. Does the drug have any activity against HIV, and is there potential for developing protease inhibitor resistance against HIV protease inhibitor drugs?

    DR. KAUFFMAN: Yes. We actually have examined that in vitro, and there is no cross-resistance with HIV. There's no effect on HIV.
    DR. VAN DYKE: Great. Super. My other question is, could you give us more clinical details on the anorectal symptoms in terms of what they were, what did they look like clinically, and how long did they last, and how much of a problem really were they for the subjects? Because they were quite common, like 29 percent.

    DR. KAUFFMAN: Yes. We first became aware of this in Phase 2 with increased reports of hemorrhoids. And, frankly, I don't think it was really hemorrhoids, but hemorrhoids are present very commonly, so when you take a look, you often will find hemorrhoids, even if they're not really related to the phenomenon.
    We have really only anecdotal reports. But I have spoken to many investigators about this phenomenon. Upon examination, there is no inflammation. Either externally or on anoscopy, there's been no inflammation. And I'll just point out parenthetically that in all of our nonclinical studies, there is no inflammation or other findings related to the colon with telaprevir.
    It's described as a burning sensation or an itching sensation. It occurs relatively rapidly after the beginning of administration of telaprevir. It actually often resolves either during the period of dosing or very rapidly thereafter. And as we pointed out, it's almost never associated with treatment discontinuation. It is an annoying side effect, but it really doesn't have any serious consequences in terms of compliance with the treatment regimen.
    DR. CARGILL: Thank you.

    Dr. Knodell?
    DR. KNODELL: I'm just trying to get a global feel for how often you have to discontinue this treatment. And I'm looking at slides C-6 and C-10. And before I ask the question, I assume that in your response rates, everybody who got telaprevir, even if they only got a couple doses and discontinued treatment, is included in your analysis?
    DR. KAUFFMAN: That's absolutely correct. It was intent to treat, and everyone who started treatment was accounted for at the end.
    DR. KNODELL: Your response rate for people who actually complete treatment is really quite a bit better than what you list.
    My question is, it looks like that you have, overall, fairly severe adverse events in about 16 percent of your telaprevir group versus 7 percent overall in your dual therapy, and that your discontinuation rate, if you look at all discontinue rates that aren't for viral failure, is about 20 percent versus 10 percent.
    Going forward, is that what we can expect,
    that probably about 20 percent of patients aren't going to be able to make it through this treatment regimen?
    DR. KAUFFMAN: Obviously, I can't speculate on what will happen in the future. We can only say what we saw in the trials. But yes, there are an increased rate of discontinuations due to telaprevir compared to the control treatment. This is due to a number of causes, but obviously, as we've pointed out, most prominently due to anemia and to rash.
    DR. CARGILL: Thank you.

    Dr. Korman?
    DR. KORMAN: I wanted to ask a question about the side effect profile. You pooled the data for the treatment-experienced and treatment-naive patients, I think; that's all that I saw. I'd be interested in knowing whether the treatment-experienced patients that have more advanced liver disease have a different side effect profile and whether the rash can be predicted by treatment-experienced patients having had skin reactions of
    either mild, moderate, or severe characteristics.
    At least in my experience, many of the patients, particularly the female patients, complain of hair loss, and I notice that that's not included. I wondered if that actually is also something that other hepatologists see, and whether you can comment on seeing that during this process. It's clearly a very disturbing symptom.
    DR. KAUFFMAN: Yes. Just overall, the adverse event profile in the treatment-experienced patients was very similar to the treatment-naive patients, and that's why we were able to pool those groups together.
    I'll ask Dr. Singhal to come up and address your other questions in more detail.
    DR. SINGHAL: So I'll just address, actually, your second question first about the hair loss.
    DR. KORMAN: Complicated.
    DR. SINGHAL: Okay. No, I just wanted to mention that the alopecia, which is actually commonly noted with peginterferon and ribavirin, we
    did not see any increased rates with telaprevir at all, which is why it was not mentioned as part of -- and it's below 20 percent, so it didn't make it to the most common AEs, either. So that's for --

    DR. KORMAN: The second question.
    DR. SINGHAL: -- the second question. And the first question is that the treatment failure and the treatment-naive patients actually looked very similar in their safety profiles. There were a few events that seemed to occur at a higher incidence in the treatment-naive population, and these are primarily nausea and vomiting. With the rash as well as the anemia, we did not find any notable differences.
    DR. KORMAN: Just quickly, about photosensitization, did you make any recommendations?
    DR. SINGHAL: Right. So we did examine that from a nonclinical perspective, the photosensitization, and we do not believe that it contributes to the rash in any great way. From a clinical perspective, the dermatology expert panel also examined this based on about 150 cases with photographs that they examined, and there did not seem to be any indication of it being photosensitive.
    DR. CARGILL: We will now take a 15-minute break. For those committee members who still have outstanding questions, we will return to those questions.
    Panel members, please remember that there should be no discussion of the meeting topic during the break among yourselves or with any member of the audience. We will resume promptly at 10:30.
    (Whereupon, a recess was taken.)
    DR. CARGILL: We're going to resume the meeting, if you could please take your seats.
    I'm going to take the chair's prerogative, as there have been a number of questions around rash and the side effect, and ask the sponsor to please show us the slides of the rash before we proceed with the FDA presentation.
    DR. KAUFFMAN: Thank you. It's coming up now.
    Can we put the slide up?
    DR. CARGILL: Thank you.

    DR. KAUFFMAN: It's a little difficult to see. It's a little dark on the screen.
    DR. CARGILL: We're going to ask for the lights to be dimmed, if you can give us one moment, so this can be seen very well.
    DR. KAUFFMAN: Sure.
    MR. TRAN: We are asking the staff to dim the lights for us. Thank you. Hang on for a moment.
    [Pause.]
    DR. CARGILL: Now that we actually have seen this rash, I'd like to just ask the committee members if they have any specific questions after seeing this, to please address them now so we can go on to the next presentation.
    DR. KAUFFMAN: If you like, I can go through the slide. It's just not quite so visible.
    DR. CARGILL: Why don't we let the sponsor go through the slide, and then if you would follow up with your question.

    DR. BIGBY: Sorry?
    DR. CARGILL: I said we're going to let the sponsor go through his slide, and then have you follow up with your question. Thank you.
    DR. KAUFFMAN: Fine. So this slide shows a series of photographs of a rash that's been associated with telaprevir.
    I'll make a couple of general points. One is that -- if I can use the word "typical," and I use it advisedly -- the typical telaprevir rash is as described. It's a primarily eczematous eruption. And the difference between mild, moderate, and severe is related primarily to the amount of skin area that's involved. The rash itself morphologically is the same; it's just the amount of skin involvement.
    As we noted, the vast majority of rashes are mild and moderate, 93 percent, and those are shown in the two panels on the left: mild, which is described as localized, as we pointed out, and moderate, which is present on less than 50 percent of the body surface area.

    It's a little difficult to see, but the severe rash, which occurs on more than 50 percent of the body surface area, represents only about 6 and a half percent of all rash events. And there's a closeup of that rash that shows you the amount of skin that's involved and the eczematous appearance.
    DR. CARGILL: Thank you.

    Dr. Bigby?
    DR. BIGBY: I think it would be informative to the panel if you could also, if you have it available, just show an example of a case that was described as a potential SJS case, and also show the panel what a patient that has DRESS looks like.
    DR. KAUFFMAN: I'm told we don't have slides available of either of those two available to show today.
    DR. CARGILL: All right. Thank you.
    We will now proceed with our presentation from the FDA. I would like to remind public observers at this meeting that while this meeting
    is open for public observation, public attendees may not participate except at the specific request of the panel.


    FDA Presentation – Russell FleischerMR. FLEISCHER: Good morning. On behalf of my colleagues on the telaprevir review team, it's a privilege to present the FDA's perspective on the telaprevir NDA.
    Our presentation will focus on efficacy issues and safety issues. We're going to review some issues related to the Phase 3 trial design. Dr. Jadhav will come up and talk about some pharmacometric analyses on the response-guided therapy for prior relapsers. Then I'll come back and talk about virology, IL28B, and discuss a few interesting and important subgroups. I'll switch to safety, talk about rash, anemia, a little bit on the anorectal disorders, cover some clinical adverse events and some laboratory abnormalities.
    So you've seen the trial designs from the sponsor, and you've seen the overall efficacy results. But each trial was actually designed to
    answer an additional question. So in the study 108, the T8 arm was included to test if a shorter duration of treatment might change the risk-benefit assessment by improving tolerability, particularly by reducing the frequency of severe rash while not sacrificing efficacy.
    In study 111, a 24-week regimen of peg and ribavirin versus 48 weeks was included to see if there was an increased efficacy associated with longer duration therapy for subjects who achieved an early response. And there was a delayed versus immediate start in study 216 to assess the effect of a short course of treatment with peg and ribavirin on the frequency of emergence of resistant strains during telaprevir exposure and to determine whether this strategy had an overall impact on efficacy.
    So on study 108 -- you've seen some of this data -- about 58 percent of subjects in the T8 and T12 arms achieved eRVR, which was undetectable HCV RNA, at weeks 4 and 12. The SVR rates for the two regimens were 87 percent for the T8 group and 92 percent for the T12 group, which was about a 5 percent improved SVR rate. The difference was numerically higher but not statistically significant, and on a formal test for noninferiority, the 8-week duration would have failed to meet our overall standard for clinical equivalence.

    In further looking at the advantages for the T12 compared to T8, as I just showed you, the SVR rate in early responders was higher at 92 versus 87 percent. For subjects who had no EVR and received 48 weeks of peg and ribavirin, the SVR rates were 62 percent for the T12 group, 55 percent for the T8.

    The breakthrough rate on peg and ribavirin after telaprevir was completed was lower in the T12 group at 10 percent, compared to 16 percent. And I'll show you a little bit later that there was a suggestion for increased responses for CT and TT genotypes with the T12 regimen. We hypothesize that the extra 4 weeks of telaprevir suppressed potential substitutions that confer resistance to telaprevir. Disadvantages, which you heard a little bit, was that there were additional cases of rash and anemia between treatment weeks 8 and 12.
    In study 111, all 540 subjects started on telaprevir and peg/ribavirin for 12 weeks. Of those, 60 percent, or about 322, achieved an eRVR and were randomized to 162 to 24 weeks and 160 to 48 weeks of peg and ribavirin. You can see from the table the SVR rates were pretty comparable at about 90 percent.

    For subjects who did not achieve an eRVR and received 48 weeks of peg and ribavirin, the SVR rate was 66 percent, which was again generally comparable to what was seen for the same population in study 108.
    In study C216, studies were randomized to either an immediate start or, as the applicant said, a lead-in. We use the term "delayed start." It's the same thing, 4 weeks of peg and ribavirin, and then initiation of peg and ribavirin -- I'm sorry, 4 weeks of peg and ribavirin, and then telaprevir was added.

    You can see that for all subjects, the SVR rates were comparable. Same for prior nulls. Prior partial relapsers -- I'm sorry. Prior partial responders and prior relapsers all were pretty comparable to each other, and when we further investigated, looking for differences, we didn't find any between the SVR rates for the delayed start compared to the immediate start for any nonresponse group or by any demographic or disease covariates.
    So I'm going to turn it over to Dr. Jadhav to talk about his analysis of RGT in treatment-experienced prior relapse subjects.
    FDA Presentation – Pravin Jadhav
    DR. JADHAV: Thank you, Dr. Fleischer.
    Good morning, everyone.
    For my part of FDA presentation, I have one key question: Is response-guided therapy for prior relapse subjects acceptable? Why are we asking this question? Dr. Fleischer, as well as the sponsor, shared with you that the response rate in treatment-naive subjects who achieve eRVR was
    greater than 90 percent whether peg/ribavirin was given for 24 weeks or 48 weeks.

    Therefore, study 111 provides us a direct within-trial comparison to recommend 24 weeks of peg/ribavirin for all treatment-naive subjects who achieve eRVR. However, to recommend a 24-week duration of peg/ribavirin for prior relapse subjects who achieved eRVR, we have evidence from cross-trial comparison.
    What I'm going to show you in the course of this presentation, that we can not only use data from cross-trial comparison to support 24 weeks of peg/ribavirin duration for this population, we can also use data from naive subjects to inform us about durations in prior relapse subjects. Also, viral dynamic modeling that sponsor has done, and I'm not going to present -- guide us in the same direction.
    For the first evidence, we look at three studies, study 106, 107, and C216, as it has been presented before. C216 was the Phase 3 study, and 106 and 107 were the two Phase 2 studies.

    These are the same data that sponsor has showed earlier, and in these studies, the prior relapse subjects were treated. So starting with the first bar in this plot, we observed greater than 90 percent SVR with 24 weeks of peg/ribavirin in prior relapse subjects who achieve eRVR. Similarly, the second and third bar shows that even with 48 weeks of peg/ribavirin in patients who achieved eRVR, the response rate was greater than 90 percent.

    The point I would like you to take from the slide is that the response rate for prior relapse subjects with eRVR was greater than 90 percent, suggesting that 48 weeks of peg/ribavirin does not provide additional benefit on SVR.
    Remember, the SVR rate for treatment-naive subjects overall with eRVR was also 90 percent, as sponsor has presented, for the same treatments, 24 weeks or 48 weeks. And please note that in all these treatment arms, telaprevir was given for 12 weeks.

    So for those of you who believe these cross-trial comparisons and similar evidence from naive, you can ignore rest of my talk.
    [Laughter.]
    DR. JADHAV: But if you don't, if you believe this is a cross-trial comparison, I would like you to provide second set of argument why we can support 24 weeks of peg/ribavirin for prior relapse subjects who achieved eRVR.
    From this evidence, I'm going to take you to treatment-naive data. So before I show any data, let me first introduce, or in fact one more time try to introduce, a concept that we need to carefully think about, treatment-naive and -experienced subjects in context of peg/ribavirin.
    Let's start with the treatment-naive subject. And when the subject is going through the first round of therapy, there are four possible outcomes. The subject might end up to be a responder, relapser, partial, or null. The treatment failures here are now called treatment-experienced, and to be very specific, these should be called peg/ribavirin-experienced.

    In HIV subjects, if this was HIV treatment, these experienced patients also means they're resistants, and they would fail under retreatment. What we are able to see from this program, as well as the program that we reviewed yesterday, that if these subjects go through the second round of treatment, the response to peg/ribavirin is not lost.
    What I'm going to show you here is data from study 108, where the treatment-naive patients who eventually ended up to be relapsers, what happened to them at week 4, we knew the response status in study 216 at the baseline, what happened to them at C216.
    So for the terminology, what I'm going to call the patient on the left-hand side to be a future relapser because at week 4 I did not know the end-of-study status. For the patient on the right side, the patient will be prior relapser because we know the status. We find that the response at week 4 for the first -- that's the first row in this table, where it's a future relapser -- is the same compared to week 4 response to peg/ribavirin for a prior relapse patient.
    So what I'm trying to show you here is whether the relapse status was known or unknown, peg/ribavirin response is not lost; it's still the same. What I'm not saying here is that week 4 response can tell you who the relapser is. That is not the point, that the patient responsiveness stays the same on the first and the second round of treatment.
    So for those of you who treat patients, think about it this way. Should the treatment approach be different, if the responsiveness PR is the same, but only that you know that they failed on the peg/ribavirin before, the point is the peg/ribavirin response is same even for these two patients.
    The point I am trying to make, as I've been repeatedly saying, the patient's responsiveness to peg/ribavirin is not lost in peg/ribavirin failures. Therefore, whether it's a known relapser or a future relapser, a patient should be treated in a similar way.

    So if you agree with that principle, let me take you to treatment-naive data and let's understand, going through the treatment-naive population.
    Here is the distribution of end-of-study outcomes for peg/ribavirin, the control arm only in treatment-naive, treated for 48 weeks of peg/ribavirin. We found 44 percent were responders, 18 percent relapsers, 19 percent partial responders, and 9 percent were nonresponders. So at the outset, our treatment-naive population is this mixed bag of future relapsers, future partial responders, future nulls, and obviously, the responders.

    Because peg/ribavirin and telaprevir treatments were randomized, this distribution must be present in the telaprevir-treated group. So in telaprevir-treated group, when we find 79 percent of responders to triple therapy of which 54 percent had SVR and eRVR and 25 percent had SVR without eRVR -- so just to clarify, that 54 percent number
    comes from 58 percent achieving eRVR, and of them, 92 percent achieving SVR. So if you take both combined, it's about 54 percent.
    Now, you've got to stay with me on this.
    [Laughter.]

    DR. JADHAV: If you think about responders, 44 percent who are responders -- and hepatologists can help me with this. If you are a responder to a dual, which is the left-hand side, you are most likely to be a responder for triple. Correct? Makes sense. Once you are a responder for dual, you're a responder for triple. So we accounted for 44 percent in the right pie chart.

    Let's think about who are these additional 10 percent who are achieving eRVR and SVR in overall population. Who is the most next likely group who we are going to give 24 weeks of peg/ribavirin because we studied them in treatment-naive population? Remember, the entire treatment-naive population is going to get 50 percent of them, SVR and eRVR, ending up to be the 24 weeks of duration.

    So who are these patients? Most likely they are the relapsers, your future relapsers. So the point I'm trying to make is the most likely group is the future relapsers. So if you are treating a future relapser without knowing who they are, should you not be treating a prior relapser because their treatment response is the same, with shorter duration of peg/ribavirin?
    Therefore, the 90 percent response rate you see in the EVR subjects from treatment-naive population, as realized from the relapser, is no artifact. In fact, I would credit sponsors one of the statements in their slide, it is not that the treatment-naive population and the relapsers are presented at 90 percent high interferon responsiveness, it's the patients with eRVR -- that's more important, the response to the current therapy -- are the ones that represent population with high interferon/ribavirin response, and the baseline status is less important than how you're responding to the current treatment than the previous.

    Again, if you're interested, I can show you. Simple mathematics can explain why both numbers turn out to be 90 percent. It's no artifact.
    So the point I'm trying to make is we can reasonably say that prior relapse subjects with eRVR can achieve SVR with 24 weeks of peg/ribavirin, about 90 percent of them. We have learned that the first and the second round of peg/ribavirin treatment does not make patients lose their responsiveness to peg/ribavirin. If that is so, the eRVR on the new treatment is more important than what happened to them in their previous round of therapy. Combined, these arguments support that prior relapse subjects can be treated with 24 weeks of peg/ribavirin therapy, given the 90 percent SVR from cross-trial comparison.
    Thank you, and Dr. Fleischer will present other analysis from FDA.
    FDA Presentation – Russell Fleischer
    MR. FLEISCHER: Okay. So I'm going to give you some virology.
    We found that almost all of the subjects who
    did not achieve an SVR and who failed on telaprevir before week 12 had a treatment-emergent telaprevir-resistant substitution, and the majority of subjects who did not achieve an SVR and failed on peg and ribavirin after week 12 or who relapsed also had treatment-emergent telaprevir-resistant substitutions. And, overall, more treatment failures were in subtype 1A subjects.
    So here are the treatment-emergent substitutions. We examined the data for these substitutions that emerged on telaprevir, and these are in subjects who did not achieve an SVR in the pooled Phase 3 studies.
    You can see there's different resistant pathways for telaprevir failures with subtypes 1A and 1B. And in the light blue are the subtype 1A failures and the most frequent emergent substitutions, and this group were the V36M, R155K, or the double of these two.
    In dark blue are the subtype 1 failures, where the most frequent substitutions were the T54A, S, V36A, A156T, S, or V. On the far right,
    we also found that about 1 percent of 1A failures had the D168N substitution, which conveys cross-resistance to other macrocyclic protease inhibitors.
    We looked at the persistence of telaprevir substitutions at the end of the Phase 3 study. So these were subjects who have failed the telaprevir-containing regimen in the Phase 3 trials. They were followed at multiple time points after telaprevir failure off-treatment to assess the persistence of the substitutions at the end of the study. After median follow-up of 45 weeks, 40 percent of the subjects had detectable resistant variants by population sequencing by the end of the study.
    On the right side of the graph, you can see that 50 percent of subjects with subtype 1A in light blue had detectable substitutions at end of study, and these were the V36M or A and the 155K. About 20 percent of subjects with subtype 1B -- I guess it's sort of more in purple; on my screen it's dark blue -- were still detected by the end of
    study, and these were the V36A, T54, but the A156T had gone away.
    Vertex has already described study 112. And in this trial or in this study, 56 subjects were followed off-treatment for persistence of resistant variants after failure on telaprevir in the Phase 2 trials. Follow-up periods ranged from 5 to 40 months, with a median of about 25 months.

    The denominators reflect the numbers of subjects with an available sample. And you can see that at six months, which is I guess the far left, grayish-looking bar, a high proportion of V36, T54, R155, and A156 substitutions persisted.

    All variants were still detectable in some subjects at 24 months, which is sort of the bluish-green. And by 36 months, shown in light green, most variants had fallen below the level of detection by population sequencing. However, the R155K variant was still detectable by population sequencing at 36 months in 3 percent of subject isolates.
    Prior null responders in study C216:
    70 percent of the subjects who did not achieve an SVR in that study were prior null responders. Here are the rates by the treatment arms, and you can see they're fairly similar. And of these prior null responders who did not achieve an SVR, 80 percent of them had treatment-emergent telaprevir substitutions.
    So the virology summary: Treatment-emergent substitutions emerged in the majority of isolates from subjects who did not achieve an SVR. There's divergent resistant pathways, depending on subtype. More failures were subtype 1A and null responders. Some substitutions can persist upwards of 2 to 3 years, especially the R155K. And we don't know yet if these substitutions will affect future treatment options.
    Nobody's interested in IL28B, and people are pretty familiar with the 60 SNP, which strongly determines the outcome of HCV therapy and three genotypes of CC, CT, and TT.
    The applicant obtained samples from about 1374 subjects from their Phase 2 and 3 trials, and a retrospective sub-study on populations from about 975 samples from subjects in studies 108 and C216 was conducted. I'm going to show you the sub-study results, but just to let you know that the treatment effects in the sub-study were generally comparable to the overall population.

    For naives, the addition of telaprevir increased the SVR rates across all three genotypes compared to peg and ribavirin alone. And you can see that there's a suggestion here that the T12 regimen may be better for CT and TT subjects compared to T8.
    In treatment-experienced subjects, again, the addition of telaprevir substantially increased the SVR rates of all genotypes relative to retreatment with peg and ribavirin alone. And here we didn't really see any major differences between the immediate and delayed start of telaprevir.

    So, in general, the results are consistent with previous reports of IL28B genotype effects on peg and ribavirin responses, and that CT and TT subjects had lower SVR rates in the peg/ribavirin
    arms, and telaprevir increased the SVR rates in both treatment-naive and treatment-experienced subjects across all three genotypes compared to peg and ribavirin.
    It's important to point out that this was a retrospective subgroup analysis. Subjects were not stratified at the beginning of their studies by genotype, and in some groups the sample size was rather small. Further, this cohort may not be fully representative of the overall population of chronic hepatitis C patients as there is a minimal contribution of samples from black subjects.
    Overall, telaprevir increased the SVR rates by anywhere from 28 to 40-plus percent compared to peg and ribavirin across a broad range of disease and demographic covariates, including age, sex, body weight, race and ethnicity, geographic location, genotype subtype, high versus low baseline RNA, and cirrhosis. But we wanted to review the results of some important subgroups of interest, including blacks, Latinos, and subjects with cirrhosis.

    So overall, about 9 percent of subjects in the telaprevir arms of the Phase 3 trials were black, compared to about 8 percent in the peg and ribavirin groups. In general, the SVR rates for telaprevir-treated subjects was 30 percent higher than those treated with peg and ribavirin.
    In the table, you can see the total -- you can see the distribution in the top half of the table by naive versus experienced, so about 11 percent of naive subjects and 4 percent of experienced subjects in the telaprevir arms were black. The overall SVR rates were, for naives, about 10 percent lower than Caucasians, and for experienced, they were very close, at 63 and 65 percent.

    This table shows the proportion of naive blacks who achieved an eRVR and their SVR compared to Caucasians. So in the top box, about 45 percent of blacks compared to 63 percent of Caucasians in the two naive trials achieved an eRVR. And of those, you can see the SVR rates were 76 percent for blacks compared to 87 percent for Caucasians.

    So overall, in experienced subjects, the SVR rates were similar to Caucasians. In naives, the eRVR and SVR rates were lower than Caucasians, which is fairly consistent with previous peg and ribavirin trials, but still looks pretty encouraging. However, for some of these groups, the numbers of subjects was again kind of small.

    For Latinos, 10 percent of telaprevir and 12 percent of peg/ribavirin subjects were Latino. In naive trials, it was 10 versus 11 percent, and in experienced trials, it was 10 versus 15. The overall SVR rate was 40 percent higher for telaprevir compared to peg and ribavirin. In treatment-naive subjects, the SVR rate was just a little bit lower in treatment-naives for Latinos, and in treatment-experienced, it actually was just a little bit higher. So it was pretty much comparable.
    Treatment-naive subjects with cirrhosis, overall, about 9 percent of the telaprevir subjects and 6 percent of peg/ribavirin subjects were cirrhotic at baseline. For the naive trials, the top part of this table shows study 108. There were 47 subjects enrolled, 26 to the T8 group and 21 to the T12 group. You can see the eRVR rate for the T8 and T12 were 42 and 43 percent, which is pretty comparable, and the eRVR to SVR rate was 64 percent for T8 compared to 78 percent for T12.
    In study 111, 61 out of 530 subjects, or 11 percent, were cirrhotic at baseline. A total of 30 of them, or 49 percent, achieved an eRVR, which was fairly consistent with what was seen in study 108. And then these subjects were randomized, 18 to 24 weeks of peg and ribavirin and 12 to 48 weeks, and you can see the SVR rates on the right, 67 percent for the 24-week group, which is sort of consistent with what we saw in study 108. But for the T12 PR48 group, the SVR rate was much higher, 92 percent. So there's a suggestion for an improved SVR with 48 weeks of peg and ribavirin. And again here, the numbers of cirrhotic subjects in some of these groups is very small.
    This is treatment-experienced subjects with cirrhosis. As the applicant pointed out, about a quarter of the subjects in this trial -- this was C216 -- were cirrhotic at baseline. The numbers suggest a substantial benefit for prior relapsers, 87 versus 13 percent. For prior partial responders, it's a moderate benefit of about 14 percent improvement. But for prior nulls, it's only about a 4 percent improvement. So it's a fairly minimal benefit. But again, in some of these groups, the numbers were rather small.

    I'll switch over to the safety review. As you heard, 3800 healthy and infected subjects have been exposed to at least one dose of telaprevir. In the Phase 3 trials, about 1800 were treated with telaprevir, a little over 1400 with the T12 regimen with peg and ribavirin for either 24 or 48 weeks, and about 364 subjects were in the T8 plus peg for 24- or 48-week group, and 493 subjects were randomized to peg and ribavirin. Because telaprevir is given for a fixed period, we primarily focused our review also on the telaprevir placebo dosing period.

    Rash and pruritus you've heard a lot about. It was identified in the Phase 2 trials. A detailed monitoring and management plan was put into place for Phase 3. You've heard about the rash special search criteria. Severe rashes and rashes that led to discontinuation were considered events of special interest. And there was a dermatology expert panel convened to retrospectively adjudicate cases.
    Rash events: 56 percent for telaprevir subjects versus 34 percent for peg/ribavirin. There were more grade 3 severe rash at 4 percent versus less than 1 percent. About 1 percent of telaprevir subjects had a serious adverse event of rash compared to no subjects treated with peg and ribavirin. The rash event of special interest was 7 percent compared to less than 1 percent.

    Pruritus by itself was reported in about 47 percent of subjects compared to 28 treated with peg and ribavirin. And when you looked at rash and/or pruritus, it was 73 percent for telaprevir versus 48 percent for peg and ribavirin.

    Rash was managed a number of ways, as you've already heard. About 7 percent of subjects actually discontinued telaprevir due to rash compared to less than 1 percent who discontinued placebo. Just under 1 percent discontinued the entire regimen compared to no subjects in peg and ribavirin.

    Use of oral antihistamines, topical steroids, and systemic steroids were higher for treatment of severe rash among subjects who were treated with telaprevir. And in this group, there were a number of subjects who received more than one intervention.
    You heard about the dermatology expert panel. And their conclusions were that this rash was clinically and histologically similar to peg and ribavirin. It was a pruritic, eczematous, spongiform dermatitis with lymphocytic perivascular infiltration. However, it was more severe and more extensive, and did occur at a higher incidence.
    The majority of evaluable rashes by the expert panel involved less than or equal to 30 percent of the body surface area, but investigators frequently estimated the extent of body surface area to be greater.
    In general, the rash improved after discontinuation of telaprevir, but it could take up to weeks to resolve. And less than 1 percent of subjects experienced suspected -- I'm going to induce another new term -- SCAR, which is severe cutaneous adverse reactions. And those are things like SJS, TEN, and DRESS. And as you heard, there were no cases of TEN. And among the cases of SJS or DRESS, there were no fatal outcomes. All subjects recovered.
    We obtained a consult from our colleagues in the dermatology division, and some of the issues that were identified include that there needs to be a distinction between severe rash and SCAR, and that distinction is important because SCAR has implications for morbidity and mortality that may not attach to severe. So, for example, someone could have severe acne, but it's not necessarily a SCAR event.

    Most SCAR events, particularly as they relate to DRESS, were suspected on case review, retrospective case review by the expert panel and not by investigators, which may actually suggest under-reporting and under-diagnosis of these events in the field. And the detection of suspected SCAR cases may be noteworthy given that SCAR are generally considered to be rare, and the sample size of clinical trials intended to support marked improvement are generally not powered to detect such rare events.
    We also feel that the expert panel review may have been biased towards characterizing only the more severe events, given the definition of "event of special interest." So the extent to which the conclusions about eruptions from the expert panel review might apply to the broader population of telaprevir-treated subjects who experienced cutaneous eruptions is unclear. And most of the ESI events had what we considered an eczematous component, so it's not clear that this translates to a general characterization of the rash as being primarily eczematous.
    So in summary, telaprevir-related rash and pruritus occur frequently. They can be severe and treatment-limiting. SCAR events occur infrequently but may be significant. They may have been under-diagnosed by investigators. It's important to recall that not all severe rash were SCAR events. And in the subjects who had SCAR events, there were no deaths. Rash may take weeks to resolve. The effect of antihistamines and steroids remain unclear. And as you also heard, the etiology of this rash remains undetermined.
    Anemia. Again, we've heard a lot about anemia. It's the most problematic side effect of ribavirin. We know the onset of rash in subjects treated with telaprevir is faster and that telaprevir increases the decline in hemoglobin levels by a gram to gram and a half above what's observed with peg and ribavirin.
    Management. Prospective management was ribavirin modifications and/or telaprevir discontinuation. Again, no dose adjustments of
    telaprevir for anemia were allowed, and as you've heard, use of erythropoietin-stimulating agents was generally prohibited. And the applicant also put together another one of these special search criteria to capture the various descriptive terms and events used by investigators to characterize anemia.
    Again, we had some challenges such as looking at differences in sex and baseline hemoglobin levels. But what we did find is that for any anemia special search criteria event, it was 36 percent of subjects treated with telaprevir compared to 17 percent treated with peg and ribavirin. There were more grade 3 events, more SAEs.
    More subjects had a shift to greater than or equal to grade 3 reductions in hemoglobin using the DAIDS grading scale. Absolute reductions to below 10 was about 37 versus 15 percent, and hemoglobin reductions to below or equal to 8.5 grams per deciliter were higher in telaprevir at 10 compared to 3 percent in the peg and ribavirin group.

    We looked for clinical events possibly associated with anemia. There were 3 events of myocardial infarction in telaprevir subjects. None of these subjects had anemia. Two had significant coronary artery disease at the time of their MI. One unfortunately died. One discontinued his study medications, recovered, and resolved. And one, who had an anterior MI, continued dosing and completed the study.
    Things like angina, dizziness, dyspnea, syncope, we didn't find any important differences between treatment groups based on the presence of anemia. There was more fatigue in subjects who received telaprevir, but it didn't appear that every one of those was associated with anemia.
    Management of anemia. Again, 4 percent of subjects discontinued telaprevir. Ribavirin dose reductions were more frequent in the telaprevir group, as were ribavirin dose interruptions. About 2 percent of subjects in the telaprevir group had to discontinue ribavirin; 1 percent discontinued all medications in the regimen. You've heard about
    the 6 percent who received the blood transfusion. And the ESA use was right about 1 percent per treatment group.
    Ribavirin modifications/interruptions really had little effect on the SVR rates. But if you had to stop ribavirin and telaprevir, there was a much greater negative effect on SVR rates.
    We looked at the outcomes of subjects who received blood transfusions, and actually 59 percent of the subjects who received a blood transfusion were able to go on and achieve an SVR. And of the 24 subjects who received an ESA in the telaprevir group, 14, or 58 percent, achieved an SVR.
    So in summary, telaprevir anemia occurs frequently. It can be severe and treatment-limiting. It's manageable through discontinuation of telaprevir and ribavirin dose reductions, interruptions, or discontinuations. I didn't show you this, but it was more frequent in females, a little bit older age people, people with lower BMI and cirrhosis, and that's basically consistent with what's been seen in prior peg/ribavirin studies. And anemia-associated clinical events were generally comparable between the treatment groups.
    Anorectal disorders you heard a little bit about, 29 versus 7 percent. The most commonly reported were hemorrhoids, anorectal discomfort, and anal pruritus. Less than 1 percent were serious, less than 1 percent were severe, and there were 7 events that led to discontinuation of telaprevir. They seem to be fairly well managed with local and topical agents, and we still don't have a mechanism for this toxicity.
    All the adverse drug reactions in the database were essentially consistent with the profile in peg and ribavirin. Events that occurred in at least 20 percent of subjects in all treatment groups included rash, fatigue, pruritus, nausea, headache, anemia, diarrhea, flu-like symptoms, insomnia, and pyrexia.
    This table shows the frequency of events that -- or the events that occurred at a frequency of at least 5 percent more amongst subjects treated
    with telaprevir. We've already talked about rash and fatigue and pruritus and anemia and the anorectal stuff. There was more GI events, such as nausea, diarrhea, and vomiting, and again, subjects didn't seem to like the taste of telaprevir, so there was more dysgeusia.
    We looked at the hematologic abnormalities, primarily grade 3 or greater, to see if there was any differences, and we found that total white count reductions were a little bit more in telaprevir. But when you look at the absolute neutrophil count reductions to grade 3 levels, it was more often in the peg and ribavirin group. Absolute lymphocyte reductions to grade 3 levels was much higher in telaprevir, at 18 versus 6 percent, and reductions in platelet count to above grade 3 levels was 3 percent versus 2 percent.
    We again looked for clinical events that might be associated with these abnormalities. And it is difficult to count clinical events based on lab values, as investigators report things
    differently and may have reported isolated asymptomatic lab abnormalities as a clinical event.
    We looked for neutropenia as a clinical event, and it was actually reported more often in peg and ribavirin subjects than in those who received telaprevir. There were two telaprevir subjects who had events of febrile neutropenia, but neither one had an infection. And there were no other cases of serious infections related to low white cell count. Colony-stimulating factor use was about 1 percent in both treatment groups.
    With regard to lymphopenia, we looked for serious opportunistic infections that might have been associated with low lymphocyte counts. There were none. Thrombocytopenia, the frequency was generally comparable.
    There was one life-threatening event in a telaprevir subject who presented with epistaxis and purpura and a platelet count of 10,000. The subject was treated with ethamsylate and prednisone, and then asked to be discharged from the hospital and was lost to follow-up. So we
    don't have any follow-up on that patient. But no subjects in any of the treatment groups received a platelet transfusion.
    Another lab abnormality we noted was uric acid elevations. They were noted in Phase 2, and in the Phase 3 trials they were substantially higher in telaprevir compared to peg/ribavirin subjects at 73 versus 29 percent. They seemed to follow a similar pattern as other labs associated with telaprevir. There was a steep increase during the first couple of weeks, levels peaked and then stabilized, and then within about 8 weeks after cessation of dosing with telaprevir, were back to levels comparable with peg and ribavirin.
    Thirteen subjects experienced clinical gout or gouty arthritis, 11 in the telaprevir group versus 2 in the peg/ribavirin group; 7 of the 11 events in telaprevir occurred during telaprevir dosing period. Only one of those subjects had a previous history of gout. Subjects generally were treated with colchicine, indomethacin, hydration, ibuprofen, and all subjects recovered. There were no SAEs and no discontinuations.

    Bilirubin elevations were much higher in telaprevir versus peg/ribavirin at 41 versus 30 percent. Elevations to above grade 3 levels, which was greater than about 2.6 times the upper limit of normal using the date scale, were 4 percent for telaprevir, 2 percent for peg and ribavirin. Again, the steepest increase very early, then there was stabilization, and by weeks 12 to 16, they were back to baseline levels.
    About 1 percent of subjects in both treatment groups had a "clinical event." Again, this could have been an isolated lab abnormality. And 14 of these telaprevir subjects only listed -- the only thing that was listed was either elevated bilirubin or hyperbilirubinemia. Thirteen of the 14 actually occurred after the telaprevir dosing period was completed.
    There were two subjects that had ocular icterus, and one of those subjects discontinued. And one of the 14 subjects had a history of Gilbert's disease and had jaundice.

    In the peg and ribavirin group, the three subjects, there were 2 with just increased bilirubin and one that had increased bilirubin with jaundice.
    So, to conclude, with efficacy, telaprevir and peg and ribavirin significantly increased the SVR rates compared to peg/ribavirin alone across a broad range of populations and demographic subgroups. In treatment-naive subjects, there was significantly increased eRVR and resultant SVR rates, which demonstrate that the duration of therapy can be truncated in early responders without jeopardizing efficacy.
    The T12 regimen was more efficacious than T8, but may be associated with an increased risk of anemia and rash. There was no apparent increase in efficacy for 48 weeks of peg and ribavirin in early responders. And peg/ribavirin for 48 weeks did appear to provide an additional benefit in subjects who had not achieved an eRVR.

    In treatment-experienced subjects, the T12 PR48 regimen was superior compared to retreatment with peg and ribavirin for all treatment groups. We found no important differences between the delayed or immediate start of telaprevir. And the T12 PR24 regimen may be acceptable for prior relapsers who achieve eRVR, as you heard Dr. Jadhav tell you.
    For the subgroups of blacks, Latinos, and cirrhosis, across all of them there were increased responses compared to peg and ribavirin. For blacks, there were lower SVR rates in naives compared to Caucasians, consistent with prior PR studies. Latinos had pretty much similar responses to Caucasians.

    And in cirrhotics, there was a suggestion of an added benefit for longer duration of peg and ribavirin in naives and a suggestion of minimal added benefit for null responders. Many of these results are encouraging, but, again, there were small numbers of subjects in some of these subgroups.

    Virology. On-treatment virologic failure is associated with the presence of resistant substitutions, more subtype 1A and prior nulls
    failed treatment. There was divergent resistant pathways, depending on the subtype. Some mutations, such as the R155K, may persist for upwards of 3 years, and so the impact on retreatment is currently unknown.
    The pharmacogenomic post hoc analysis of IL28B data suggests improved SVR rates for all genotypes treated with telaprevir, but may need prospective confirmation with a more representative population.
    Safety. Rash and pruritus can be severe, treatment-limiting, but manageable. Not all severe rash were SCAR events, and SCAR events did occur but there were no deaths.

    Anemia can also be severe, treatment-limiting, but also manageable with telaprevir discontinuation and/or ribavirin modifications. We didn't really see a signal for increases in clinical events related to anemia. Anorectal disorders occur frequently, primarily are bothersome and appear manageable.

    We saw a few more GI events of nausea,
    vomiting, diarrhea, and dysgeusia with telaprevir. There was a low frequency of other hematologic abnormalities. Elevated uric acid levels, possibly related to anemia-related red cell breakdown. And the elevated bilirubin levels may be partially explained by a similar mechanism, but we're still looking at them to make sure we don't miss something else. And that is it.
    Thank you.

    Clarifying Questions from Committee to FDADR. CARGILL: Thank you.

    Clarifying questions from the committee for the FDA? We'll start with Dr. Clay.
    DR. CLAY: I guess the first is just a comment. I have a hard time hearing presentations about dermatology, and your making an abbreviation of a multitude of symptoms and calling it SCAR, because in my mind, a scar is a certain dermatological condition. So that's neither here nor there.
    I have a question that relates to how well the study was blinded. These were blinded studies done?

    MR. FLEISCHER: 108 and C216 were blinded; 111 was open label.
    DR. CLAY: Okay. So my question to that is, you're conducting a blinded study in which you had a significant difference in rash from one group to the next. So I guess that's okay if nobody knew that there was the likelihood of a rash occurring.
    I was just curious, when the sponsor submitted their material, did they discuss in there how it was managed at the investigator level?
    MR. FLEISCHER: The rash?
    DR. CLAY: No. The determination as to whether the rash was possibly or probably related to study medication.
    MR. FLEISCHER: I think that was in the rash management program.
    DR. CLAY: Okay. And that was --
    MR. FLEISCHER: It was in the protocol describing how rash should be assessed and managed, that was given to every site, yes. But Bob can --
    DR. CLAY: No, no. I'm not asking about
    management. I'm asking about the determination in the reporting to the sponsor as to whether or not the investigator felt the rash in and of itself was related to study drug.
    DR. KAUFFMAN: Yes. During the blinded phase, of course, no one knew what the treatment assignments were. The investigators used their judgment to decide whether it was related or not to the treatment that was occurring.
    DR. CLAY: But rash was included in the investigational drug brochure provided to the investigators when they were considering taking on this study?
    DR. KAUFFMAN: Yes. And by Phase 3, there was general acknowledgment that rash was associated with telaprevir, and it was certainly in the investigators brochure.
    DR. CLAY: Okay. That's fine. We've been through efavirenz. We understand how to deal with that, and nevirapine and the others.
    My next question actually relates to a different way -- maybe it also relates to blinding
    as well. We talked about an increase in bilirubin within the first two weeks in a fair number of patients. When that sample is drawn and separated, as you need to separate out your samples in order to do viral loads, you're going to see a nice, pretty, yellow color.
    I'm wondering, did your placebo color your plasma yellow?
    DR. KAUFFMAN: I'm not aware whether it did or not.
    DR. CLAY: Because you're going to spin it down in a serum separator tube to send it off, and you're either going to have a yellow tube or not. If you did not blind -- or unblind and mandate unblinding at the site level for your lab processing personnel, I'm questioning the validity of your blindness.
    DR. KAUFFMAN: I mean, I think in most cases the bilirubin levels were not markedly elevated. Therefore, it's highly unlikely the plasma would actually be icteric.
    MR. FLEISCHER: Yes. Only about 4 percent
    were above grade 3, so the majority of those were grade 1 and grade 2. And so --

    DR. CLAY: Yes. Icteric is a clinical presentation, as I understand it. Now, my disclaimer, I'm just a pharmacist. But I have processed atazanavir drug before, or atazanavir-containing plasma before, and the patient didn't have to be clinically icteric in order for me to see the yellowing of the serum.
    MR. FLEISCHER: The other thing is, like I said, some of these were individual spikes. Not everybody fit this pattern. Some were asymptomatic. They presented. They got a lab. It was a grade 2 level, and it was called hyperbilirubinemia. And then it went down at the next lab. So we can't say that everybody went up, persisted and then went down. They were all over the place.

    DR. KAUFFMAN: Also, bilirubin levels were also elevated in the placebo subjects, as you saw. It's not very likely that someone, without knowing what the treatment assignment was, would really be
    able to assign a treatment assignment based on the color of the plasma because elevations occurred in the placebo subjects as well as in the telaprevir subjects.

    DR. CLAY: Fair enough, 40 percent versus 20 percent.
    DR. KAUFFMAN: Yes. But those are both high, so in any individual case, it would be hard I think for someone to ascertain that. And I think, also, just to go back to your question about rash, as you saw, the rates of rash were quite high in the control arm as well as in the telaprevir arm. True, they were higher. But overall across the study, rash was occurring quite frequently.

    Therefore, again, because of that, we think it's less likely that a patient could be unblinded as to their treatment assignment just by the occurrence of a mild or moderate rash because 30 percent of the control patients had mild or moderate rash. So we felt that, as best we could, that the blind would be maintained.

    DR. CARGILL: Dr. Lewis, you wanted to join in?
    DR. LEWIS: I was just going to point out that same fact, that we really did not think the blinding was affected because the rates were high in both arms. In an individual patient, you wouldn't be able to predict which event was related to which drug.
    DR. CLAY: And my final question relates to the statistical presentation that was up there. I'm not a statistician, but statistically, everyone who eats pickles dies. So I'm always questioning statistics.
    You made a comment in your presentation, and it was that you assumed that the response for triple would be identical to the response for double. Does that take into account -- and I apologize if you said it and I missed it -- that you may have a change in dropout rate due to that third drug administration in a triple versus a double?
    DR. JADHAV: I mean, it completely doesn't
    take that into account. But the point is --
    DR. CLAY: So no?

    DR. JADHAV: Yes. No.
    DR. CLAY: Okay. Does that invalidate your model?
    DR. JADHAV: No. It's not a model, actually. It's not even any statistics. It's actually --
    DR. CLAY: You had to calculate it for about 15 minutes, so I'm really hoping it was important.
    DR. JADHAV: Yes.
    [Laughter.]
    DR. JADHAV: So while it does not per se fully account for everybody in that group, even if, say -- in fact it validates more than even the 44 percent of those responders that I talked about, even if 100 percent of them don't respond, that 44 percent goes down. So actually, that difference you see increases.
    So 100 percent is actually -- if you think very carefully, it's a simple multiplication is the worst case. If patients drop out and don't
    complete, the difference you see in the telaprevir arm of going from 54 to 44, which was 10 percent here, assuming 100 percent, that difference will increase.
    So while it does not account to me, it's actually a worst-case estimate of who are those additional 10 percent patients. This makes sense because we have the SVR data from both arms. So even if I stake out, I don't have in front of me what's the percentage of dropouts due to, say, adverse event in telaprevir, the argument becomes even more convincing because the difference will become more than 10 percent. So those additional patients are the future relapsers that we are treating.
    Does it make sense? There's no statistics here. Let me be very clear. It's a simple -- there are three steps: multiplication, addition, and division, and that's it.
    [Laughter.]
    DR. CLAY: And in the right combination, sir, that is statistics.
    DR. CARGILL: I think we got this. We'll move on.

    Ms. Dee?
    MS. DEE: Okay. I'm not sure if this is for the sponsor or for the agency. I know I heard somebody mention prior rash in patients before, but I didn't really hear an answer about whether people who had had a rash situation in the past might be more susceptible to getting a rash with this drug.
    DR. CARGILL: I think the sponsor can answer that.
    DR. KAUFFMAN: I'll ask Dr. Singhal to come up and talk about that.
    DR. SINGHAL: We did not collect information on baseline experience with rash. For example, in the treatment failure trial, C216, we did not exclude any subjects who had previously had rash.
    However, when we collected the data on the rash events that patients experienced during the study, we did collect information as to whether they had had an event previously. And the rates that we observed in C216 were really no different
    from rates in the 108 study. So we don't believe that there's an impact on prior rash. But to be really sure, we are collecting prospective information in ongoing studies, so we are looking at that in further depth.
    I also just wanted to address a comment that was made earlier with regards to blinding. And I just wanted to say two things. One is that in our Phase 3 program, we collected rash data as events of special interest, as Dr. Fleischer pointed out, and these were collected in a blinded manner.
    So if a physician felt that he needed to discontinue telaprevir because he had assessed it at greater than 50 percent body surface area, we simply collected that data and accepted it for what it was. We did not unblind any patients. And the DEP's review was a complete blinded review of 221 cases. There was no unblinding. The unblinding was done only at the end of the study.
    MS. DEE: And I have a quick EPO question.
    DR. CARGILL: I need to give a clarification instruction to the committee.

    This is the FDA Aquestion period, so please make sure that your questions are directed to the FDA. We will have a chance to go back to the sponsor.
    MS. DEE: All right. This is for the FDA.

    On the anemia stuff, the sponsor provides -- I don't want to drill down too far -- data that talks about days to onset and the weeks it was stable, became stable, the weeks that it started to reverse back to baseline. Were you able to confirm that?

    MR. FLEISCHER: Yes.
    MS. DEE: Good. And this is for the agency as well, I guess more of a statement rather than a question. But all of these questions that we've been having about the number of blacks in trials, it's giving me deja vu. And I'm just hoping that we're not going to go through this again with these new drugs.
    We need black people in these studies. There are plenty of black people with HCV, and unlike HIV, there's an absolute difference in what happens in people of color, Latinos as well. I
    just really think that it's time for the agency to consider, like we saw yesterday, a cohort 1 and a cohort 2.
    It's like the dentist. It's painful at first, but it's always worse if you wait. I mean, half of the questions that we have are like what do we do about these situations for black people that we don't have enough numbers to give us direction? So just hopefully it'll be different this time.
    DR. BIRNKRANT: We agree with your comments. We, too, have discussed the situation with sponsors at very early phases of drug development, and then we're presented with a marketing application.
    So on one hand, we could hold up a marketing application while we wait for more data in certain subgroups; or we can act on the marketing application and perhaps ask for additional data postmarketing, given that these are serious and life-threatening diseases.
    But maybe if we say it publicly more and more, that the message will get across because we too share your frustration. And we also want to do
    what's best for the public as well. So we hope all of the companies who are here and are listening to the webinar, pay attention to this situation.
    DR. CARGILL: Thank you. And with that, we will turn to clarifying questions for the FDA.
    Dr. Roland?
    DR. ROLAND: Yes. I have a question about the creative approach to the prior relapse patients and response-guided therapy. I just wanted to clarify because it gets very confusing following these discussions, and my best understanding is that the data that you presented on actual subjects is the Phase 2 data.
    It makes me really wonder why the Phase 3 studies, if there were compelling Phase 2 data – granted about 50 people -- why the Phase 3 studies didn't address that question specifically.
    So I just want to confirm that my understanding is clear. And if you have any insight about why the Phase 3 study didn't address it specifically, I'd appreciate that as well.
    MR. FLEISCHER: The data was from Phase 2.
    But I think, as the sponsor said, they didn't have it all by the time they initiated their Phase 3 trials. So there wasn't sort of this expectation that it would work, and they didn't propose to study it. So this is sort of a -- yes. That's it.
    DR. CARGILL: Thank you.

    Dr. Ellenberg?
    DR. ELLENBERG: Yes. Dr. Fleischer, you confirmed what the sponsor said about the lack of impact of the dose reductions of ribavirin on the sustained response. You did also comment that the people who had to have it stopped did have a somewhat lower response rate.
    I'm interested in whether you looked at this for the people who stopped treatment for the rash. There was a substantial proportion who had to terminate telaprevir treatment because of rash. I could not -- there was a plot that showed the onset, but that was for all rashes. I don't know whether the more serious ones occurred somewhat later. I don't know how many weeks of therapy they didn't get.
    So that's my first question, is can you comment on the impact on response rate of the termination of therapy for rash?

    MR. FLEISCHER: I can't. I don't know if the sponsor can.
    DR. KAUFFMAN: One part of the question is, yes, it's true. Patients with severe rash generally tended to occur later during the 12-week period of telaprevir, so the impact on potential efficacy would have been less.
    For example, when you look at the 8-week telaprevir arm, you can see that although there is a decrement in SVR compared to the 12-week arm; still they have a 72 percent SVR rate. So even if patients receive 8 weeks of telaprevir, they still gain substantial benefit from it.
    DR. ELLENBERG: Okay. Of course, I don't know whether there's something special about the patients who had this severe rash. Maybe after lunch we can get that information about what actually happened to the ones who terminated in terms of their final outcome.

    DR. KAUFFMAN: Their outcome?
    DR. ELLENBERG: Yes.
    DR. KAUFFMAN: We can try to get that.
    DR. ELLENBERG: The other thing that I wanted to -- I'm not sure whether it's a question or a comment. But it has to do with this attempt to understand about what an RGT would be in the relapsed patients. And I'm just concerned that the presentation that we got seemed to make a very strong assumption that numbers like 54 percent and 44 percent were in fact truth.

    I haven't seen very many confidence intervals in these presentations. But I don't -- you know, 54 percent here could be 49 percent at another time, and the 44 percent could be 48 percent another time. And I can't say I really totally followed this presentation, but at this point I don't really find it very convincing.
    I would think that some good use of statistics might be helpful in terms of seeing what the effect of the variability on these estimates would do to the conclusions. You've lots of good
    A Matter of Record
    (301) 890-4188
    191
    statisticians over there. I know that.

    DR. CARGILL: Dr. Ghany?
    DR. GHANY: Yes, thanks.
    I have two questions. The first relates to anemia, and it's for the agency.
    Did the agency look at classifying the severity of rash using the investigators' adjudication of severity? And how does that change the percentage of people who had grade 3 severe rash, instead of using the DEP's adjudication?
    MR. FLEISCHER: We were using the investigators' assessment for our assessment of -- in the table, for events, those were investigator assessments.
    DR. GHANY: So it wasn't the DEPs, then?
    MR. FLEISCHER: No. No.
    DR. GHANY: Okay. And the second question, I don't know whether this is more for the agency than the sponsor. So we were told that despite ribavirin dose reduction in almost a quarter of patients, it had little effect on SVR. So was any analysis done or consideration taken to starting
    the ribavirin dose at a lower dose during the triple therapy period?

    MR. FLEISCHER: Prior to the study initiating, about starting with a lower dose?
    DR. GHANY: Starting with a lower dose.
    MR. FLEISCHER: No. I don't recall having any of those discussions.
    DR. CARGILL: If the sponsor could address that?
    DR. KAUFFMAN: Just make a quick comment that, obviously, ribavirin is an approved drug for hepatitis C. In general, the advice we gave for physicians was to follow the label instructions for both peginterferon and ribavirin. Therefore, it was started at the usual dose based on weight and was adjusted according to the ribavirin package insert. So that's all we can say.
    DR. CARGILL: All right. Thank you.
    Dr. McGovern?
    DR. MCGOVERN: I'm going to hold my question for the sponsor.
    DR. CARGILL: Thank you.

    Dr. Van Dyke?
    DR. VAN DYKE: Yes. Regarding the differential response by race and gender, I wonder how much of that can be accounted for by the IL28 genotype. And I also wondered, is there a relationship between the IL28 genotype and cirrhosis? Could that be a confounding variable there as well?
    DR. PACANOWSKI: Mike Pacanowski, genomics group.
    Concerning the IL28B relationship with race, yes, it is. The T allele is more common in black subjects. In these trials, there were very few black subjects enrolled overall, let alone included in the generic sub-studies. So it's close to impossible to make any assessment concerning that.
    We expect the IL28B genotype to be independent of sex or gender. So, again, there wouldn't be any explanation of the response there, either.
    DR. VAN DYKE: Cirrhosis?
    DR. PACANOWSKI: Cirrhosis? So for the treatment-naive trial, we did not have complete clinical data because it was a de-identified data set because it was voluntarily collected generic information. So we were not able to compare things like cirrhosis across the different genotype groups.
    I do believe in the treatment-experienced trial, C216, that cirrhosis rates were similar across the genotype groups, but I'd like to check that, or perhaps the sponsor can clarify.
    DR. CARGILL: Thank you.

    Dr. Knodell?
    DR. KNODELL: I have two questions. The first one relates to this ribavirin. I think this is really an important clinical point.
    When we make dose reductions with ribavirin, we try and do it later in the treatment. If you can carry these people out 16 or 17 weeks and then cut your dose, they seem to do better. There's no question that ribavirin reductions decrease SVR.
    Now, you made a statement, Dr. Fleischer, that there was no decrease in SVR with ribavirin
    decreases or reductions, but we didn't see any data to support that. And when I look at their anemia timeline, the timeline of anemia is fairly quick. It's during the first 4 to 12 weeks of treatment than both groups. And the majority of the anemia is due to ribavirin. There is an incremental increase with the telaprevir, but certainly ribavirin would probably be the drug that we'd want to start reducing rather than stopping the telaprevir, if I'm a clinician out in the real world taking care of these patients.
    So I wondered if you could expand, what is your basis for saying that ribavirin reductions did not decrease SVRs? And what was the timing of those ribavirin reductions when they were made?
    MR. FLEISCHER: I'd have to go back and look at the timing. But we found -- let's see. It says, "Among subjects who underwent a dose modification or interruption of ribavirin, a temporary interruption, the SVR rate went down to about 73 percent in naives and about 77 percent in experienced subjects overall."
    So there was a little bit of a -- I forget what the --
    DR. KNODELL: We didn't see that data.
    MR. FLEISCHER: I know. I have it in case you want it.
    DR. KNODELL: All right.
    MR. FLEISCHER: But just doing things like reducing the dose of ribavirin, interrupting it or discontinuing, can decrease the SVR rate from about 75 percent to about 40 percent in naives, and from about 65 percent to about 20 percent in treatment-experienced subjects.
    So taking the ribavirin away had a much bigger impact than trying to keep some ribavirin on board. And we saw from the Phase 2 trials, where there were ribavirin-sparing arms, that they didn't really actually respond better than control, peg/ribavirin arms. So I think there was an attempt to try to keep patients on ribavirin as best as possible. But it is what it is.

    DR. KNODELL: My second question relates to cirrhosis. The sponsor, in data that they provided, said that they only -- the accession was for cirrhotics A and B because there was decreased telaprevir exposure, 15 percent in cirrhotic A, class A, Child-Pugh class A, and 53 percent in Child-Pugh class B, and not to use this drug in Child-Pugh class C.
    Now, we've not had any breakdown in your data on cirrhosis, which class these were. So my question is, generally, if you've got a drug that's metabolized by the liver, when you've got liver disease, you have increased exposure. Unless this is a pro drug, telaprevir, which I don't believe it is, I'm not quite sure why there should be decreased exposure in cirrhotics.
    Then my question is, do any of your cirrhosis data that we have seen today break the cirrhotics down as to whether they're A, B, or C, or can we just assume that there weren't any class C cirrhotics in the trial?
    MR. FLEISCHER: They were all supposed to be -- they were all compensated, and they all looked like they had normal synthetic function. So
    we didn't parse them out by A, B, or C. They were all supposed to be pretty well compensated.

    DR. ROBERTSON: Just to address the decrease in exposure, we have seen this with other hepatically metabolized drugs, but not typically. You're right. Some hypotheses might be related to protein binding, less protein binding in these more severely impaired patients, though in this study there was some problem with assessment of protein binding so we can't use that -- but that's a hypothesis -- as well as decreased absorption might be another contributor to the lower exposure.

    DR. KNODELL: So that's just AUC? When you say "exposure," you're looking at AUC?
    DR. ROBERTSON: AUC. Exactly, yes.
    Oh, sorry. Sarah Robertson, FDA.
    DR. CARGILL: Thank you.
    Dr. Strader?
    DR. STRADER: I want a clarification question answered on the rash. I keep hearing that there was a whole bunch of people who discontinued as a result of rash, but I see in your slide
    number -- hang on -- 35, it says less than 1 percent of patients in the telaprevir PR group discontinued treatment because of rash. Is that correct?
    MR. FLEISCHER: That was less than 1 percent who discontinued the entire regimen. But 7 percent stopped telaprevir and may have continued on peg and ribavirin.
    DR. STRADER: And ribavirin. Okay.
    My last question, you talked about dermatologic issues that needed to be identified, and you said distinction between severe rash and SCAR, but you don't tell us what that distinction should be, how do you define this SCAR. Because if you don't tell us, how are we going to know?
    Have you decided yet, or it's something --
    MR. FLEISCHER: Well, SCAR events, maybe Dr. Bigby can say something. But SCAR is sort of a specific definition that's associated with morbidity and mortality. Things like the DRESS is a SCAR. SJS is a SCAR. TEN is a SCAR.
    DR. STRADER: But there have to be some
    distinctions, something that you find, bullae, something that tells me this is SCAR as opposed to just a maculopapular rash.
    DR. CARGILL: I think perhaps we can ask Dr. Bigby to clarify that for us.
    DR. BIGBY: I think that the rashes that would be considered severe and sometimes life-threatening would be toxic epidermal necrolysis; Stevens-Johnson syndrome, which most people think those two are related and only depend on extent.
    The big thing about those two are that they lead to epidermal detachment, so that if there's more than 30 percent epidermal detachment, you would call it TEN. If the epidermal detachment is less than 10 percent, it's SJS. And if it's somewhere in between, there's an overlap syndrome.
    Then patients that have what's referred to here as DRESS have actually an exanthematous eruption that usually covers more than 50 percent of the body, associated with eosinophilia and systemic symptoms, the most common one being fever. And they can also have internal involvement of
    other organs like liver and kidneys. It, too, has a mortality associated with it.
    So those are the ones that I think he is referring to when he says SCAR.
    DR. STRADER: So anything else is a severe rash. TEN, SJS, and DRESS, and the overlap thing would be considered SCAR, but everything else would be considered regular rash or severe rash?
    DR. BIGBY: That's usually the case.
    DR. STRADER: Okay. Thank you.
    DR. CARGILL: Thank you, Dr. Bigby.

    Dr. Korman?
    DR. KORMAN: I have a question for the agency. In slide 54, you conclude that the rash is severe, treatment-limiting but manageable. And when I get to look at slide 39 -- and remember, a lot of us are gastroenterologists, hepatologists as well; we're good at recognizing polyps, but we may not be that good at recognizing rashes -- the rashes are severe and treatment-limiting. They're under-diagnosed by the investigators. They take weeks to resolve, which I assume produces some
    morbidity. We don't know how to treat it; antihistamines and steroid therapy is unclear. And we have no idea what causes it.
    So I don't know how that manageable fits in to that slide because we don't have MGH dermatologists available on-call when we confront these things. So I'd like to get some comment about that manageability issue.
    MR. FLEISCHER: Well, we're working it out. But if nothing else, you can stop the telaprevir. So that's pretty manageable. If you can get them out to 8 to 12 weeks, you've got a pretty good chance of either completing therapy or still having a pretty good SVR rate. If it happens really early and you have to stop, it doesn't say you have to stop peg and ribavirin.
    So from a very sort of simplistic approach, stop the telaprevir. For more convoluted approaches, we're trying to figure those out with our derm consultants, and how would we convey that kind of a message in a label for a clinician to say, okay, if this rash hits this much, these are
    the options available to me to manage it. And we have some ideas.
    DR. KORMAN: It's clear when the hemoglobin reaches a certain level or where the neutrophil count reaches a certain level; it's not clear about rashes. And "tough it out" isn't something that patients want to hear.
    DR. CARGILL: Thank you.

    Dr. Lewis?
    DR. LEWIS: I would just point out that even though the dermatology expert panel thought that there may have been some under-diagnosis of these cases of DRESS or SJS, they were in relatively -- they were very rare. There were single digits out of a program of 3800.
    While that's not insignificant, if it was under-diagnosed, the management plan that was provided in the Phase 3 protocols seemed to allow patients to recover without serious sequelae, and in many cases complete their treatment. So the management plan that was used in the Phase 3 program appeared to be effective.
    DR. CARGILL: Thank you.

    Dr. Bigby?
    DR. BIGBY: My question for the agency is, were you given the details of the three suspected cases of SJS and the 11 suspected cases of DRESS, and did you come to any conclusion about whether those diagnoses were correct? And of the ones that were correct, were they drug-related?
    DR. CARR: I'm Brenda Carr, dermatology. And we were in agreement with the dermatology expert panel in their assessment of the suspected cases of SCAR. The cases were generally considered treatment-related, is my recollection, except for one case of Stevens-Johnson syndrome, which occurred, I believe, 11 weeks after treatment was discontinued, so was therefore not thought to be telaprevir-related.
    DR. BIGBY: So that means that -- so two of the suspected cases of SGS you concluded were cases of SGS, and that they were related to telaprevir?
    DR. CARR: I believe one case was definitive. And again, that was the case that
    occurred some weeks post-treatment. I don't recall whether the other two were possible or probable.
    I'm sorry. Could you repeat the question again?
    DR. BIGBY: So there are listed three suspected cases of SJS.
    DR. CARR: Yes.

    DR. BIGBY: So the first part of the question is, did those patients have SJS or not?
    DR. CARR: Yes. They were suspected to. The adjudication was based on a review of photographs and not on actual assessment of the patients or subjects. So based on a review of their photographs, conclusions were reached that the subjects were suspected to have Stevens-Johnson syndrome.
    DR. BIGBY: And how about for the 11 cases of DRESS? Your conclusion was that they actually did have DRESS?
    DR. CARR: They were suspected to have had DRESS, not definitive. Not all of them were definitively determined to have DRESS. And I could
    follow up with you on specifics of the breakdown.
    DR. CARGILL: If you could do that, that would be appreciated.
    We will now break for lunch. We'll reconvene again in this room in one hour from now, which is at 1:00 p.m. Please take any personal belongings you want with you at this time. The ballroom will be secured by the FDA staff during the lunch break.
    Panel members, please remember that there should be no discussion of the meeting during lunch amongst yourselves or with any member of the audience. Thank you.
    (Whereupon, at 12:03 p.m., a luncheon recess was taken.)
    A F T E R N O O N S E S S I O N
    [1:00 p.m.]

    Open Public Hearing
    DR. CARGILL: Good afternoon.
    Both the Food and Drug Administration and the public believe in a transparent process for information-gathering and decision-making. To ensure such transparency at the open public hearing of the advisory committee, FDA believes that it is important to understand the context of an individual's presentation.
    For this reason, FDA encourages you, the open public hearing speaker, at the beginning of your written or oral statement to advise the committee of any financial relationship that you may have with the sponsor, its product, and, if known, its direct competitors. For example, this financial information may include the sponsor's payment of your travel, lodging, or other expenses in connection with your attendance at this meeting.
    Likewise, FDA encourages you at the beginning of your statement to advise the committee
    if you do not have any such financial relationships. If you choose not to address this issue of financial relationships at the beginning of your statement, it will not preclude you from speaking.
    The FDA and this committee place great importance in the open public hearing process. The insights and comments that you provide can help the agency and this committee in their consideration of the issues before them.
    That said, in many instances and for many topics there will be a variety of opinions. One of our goals today is for this open public hearing to be conducted in a fair and open way where every participant is listened to carefully and treated with dignity, courtesy, and respect. Therefore, please speak only when recognized by the chair. Thank you for your cooperation.
    One other note, please. Because we have a number of speakers, your time will be 4 minutes. Your microphone will be cut if you exceed that time. Thank you.

    Martha Saly?
    MS. SALY: Good morning -- good afternoon, I guess. Thank you for inviting me to speak once again today. As you might remember, I spoke yesterday, and I'm the director of the National Viral Hepatitis Roundtable, which is a coalition of about 175 public and private organizations that are dedicated to the elimination of viral hepatitis in the United States. Vertex Pharmaceuticals is a member of NVHR, and they have been a funder of NVHR, but they did not pay me to attend these proceedings today.
    Dr. Cargill, yesterday I was really touched by your statement about bringing new hope to your patients. And I really wanted to speak again today from the point of view of being a patient and talk about the many patients who have come to me looking for new hope in treatments because treatments have previously failed these patients.
    So I wanted to just talk about how -- I understand that these treatments are still going to be difficult for patients. They are still going to
    be based on interferon and ribavirin, and that is a concern to me. But I still think that the new treatments will create a real watershed event in this country, and it will be monumental what we can do with these treatments available to the patients who have been failed before by our efforts.
    Dr. Friedman, you also noted yesterday that someone might need the wisdom of a Talmudic scholar to be able to prescribe these treatments, and I realize that that's also a concern of this committee. I would ask you if that might have also been the same in the past with HIV treatments, and yet many dedicated providers stepped up and started treating their patients with HIV. And in fact, many of those providers went on to be our champion providers for patients with hepatitis C.
    I think that that is going to be quite possible for the many dedicated providers who want to do something for their patients who have been suffering with hepatitis C for far too long.
    I think that if the influx of patients is anywhere near as great as we expect, that as the
    community works to amplify awareness and screening efforts in conjunction with the new therapies, it will be very imperative for the labeling of these products and the medical provider education materials to be carefully constructed in order to make prescribing of these drugs an option for a wide group of providers and their patients, and not just a handful of specialists that we've been hearing from here.
    So I want to thank you again for allowing me to speak, and I am going to step back and let the other speakers take their turn. Thank you.
    DR. CARGILL: Thank you.


    Tracy Swan, Treatment Action Group.
    MS. SWAN: Hello, everybody. Thank you for the opportunity to speak. I do not have a financial relationship with the sponsors or the competitors, nor did anyone pay for my travel other than the organization that I work for, which is a nonprofit.
    I'm very excited about these drugs, the one we discussed yesterday and today. I think they
    open up a lot of new opportunities. I do want to say, okay, a rash is a scary new territory for people who are used to thinking about a liver. But people have had years of experience with interferon, which is also pretty nasty, and I feel like it's really a question of partially the labeling, partially pictures and other things that make it easier for both patients and providers to assess the risks and benefits of how to use a drug and that education is going to continue to be extremely important.
    So these are the key issues I'd like to talk about. Yesterday you saw the same grim scenario that patients face, and I'd like to say there's a lot of information on drug-drug interactions with telaprevir. There's a lot of information in people with -- or some information in people with hepatic and renal impairment.
    I'm disappointed that studies for more desperately ill patients with really urgent need couldn't have begun a little earlier, given that Phase 3 started in 2008 for this drug. I'm also,
    as has been mentioned, disappointed by the low enrollment of both populations among whom hepatitis C is highly prevalent and populations that really need more effective therapies.
    There are some clinical issues. The long-term clinical implications of protease inhibitor drug resistance in hepatitis C are largely unknown. And instead of looking at what might be in someone's blood years after they stopped treatment, I think it would be quite helpful to apply the pressure of the drug again in the context of peg/ribavirin in people who are in early monotherapy studies and only got telaprevir who have a resistance to see what happens when they get retreated with standard of care. That's a really important way to get there.
    There may be a possibility to use this drug twice a day. I know there are studies that are looking at that. It would be great to resolve that quickly because it's a lot easier to take drugs twice a day than three times a day.
    Since the SVR rate among prior null
    responders with cirrhosis is low -- these are people who are pretty desperate -- it would be great to have early predictors of response or nonresponse, or any additional information to help people guide their decision-making process about treatment.
    Stopping rules, which hasn't come up yet, I think should be really explained clearly to patients and providers, especially in the absence of a treatment guidelines panel. And I hope that providers beware of drug-drug interactions, and we've heard a lot about the importance of rash management.
    Quickly, I know that there was some talk about post-approval studies. In particular, cross-country multi-DAA trials in hard-to-treat populations would be great. I think there is pretty urgent unmet need, and we've already got proof of concept that this can work.
    Thank you very much.
    DR. CARGILL: Thank you.


    Lorren Sandt, executive director, Caring
    Ambassadors Program.
    MS. SANDT: Good afternoon. Thank you for having me here today. Lorren Sandt with the Caring Ambassadors Program, Oregon City, Oregon, and I'm the chair of the National Viral Hepatitis Roundtable.
    Thank you for your presentation today. I do want to say to the sponsors that that was a very clear, concise presentation, and really, really appreciated it. It showed the care that you put into that. Thank you.
    I do have the same concerns that I had yesterday about adherence. A three-times-a-day drug at q8 hours with food with fat, while you're having diarrhea, nausea, and vomiting, and a little bit of brain fog to go along in there, I'm concerned about adherence. And so again I'm going to ask that the FDA use as strong language as possible about making sure that there are tools in place for people to actually have good adherence.
    I also am concerned about the lack of providers who are going to want to pick up this
    treatment. So, again, that's why I thank the sponsors for being so clear about the protocol because I think the clearer the protocol, the more doctors who are going to be willing to treat hepatitis patients. And we do have a lack of hepatitis doctors treating patients.
    I don't know if the panel is aware, but many patients have to wait six months to see a hepatologist, and often they don't even get to see one unless they're in stage 3 or stage 4 fibrosis. So we're talking about the most difficult-to-treat patients, and we're holding back patients from treatment that really could be treated more successfully and with shorter duration if we had more doctors treating. So, again, the more clear and the more concise you can make the directions to physicians, I think we'll have more physicians treating.
    I do understand that some drug-drug interaction studies are going to be taking place with milk thistle, so I'm not going to make that comment today. But I think all alternative
    medicines that are common, such as St. John's wort and milk thistle, need to be studied, and that also needs to be in the label so that the 40 to 70 percent of patients who are taking those herbal therapies are understanding what kind of implications it can have if they continue to take those while they're on these treatments.
    Other than that, I just would like to say that I do endorse the approval of both of these products. Yesterday you voted for one, but really looking forward to seeing more options for patients, and seeing those hundreds of thousands of people that are treatment-experienced cured, and also getting those people, those millions of people, that are undiagnosed cured.
    Thank you very much.
    DR. CARGILL: Thank you.

    Michael Ninburg?
    MR. NINBURG: Good afternoon. Thank you for having me back here this afternoon. My name is Michael Ninburg. I am the executive director of the Hepatitis Education Project in Seattle,

    Washington. Our organization has received support from the sponsor, and I do have experience with a product of the sponsor. I was a patient in one of the sponsor's Phase 3 trials.
    I'll make my comments brief today. I was very encouraged by the unanimous vote yesterday to recommend approval of boceprevir, and I fully expect the same today. Choice is good for patients, and as was mentioned earlier, this is indeed a watershed moment in the history of treating hepatitis C.
    A couple of things that I would like to say. Again, the treatment is not going to be easy. It's going to become a much more complex regimen. We heard today about mitigating side effects. Dr. Korman mentioned the fact that those who are treating hepatitis patients are not dermatologists.
    I was very encouraged to see from the sponsor the plans for provider and patient education. That's going to be absolutely crucial. Adherence to therapy and mitigation of side effects is going to be the only way that patients are going
    to have the best chance that they can of getting cured.
    I will end on the importance of getting these drugs and new drugs that are in development to those who most need them. And this will be more for the folks at FDA, but if we could get a bit of a push for an expanded or early access program for those who are most at risk of decompensating and getting to some of the more serious sequelae of this disease, that would be something that will save lives.
    Thank you.
    DR. CARGILL: Thank you.

    Jules Levin?
    MR. LEVIN: Let me try and be brief. I have six things I want to mention.
    First, it's been mentioned a couple of times already, the rash. I'm concerned about the rash, and so the education for clinicians and patients are really crucial here with regards to the rash. And I just want to give two examples of my concerns about the rash.

    One, what happens if a patient, who knows nothing about rash -- because I think the rash is new to this field of hepatitis; we're all trying to understand it better. So what if a patient gets a rash on a weekend? What are they going to do? If they get nervous, they may stop telaprevir without knowing that they probably could have waited till Monday to go see their doctor.

    Let's just say that their clinician that they go see on Monday is not adequately knowledgeable about the rash. It's one thing if you're going to see Ira Jacobson at Cornell and he sends you up to 3 to see dermatology. What if you're in some small clinic in some far reaches of Los Angeles or New York or somewhere, and they don't have a dermatology clinic to send you to? So how about educating all these potential clinicians or prescription writers about this?

    Now, I understand that -- I'm sure Vertex is thinking about this and understands it. But I think thinking about it and being successful at it are two different things. So rolling out an
    adequate education program for patients and clinicians is really crucial.
    I want to mention -- early access has been mentioned. I think this needs to be expanded on. We're talking about people who have advanced disease, who have not decompensated yet, but people who have advanced compensated cirrhosis.
    We really need a study, not access, not expanded access in the mold of HIV, but a well-controlled, monitored study where adequate PK has been done previously. And these patients need two or more oral therapies with or without peg/ribavirin because, obviously, most of them have already been through peg/ribavirin, so they're going to need at least two orals.
    That's what we're talking about, and the FDA had a hearing about this last year. And I haven't heard enough from the companies in terms of addressing this and planning studies. So if the companies are planning studies to address this, and if FDA is supportive of this -- because I'm not convinced that they are -- I'd like to hear about
    this, that this is in the plans, we're thinking about it, and we're planning this.
    I want to mention also the issue of -- so I think that I support the indication for null responders here today, the 33 percent. I think the prospective data supports that. And one reason I think I support the indication is because without an indication, we may not get payers behind it for all the patients that do need it.
    But we do need adequate education because some patients can wait and some patients may not be able to wait. So I think we need to explain this, so not every null responder needs to go on therapy; they can make a thoughtful decision about whether to use the drug or to wait.
    I want to talk about resistance for a minute. So I think it was clear today in the presentations that with clonal sequencing, resistance seems to go down. But even in the FDA presentation, the 155K after 36 months was still there in 3 percent of patients. And so the conclusion from both Vertex and the FDA was, we
    don't know what this means clinically. But what I'm worried about is that outside of this room, at conferences and other places, you hear thought leaders say, we can retreat patients successfully with protease inhibitors.
    I'd to put a stop to that. I'd like a more thoughtful discussion about that to go on, that we really don't know the answer. One of the answers would be to try and design, perhaps, a retreatment study that would be ethical for patients.
    So I guess lastly I just want to say patient support services. You know, in HIV we have the Ryan White Care Act, and we have nothing like that in hepatitis C, and I don't think that there's anything going to happen in the near future.
    So I'm very worried about it because, as has been mentioned by some of the speakers here, and we all know, these treatments are going to be very complicated, and there are no real support services for patients. Maybe in some of the major clinics, like at Cornell or Mt. Sinai in New York, they have some supportive services. But outside of that,
    there's very little patient support services.
    I don't know what the FDA can do about this, but I just wanted to mention that today because there are companies here today, and the companies are going to have to pick up the slack here to try and provide this because there's very little government support at this point.
    Thank you.
    DR. CARGILL: Thank you.

    Colin Schwartz?
    MR. SCHWARTZ: Good afternoon. My name is Colin Schwartz. I am with the National Alliance of State and Territorial AIDS Directors, or NASTAD. NASTAD represents the health departments, HIV directors, and the viral hepatitis coordinators. NASTAD does receive funding from Vertex but did not support my travel. Yesterday you heard from my colleague, and I have similar remarks today.
    While not necessarily in the purview of the Antiviral Drugs Advisory Committee or the FDA, I would like to use my time today to urge my colleagues in the room to consider some overarching
    comments, overarching issues with this new, important drug.
    Specifically, I would like to remind all of us that given the lack of public health infrastructure for viral hepatitis and the many barriers to care for persons living with chronic hepatitis C, a cure for hepatitis C is not enough unless we do two things.
    First, a cure must be affordable and its coverage comprehensive. The cost for this new treatment will be added onto the current standard of care. Further, we need a robust patient assistance program that provides affordable coverage for ancillary costs such as PCR, liver biopsies and liver tests, management of treatment side effects, and counseling and risk reduction services, especially for those who are uninsured or underinsured and are hard to reach by the healthcare system or do not readily access the healthcare system, even under health reform. We recommend that such a patient assistance program target these populations.
    Second, a cure must rely on infrastructure, with the government and the industry's support, to increase funding for screening and testing, increase staff capacity in medical settings, and increase the number of educated health providers in order to identify those who need treatment in the first place.
    In order for this new treatment to work, new diagnoses are required. Unfortunately, 75 percent of Americans living with chronic hepatitis C do not know it, or roughly 3 million. The Institute of Medicine recently reported a lack of knowledge and capacity among providers to identify infection and deliver expert care and a lack of knowledge among the public, most importantly, among populations at risk.
    Even with this new treatment, low provider awareness will continue to lead to lower than anticipated utilization of this drug and misdiagnoses. And we all know the numbers of specialists, such as infectious disease doctors, hepatologists, and gastroenterologists, are
    limited. Further, low public awareness will continue to lead to misinformation and missed opportunities for prevention and treatments.
    I would like to close by saying it behooves all of us to work together -- industry, the community, and the federal government, which is coming out with its first-ever viral hepatitis action plan in two weeks from today -- and to work quickly, because if we do not, we will not capture the 75 percent of those who are living unknowingly with hepatitis C.
    We will not prevent the costs associated with liver cancer caused by hepatitis and the care and treatment of persons living with chronic hepatitis, which is estimated to increase by over 85 billion by 2024. And we will not prevent progression to liver disease and liver cancer among the estimated 4 million people living with HCV.
    CDC's Division of Viral Hepatitis has said that if we have the capacity in this country to effectively combat this epidemic, we could increase diagnosis rates from 30 to 40 percent up to 75 to
    80 percent, avert 227,000 cases of end-stage liver disease and 11,000 liver transplants, and reduce new infections by 50 percent by 2020.
    All of these estimates were made presumably based on the current standard of care. Think of what we then could accomplish if we invested in public health infrastructure with this new cure. Thank you.
    DR. CARGILL: Thank you for your comments.

    Paul Brayshaw?
    MR. BRAYSHAW: Good afternoon. My name is Paul Brayshaw, and I'm here on behalf of people with bleeding disorders. And I have no conflicts to report, although I have to say I am conflicted. Among people with bleeding disorders, there's a crisis. Without access to therapy, so many of the people that are affected are going to either face a liver transplant or die. We've had these illnesses, hepatitis C for a long time, but we're also confronted by HIV and other comorbidities.
    So we salute the FDA for the early approval or the potential approval for telaprevir and
    boceprevir, but we do seek other urgent developments to access new and far better treatment options.
    As I mentioned the other day, nearly the entire cohort of people with hemophilia over 30 are infected or were exposed to hepatitis C. As a result, we have a long-term illness, upwards of 30 years. Because of the problems with bleeding or our ability to stop bleeding, those issues could be further exacerbated by the products being considered, as well as hepatitis C or HIV medications and HIV in and of itself.
    We would like to encourage the FDA to support advertising or to support advertising and educational materials that indicates difficult-to-treat populations for people who might have these types of bleeding issues, with adequate warnings indicating the treatments that could potentially lead down this road.
    One other issue I mentioned the other day, too, is that we want the FDA to recognize the cost of therapies for hemophilia and other
    comorbidities. On average, hemophilia costs 100- to $150,000 a year. So that cost, plus the other costs associated with the treatments for hepatitis C, could potentially cause some adherence issues or access to therapy issues.
    So our lives basically depend on access to these therapies, and we hope that the ability to -- the information available for treaters is going to allow us appropriate access.
    Finally, I just want to say that we appreciate the guidance and support of the FDA in pursuing small group trials, and we just would like to find additional -- or we seek additional efforts to provide that further guidance so we can get access to those trials within small groups or other difficult-to-treat populations.
    Thank you.
    DR. CARGILL: Thank you for your comments.
    Sonia Spangenberg?
    MS. SPANGENBERG: Hello. I'm here just on my own behalf as a telaprevir trial patient, and also as a R.N. who has cared for, in ICU
    A Matter of Record
    (301) 890-4188
    231
    environment, patients with end-stage chronic liver disease and liver transplants.
    When my husband and I were diagnosed in 2008, it was really very devastating because I knew the potentials of this disease. When I started researching clinical trials and I came across telaprevir, what I read -- at this point they were doing stage 3 trials -- gave me incredible hope. And what I've seen as far as the outcomes from the stage 3 has been just reinforcing everything that I had read.
    What I experienced during the trials, I experienced every single one of the side effects that were mentioned today. I did have the skin rash reaction, and I ended up having to stop telaprevir at week 7. My husband and I were both early responders and we both have maintained SVR, and I am very grateful for having had the opportunity to be able to try this drug.
    I do have serious concerns about how the providers out in the real world are going to use the information that you have in regards to
    observing signs of the development of the skin reaction and progression of the skin reaction.
    It is a little bit vague, and it can become very serious in a small percentage of the people. And because it's such a small percentage, I think it has the potential for kind of getting overlooked, and somebody might fall through the cracks.
    So I would really like to encourage more development of more precise identifying protocols and treatment protocols and really look at how that's going to translate out in the real world because I've worked in the hospital environment, doctor's office environment, a clinic environment. I know the reality of the time that these providers have to learn about this information, much less do the assessments.
    So, obviously, we also need to educate patients. We need to provide more pictures, I think, that would clarify. And that was just an overwhelming sense that I got from everybody so far today that it's just a little too vague. And so that's the one thing.
    Otherwise, I heartily endorse this drug approval. I think it's just too beneficial to snag up on something like that. And I think that's it. I had no financial strings, by the way.
    DR. CARGILL: Thank you for your comments.

    Michael Casey?
    MR. CASEY: Hi. My name is Mike Casey. I have no financial relationship with Vertex. I'm just here on my own.
    I was a patient in a phase 3 trial with telaprevir, and I had a sustained response. I had been under four different treatments before, had failed every one by week 12. And I have type 1A with cirrhosis, and I had given up hope.
    I'm at gratitude to Vertex for offering a trial to me. And Dr. Younossi at Fairfax Hospital, I couldn't be happier. Thank you very much.
    DR. CARGILL: Thank you so much for your comments.
    KellyAnn Mann Hester?
    MS. HESTER: Good afternoon. I want to
    A Matter of Record
    (301) 890-4188
    234
    thank you for this opportunity to speak with the panel. And I have no financial disclosure, but I was a patient on Phase 3 of the null responders.
    I'm the mother of four children and the grandmother of five, and so, obviously, on December 12th of 2009 when I had my hepatitis C fully relapse, obviously fully responding, I was excited and very scared. But I would like to go back to the years when I did all the different treatments that came along.
    I was diagnosed in November of 1993. A well-meaning physician at the time told me I would not see my son graduate from high school. He had just started kindergarten. I walked out of that doctor's office that day and told my family that wasn't an option, obviously, and made up my mind on that day that I would have to try every treatment that came down the system for hepatitis C so that I could be with my child to see him graduate from high school.
    Being a relapser is a very difficult situation. I have done every type of -- when I was
    diagnosed in 1993, it was still non-A, non-B for most physicians, and there was no treatment. I've done interferon, interferon with ribavirin, pegylated interferon with ribavirin at half strength, pegylated interferon with ribavirin at full strength, and now with the telaprevir. Telaprevir gave me the clear. I have no reason to think that the telaprevir wasn't the answer.
    I did get the telaprevir rash. I had it on my hands, my legs, and my feet. But it was very little consequence to me, and would it have made a difference if you were telling me it would save my life? Absolutely not. I would do everything in my power to save my life.
    My physicians played a huge role in my ability to get treatment and be treated because two physicians had to step aside and say, you need to go to somebody else. You need to go be in this study, for me to get the treatment that made me so that I am virus-free at this moment.
    I'm not even sure that I can explain to you that this saved my life. I think I was on a path
    that was going to lead me to a place that -- I was in a place where I was living to die. I knew that -- or in my mind had thought this was going to be the thing that would take me. And I had already accomplished my goal, which was to see my son not only graduate from high school but from college. And so I had no hope left, that I thought I was going to live with this disease to die.
    Now I'm living until I die, which is a whole new concept for me, because now I have many windows and many avenues available to me that I did not have before. So I wholeheartedly am asking you to please approve this drug for the general public so that other people can tell my story at some point in time.
    I thank you very much.
    DR. CARGILL: Thank you for your comments.
    DR. BIRNKRANT: I thought before we would go to the questions that it would be helpful for the panel or the committee if Vertex would present their rash management plan that was used in the clinical trials to help answer some questions that
    it seems as though the panel still has.

    DR. KAUFFMAN: Yes. Thank you, Dr. Birnkrant.
    Just following up on some of the discussion right prior to lunch, I'm actually going to ask Dr. Stern to come up to provide some comments related to the severe cutaneous adverse reactions that were being discussed at that time. Then we will show the rash management plan and discuss that for the committee. Thank you.
    DR. STERN: Thank you. I'm Robert Stern. I'm at the Beth Israel Deaconess Medical Center and Harvard Medical School. I was the head of the dermatologic evaluation panel for Vertex, and I'm here today as a consultant to them.
    So first I think it might be useful for me to describe a little bit of our activities. There are three of us, Jean Claude Roujeau, Maja Mockenhaupt, and myself, who reviewed a total of 221 cases of rash. And those were all materials, including photographs on approximately two-thirds of those cases that were either of particular
    interest or led to stopping of any of the medications. Our review, in addition to photographs, included biopsy specimens when available, which was in about 50 percent of eruptions that were in Phase 3 that ended in -- that led to discontinuation, and also the clinical records in terms of laboratory studies, dermatologic examinations.
    At the time of review, we were blinded in any of the blinded studies as to whether patients were on triple therapy or receiving placebo plus peginterferon and ribavirin.
    We reviewed the materials separately, came to our separate conclusions about extent, morphology, the extent to which it might in fact represent a SCAR, that is, Stevens-Johnson, TEN, or DRESS, drug rash with eosinophilia and systemic symptoms. And then periodically, the three of us came together and came to consensus. And I would say that in the great majority of times, there was a high degree of agreement even before we discussed consensus.

    In order to be able to give some idea of the likelihood of reactions, we used two separate standard algorithms that are used in case control studies, one for SJS and TEN and the other for DRESS, with "definite" meaning extremely likely that it was a case, "probable" meaning more likely than not, "possible" meaning not very likely to be a case but one could not exclude it. And we tried to be as sensitive as possible at the cost of specificity.
    So with respect to Stevens-Johnson syndrome, there were three cases that rose to the level of any suspicion, one which we believed was a definite case. And this was a gentleman who was then 11 weeks past using telaprevir and had just started a drug between 1 and 2 weeks before that has been at least associated in the literature with causing Stevens-Johnson syndrome.
    The second was a case we thought to be probably Stevens-Johnson syndrome, more likely than not, and that was a person on triple therapy in the relevant time window. The third was a possible
    case, not very likely, but we could not exclude that possibility, also on triple therapy.
    With DRESS, it's a great deal more difficult because, as Dr. Bigby pointed out, 50 percent -- in general, when you think of DRESS, you think of someone having 50 percent of their cutaneous surface covered with rash. And in fact, relatively few people had that. But we looked at anyone with extensive eruptions or any eruption and looked for eosinophilia and fever.
    Absolute eosinophilia was rare, and because of the leukopenia, we decided to use a more sensitive criteria, which is percent eosinophils, and seeing if there was a substantial change from baseline, even though, to my recollection, none or at most one of these people had absolute eosinophilia at the time.

    All of our DRESS cases had eosinophilia and fever. But in fact, by our criteria, there was one definite and two probable cases, which I think one should look at as highly likely to be cases. And there was the question of diagnosis at the site.

    Two of the three had been listed as DRESS at the site, and one was just listed as rash at the site that we detected. There was one case of DRESS that the investigators indicated was such that we in fact excluded based on our review.
    So I think if you want to put this into clinical perspective in terms of severe cutaneous adverse reactions, I think if I were thinking of this, particularly with Stevens-Johnson, I would think it is likely that this drug will have a risk profile not unlike that of Bactrim.
    DR. KAUFFMAN: Thank you.

    So I wonder if I could bring up the rash management slide. And, Dr. Singhal, do you want to talk about that? Thank you.
    While it's coming up, I should just make a point. One of the comments from the committee earlier was that hepatologists and others taking care of patients with hepatitis C are not exquisitely knowledgeable in dermatology. When we designed the rash management plan, we took that into account, and we did not -- we wanted a plan
    that would be practical for people who were not exquisitely knowledgeable in dermatology to be able to use.
    Whether or not the correct dermatologic diagnosis is made, the plan provides an algorithm for stopping treatment that can be followed by anyone based on the skin area, whether or not a label of DRESS or just an eczematous rash or whatever else is put on it.
    That's what we mean by being practical and manageable, in that anyone can use it. You don't have to have a specific dermatologic diagnosis. The criteria for stopping are directly in the plan, as you'll see. So if we can bring that up.
    DR. SINGHAL: Thank you.

    If you could put CS-16 up? So whilst it's coming up, I'm going to take you back to my core safety presentation, where I spoke about the rash management program within the Phase 3 program. I want to start off by saying that it's a very simple approach, and investigators were able to follow it. I do appreciate the comments that were made today
    by the patient advocates as well as the committee, and please stop me if there are any questions at any time.
    The first point that we were very clear about was that general principles for rash management should be followed. We also required monitoring for all rash events. What did this exactly mean? It meant that patients were to be told that the rash was possible -- informed consent had this information very clearly stated -- as well as physicians were made very much aware that a rash was possible.
    They were also made aware of the fact that more often than not, they would be mild or moderate, but that all rash events needed to be monitored for the possibility of progression, which, albeit small, it definitely existed.
    In addition, we provided a very simple algorithm where we said that if it was a mild or moderate rash, no study drug discontinuation was required. If it was a severe rash -- and we provided descriptions of each of these, so a mild
    or moderate rash was either a localized rash or a rash that involved less than 50 percent of the body surface area, and while it may be disturbing, didn't keep the patient up at night or was not associated with extensive symptoms. That did not require any discontinuation.
    Moving on to the description of a severe rash, I'd like to spend a minute here because this is really broken up into the two categories that I spoke of earlier, which I think have been raised as very specific points today.
    One is a severe rash that involves an extensive body surface area and could keep the patient up at night, maybe associated with extensive pruritus, but in general is not a life-threatening reaction such as Stevens-Johnson syndrome or DRESS, as Dr. Stern just described. So, in a sense, it didn't have any mucosal involvement or it didn't have any lesions that may look like target lesions. That type of severe rash just required telaprevir to be discontinued.
    Physicians were given more guidance on this,
    where we asked them to monitor these patients very closely, and if they felt within 7 days that the rash had not abated after stopping telaprevir, they could think about stopping ribavirin and/or peginterferon. So that was the instruction for severe rash.
    Rare severe skin reactions, which are actually included in the label for peginterferon and ribavirin, certainly were the ones that would require all study drug discontinuation and all medication discontinuation, because there are several drugs that can cause these reactions.
    So that was, in essence, the rash management plan within the Phase 3 program. Within the Phase 3 program, we also built in within the rash management plan an extensive data collection plan. And that is exactly what you have heard from Dr. Stern with regards to the materials that were evaluated because we wanted to evaluate the severe reactions as well as the severe rash very comprehensively.
    So I'd like to stop there. That was the
    rash management plan in Phase 3. And the purpose of today is to translate this into the real world, which we realize and are committed to entirely.
    So I'd like to put up RM-15, please. This is a very high-level management approach for what we are thinking about and planning for rash management within the real world and hopefully after the drug is approved. So the first would be very specific language in the label. This would be warnings for severe rash, but also, most importantly, to distinguish the severe rash from the severe cutaneous adverse reactions, and also to provide the information on the mild and moderate rash and alert prescribers to monitoring. We would also provide all the rates that have been observed during the Phase 3 program, and comparative rates with placebo.
    The second point, which cannot be overstated here, is a very robust communication plan to patients and healthcare providers. And how are we going to do this? It's going to be a very robust medical education program. For patients, we are
    going to inform patients that they need to connect with their healthcare providers, that I think a point was raised a few minutes ago about discontinuing. This is a very specific point that we will be stressing in education and all materials that are being prepared. They should not discontinue telaprevir on their own.
    Healthcare providers similarly will know when to connect with their patients and what to look for, and to follow a very simple plan which says that if it's severe or you see any of these reactions, discontinue telaprevir.

    In addition, to provide additional resources and support, we are preparing for a nurse hotline so that there will be the resource for physicians and prescribers who may not be familiar or may not be comfortable treating rash on their own.
    Finally, I want to stress the point of medical education, which is going to be performed through multiple forums. And this is going to be via continuing medical education, our medical-to-medical-science liaisons, the field force, and a
    lot of resources for patients and physicians which are going to help to communicate -- help patients and physicians and healthcare providers connect with each other.
    So that's, in essence, the translation of the rash management plan from Phase 3 into the post-approval setting. And I'm happy to address any questions.
    Questions from Committee to Sponsor and FDA
    DR. CARGILL: Thank you.


    I think at this point I would like, in the interest of time, for the committee to return to the questions they have for both the sponsor and to the FDA. And I will continue to take those names in sequence.
    Dr. Giordano?
    DR. GIORDANO: My question was addressed last session.
    DR. CARGILL: Thank you.
    Dr. Ellenberg?
    DR. ELLENBERG: Yes. In the briefing document, the sponsor indicated that virologic
    outcome was assessed in what seemed to me quite a wide visit window. For subjects who had a treatment duration of 24 weeks, the visit window was week 32 to week 78; and for subjects assigned to 48 weeks of treatment, it was week 56 to week 78.
    I'm not sure I'm understanding this. But I just wanted to make sure that it couldn't be the case that some people waited much longer than others to have their viral load assessed, and if that wasn't balanced on the treatment arms, there could be a bias, and that some people were looked at, at a much longer time after they had treatment than others.
    DR. KAUFFMAN: Yes. So patients did come back at specific times to have their HCV RNA assessed, as we indicated. And we actually had very, very good adherence with that schedule. There was very little missing data. And we continued to follow patients out to week 72, even those who had stopped treatment at week 24 in the short duration arm so that we had common assessment points in the protocol for everyone for their HCV RNA, and we had very, very good ascertainment out to week 72.
    The way the analysis was done was that the last visit in that window was the one that was generally used for the adjudication of detectable or undetectable for the assessment of SVR. And the windows obviously varied depending on whether patients had stopped at 24 weeks or 48 weeks.

    DR. ELLENBERG: And can you say anything about the comparison of the average window time on the treatment arms? It sounds like the time of what you used in the analysis varied. It was the last one. So for one subject, it might have been 72 weeks and for another subject it might have been 58 weeks.

    DR. KAUFFMAN: Yes. But in general, the assessments were really done really very close to the nominal time that they were supposed to have come in.
    I can ask Dr. Sankoh to come up and expand on that further, if you like.

    DR. SANKOH: Hi. A.J. Sankoh, Vertex Biometrics. Yes. As Dr. Kauffman indicated, we actually have a common assessment for all the subjects. We realized that, yes, as you were saying, for those who stopped treatment at 24, compared to those who stopped treatment at 48, the duration of antiviral follow-up was different.
    So we followed everybody to week 72. Even for those who stopped treatment earlier, they're not finished, they're assigned treatment regimen. And there was a very small loss to follow-up of all subjects, less than 1 percent, actually, in all the treatment groups. So the assessment was really intentionally done that way to make sure we have consistency in the way we assess.
    DR. ELLENBERG: Okay. So the way I'm reading it I think then is incorrect. You're telling me that almost everybody on all the arms were in fact assessed at approximately the same time?
    DR. SANKOH: Yes. Yes.
    DR. ELLENBERG: Thank you.
    A Matter of Record
    (301) 890-4188
    252
    DR. CARGILL: Thank you.
    Ms. Young?
    MS. YOUNG: Yes. I was wondering if the sponsor could summarize the protocol for anemia management. Is it a very clear-cut protocol that is going to be easy to understand and follow? Thank you.
    DR. KAUFFMAN: Yes. As I mentioned, it was primarily based on the ribavirin package insert. I'll ask Dr. Singhal to come up and talk about that.
    DR. SINGHAL: Here again, I'd like to go back to my safety presentation. CS-26, please.
    Anemia management was included in the protocol very specifically, and it required prescribers to follow the ribavirin dose modification per the ribavirin label. However, because we were aware from the early Phase 2 data that telaprevir can cause an additional gram or 1.5-gram decrease, we did specifically mention that telaprevir discontinuation could be considered per clinical judgment, and investigators were aware of
    this based on their data that was included in the investigator brochure in the guidance to the investigator section. However, we pointed out that telaprevir could not be reduced and restarted, just like for rash.
    Discontinuation. If the investigator followed the ribavirin label, which led them to discontinue ribavirin, they needed to also discontinue telaprevir because we were aware that telaprevir could potentiate the anemia, and we did not want patients to continue if the anemia had gotten to 8.5 or lower, where ribavirin needed to be discontinued. So we wanted to take a conservative approach. And that was, in essence, the management. Monitoring was done at regularly scheduled visits for hemoglobin.
    Here again, maybe if I could translate this on, to what is going to happen beyond the Phase 3 program and how we are going to build on this for the future.
    If I may have that?
    DR. CARGILL: If you can be succinct.
    A Matter of Record
    (301) 890-4188
    254
    DR. SINGHAL: Yes. Sure. Please, can I have RM-33?
    It's again going to be a three-pronged approach, with the labeling very specifically giving the information on anemia, alerts to monitoring hemoglobin, providing the details of what is the pattern of hemoglobin decrease and recovery, as well as the details of what was the use of ESA during the clinical trial program -- because it is not an FDA-approved drug for hepatitis C-related anemia -- and ribavirin dose modification.
    In addition, we'll be reaching out to patients, healthcare providers, with the same information, very much like the rash, with the alerts to monitor and also to use clinical judgment and discontinue per the same algorithm. And medical education, the nurse hotline, the websites available with this information, and reach-out programs will be identical, just like us.
    DR. KAUFFMAN: Thank you.

    DR. CARGILL: Dr. Clay?
    DR. CLAY: Put your slide back up because I guess I have a practical question.
    We have two drugs that are going to be used to treat hepatitis C, pegylated interferon and ribavirin. We now have a choice, potentially, of what the third drug will be. Both of these third drugs have shown that they can cause anemia.
    Historically, when a person developed anemia during the course of therapy for hep C, you reduce the ribavirin, potentially interrupt therapy. You have made it very clear if you interrupt ribavirin you must also stop telaprevir.
    Dr. Friedman is seeing somebody in clinic. His patient has shown up with anemia. How can he tell if the anemia is from ribavirin or is it from telaprevir? Because we can discontinue our telaprevir and perhaps switch to another drug that has been shown to be effective in combination with pegylated interferon and ribavirin.
    My question specifically gets to, can you tell me what exactly the anemia looks like with respect to telaprevir? Can we distinguish the
    anemia of telaprevir from the anemia of ribavirin? Because it directly affects how we're going to manage the person in front of us.
    DR. KAUFFMAN: So I guess I'd answer that by saying that the hemoglobin drops that are seen --

    DR. CLAY: What I'm talking about here is sideroblastic anemia with erythroid hyperplasia, vacuolization of erythrocytes, increased serum iron concentrations.
    Do you have that information that a clinician could then look and say, "Oh, gee, that's not present, so it must be from the ribavirin. But, oh, wait, that is present, so let's switch over and let's use perhaps the other third drug"?
    DR. KAUFFMAN: Yes. So I guess I have to answer that question by saying that, of course, because both telaprevir and ribavirin are obligatorily given together, and whatever anemia occurs is an anemia related to the regimen, the data that we have in our program really cannot answer the question as to which specific component is related to the anemia in any given patient.
    All we can say is that on average, there's a 1- to 1 and a half gram further drop in hemoglobin as compared to peg and ribavirin alone, and that it can be managed by ribavirin dose reduction.
    DR. CLAY: And you had --
    DR. KAUFFMAN: I understand your point, but we have -- I have no data.
    DR. CLAY: You had zero occurrence, zero change in hemoglobin in your normal, healthy studies when you administered it for 12 weeks? So that's kind of the life cycle of the red blood cell.
    DR. KAUFFMAN: No healthy volunteers received the drug for that long. And all healthy volunteer studies are confounded a little bit by the amount of blood drawing that occurs in clin pharm studies, and so they're almost always drops --
    DR. CLAY: Yes. You know what? You pay them enough, they'll get it drawn.
    DR. KAUFFMAN: Pardon me?
    DR. CLAY: You pay them enough, they'll take
    that blood.
    Okay. So you don't have that information?
    DR. KAUFFMAN: We do not.
    DR. CLAY: We cannot tell what's causing it, so we don't know if it's going to be safe to switch someone from telaprevir to another third drug. Are we going to have to --
    DR. KAUFFMAN: We simply don't have that information at this point.
    DR. CLAY: Okay. I do have a second question.
    DR. BIRNKRANT: That's been brought up twice now, about switching from one to another. But there's no data with regards to that.
    DR. CLAY: I was proposing a hypothetical, that in the future, if there is a third agent, we would like to know which one is causing the anemia.
    DR. BIRNKRANT: Right. But that I think would require additional data, which we don't have at this point.
    DR. CARGILL: Dr. Roland? Dr. Ghany?
    DR. GHANY: Yes. Thank you.

    So I wanted to get back to the issue of a ribavirin dose reduction. And either the FDA or the sponsor could explain to me the quantity of ribavirin dose reduction that we've been talking about. Is this 200? 400? 600? And the timing of the reduction, and then how many individuals restarted back at their prior dose? So what was the average ribavirin dose in the first 12 weeks during triple therapy?
    DR. KAUFFMAN: So the dose reduction that was recommended and what is in the ribavirin package insert is to reduce the dose in half. It depends on what weight the patient is and which dose they're taking. But the initial dose reduction is in half. And then if there's a failure of response, the subsequent reduction then is to discontinue ribavirin, and that is what was followed.
    We understand in the community there are many ways that ribavirin is being managed. But in the protocol, we had to stick with the dose reduction scheme that was actually in the ribavirin
    A Matter of Record
    (301) 890-4188
    260
    labeling.

    DR. GHANY: So do we actually know what the average ribavirin dose was in the first 12 weeks?

    DR. KAUFFMAN: I don't have that information offhand. I can see if we can get that, but I don't know that.
    DR. GHANY: Okay. And I do have a second question, if I may, and it pertains to the design of study 111. I think that's ILLUMINATE.
    DR. KAUFFMAN: Yes.
    DR. GHANY: I wondered, why did the randomization occur at week 20? And was that because some individuals, after telaprevir was discontinued at week 12, did they have virological relapse? And if so, were those patients excluded from the analysis? How was that handled?
    DR. KAUFFMAN: Yes, you're correct. The randomization took place at week 20. If you remember, the study was really designed for a very specific purpose, and that was really to examine whether there was any benefit to extending treatment to 48 weeks compared to 24 weeks in those
    who had a rapid virus -- an eRVR.
    Therefore, we wanted to maximize the SVR rates in those randomized arms to reduce any noise that might occur by discontinuations and other things. So we randomized relatively late in the treatment period.
    We certainly accounted for every patient who entered the trial. The overall SVR rate of 74 percent that we showed includes everyone, no matter what happened to them, whether they're randomized, not randomized, assigned to 48 weeks, or had dropped out prior to that. Everyone was accounted for at the end of the study, and that's the 74 percent SVR rate.
    Of course, in those who discontinued earlier, if you have the study design in mind, you remember that lower arrow. Those patients, of course, because they discontinued early for any number of reasons, had a lower SVR rate than those who didn't. Everyone was accounted for in the end. That's why we presented that overall rate, to give you an idea of sort of what the overall outcome was
    in the entire trial.
    DR. GHANY: But patients between week 12 -- dropouts between week 12 and 20 were not counted in the rates, the final rates, at 24 and 48. Is that correct?
    DR. KAUFFMAN: Those who dropped out between week 12 and 20 were not randomized. They are accounted for in the analysis later, but they were not part of the randomized group. Patients had to be on treatment at the time of randomization in order to be randomized.
    DR. GHANY: And do we know what that number was?
    DR. KAUFFMAN: The number who dropped out?
    DR. GHANY: Yes, between week 12 and 20.
    DR. KAUFFMAN: It was about 5 percent. The eRVR rate in the study was 65 percent, and about 60 percent of patients were randomized. It was about 5 percent who didn't make it from week 12 to week 20.
    DR. GHANY: Thank you.
    DR. CARGILL: Thank you. I'm going to ask
    the panel to be mindful of our time so that our questions can be succinct.

    Dr. Knodell?
    DR. KNODELL: Just one quick comment and then my question. My comment is that the cirrhotic patient population are the people that most need your treatment. Relapsers responded really quite well to treatment, but we know that they're pretty easy to treat anyway.
    I'm wondering whether perhaps some additional studies with telaprevir might be worthwhile at a higher dose and that perhaps the poor results in cirrhotics were due to low drug bioavailability.

    My question relates to one that I'm surprised hasn't been raised before, and that's to the information. You haven't provided as much information on drug-drug interactions. You gave us an intriguing slide, CO-5, that said a bunch of studies had been performed, and in the information you provided to us pre-meeting there was a statement that methadone concentrations dropped off in people that were dual exposed to telaprevir.
    Can you comment more on your drug-drug interaction studies? I'm sure you've got that information.
    DR. KAUFFMAN: Did you have a specific study in mind? Would you like to see the -- you mentioned methadone. Is that one you'd like to see? Otherwise, there are a large number of studies. We can certainly go through them. But if there are specific ones you're interested in, we can focus on those.
    DR. KNODELL: Well, I don't know that the committee needs to see
    everything. But, yes, I mean, methadone is -- certainly in the VA, it's a commonly used drug in hepatitis C patients. Pick a couple that are most likely to be used concomitantly with patients with hepatitis C.
    DR. KAUFFMAN: Fine. So I'll ask Dr. Garg to come up and talk to you about the methadone interaction study.
    DR. GARG: So if I could have slide DI-38. While the slide is being put up, I would just like
    to tell you about the design of the study.
    This was a study done in healthy subjects -- well, non-HCV subjects who were on methadone for a few weeks, at least. I think it was more than a couple of months. And they were stabilized -- and they were on an individualized stable dose of methadone. And then their PK levels of methadone were taken before starting telaprevir as well as after starting telaprevir.
    This slide shows that -- in the green is the individualized methadone and the red is the individualized therapy of methadone with telaprevir. As you can see, there's about a 30 percent drop in the AUC of methadone. This is for the total drug, however.
    We actually examined the unbound levels of methadone because we thought that they might be approaching binding displacement, and we actually found that the unbound levels of methadone were actually similar. So it also accounted for the fact that there was no increased withdrawal symptoms in the subjects that had a lower level of
    methadone.

    So we believe that this interaction was actually not clinically significant because the protein binding displacement reduces the total levels of methadone but not the unbound levels.
    DR. KNODELL: Do you have anything on midazolam?
    DR. GARG: So in the case of midazolam, we studied, again, midazolam before starting telaprevir and then, at steady state of telaprevir, a single dose of midazolam, 2 milligrams orally as well as with IV midazolam. And for that, we found that oral midazolam levels were increased roughly about ninefold with telaprevir. So we concluded that telaprevir was a potent inhibitor of CYP3A based on that.
    DR. CARGILL: Thank you.
    Dr. Friedman?

    DR. FRIEDMAN: My question is also to the sponsor, and it's about the rash management program. Will that include specific recommendations regarding the use of antihistamines, topical and systemic glucocorticoids, and referral to a dermatologist?

    DR. KAUFFMAN: I'll ask Dr. Singhal to come up and address that.
    DR. SINGHAL: While the use of treatment used during the Phase 3 program was collected and we have this available, we are unable to assess whether they were effective because, as you know, 93 percent of subjects don't go on to progress. So we're not sure whether it was the treatment or they would have resolved on their own.
    So we don't have -- we will not be including specific recommendations. However, good skin care practices will be included. And we will have a five-step approach where the first step will be, really, patient awareness. Clinical diagnosis will be step 2. Assessment of severity, management as per general principles of management, and we do believe that it is an eczematous rash for which, in general, topical steroids and good skin care practices do help to keep patients on.
    I think one very important point is it's a
    12-week duration, and it's the first 12 weeks. And we do know that the rash improves significantly as soon as telaprevir is stopped. So we do believe that once telaprevir is completed, with good skin care management, this will be really very manageable.
    DR. FRIEDMAN: Thank you.
    DR. CARGILL: Thank you.
    Dr. Strader?
    DR. STRADER: This is just a clarification question on the rash. I think it was Dr. Stern, if I have the name right, who presented the information on the severe, the Stevens-Johnson, the DRESS.
    But my confusion is the use of the terms "definite," "probable," and "possible." Are you talking about the possibility that this was a rash or the possibility that it was associated with telaprevir? So there's a difference between those, if you understand what I mean.
    DR. KAUFFMAN: It was definite, possible, or probable related to that specific diagnosis. It
    had nothing to do with the relationship to the drug. Because most of them occurred in telaprevir, we considered that the relationship was certainly at least possible --
    DR. STRADER: Possible.
    DR. KAUFFMAN: -- related to telaprevir.
    DR. STRADER: Okay. Thank you.

    DR. CARGILL: Dr. Giordano?
    DR. GIORDANO: We haven't heard anything about futility rules in any of these trials. It was brought up by the public comments. Were there any futility rules in these trials? And if not, were there any post hoc analyses that would suggest that there could be?
    DR. KAUFFMAN: Yes. Each of the trials included stopping rules for virologic failure and also the typical quote unquote "futility rules" at weeks 12 and 24 that are generally applied for current standard of care for hepatitis C.
    In the case of the treatment-naive studies, the virologic rule was those patients with an HCV RNA above 1000 international units per mL at week 4
    were considered to have either incipient breakthrough or failure. And those patients were recommended to stop telaprevir, but they could continue peg and ribavirin. Those who had detectable HCV RNA at week 24 had to stop, as they are in the current rules.
    In the treatment failure study, there were slightly different stopping rules. The threshold was somewhat lower. And this is something that we'll be working with the agency on to devise appropriate stopping rules for telaprevir.
    We're obviously looking back now at the Phase 3 data to learn more about what the rules perhaps should be, and we'll be certainly interested in talking to them about what to include in the labeling.
    I think the issue perhaps is most acute in the null responder group, where we know that virologic failure occurs most frequently, and we certainly will be looking at futility rules there to be able to identify patients early who have no chance of achieving an SVR, therefore preventing
    them from being exposed to the regimen for the full length of time.
    Obviously, we can identify many of those patients quite early on in treatment, and therefore, they don't have to be exposed to the drug for very long. And again, we'll be working on those rules with the FDA. We've had some preliminary discussions with them and will obviously continue after this meeting.
    DR. CARGILL: Thank you.

    Dr. Bigby?
    DR. BIGBY: Actually, this is a point of clarification for the committee. I don't know, it's not really a question. So if you want to do it later, I can do it later.
    DR. CARGILL: Okay. Then why don't we do that after the break.
    Ms. Valbh?
    MS. VALBH: During the FDA presentation, a comment was made that baseline hemoglobin was hard to assess. And so my question is, was it not collected in any of the trials? Was the baseline
    hemoglobin not collected?
    MR. FLEISCHER: No. It was clinical anemia. We have hemoglobins on everybody at multiple time points.
    MS. VALBH: At multiple time points?
    MR. FLEISCHER: That wasn't the issue, no.
    MS. VALBH: And then for the sponsor, there's a lot of guidelines on how anemia should be managed and how often the labs should be done. But in clinical practice, it's a little grey. And even now with ribavirin, it's a bit grey, and especially for those clinicians who don't really treat hepatitis C patients that often. So they look mostly towards what was done during the clinical trials.
    During all your clinical trials, how often was a CBC assessed? Was it every 2 weeks? Every 4 weeks? Once a month? Can you provide some clarification on that?

    DR. KAUFFMAN: In the trials, certainly during the first 12 weeks, which was the period of telaprevir administration, patients were seen no
    less often than every 2 weeks, and so we have very frequent hemoglobin measurements during that time.

    Again, in thinking forward, we're looking at those data to try to determine what the most informative time points would be to make recommendations once the drug is approved, what the labeling should recommend in terms of monitoring. And, again, that's something I think that we in the FDA will deal with. But we're going to look back at our data to try to find the most appropriate time points for monitoring.
    Since for eRVR status HCV RNA has to be assessed at week 4 and again at week 12, those are obvious points at which time we would look at them, but there likely will be points in between as well.
    MS. VALBH: Okay. Thank you.
    DR. CARGILL: Thank you.

    Dr. Clay?
    DR. CLAY: I'm going to go back to a point I raised much earlier in the day, and it relates to your pyrazinoic acid level concentration. That's part of your metabolite.

    I understand that you performed some pharmacokinetic assessment during a phase 2 study that involved 17 individuals. My question relates to the fact that the more I look at the side effect profile of this, it really looks like pyrazinoic acid is causing a great deal of this.
    The clinical implication of that is you did not perform PPD baseline on people at the start of the study. People who have hepatitis C are also the same population that are at high risk for tuberculosis. Obviously, we would be concerned if someone were to receive pyrazinoic acid by itself for the duration of period of hepatitis C therapy.
    So I guess I need to see the pyrazinoic acid assay results. And if you could show me that -- because, granted, I don't know all there is out there, by all means. My children tell me that all the time. If you could tell me what assay you use because I didn't think there was an assay for pyrazinoic acid; that there was an assay for pyrazinamide, but pyrazinoic acid is measured through 24-hour urine collection of
    5-hydroxypyrazinoic acid.
    DR. KAUFFMAN: So I'll ask Dr. Garg to come up. I actually wanted to correct something that I told you earlier. It turns out that patients with tuberculosis were excluded from -- anyone with active tuberculosis would have been excluded from the trials.
    DR. CLAY: I was a little surprised, yes. That's generally a standard enrollment criteria.
    DR. KAUFFMAN: Yes. That's correct. So I'm sorry I misspoke at that point.
    I'll ask Dr. Garg to come up to talk about pyrazinoic acid assay.
    DR. CLAY: Thank you.
    DR. GARG: So we did have a pyrazinoic acid assay, which was an LCMS assay. And we used the standard pyrazinoic acid as a reference standard for that.
    There are at least two papers that I'm aware of -- And, actually, those are just the papers that I have -- which have pyrazinoic acid PK described in there. The maximum concentrations of pyrazinoic
    acid in the plasma was reported from about 5.5 micrograms to 67 micrograms per mL. These were in healthy volunteers, subjects taking about 35 milligrams per kilograms of dose of pyrazinamide. And they measured both pyrazinamide and pyrazinoic acid in that. And the levels that we observed, the maximum levels that we observed, on an average was about .4 micrograms per mL.
    So that's why I say it was about 14-fold to more than 150-fold greater in the subjects that took pyrazinamide as opposed to telaprevir at steady state of telaprevir.
    DR. CARGILL: Thank you.

    Dr. McGovern?
    DR. MCGOVERN: I'd just like to ask some questions about the rash again.
    In terms of the information you gave to us, you said that the median onset of rash was about 25 days, and about a third of those occurred in the first month. I just wanted to know, was there any information that drug rashes that ended up being severe, did they have a different tempo? If a rash
    develops that's mild to moderate, was there a certain window of time that let's say that rash stayed mild to moderate, that severe rash didn't seem to occur?
    I guess I'm just trying to think about this issue of monitoring closely how often, when can I start to relax with those patients according to the data you already have.
    Then, as a non-dermatologist, when you say less than 50 percent of skin involvement, is that just by gross observation or do you have any more information to share with me about that?
    Finally, in terms of the Phase 2 and the Phase 3 trials, 5 percent discontinued in the Phase 2 and then it was 0.8 percent in the Phase 3. What do you attribute that to? Is that due to your algorithm? Thank you.
    DR. KAUFFMAN: Yes. I'll ask Dr. Singhal to come up. But to your last question, yes. We believe that the difference in discontinuation of all drugs is in fact the algorithm. Since in Phase 2 we recommended all three drugs be stopped,
    in Phase 3 we refine that so that telaprevir is stopped for mild to moderate rash and PR can continue. That was not the case in Phase 3. We just didn't have enough experience with rash at that point to have been so discriminating, shall we say.
    Dr. Singhal?
    DR. SINGHAL: With regards to your first question, severe rash in general, the 4 to 5 percent of subjects who experienced severe rash across the Phase 2 and 3 program, about two-thirds of those at least were reported as being severe at the outset. And then about 30 percent were reported -- 30 to 35 percent were reported as being progressive rashes. So this was progressive grade 1 to grade 3, and also progressive grade 2 to grade 3. But it was a third.
    With regards to time to onset, we examined this with regards to any rash as well as by grade. And as you very rightly pointed out, about 50 percent of all rashes that occur on telaprevir combination treatment occur in the first 4 weeks.
    And if you break that down, as I showed with the Kaplan-Meier earlier today, they really occurred -- many of them start occurring in the first week.

    However, rash has been noted at any point until the 12-week period. But the median time to onset of a severe rash is approximately 7 to 8 weeks. So 50 percent of them can occur after week 7, which is a reason to be vigilant up to the end of the dosing period.
    With regards to your second question about body service area, I just wanted to reinforce that the body surface area, when we spoke of the 50 percent body surface area, we meant involved body surface area. So not just by saying that if the trunk is involved, you have some rash on it -- it's not tantamount to 18 percent, for example, by the Burns Rule of Nines. It should be really the entire trunk being involved to be counted as 18 percent.
    However, because our derm expert panel noted that investigators tended to overestimate rash body
    surface area, and we would want to prevent patients from unnecessarily stopping telaprevir, we are giving this special consideration in the educational documents that we are creating.
    So our plans are, and they're currently underway, to create illustrations of body surface area, to have photographs that will enable prescribers to distinguish 50 percent from localized or from above 50 percent.
    DR. MCGOVERN: And just one quick thing. Are you going to have a patient handout at the time you give out the prescriptions, if this drug is approved today, that physicians can give out to patients so that they understand the issue of rash?

    DR. KAUFFMAN: Yes. We do plan to do that. We sort of have in mind a starter kit that would include information like that when patients get their first prescription.
    DR. CARGILL: Thank you.

    Dr. Van Dyke?
    DR. VAN DYKE: My question was already addressed. Thanks.
    DR. CARGILL: Excellent.
    Dr. Ellenberg?
    DR. ELLENBERG: Yes. My question was just asked.
    DR. CARGILL: We're zooming here.
    Dr. Murata?
    DR. MURATA: I'm just simply requesting a clarification.
    I understand from the sponsor's consultant, Dr. Stern, that there were a number of biopsy specimens that were submitted for outside consultant review. Were such skin biopsies of rashes mandated by protocol or performed at the discretion of the site investigator?
    DR. KAUFFMAN: I'll ask Dr. Singhal to discuss that.
    DR. SINGHAL: For purposes of the Phase 3 program, this was a very specific requirement. And if I can elaborate further, there were three Phase 3 trials. The ADVANCE study, which was in treatment-naive subjects, required all patients to have a dermatology consult with a rash that
    required discontinuation or was grade 3 or severe, to have a dermatology consult at the site as well as have photographs taken, which were stored in a central repository, and have a skin biopsy performed with four unstained slides.
    Then those unstained slides were sent to a central histopathology at the Beth Israel Deaconess here, who was part of our derm expert panel, who stained those slides and read them and gave us his consistent, sort of central read on every single biopsy. That was for study 108.
    From study 108, we had about 56 subjects who experienced an event of special interest. And from those, we collected biopsies in 37. So we did have a substantial number.
    Now, in addition to that, we also collected all biopsies that were collected or performed for 111 and C216, which were not stained by our central reader but were read and read out by him, bringing it to a total of about 82 biopsies.
    DR. MURATA: So if I can just ask for one further clarification.
    Given those numbers and descriptions, what percentage of the Phase 3 studies actually received the biopsies?
    DR. SINGHAL: What --
    DR. MURATA: Those who met the criteria for a dermatology consult?
    DR. SINGHAL: Yes. So in the Phase 3 program, the event of special interest, so a rash that was grade 3, serious or causing discontinuation of any drug, was reported in about 7 percent of all subjects. And from those, we received biopsies of close to 70 percent of all of them.
    DR. CARGILL: Thank you.

    Ms. Dee?
    MS. DEE: Hopefully this will be quick. The one-dose kidney impairment study, do we know what happened with that? It seemed like there was a question in the briefing document as to whether that would be sufficient.
    DR. ROBERTSON: Yes. We did have a question, and a lot of it had to do with the dose
    proportionality and linearity and time-dependent PK, which is a little different for telaprevir. But upon further discussion, discussion with the sponsor and some internal analyses, we're content with the single-dose study, and we believe it can be dosed full dose in all stages of renal impairment.
    MS. DEE: Great. Thanks.
    The other thing that was very puzzling to me, in your briefing document on page 11 it talked about illicit drug users, and it mentioned the interaction with methadone and said that 11 illicit drug users were in Phase 3. I can't believe there were only 11.
    Am I misunderstanding something about that? It said, because of these methadone interactions, they were careful about allowing drug users into the trial, but they did allow 11 in. In other words, is that possible, that only 11 drug users were in all these Phase 3 trials?
    MR. FLEISCHER: Well, it was a pretty strict exclusion criteria, unfortunately, and it
    was -- I'll tell you what we found.
    So, basically, "Subjects abusing illicit drugs, narcotics, or a controlled substance, or alcohol, or who had a history of illicit substance or alcohol abuse within two years of prior screening were to be excluded; unless subjects who had a history of abuse of illicit drugs or alcohol and an incidence of abuse within two years prior to screening, or subjects who had a history of abuse of narcotics or other controlled substances, known by the investigative site and considered to be a good candidate, were allowed in."
    So that's why -- essentially, those 11 got in because they appeared to be good candidates for going onto a clinical trial. Unfortunately, the rest were excluded.
    MS. DEE: So in the discretion of the investigators, all these people that were in these trials, people that were off drugs for a two-year period, only 11 were considered reliable enough to be in these trials?
    I guess I should ask you that.

    DR. KAUFFMAN: The inclusion/exclusion criteria, it was obviously up to the investigators to decide who to enroll in the trials among those patients on stable methadone who were considered good candidates that were not excluded, and the 11 are those that were chosen by the investigators to enroll.
    DR. CARGILL: Thank you. We're going to continue.

    Dr. Clay?
    DR. CLAY: Convince me the statement, "This drug is not approved for use in people with gout," doesn't need to be in your package insert. I didn't see any interaction studies with gout medicines. Colchicine is no longer available, and you used it -- well, the investigators used it to treat. So convince me that doesn't need to be in your package insert.
    DR. KAUFFMAN: Sorry. Could you just repeat the first part about --
    DR. CLAY: Sure. "This drug is not approved for use in people with gout." It's not approved
    for the use of hepatitis C in people who have gout.
    DR. KAUFFMAN: Yes. I think that, you know, as you saw, there are some elevations of uric acid. I think that would really be something that would be left to the investigator's discretion or in the future prescriber's discretion as to what the potential risks are.
    DR. CLAY: So is your drug safe to use in people who have gout?
    DR. KAUFFMAN: We don't have much experience with it in the clinical trial, so I can't really specifically answer that question.
    DR. CLAY: Okay. Thanks.
    DR. CARGILL: All right. We will now proceed to our break. We will return here at 2:40. It's not going to be 15 minutes; it'll be 10 minutes -- yes, I'm being the Scrooge -- 2:40, so that we can proceed directly to the questions.
    Again, a reminder to the panel: You may not discuss this meeting topic among yourselves or with members of the audience.
    (Whereupon, a recess was taken.)
    DR. CARGILL: We will now begin the panel discussion portion of the meeting. Although this portion is open to public observers, public attendees may not participate except at the specific request of the panel.
    Before I proceed further, I'd like to ask Dr. Birnkrant to make a few remarks.

    Charge to Committee
    DR. BIRNKRANT: I want to thank the committee for their challenging questions and their discussion. We greatly appreciate that.
    Before we go to the questions, I just wanted to remind everyone that we would be most appreciative if you would refrain from comparing the two products that have been discussed over these two days. So today we're focusing on telaprevir. Thank you.
    Discussion/Questions to Committee
    DR. CARGILL: Thank you, Dr. Birnkrant.
    We are going to read the questions and then have discussion after them. Our first question is this: Rash associated with telaprevir use was
    A Matter of Record
    (301) 890-4188
    289
    common and sometimes severe and treatment-limiting, and anemia was more frequent and more severe in patients treated with telaprevir.
    Please comment on the safety profile of telaprevir, focusing on the increased frequency and severity of rash and anemia when telaprevir is added to pegylated interferon and ribavirin. Do these adverse events affect your risk-benefit assessment, and if so, how?
    I'm going to take the chair's prerogative to ask Dr. Bigby to speak first.
    DR. BIGBY: Yes. Thank you very much. I just wanted to clarify a couple of things about what has been referred to as severe skin reactions.
    So SJS and TEN, like I said before, are similar conditions. They generally occur in -- one case in something like 250,000 to a million exposures. In the case of this drug, you've seen two, maybe three cases in Phase 3 trials that had around 2,000 people. So it's much more common than in drugs in general.
    For DRESS, the data is not as good, but I'd
    say it's somewhere in the order in general of a case in 5- to 10,000 exposures. And again, with this drug you see 10 or 11 cases in Phase 3 trials of about 2,000 patients.
    The third thing is that I actually was given a list of the reactions and the descriptions. And an additional severe reaction that wasn't mentioned yesterday in our discussion is AGEP, which is acute generalized exanthematous pustulosis, which again is usually a very rare entity. And it was mentioned, I think, at least once in Phase 3 studies.
    So I think you will be seeing severe adverse reactions with this drug, probably more than with many or any existing.
    DR. CARGILL: Dr. Roland?
    DR. ROLAND: I think it's clear that for both of these adverse events, that it does affect the risk-benefit ratio, but I don't think that the risks outweigh the benefits, at least as far as we know them right now. I would like to think a little bit more about how to try to minimize the
    risks and how to understand more completely in the real world what the real incidence of especially the skin reactions are.
    So I don't really know what the precedence is in a situation like this for mandated reporting, some sort of registry. I'd also like to think more about how we use 21st century communication mechanisms with a nurse hotline. So being able to send photographs, being able to have real-time expert consultation, would make me feel more comfortable.
    DR. CARGILL: Thank you.

    Dr. Friedman?

    DR. FRIEDMAN: I have a question for Dr. Bigby. The reactions of telaprevir were compared with those of Bactrim, which is an approved drug. And I wonder how your calculated figures as to the frequency compares with what we know about Bactrim.
    DR. BIGBY: Yes. I don't know how Dr. Stern could have made that statement because I think that the -- at least -- you don't usually pick up an
    ASJS/TEN/ DRESS signal when you do a trial of 2,000 patients. And I don't think that if you look back at Bactrim trials, you'd find it of that frequency at all.
    DR. CARGILL: Dr. Strader?
    DR. STRADER: I agree with Dr. Roland that these side effects do appear to be more frequent. However, I think that the sponsor was very diligent in trying to characterize the symptoms and come up with plans for monitoring patients and for managing them.
    As a practicing hepatologist myself, I think that I would not be terribly concerned about the management of these things. I think it is important, though, that we somehow make sure that -- with respect to the rash, that dermatologists may be involved. If there's some concerned, 21st century communication notwithstanding, I think it's better to actually have your eyes on the patient as opposed to being sent a photograph. But I think it's important that perhaps that happened, but I do not think that
    A Matter of Record
    (301) 890-4188
    293
    these would preclude my saying that the drug should probably be approved.
    DR. CARGILL: Thank you.
    Dr. Van Dyke?
    DR. VAN DYKE: Yes. My concern is that rash is obviously common enough that I think clinicians are going to see it regularly. And my concern is that they will begin, over time, to not appreciate the potential concern that the rash may have. So I think it's going to be really important to keep reinforcing the message to clinicians that although it's very common, it can also be very serious.
    Another drug, abacavir, comes to mind. This is a drug we use for HIV therapy that is commonly used. And I think the message there is that there's a hypersensitivity reaction that occurs with every challenge that is extremely serious.
    I think that message has gotten out pretty well. I think that's a good example of a situation where there is a very serious, life-threatening adverse event that could easily be overlooked with the way we manage most drugs in this day and age.
    And I think a concerted effort was made to get that message out, and I think it's worked very well.
    So I think there is an example of how to do this, but I think it's going to take a lot of education of providers as well as patients to make them appreciate that.
    DR. CARGILL: I'm going to take the chair's prerogative and jump in here as well. We also in HIV use a drug called nevirapine, and we all certainly know very well the rash consequence of that drug, which can also be severe.
    So I think that it's going to require that we be vigilant and pay attention to that. But we're not unaware that in and of itself, HIV infection, as an example, can also just de novo present with a severe rash.
    Dr. McGovern?
    DR. MCGOVERN: All the points I was going to make have been made. Thank you.
    DR. CARGILL: Dr. Ghany?
    DR. GHANY: Yes. I agree with everything that's been said by the other speakers. I am
    concerned about the risk of the rash and the anemia.
    But I'd like to just reflect a little bit about when we first started using interferon and ribavirin, hepatologists had to become psychiatrists and hematologists, dealing with depression and anemia. And we learned how to do that, but actually it took a while.
    So my comment is to the sponsor. I'd strongly encourage the -- first I'd like to congratulate the sponsor on what they've done. I think they've done a tremendous job. But I would encourage them to try to disseminate the management of rash and anemia to expedite this so that we don't have to go through a very long learning process before we learn how to deal with this.
    DR. CARGILL: Thank you.

    Ms. Dee?
    MS. DEE: A number of things have already been said. But I feel a lot better about the anemia concerns that I had. It seems like that they are manageable. I didn't see neutropenia
    issues, and that made me feel fairly relieved. And the idea that the FDA confirmed that between 16 and 18 weeks, people start back to baseline was very helpful for me.
    As far as the rash, it's funny. I thought the sponsor did a good job, too, in the assessment and management plan. The discontinuation rates between the second and Phase 3 trials I think is telling and important.
    But I've also had one of these serious Phase 3 rashes -- I mean, stage 3 rashes. And it was interesting to me to see that the rash that I had was not near -- let's put it the other way. Telaprevir rashes are not near as serious by the pictures as the rash that I had.
    So I'm more wondering about that patients be just as educated on not stopping instances where the rash may be manageable. So I think it was helpful to me to see the slides from the sponsor about what kind of educational sorts of things they're going to do to make sure that that message gets out as well.
    DR. CARGILL: Thank you.

    Dr. Ellenberg?
    DR. ELLENBERG: I think that it would be unusual to have an effective treatment for a chronic, life-threatening disease that didn't come with some extent of serious toxicity. I suppose there are some examples, but I think they must be rare.
    I think that what we've heard today are certainly well within what we might expect for an effective treatment for a serious disease. So I think that while they contribute to risk-benefit assessment, they're not risks that overrun the benefits by any means.
    I assume that the sponsor is doing/has done analyses to try and understand which patients might possibly be more at risk for the serious rashes. If there's any way to learn and understand who might be at greater risk, that could help physicians in making treatment decisions and it could also help with patient education in terms of emphasizing what to watch for.

    DR. CARGILL: Thank you.

    Dr. Korman?
    DR. KORMAN: Just the following comments.
    A clinical trial requires education and certification and has a defined protocol. Clinical practice requires only a license. So the answer to that question, do these adverse events affect my risk-benefit assessment, is it depends on how much anxiety is generated by that adverse event. And that anxiety is going to be dependent on my sense of security in managing that adverse event. And the more I understand it, the more I have real algorithms, not generalities, like good skin care. I mean, I don't know what good skin care is other than washing your face a couple of times a day with a non-deodorant soap.
    So I think it's an obligation to really pay a tremendous amount of attention to the educational process, and not just by having some focus groups but by really using educational technology to make sure right up front that the management is as effective as possible so that as many patients who
    need to be treated can be treated. Then my risk-benefit assessment and my anxiety level will go down; maybe not as high as being at this meeting, but you never know.
    [Laughter.]

    DR. CARGILL: Ms. Valbh?
    MS. VALBH: Rash and anemia are of a concern, but I think that the sponsor has done a wonderful job in outlining how it was managed. And I would ask that the FDA, when they do label for this product, if it's approved, that there's tighter guidelines on how rash is managed -- and I'm talking about details, not general guidelines -- and also tighter guidelines on how anemia is managed so that in clinical practice, everybody knows what needs to happen and at what time points.
    I think that the benefit of telaprevir speaks for itself, and I think that it would be a wonderful addition and hope for the patients that are not currently treated or who are previously treated and not responded.
    DR. CARGILL: All right. Ms. Young? I was about to summarize.
    MS. YOUNG: No. I'd just like to thank the FDA and the sponsor for designing some very good studies here that help us to determine the effectiveness and the risks.
    Given the complexity of this course, I would like to measure in the end, it's how many patients completed successfully. So I would like monitoring, some kind of registry in terms of adverse events, maybe even dropout rates and such, so that this regimen can be improved over time.
    Resistance also I think is worth following as we go along so that it won't affect patient treatment options in the future. Other than that, it's a wonderful addition to the regimen. Thank you.
    DR. CARGILL: All right. So if I can summarize, we began our discussion of this question with Dr. Bigby giving us some background numbers in terms of how often some of these skin events that we've been discussing occur.
    So, for example, in the case of Stevens-Johnson syndrome and TEN, or toxic epidermal necrolysis, we heard figures in the range of 1 in every 250,000 to a million in terms of background; in terms of DRESS, which is the eosinophilic and systemic symptom syndrome, again about 1 in about 5- to 10,000.
    However, in the face that we also heard from the committee that there was an expectation that when there is a drug that's going to be coming forward for a life-threatening illness, that there would be side effects, and this would be part of the risk-benefit profile, and that while there is concern about risk, the overwhelming theme of the committee discussion appeared to be that the risk did not outweigh the benefit.
    There were references made to several antiretroviral agents used in the care of patients with human immunodeficiency virus such as abacavir and nevirapine where certainly rash is not only not uncommon but anticipated, and we have learned how to manage.
    I appreciate the comments from our hepatologists who say that they have had to learn how to manage neuropsychiatric disorders, with some tutoring. And I think we heard, repeatedly, messages for strong provider as well as patient education so that this can be accomplished and individuals can know exactly what they're looking for, and that this be detailed guidance as opposed to general guidance.
    We're going to discuss question number 2. And I just want to say that before we discuss this, I just want to give you an overview, as I did yesterday, of our electronic voting system.
    We will be using the new electronic voting system for this meeting. Each voting member has three voting buttons on your microphone, "Yes," "No," and "Abstain." Once we begin the vote, please press the button that corresponds to your vote. You will have approximately 20 seconds to vote.
    After everyone has completed the vote, the vote will be locked in. The vote will then be
    displayed on the screen, and I will read the vote from the screen into the record. Then next we will go around the room and each individual who voted will state their name and vote into the record as well as the reason why they voted as they did.
    So question number 2 states: Considering the overall risks and benefits, do the available data support approval of telaprevir for treatment of treatment-naive and treatment-experienced patients with chronic hepatitis C genotype 1 in combination with pegylated interferon and ribavirin?
    Dr. Friedman.
    DR. FRIEDMAN: So considering where we started with non-A/non-B hepatitis, I think that it's a stunning achievement that we will be able to cure nearly 80 percent of naive patients and probably the same number of relapsers, two-thirds of whom will only have to take treatment for 24 weeks. We've almost completed a transformation of genotype 1 into genotypes 2 and 3, which I think is just a remarkable success story.
    The other important aspect of this drug is that the protocol for using it is relatively simple, and it's familiar because what we're doing is basically using the same milestones we've used for peginterferon and ribavirin and grafting the new protocol onto that. And the drug that is causing us concern in terms of its side effects is only used for 12 weeks. And there's some wiggle room because if you have to stop it a little early, you might not sacrifice success.
    So I think there are so many positive aspects of this drug, and for those of us who have been in the field, this is a very exciting moment.
    Rash is not unfamiliar to people who have taken care of patients with hepatitis C. Of course, there are some cutaneous complications such as cryoglobulinemia. But ribavirin also causes a rash, and occasionally it can be nasty. So I think we actually have a head start in dealing with what will be a principal complication.
    I think that not only do we have to educate ourselves and our patients, but I would enlist the
    dermatology community. I find that to take care of all the complications related to peginterferon and ribavirin really requires a team effort. It's a multidisciplinary effort. And it goes well beyond the conventional subspecialists when you think about issues such as adherence, psychiatric care, drug use, and so on.
    So I would line up the dermatologists to enlist their cooperation in the care of these patients. Many of us have already done that. But I really think this is a stunning success, so I wholeheartedly endorse approval of this. And I will not repeat my comments when it comes time to vote.
    [Laughter.]
    DR. CARGILL: Thank you, Dr. Friedman. I must say when I listen to your comments, I feel like I'm in a time warp, that I came to these meetings with one set of options, and depending on how the day goes, I'm sort of pinching myself and saying, is it really possible that I'm looking at numbers like this, because it's unbelievable.

    Dr. McGovern?
    DR. MCGOVERN: I'd like to comment that I think the sponsor has had a very clear presentation today. I understood the rationale for every single Phase 3 trial. There was a logical progression. The questions that were being posed were answered.
    I think that, all along, they've been very forthcoming on resistance data, the drug-drug interaction data. They didn't present it here, but there was data on coinfection at an HIV meeting, and I was very grateful that they've already begun to address this and very exciting results there.
    In terms of the safety, I commend them on restricting the use of EPO because we understand what the effect of telaprevir is and using dose reductions of ribavirin that we're familiar with.
    In terms of the rash, again I commend them also on trying to dig in and understanding the rash. I just really encourage them to follow some of the recommendations that my colleagues here made, Dr. Roland and Dr. Strader, about having a hotline, having postmarketing monitoring, having
    the patient handout that they're already planning on doing, things like that. I think they've done a tremendous job today.
    DR. CARGILL: Thank you. Some of the things that were raised we'll come back to in the other questions.
    So now we are ready to vote. If there is no further discussion, we will now begin the voting process. I would ask that you please press the button on your microphone that corresponds to your vote. You have approximately 20 seconds to vote. And press the flashing button -- you should have flashing buttons on your screen now -- firmly. After you've made your selection, it will continue to flash. And if you're unsure of your vote, please press the button again.
    [Vote taken.]
    DR. CARGILL: The voting result is as follows: Yes, 18, zero noes, and zero abstentions.
    All right. I will ask you -- we'll begin, starting with Dr. Murata. If you would state your name into the record, your vote, and why you voted
    as you did, please. And we will continue forward through Dr. Friedman and around.

    DR. MURATA: Yoshi Murata. I voted yes. I was favorably impressed with the efficacy data, given the limitations of the existing ribavirin and pegylated interferon alone.
    The safety concerns have been noted by myself and other colleagues on the panel today, but with the appropriate monitoring measures, as has been suggested by the sponsor, and with input from the committee and final decision with the agency, I believe that the risk-benefit profile is favorable for approval.
    DR. FRIEDMAN: Lawrence Friedman. I voted yes.
    DR. BIGBY: Michael Bigby. I voted yes because I think the benefits for outweigh the risks. And I'm pretty sure that there will be an adequate warning about the development of rashes.
    MS. YOUNG: Kathy Young. I voted yes because of the added benefit of telaprevir, and it was proven with the statistics that we have.
    DR. GIORDANO: Tom Giordano. I voted yes. Clearly, the benefits of this treatment outweigh the risks, and the risks are manageable. And I look forward to using this. There are a lot of patients waiting for this.
    DR. VAN DYKE: Russ Van Dyke. I voted yes. I think we've now entered the era of specific antiviral therapy for hepatitis C. I think that's a tremendous advance, with future advances to come. I think along with that come challenges such as, particularly, viral resistance, which we'll learn how to deal with. But I think we're heading down a very important road.
    DR. STRADER: I'm Doris Strader. I voted yes. I would like to congratulate the sponsors on a very well-done and thoughtful presentation. I would like to thank you as a hepatologist for the elegant simplicity of the regimen, taking advantage, as Dr. Friedman said, of already well-known landmarks and making the telaprevir dosing very short so that if there are any adverse events, that they can be reasonably well-managed. Thank
    Ayou again.

    DR. CARGILL: My name is Victoria Cargill, and I voted yes for all the aforementioned reasons. And I specifically want to thank the sponsor for not only an elegant presentation and for clarity, but even if someone has concerns about the number, you offered people who are injecting the opportunity to have access to this, and I thank you.
    DR. CLAY: Patrick Clay, and I voted yes. The benefits far outweigh the risk in this, and it is yet another step. And that's all it is. There's still a long way to go. This is a marathon, not a sprint.
    DR. ELLENBERG: Susan Ellenberg. I voted yes. I think it's a pleasure to see an application where the efficacy is very clear and large and undeniable, and where the risks, while real, seem clearly to be manageable.
    DR. ROLAND: I'm Michelle Roland, and I voted yes. I have some concerns that we'll discuss later about some of the specific populations. But
    several times in the last couple of hours, I've gotten incredibly choked up about the moment that we're in right now for both personal and professional reasons. And I wanted to really acknowledge the community folks who are in the room.
    DR. MCGOVERN: My name is Barbara McGovern, and I voted yes. I think this is a long-awaited landmark day for our patients with hepatitis C. And I want to thank the sponsor for taking on this challenge and also not only for addressing patients with chronic hepatitis C, but also the harder-to-treat patients such as the null responders and coinfected patients as well. Thank you.
    MS. DEE: Lynda Dee. Yes, for all the reasons that everybody's said. You know, there's a discussion often among activists about whether drug companies should ever get As; we do a report card in one of my groups. And I can't forget about the small numbers of blacks and that 11 of illicit drug users. But I won't say that this is an A, but it's very close to it.
    I'm very grateful for such clear data, for such concise rules about how to do this, such manageable toxicities. It was a great, really, really excellent application.

    DR. GHANY: I'm Marc Ghany, and I voted yes, again, not to flog a dead horse, but because of all the reasons that have been said. And I too would like to take the opportunity to congratulate the sponsor on a really well-done and thought-out development program. It's made my job in approving this drug a lot easier.
    MS. VALBH: Pritybala Valbh, and I voted yes, for all the reasons that everybody had mentioned. We're entering a new era, and telaprevir is a huge advancement in treating hepatitis C.
    Thank you to the sponsor for a wonderful presentation. It was extremely clear. You did extensive drug interaction studies. You did not muddy the water with the use of EPO in your trial. And it was very easy for me and convincing that the efficacy was there for the treatment of these
    patients.
    DR. CONNICK: Liz Connick. I voted yes, for the same reasons everybody else stated. The tremendous clinical benefits outweigh the very concrete but manageable risks.
    DR. KORMAN: Louie Korman. I voted yes because this is an important advance. I'm happy to take care of these patients, and I'm even happy to become a part-time dermatologist.
    [Laughter.]
    DR. KNODELL: My name is Robert Knodell. I voted yes. I think that the advent of direct antiviral treatment for hepatitis C will probably do for this disease what cimetidine did for the management of esophageal reflux and peptic ulcer disease.
    DR. CARGILL: Thank you very much, members of the committee.
    We'll now turn to our third question: Please comment on the strength of the evidence to support response-guided therapy with telaprevir in combination with pegylated interferon and ribavirin
    for the following patient groups. And you see there treatment-naive and prior relapsers.
    Dr. Clay? Yes, Dr. Giordano?
    DR. GIORDANO: There's clearly strong evidence for response-guided therapy in treatment-naive patients. The prior relapsers' case, essentially made by the FDA, again I think that was an interesting argument but not necessarily the kind of data that is going to convince me that it's safe to use in patients.
    Fortunately, however, the sponsor had Phase 2 data. And while the sample sizes were small, I think that the hypothesis and the biological plausibility are very strong. And even sample sizes of 100 patients are enough to convince me that there's enough evidence in prior relapsers that response-guided therapy is appropriate in this situation.
    DR. CARGILL: Thank you.
    Dr. Strader?
    DR. STRADER: I agree. I think that in treatment-naive patients, there's adequate evidence
    that response-guided therapy is efficacious and beneficial. What is a little bit of a departure for me is in the prior relapsers. Even though there weren't a lot of patients involved, to me it makes sense that if someone that I've treated before who had a response had complete undetectable HCV RNA at week 4 and then relapsed, that they may do this again, and that response-guided therapy in those patients may be beneficial; and, again, the FDA's very complicated explanation notwithstanding.
    So I think that just in -- I would encourage you to abandon that. You know, it's very confusing. It's very confusing. I think that common-sense explanations probably work better. I understand addition, subtraction, multiplication, division, but still it's very confusing.
    So I think that even though normally I like to see a lot more data involved with respect to prior relapsers, it makes sense to me that those patients would probably be very similar to treatment-naive, and I don't really have a problem concerning response-guided therapy in those
    Aindividuals.
    DR. CARGILL: Thank you.
    Dr. Roland?
    DR. ROLAND: I have concerns about response-guided therapy in the subpopulation of treatment-naive individuals with cirrhosis. And I also have concerns about the sample size for the prior relapsers in the Phase 2 studies, which I think was closer to 52 than 100.
    So not comparing the drug to yesterday but similar to the conversation yesterday, this may be a place where the label needs to be nuanced, where patients and providers together will need to make individualized decisions about what they value more, given the uncertainty.
    DR. CARGILL: Thank you.
    Dr. Knodell?
    DR. KNODELL: Well, I would like to weigh in, in favor of response-guided treatment for prior relapsers. I think that the sponsor has significant positive data that directly relates to this question.
    I did have a little -- I don't see the category up there, but my notes say that cirrhotics should be treated with 48 weeks of treatment whether they have an early or an extended rapid viral response or not. The numbers were pretty small, but at least I did make that -- or jump to that conclusion with regard to that treatment group.
    DR. CARGILL: Thank you.
    Dr. Connick?
    DR. CONNICK: I agree with Dr. Knodell that I think that there are reasonably convincing data that the relapsers can be treated with response-guided therapy.
    I also think, now that it's public information, that some of the data yesterday also support that approach in that I think there were clearer data presented yesterday showing that relapsers behaved similarly to -- when they are retreated to -- and that they can be effectively treated with shorter courses of therapy.
    So, I mean, to subject those people to an
    extra six months of interferon and ribavirin seems a little extreme. And perhaps closer monitoring so there's a larger group post-approval could provide even stronger and clearer data. But I think it's very reasonable.
    DR. CARGILL: Thank you.
    Dr. Ghany?
    DR. GHANY: Yes, thank you. So I also agree with what everyone said already. I think the data is quite strong for the treatment-naive population that response-guided therapy is the way to manage that group.
    I think the data is less strong in prior relapsers. My concern is really for prior relapsers who are difficult to treat, who have poor baseline characteristics, so those who are overweight, high viral load, cirrhotic, diabetic, and so forth. So I'd like to see some more studies on response-guided therapy in that particular group of individuals. But I do think for relapsers who have favorable pretreatment parameters, a shortened course of therapy is likely to yield very high SVR
    rates.
    DR. CARGILL: Thank you.
    Ms. Dee?
    MS. DEE: Most everything has been said. I think that -- not to compare yesterday's drug, but when I was reading both of these briefing documents, it became clear to me that -- I don't know. It demonstrated for me that there was a similar virologic response in relapsers with naives in that they behave in a similar way. So I think that response-guided therapy would be indicated for prior relapsers.
    You know, the nuances about whether or not you have -- what your profile is, whether you're older, all the other considerations, might be something that the agency might want to put in the label.
    DR. CARGILL: Thank you.
    Dr. Ellenberg?
    DR. ELLENBERG: I think the data are pretty clear for the treatment-naive population. For the prior relapsers, my intuition, looking at the data,
    A Matter of Record
    (301) 890-4188
    320
    is similar to what most people have said, that it's probably going to be okay. However, I think the data are really pretty sparse. I think we're drawing from several different sort of smallish data sets that are all kind of suggestive.
    I feel like there's maybe a greater than 50 percent chance that prior relapsers can do fine with RGT, but I'm not sure it gets to the level that I would say the FDA should bless the RGT as clearly the preferred regimen in this group. I just do not think the data are strong enough.
    I don't see data that suggest that it's not going to work. But I think more data would be necessary before it can be documented that this is the optimal regimen. And I would hope that this can be an individualized decision.
    DR. CARGILL: Thank you.
    Dr. Korman?
    DR. KORMAN: I think the data are compelling for treatment-naive, and they're pretty compelling for prior relapsers. It would be a lot easier to convince a patient that's gone through 48 weeks of
    A Matter of Record
    (301) 890-4188
    321
    therapy and relapsed to undertake therapy for 24 weeks. And I think the decision can be individualized depending on the underlying severity of the disease in that patient. More advanced disease may give you the option of treating for a bit longer.
    I also like the FDA concept. I know you didn't want to use the word "model." I like the idea of past performance as a predictor of future performance, even though on all my investments, there's a disclaimer.
    [Laughter.]
    DR. CARGILL: Dr. Friedman?
    DR. FRIEDMAN: I'm convinced.
    DR. CARGILL: I'm stunned, Dr. Friedman, almost rendered speechless.
    [Laughter.]
    DR. CARGILL: Well, to summarize this question, I think that we heard a ringing endorsement support for using response-guided therapy for treatment-naive patients. I think the committee was very clear.
    A Matter of Record
    (301) 890-4188
    322
    It's less clear that we have that same sort of strong endorsement. I think the majority opinion on the committee appears to be that there is evidence to support that for prior relapsers. But this is where the nuances begin. We're hearing concerns about patients who have cirrhosis. We're also hearing concerns about patients who may have other underlying characteristics that may make them have markers of concern such as increased body weight, higher viral loads, older age, and that the option of being able to have some nuance or some individualization of the therapy may be the most helpful.
    I think we also heard some consistent discussion about it would be nice to have stronger studies and more data, that the data may not be as robust as we would like for this particular subset of the question.
    We're moving on to question number 4: Please comment on the strength of the evidence to support a recommendation for use in specific populations, including but not limited to blacks, African Americans, and patients with cirrhosis. What if any additional efficacy or safety data are needed for specific populations?

    Dr. Clay?
    DR. CLAY: So there's two. One would be patients with gout. The other would be people with tuberculosis. The changes in mean uric acid levels in people taking telaprevir were 2.5 as opposed to 0.6 in the placebo groups. I think they need to conduct studies that are going to help us better understand if this drug can be used in people who are even being treated right now for gout because there are no longer drugs available on the market for the acute outbreak of gout.
    The second group is -- and, I know, I'm driving people crazy with this -- is this tuberculosis issue I have because of the minor levels of pyrazinoic acid that you detected.
    The beauty of this drug, which makes it a very eloquent drug, is that there are multiple ways for the body to break this down. One way is through the cytochrome P450 pathway. The other way
    is through other pathways. When they did their pharmacokinetic studies and administered this drug, with drugs that block that cytochrome P450 pathway, they did not get appreciable or really significant increases in telaprevir, and that's great. That's because the drug was then shunted through the other pathways. It was the other pathway that then caused the metabolite, pyrazinoic acid, to be prevalent or be present.
    I guess they may have already done this and we simply didn't have access to this data, but I would like to know, in the presence of cytochrome P450 inhibitors, when the pathway is shunted toward increased development of pyrazinoic acid, that there is not sufficient levels of pyrazinoic acid present such that if a person may have an untreated case of tuberculosis, that tuberculosis is not being exposed to subtherapeutic and therefore resistance-inducing levels of pyrazinoic acid.
    Thank you.
    DR. CARGILL: Thank you.

    Dr. Roland?
    DR. ROLAND: While the numbers are small for both blacks and people with cirrhosis, I think that the data are definitely adequate to suggest that there's efficacy. But I do need to really stress that I think with the previous question, that lumping all treatment-naive subgroups together and saying that response-guided therapy, that the evidence supports that, it just doesn't for the people with cirrhosis. So I feel like I need to say that again.
    DR. CARGILL: Dr. Friedman?
    DR. FRIEDMAN: So here the evidence isn't quite as strong as we would like it to be, or not strong at all in some cases. So we really do need more data. And specifically, I think we need to know what role IL28B testing in African Americans will play in the decision-making and in the use of response-guided therapy.
    I have a particular concern about the cirrhotics who have been null responders to peginterferon and ribavirin. There we really don't have any significant data, and what little we have
    suggests that there was no difference when they were randomized.
    One of my colleagues was quoted in the Boston Globe this morning saying that we had warehoused those patients. I think what he meant was they're waiting in the wings for the next generation of therapy. They're sort of on a wait list. I said yesterday, we have a reservoir of these patients, and we really need to know if the null responder cirrhotics are going to response to retreatment with a three-drug regimen.
    DR. CARGILL: Thank you. And thank you to the panel.
    I think what we heard in response to this question -- oh, sorry?
    Oh, okay.
    DR. CONNICK: I brought this up earlier but wanted to mention again, people over 65 in both the naive and the experienced studies did not demonstrate great benefit. And I think that's an important question to be addressed in future studies as well, whether those populations really
    do or don't benefit.
    DR. CARGILL: Dr. Strader?
    DR. STRADER: I just wanted to mention this again. I know this was brought up earlier about the number of African American patients in studies up front. I realize that when we're talking about marketing studies, we want to get international involvement, and so that may limit the number of African American patients that we see in the beginning. Then we end up approving a drug for the majority population, but then asking the minority population, which makes up a very large number of patients with hepatitis C, to wait until we can go back and look at them.
    I think it would be a good idea to try to do that up front, especially to try to characterize patients in the U.S. who may make up a large portion of the population that we're looking at. So I just wanted to make that plug again, that we try to encourage sponsors to consider including minority patients up front as opposed to considering them an afterthought after the drug has
    been approved.
    DR. CARGILL: Thank you.
    Dr. Giordano?
    DR. GIORDANO: I'd like to second Dr. Strader's comments. And to go back to the previous question, just for the record, I did overstate the number in the Phase 2 studies. It was actually 67 participants in the Phase 2 studies where response-guided therapy was tried in prior relapsers. Of those 67, 52 were eligible by their early virologic response for short therapy, and 49 of the 52 were successfully treated with a short course of therapy. So just for the record.
    DR. CARGILL: Thank you for that clarification.
    Ms. Young?
    MS. YOUNG: Yes. I'd like to second the suggestion for more studies in terms of the cirrhotic patients as well as the 65 and over and those who perhaps could be rescued from transplant, you know, further deterioration. I think that those would be good groups to study.

    DR. CARGILL: Ms. Dee?
    MS. DEE: I'm getting older by the day, so 65 is closer. And I think that it's important that -- I think Medicaid doesn't cover people over 65 for this. So I think that's an important consideration.
    I agree with what Dr. Strader said. Obviously, I've said that numerous times. But I'm just really not convinced that there's anything but trends in the cirrhotics and in the people of color in these trials because I'm not sure there's just enough patients to say much of anything other than it looks like this or it looks like that.
    DR. CARGILL: We'll try and summarize number 4 again. I think the FDA has heard fairly clearly from the committee several points, including needing information for several types of special populations; individuals with gout, individuals with tuberculosis. And I think that Dr. Clay gave a fairly eloquent description of why, that basically if you shunt into the pathway because you've blocked off cytochrome P450, will
    this be a source of concern because of the change in levels of pyrazinoic acid?

    Certainly the question for efficacy for African Americans and blacks, as well as cirrhotics, and we've heard multiple times about the need for inclusion of these populations given their over-representation in hepatitis C populations.
    Similarly, the concern about null responding cirrhotics and what do the data say for them and can we do better, as well as the individuals over 65. And I would just also like to add that as we looked at those confidence intervals, I would remind you that 65 was one of the ones that crossed over zero.
    Number 5. I can hear the committee sharpening their pencils since they really dug into this yesterday. So I'm on it, guys.
    [Laughter.]

    DR. CARGILL: Are there any other postmarketing studies you would like to see conducted to further define risks or optimal use of
    telaprevir?

    Dr. Van Dyke?
    DR. VAN DYKE: I think a critical question is the implication of resistance. I think this becomes a huge issue when you only have a limited number of drugs, and it informs -- I think the question is about retreatment, the efficacy of retreatment, or cross-resistance to other medications, is really important.
    I think it informs us in terms of who we should attempt to treat because if you're going to attempt to treat someone with a relatively low likelihood of success and therefore a relatively high likelihood of developing resistance, you need to know what the implications of that resistance is. Does it prevent any additional therapy, or does resistance actually fade and you might have an additional chance?
    So I think this is one of the new challenges we have to face in this disease.
    DR. CARGILL: Ms. Valbh?
    MS. VALBH: I too would like to see some
    more detailed studies on resistance. And then, as I said yesterday, I would like to see studies on the nonresponder population for other genotypes other than 1. I think that this product may have a place in therapy for other genotypes as well.
    DR. CARGILL: Thank you.
    Dr. Roland?
    DR. ROLAND: Also similar to the comments yesterday, I think it's going to be very important for us to understand the effectiveness of this product in the real world, particularly looking at adherence and clinical management of rash and anemia.
    DR. CARGILL: Dr. Clay?
    DR. CLAY: I'm just going to kind of look over here because they may have already done it and I didn't see it. Grapefruit juice interaction studies because you're a PGP drug? Yes, those have been done?
    DR. KAUFFMAN: Grapefruit juice?
    DR. CLAY: Grapefruit juice; your p-glycoprotein -- because in your information that
    you submitted, you talked about how your PGP was primarily in the gut and not systemically.
    So I'm curious if you've looked at grapefruit juice impact of this or other uptake inhibitor studies within the gut. And if not, I would certainly recommend that to be a postmarketing. But if you've done it, that's great, and that's it. Thank you.
    DR. CARGILL: Thank you.

    Dr. Ghany?
    DR. GHANY: Yes. The sponsor actually has a number of studies underway in populations that I think definitely need to be addressed. So, in addition to that, I'd also like to reiterate the point that Dr. Van Dyke raised, that we need to be careful about the development of resistance.
    I would like to make sure that we continue -- that the sponsor continues to collect data on the evolution of resistance and the clinical implications, particularly, as Dr. Friedman raised, the null responders. And I think in this group of patients we clearly need to
    individualize treatment.
    In addition to that, I'd like to see other strategies developed to manage anemia and also understand the development of a rash with these agents because we don't have a clear treatment plan for dealing with -- specific treatment plan for dealing with rash.
    DR. CARGILL: Thank you.

    Ms. Dee?
    MS. DEE: Okay. So we didn't say anything today about people with bleeding disorders, so I'd like to see some work in that arena.
    As far as methadone, I know the study was done. I know also in HIV and some of the drugs, we've done work that increases the dose of methadone so that people are able to use it with HIV protease inhibitors. So I wonder if that's something that is possible with this drug.
    Also, the HIV protease inhibitors, now, I know that we saw data at CROI that talked about reactions. But I think that the appropriate doses of HIV protease inhibitors with telaprevir needs to
    be established. I mean, we know that we can use it with efavirenz and atazanavir with a ritonavir boost, but that's pretty limiting for people with HIV. And I think it's very important to that community to know a little bit more about how this drug works with HIV protease inhibitors and whether there can be a dose adjustment that would make it possible to use them.
    Oh, and again -- I'm so sick of hearing my own self -- great sorts of studies for more information about cirrhotics and people of color.
    DR. CARGILL: Dr. McGovern?
    DR. MCGOVERN: I like the lineup of studies that the sponsor has listed for us earlier today. I'd like to echo the importance of studying African Americans further. I would like to see like an interaction of telaprevir with IL28, trying to see if we can use IL28 to help us decide who can get shorter therapies versus longer therapies.
    I know that they are planning the BID versus TID studies. I know there's some published data by Marcellin on this. But I would also like, if they
    are in the process of doing another larger study of that, if they could couple it with adherence since we know in HIV the smaller number of doses a day correlates with better outcomes. Thanks.

    DR. CARGILL: Dr. Friedman?
    DR. FRIEDMAN: So in addition to what's been mentioned, one thing that concerned me is the oral contraceptive issue. And I think we need to learn more about that and whether that's going to be a problem.
    Another thing we haven't talked about -- and I don't know how important it's going to be; it may not be, but there are different formulations of peginterferon, and I think different formulations were used in different trials for different drugs. And they're associated with different dosing of ribavirin.
    So I don't know if any of that's important, but at some point that probably ought to be addressed.
    DR. CARGILL: Ms. Young?
    MS. YOUNG: I think it would be helpful to
    have some monitoring of the prior relapsers to perhaps come up with some protocols that will help risk stratification and benefit stratification for that group.
    DR. CARGILL: Come on. You've got 20 more studies to come up with.
    Dr. Korman, I knew you wouldn't fail me.
    DR. KORMAN: Yes. I can always come up with something.
    It might be useful to actually do a study of the educational benefit of some of these interventions to look at compliance and conformance and sustained virologic response to see if some of the approaches that are being recommended actually make a difference in terms of educating the community, because, again, you require education of clinical investigators to participate in a study. You can't require education of the clinical practitioner.
    So some of these questions would be useful. They're not, strictly speaking, medical trials. But these are important in, really, effectiveness
    in the community.

    DR. CARGILL: Dr. Friedman?
    DR. FRIEDMAN: Now, you see, Dr. Korman is a gastroenterologist. And I was sure he was going to want to study the anorectal complications of telaprevir –
    [Laughter.]
    DR. FRIEDMAN: -- because he does 20 colonoscopies --
    DR. KORMAN: I'm becoming a dermatologist, remember.
    [Laughter.]
    DR. CARGILL: Okay. I think, on that note, we'll summarize. I can see it's degenerating here.
    Well, it wasn't quite as robust a list as earlier. But I think that we still have a number of things for our sponsors to consider. So, again, I will say, as I said yesterday, I think you all are going to be in business for a long time, looking at this.
    Studying the implications of resistance, we heard that from several panel members it's going to
    be a big issue, particularly cross-resistance to other medications. I would add to that class-wide resistance that appears to be conferred by some of these agents, and also the question of retreating those who have had failures; especially who do we treat, and those who have a low likelihood of a sustained virologic response, but then may be trading that off for a higher likelihood of resistance, what does this mean? Studies on nonresponders to peg/ribavirin for other genotypes besides genotype 1.
    In addition, we heard some concerns about what is going to be the effectiveness in the real world, especially when we have to look at how individuals will be asked to manage anemia and rash, which is a known side effect. Dr. Clay raised the issue of the impact of grapefruit juice and whether or not those studies have been done.
    Again, more concerns about the need for more participation in studies of those who have been underrepresented and especially take a look at studies of those who have bleeding diatheses.

    Certainly trying to look at the appropriate dose of telaprevir with a number of different HIV drugs, and the interaction and impact of IL28B, and will this be helpful and predictive?
    The sponsor already has studies ongoing and is going to be doing more in several of these directions, but the BID versus TID dosing and an opportunity to look at adherence at that point, and does that make a difference, particularly given the experience we've had with HIV therapies.
    The oral contraceptive issue has been raised before. But, then again, how does this play out in real time and what is this going to mean in terms of adherence? And I would also say in terms of how our patients respond or want to be engaged in this therapy. Monitoring of prior relapse and risk-benefit in those individuals who have relapsed.
    Finally, taking a look at the educational benefit of interventions to try and manage some of the side effects, such as rash and anemia, and certainly also looking at the anorectal complications.

    So I think at this point I'll give it back to Dr. Birnkrant for last words from the FDA.
    DR. BIRNKRANT: Well, thank you very much. Well, given that we have such an esteemed panel, we have actually one more question before you go. Don't worry, you don't have to vote. It's okay.
    But thinking about phrases we've heard based on the data we've seen over the last two days, that this a game-changer, a paradigm shift, and a new era all rolled into these drugs added to standard of care, can you comment on the impact of these drugs on current and future clinical trials with regard to standard of care or control arms?

    DR. CARGILL: Dr. Clay?
    DR. CLAY: I think currently, until these -- if these drugs get approved, of course, they become part of the treatment guidelines. And I think what these drugs reflect is another arm, a subgroup of a study that an individual could be randomized to. As newer drugs are moving forward, if the FDA is going to continue to mandate comparison to pegylated interferon and ribavirin,
    then that is your control as of now. But moving forward, you add these newer agents in as comparator groups.
    DR. CARGILL: Dr. Roland?
    DR. ROLAND: What a great question. I think it depends on what the question is that you're looking at. But if you're looking at efficacy questions, I think this is going to have to be the new standard of care in a new noninferiority design.
    If you're looking at a reduction in adverse events, then you're also probably looking at these being the comparator arm, whether it's a noninferiority design or a superiority design. Those are very important questions because otherwise we get presented with data that are meaningless to us, really.

    DR. CARGILL: Dr. Strader?
    DR. STRADER: I agree this is a very good question. And part of it I think may depend on IL28B information because if it turns out that if you're a genotype 1 person and you have a CC
    genotype, it doesn't matter whether you get two or three drugs. That changes things just a little bit.
    But I do think that the three drugs are the sort of new standard of care. I have to admit myself I was hoping that we could eliminate pegylated interferon and ribavirin as part of the lexicon because it is a difficult treatment and it is wrought with a bunch of side effects. But it appears that as of April 28, 2011, it's still part of the treatment lexicon, and these three drugs may be the standard of care.

    DR. CARGILL: Ms. Dee?
    MS. DEE: I think it's probably no longer ethical to have a PR-only arm. I think that Dr. Roland kind of alluded to that. I also think it's time for us to look at double DAA arms. I mean, I think the community's dying for that. I think we've learned a lot from HIV. We've seen in that arena that that's the way to go. And I think it's obviously going to be the way to go with this, yet another virus.

    I think that it would also be important -- and I'm not sure, maybe early access. I think we scared the investigators with that term. I think they are looking at that as an HIV thing. And I think that, maybe not in pivotal trials but in conjunction with your pivotal trials, it may be a good idea to try and look at underrepresented populations or the cirrhotics and, you know, the cases that need early access, but in the format of a clinical trial as opposed to just giving it out like we did in HIV.
    So I don't know if that makes sense, but I think we're going to need to hash out how we can get these drugs to people earlier but also do it in a safe way.

    DR. CARGILL: Dr. Camardo?
    DR. CAMARDO: Finally, there's an industry-wide question I can answer or at least have an opinion on. But this is really critical because we are always faced with what is the standard of care. And we in general -- and I see that this committee -- are always much more comfortable with
    superiority efficacy studies than we are with noninferiority kinds of studies, and we seem to be becoming less comfortable with those.
    So it's very hard for a sponsor to go into a program and not know that when they come to this committee, that they're going to actually have a reference that people accept, and that is really critically important. I don't have an answer for it. I just know it's a big question for us now.
    Five years ago, six years ago, everybody knew what to compare the new therapies to. Tomorrow it's going to be a problem. And sometimes you have to make decisions in advance about what that's going to be and how you're going to evaluate it. But I don't feel like noninferiority is going to be everybody's favorite way to evaluate unless we can come up with some acceptable guidelines.

    So it is a really important question, so I'm glad you asked that.
    DR. CARGILL: Dr. Giordano?
    DR. GIORDANO: I'm not sure we're ready for noninferiority yet. I think there is a lot of room
    to improve in side effect profiles, and so there are certainly opportunities there.
    To me, one of these approaches is the new standard of care as soon as they're FDA-approved and available. I agree. I think it would be unethical to devise a control arm that didn't include a protease inhibitor.
    I also think the next thing that's going to change management is what Dr. Strader said, the IL28. We have to factor that in. And in that case, maybe you could have control arms that would spare a protease inhibitor. But, in general, yes, I think this is the new standard of care and has to be -- any trials have to take that into account.
    DR. CARGILL: Thank you.

    Dr. McGovern?
    DR. MCGOVERN: I think most of the things I was going to say have already been covered. Yes, I agree that this will be the new standard of care. I agree that we have room to go in terms of side effects. There is also going to be probably different dosing intervals that will be offered by

    other drugs. I agree that prospectively we should be looking at IL28 and how it factors in. So I'm just summarizing all those good ideas.
    DR. CARGILL: Dr. Friedman?
    DR. FRIEDMAN: So I agree. Once approved, three-drug therapy will become the standard of care. It's going to complicate how we do trials going forward. First of all, the regimens are more complicated than they used to be.
    Secondly, there's this group of HCV patients, 75 percent, that have not been identified that we're going to need to identify to get them into trials if we want to keep studying naive patients. Otherwise, we're going to enrich our studies with resistant patients, and they're going to become increasingly resistant to multiple agents. So I think it's critical that we have public health efforts to identify patients with hepatitis C.
    I think the ultimate goal is to develop a regimen that doesn't include peginterferon. As problematic as the side effects are with the newer
    agents, at the heart of the problem of treating hepatitis C patients is that interferon is such a tough drug to have to take. So I think, at least as an intermediate goal, that ought to be a priority.

    DR. CARGILL: Dr. Clay?
    DR. CLAY: So let me propose this to you. When you begin your trials, whatever they look like with the new drug, a person doesn't qualify to participate in the trial that's being assessed and that is going to be submitted for safety and efficacy because of prior treatment, but they have no other options left. You allow that individual to take part in maybe not expanded access, but a compassionate use protocol.
    The benefit to the sponsor is you cannot hold them accountable for safety because, really, they're going to have such a confusing noise in their dataset that you'll never really know if the drug was doing it. But you let them take credit for the efficacy because that patient had nothing left. And you allow them the experience, just like
    they do in oncology. "You know what? Let's go ahead and try it." So you give them credit for the efficacy, but you don't penalize them for the safety.

    DR. CARGILL: Dr. Ghany?
    DR. GHANY: Actually, most of what I was going to say has been said. But I wanted to raise two other points, and that is, some of the newer agents are going to be active against other genotypes. I don't know how you're going to factor that in because these drugs are really only -- these two drugs that we talked about yesterday and today are only effective against genotype 1. So I think that's something that the FDA is going to have to struggle with again.
    I also wanted to comment about the role of IL28B. I think, with the newer drugs that are more potent, the issue of interferon responsiveness is going to become really less of an issue. And I don't think IL28B is going to factor into future treatments once we have potent combinations of DAAs.
    DR. CARGILL: Thank you.

    Dr. Roland?
    DR. ROLAND: So your question raises another question in my mind, and this isn't my area of expertise, so I have a question to help me understand this. But are there other ongoing studies where pegylated interferon and ribavirin are currently the control arm? Because if there are, then you've got some serious ethical issues to deal with, with ongoing studies right now. Sorry.
    DR. BIRNKRANT: We can end the meeting now.
    [Laughter.]
    DR. BIRNKRANT: Well, this is why I brought it up because we're facing this with recent proposals and future proposals. But the future is here. It's not really that far away, and we have to make some tough decisions. And that's why I wanted to get some input, so that when we go back to companies, we can state the obvious but then have the backing of this committee.

    DR. CARGILL: Dr. Knodell?
    DR. KNODELL: I have something that hasn't
    been said that may be of some help. I think it depends on the mechanism of the drug that you're looking at. If you're looking at another protease inhibitor, then I think your comparison group has to be against this triple drug regimen.
    Now, there's a fair amount of side effects associated with the protease inhibitors. If something else came along -- we'll say a polymerase inhibitor -- and had a much better safety profile, then a noninferiority trial with a lot less side effects might be reasonable.
    So I would urge you, when you're thinking about this, to see whether or not -- what the mechanism. If you've got a different mechanistic drug, then a noninferiority trial may be reasonable. But otherwise, probably not.
    DR. CARGILL: And the last comment will go to Dr. Murata.
    DR. MURATA: I will just make a conceptual comment, in that I fully understand from the hepatologists in the panel, such as Dr. Strader and Dr. Friedman, about their comments on eliminating,
    down the road in terms of interferon and ribavirin, combination therapy as at least a baseline arm.
    Down the road, in my opinion, a critical study would be at what stage do you actually eliminate that arm in terms of substituting one or more new agents for the interferon and ribavirin arm? So this is essentially a conceptual wording of what Dr. Knodell explicitly said in terms of mechanism of action.

    DR. CARGILL: Dr. Birnkrant?
    DR. BIRNKRANT: Well, I appreciate everyone's responses. And they are basically aligned with what we were thinking, but we just wanted to get some feedback because it is an extremely important situation that we're all facing at this point in time.
    I want to thank everyone for this exciting two-day meeting. I haven't been this excited at work in a long time.
    [Laughter.]
    DR. BIRNKRANT: Again, thank you very much. I also want to acknowledge and thank the open
    public speakers. We heard you quite clearly. Thank you very much.
    Adjournment
    DR. CARGILL: Thank you. The meeting is adjourned, and we can all let out a collective "Woo-hoo!"
    (Whereupon, at 4:02 p.m., the meeting was adjourned.)


    Boceprevir


  • Transcript for the April 27, 2011 Meeting of the Antiviral Drugs Advisory Committee8  (PDF - 671KB)




  • FOOD AND DRUG ADMINISTRATION
    CENTER FOR DRUG EVALUATION AND RESEARCH (CDER)
    ANTIVIRAL DRUGS ADVISORY COMMITTEE (ADVAC)
    Wednesday, April 27, 2011


    Today's agenda involves the discussion of new drug application, NDA 202-258, boceprevir, manufactured by Merck, with a proposed indication for the treatment of chronic hepatitis C genotype 1 infection, in combination with peginterferon alfa and ribavirin, two medications approved for treatment of hepatitis C infection, in adult patients with compensated liver disease who are previously untreated or who have failed previous therapy. This is a particular matters meeting during which specific matters related to Merck's boceprevir will be discussed.

    DR. CARGILL: We will now proceed with the FDA opening remarks from Dr. Debra Birnkrant. I would like to remind public observers at this meeting that while this meeting is open for public
    observation, public attendees may not participate except at the specific request of the panel.

    Introduction/Background
    DR. BIRNKRANT: Good morning. I would like to welcome everyone to our two-day advisory committee meeting on direct-acting antivirals for the treatment of chronic hepatitis C. This is a very exciting time because many new therapies are being developed for this debilitating disease. Over these two days, you will hear about two new treatments for chronic hepatitis C that are members of the protease inhibitor class and were studied in combination with pegylated interferon and ribavirin.
    Why do we need new therapies and new treatment strategies? Because there is a significant disease burden, and current therapy has its limitations. Let's start with the disease burden.
    Chronic hepatitis C is a global problem, and it's estimated that up to 180 million are infected worldwide. Chronic hepatitis C is also a domestic
    problem and upwards of four million of the U.S. population is chronically infected. Notably, of the 5.6 million veterans in Veterans' Health Administration care in 2008, 2.6 percent had a diagnosis of chronic hepatitis C.

    Importantly, the incidence of infection in the United States is decreasing, but, unfortunately, chronic hepatitis C-related complications are increasing, such as cirrhosis and hepatocellular carcinoma, with the aging of the infected population. So we expect more liver-related complications in the next 10 to 20 years. And I'm sure the audience is already familiar with the fact that chronic hepatitis C is the most common reason for liver transplantation.
    To set the stage for the next two days, I'd like to introduce you to some definitions, starting with the patient populations and how we described them in the trials.
    We have the naive population, who received no prior therapy for hepatitis C; the null responder, who achieved less than a 2 log reduction
    in hepatitis C RNA at week 12 of combination therapy with pegylated interferon and ribavirin; partial responders, who had a greater than or equal to 2 log drop in HCV RNA at week 12, but did not achieve HCV RNA undetectable at the end of treatment with the pegylated interferon/ribavirin regimen; and, relapsers, where you have hepatitis C RNA undetectable at the end of treatment, following treatment with the pegylated interferon-based regimen, but HCV RNA becomes detectable within 24 weeks of treatment follow-up.

    These definitions are described in our recently published draft guidance for drug development for hepatitis C products, and they are aligned with professional societies' guidance, as well.
    Let's turn to some definitions that you'll be hearing over the next two days. Rapid virologic response, or RVR, is undetectable HCV RNA at week 4. Why is this an important point? RVR is highly predictive of achieving a sustained virologic response.

    Extended RVR, or ERVR, undetectable HCV RNA at weeks 4 and 12. What is sustained virologic response or SVR-24? It's a validated endpoint used in clinical trials and in practice. It's defined as an absence of detectable RNA in serum six months after completing therapy. It's the best indicator of successful therapy of chronic hepatitis C, whose goals are to have fewer liver-related complications, less progression to hepatocellular carcinoma, and fewer deaths.
    What's standard of care? Currently, the standard of care for treatment of chronic hepatitis C in naive subjects is pegylated interferon with ribavirin. In general, the treatment duration is 48 weeks for genotype 1 and 24 weeks for genotypes 2 and 3. On average, response rates are approximately 50 percent, with a wide range of 20 to 80 percent. A response rate of approximately 50 percent means 50 percent don't respond.
    Why is there such a range? It depends on multiple factors, some of which are genotype, IL28B status, race, gender, viral load, et cetera.
    I wanted to remind you at this point, as we begin our discussions for the next two days, that already with the approved regimens, there are significant toxicities seen.

    What is response-guided therapy?

    It's a treatment algorithm, individualizing treatment based on neurologic response. Why is it important? Because the goals are to shorten therapy, if possible, in those who exhibit favorable viral kinetics and to identify subjects who are unlikely to have a response, thereby limiting side effects and cost.

    I'd like to now introduce a concept that you'll be hearing about today and tomorrow that was developed by our pharmacometrician group. It has helped us determined various durations of treatment for different populations.
    It's based on a premise that the treatment-naive population already contains subpopulations of each possible PR responder group. Data for how treatment-experienced patients may respond are
    within data from treatment-naive patients.
    Our pharmacometrician group found that prior non-responders demonstrate similar virologic response at week 4 of initial or subsequent PR treatment. The analyses that we will present support that early virologic response may be more important than previous exposure to pegylated interferon and ribavirin in determining outcomes.
    I showed this slide back in 2006 when we had a two-day advisory committee meeting on drug development for hepatitis C products. What was true then is true now. The pipeline is still full, and over these two days, we'll be hearing about the two products who are the first from that pipeline to emerge.
    Lastly, I'd like to end with this slide. These are milestones in the therapy of chronic hepatitis C, and I'm sure the majority of you have seen this slide in one format or another because it comes from a compilation of various authors and investigators' data.
    What we have here are the approved
    medications for treatment of chronic hepatitis C over a timeline from the early '90s to the future. On the Y-axis, we have sustained virologic response. As you can see, in the early '90s, with interferon either 6 months or 12 months, response rates were quite low. As we added ribavirin, the response rates increased to 34 percent, on average. When we added pegylated ribavirin, response rates jumped to approximately 55 percent.
    The future does look quite promising and now it's possible that we have drugs that can increase the SVR rates to more than 30 percent above what standard of care yields today.
    Before I close, I would like to thank our FDA review teams who worked tirelessly on these complex priority applications. I would also like to thank the companies for developing these new products and conducting the clinical trials. And I would like to thank the patients who enrolled in these studies for their contributions to finding better therapies for the treatment of chronic hepatitis C. Thank you very much.

    DR. CARGILL: We will now continue with FDA remarks from Dr. Jeff Murray.
    DR. MURRAY: I'm going to continue with some opening remarks focusing on today's topic, which is for consideration of a new hepatitis C virus protease inhibitor called boceprevir, proposed trade name Victrelis, sponsored by Merck & Company.
    So this will suffice as some opening remarks on today's topics but also cover the charge to the committee, which usually occurs later in the agenda. When you get to that part, you can just cross it off now, and we'll be ahead.
    So I'm going to briefly summarize the agenda and questions, and I want to summarize the committee's role in these proceedings. Then I'm going to offer some remarks about balancing risk and benefit, which is, in essence, the primary mission of this committee today.
    I will briefly summarize the clinical benefit of SVR, as Debra kind of already did, and then speak to the need to consider risk-benefit for each of the issues outlined in the committee
    questions. I'll then briefly remark on considerations for requesting post-marketing trials and studies.

    So after my remarks today -- this is a brief summary of the agenda -- there will be presentations by Merck, time for some clarifying questions, presentations from FDA. We'll have lunch. There's a open public hearing. People have already signed up. It lasts about an hour. Then we'll have more questions from the committee to the presenters, and then the committee will be asked to address the questions that FDA has posed.
    A shorthand version of these question, number one, we're going to ask the committee to comment on safety. The second question will be an overall risk-benefit decision and vote yes or no for marketing approval. The third question involves including null responders in the indication. The fourth question is multipart and addresses response-guided therapy, the best duration for various subgroups, including treatment-naive late responders, black patients,

    patients with fibrosis/cirrhosis. The last question will be what types of post-marketing studies/trials should be done.
    So the Antiviral Drugs Advisory Committee has members and consultants of various backgrounds, and, by and large, our committee members were chosen based on their expertise to address our most common products, which are HIV and hepatitis C drugs currently. So the committee now consists of a union of hepatologists and infectious disease experts, particularly experts with HIV background.
    I'm going to say that lessons learned from HIV may be helpful and offer some grounding in decision-making. However, I caution us all not to over-generalize from one virus to another. As an example, let us consider the term, again, "treatment-naive." I bring this example up because this issue will be a recurring theme for both days of this meeting.
    When we think of the term "treatment-naive" for HIV and antiretrovirals, we think of a population that is largely homogenous and consists
    mostly of patients harboring wild type virus in which the vast majority is expected to respond. But treatment-naive for hepatitis C and interferon and previous interferon treatment is quite different. The majority of HCV naive patients are expected not to respond. And this population has heterogeneity, and this is an important concept. Treatment-naive patients consist of people who are already intrinsically responders, relapsers, partial responders, and null responders, with various degrees of interferon responsiveness.

    So after the committee discusses the safety of boceprevir in question 1, they'll be asked to weigh the overall risk-benefit of adding a direct-acting antiviral, in this case, boceprevir, to peginterferon and ribavirin. And you'll have to consider gains in SVR and weigh that against the addition of toxicity to peginterferon and ribavirin.
    To help with the benefit side of the equation, I'm providing some background on the use of SVR as a primary endpoint. In short, FDA views
    sustained virologic response as not only a virologic measure but also as a clinically validated endpoint, and this is stated in our draft guidance that was released last year.

    In a great review by Pearlman and Traub, recently published in Clinical Infectious Diseases, they summarize multiple studies which support this. There have been multiple -- in fact, 19 cohort studies comparing the outcome between SVR and non-responders which showed a variety of favorable outcomes for those who had achieved SVR compared to those who did not, including decreases in decompensated liver disease, decreases in hepatocellular carcinoma, decreases in mortality, and even reductions in things like diabetes and hyperglycemia.

    The largest, most recent cohort was the VA cohort in over 16,000 HCV infected patients with high rates of co-morbidities, including smoking and diabetes. All were treated with a peginterferon/ribavirin regimen. Over 7,000 patients had SVR, and, notably, there was a
    significant and a substantial reduction in overall mortality with an adjusted hazard ratio of .67 or 33 percent reduction in overall mortality over mean follow-up for four years.So I think this cohort and others tells us that treatment of hepatitis C and achieving SVR translates into substantial clinical benefit.
    When drugs are in development, FDA has to balance information, getting complete information with availability. There is a tradeoff between getting a drug to the market sooner, particularly one for treatment of a serious illness and a tradeoff of having complete information of every patient subgroup or potential drug interaction that one might want to investigate. In fact, some have suggested that the DAAs coming before the committee these two days should have been made available sooner after Phase 2.
    We thought a Phase 3 program was necessary, and that's what we're here to discuss today. I think what we've discovered is that a good Phase 3 trial answers a lot of questions, but it also
    generates a lot of questions, and I think that's the case here today given the complexity of hepatitis C therapy.
    In these Phase 3 trials, we see that there are many baseline and on-treatment factors that play a role in the primary outcome and for selecting the optimal duration of therapy. These include previous response, null, partial relapse, race, presence of cirrhosis, week 4 response, just to name a few, and these factors also have effects on one another.

    It's difficult to design a Phase 3 trial from the outset that can address each and every one of these potential factors and its effect on outcome. Sometimes we don't learn the importance of any one of these until we see the results and analyze them from the Phase 3 trials.
    So the hardest question proposed to the committee today involves selecting the best duration of treatment for certain important subgroups. To make this risk-benefit decision, the committee will need to balance optimizing SVR while trying to minimize additional toxicity.

    Prospectively defined and statistically powered comparisons are not available for all subgroups of interest. Despite this fact, FDA still needs to make labeling recommendations in the face of a degree of uncertainty and in the setting of incomplete data.
    Thus, we are asking the committee to answer difficult benefit-risk decisions based on the totality of evidence, including randomized comparisons and some retrospective analyses. We are asking this question, importantly, with patient interest in mind. Patients need not and should not be exposed to additional toxicity if there is a low likelihood of additional benefit.

    Finally, I would like to make a few comments on post-marketing studies in order to give the committee some background parameters on what is generally expected under the regulations. First, some semantics.
    The term "trial" is used when asking for investigations that are clinical and prospective,
    generally. The term "studies" refers to nonclinical investigations or observational clinical cohorts.

    Next, post-marketing investigations can be divided into requirements or commitments. Requirements fall under the following categories: pediatric trials under the Pediatric Research Equity Act; accelerated approval to confirm clinical benefit, which doesn't apply here because we're talking about an endpoint we consider to be validated for that purpose; but probably the most pertinent for today are required studies to address safety issues. These became requirements under the Food, Drug Amendment Act, the FDA Amendment Act of 2007.

    Safety investigations that can be required under FDAAA include trials to investigate adverse reactions. Some examples would be signal confirmation and evaluation, frequency of the adverse event severity, risk factors, and management. Safety issues can also be drug-drug interactions, and, also, drug resistance is
    considered a safety issue. And investigations can be trials or they can be studies.

    Post-marketing commitments, on the other hand, are voluntary agreements from the sponsor and usually include investigations to evaluate new indications, different aspects of efficacy, or exploratory studies based on theoretical hypothesis or long-term follow-up without specific safety concerns in mind.

    For both cases, post-marketing trials should be individual investigations that can be conducted in a specified time frame period and have a discreet starting and stopping point. Long-term follow-up investigations without a specified stopping point or analysis point are discouraged. We want these investigations to be delivered in a specified time frame and to be considered for labeling as appropriate.
    With these remarks, I will turn the agenda back to the chairperson, and I thank you very much for your attention.

    We will now proceed with the sponsor presentations.

    We will now begin the sponsor presentations.
    Sponsor Presentation – Laurie MacDonald

    DR. MACDONALD: Thank you, Dr. Cargill.
    Members of the committee and the FDA, good morning. I'm Dr. Laurie MacDonald, and I'm an infectious diseases physician in the Regulatory Affairs Department at Merck. It's our pleasure to bring you the data on boceprevir for the treatment of chronic hepatitis C infection.
    After this brief introduction, Dr. Jan Albrecht will present the clinical efficacy data on boceprevir, and then Dr. Clifford Brass will present the resistance and safety data. And finally, Dr. Keith Gottesdiener will present a summary of the benefit-risk.
    There's a substantial unmet medical need for
    new therapies to treat chronic hepatitis C infection. Chronic hepatitis C is a major public health issue in the United States. It's estimated that 3.2 million Americans are chronically infected with chronic hepatitis C.
    In the U.S., infection with hepatitis C virus genotype 1 is the most common and is also the least responsive to approved therapies. African-Americans and persons with liver cirrhosis have even lower response rates. Hepatitis C is a leading cause of liver cirrhosis and hepatocellular carcinoma and is the leading indication for liver transplantation.
    There are limitations to current chronic hepatitis C therapies. The goal of treatment is sustained virologic response, which is defined by undetectable HCV RNA at 24 weeks after the end of treatment. Currently, the recommended treatment for chronic hepatitis C genotype 1 infection is a 48-week course of peginterferon and ribavirin. Treatment is often poorly tolerated because of side effects that may prevent patients from completing a 48-week course of therapy.

    Only about 40 percent of patients with genotype 1 infection achieve a sustained virologic response with the current standard of care. The development of orally administered direct-acting antiviral agents is ushering in a new era in the treatment of chronic hepatitis C infection.
    The mechanism of action of boceprevir is inhibition of the NS3 protease. Hepatitis C is a single-stranded RNA virus. The genome codes for the synthesis of a single large polyprotein. The HCV NS3 protease is responsible for cleaving the polyprotein into functional viral proteins. Boceprevir is a protease inhibitor of the ketoamide class. It potently inhibits viral replication by binding reversibly to the active site of the HCF NS3 protease.

    The safety and efficacy of boceprevir has been evaluated in an extensive Phase 2 and Phase 3 clinical development program, which is briefly summarized on this slide.
    RESPOND-1 was an early Phase 2 dose-finding
    study, enrolling patients who had previously failed peginterferon and ribavirin treatment. SPRINT-1 was a Phase 2 study enrolling treatment-naive patients and was the first study to randomize patients to proposed clinical dose of 800 milligrams three times a day.
    RESPOND-2 and SPRINT-2 are the two pivotal Phase 3 studies that demonstrated the safety and efficacy of boceprevir administered with peginterferon alfa-2b and ribavirin in previous treatment failure patients and in treatment-naive patients.
    Additionally, a Phase 3 study has recently been completed confirming the efficacy of boceprevir when administered with peginterferon alfa-2a and ribavirin.

    There are several ongoing studies, one enrolling HIV/HCV co-infected patients and one comparing the use of erythropoietin versus ribavirin dose reduction alone for the management of anemia.
    Finally, there are two rollover studies, one
    providing boceprevir to patients who failed to achieve a sustained virologic response in a peginterferon and ribavirin control arm of a previous boceprevir study and a long-term follow-up study evaluating the durability of response. Our presentation this morning will focus primarily on the results of the two pivotal Phase 3 studies.
    In the pivotal Phase 3 studies, boceprevir has been shown to have a favorable benefit-risk profile. Boceprevir produced superior sustained virologic response rates compared to standard of care in both treatment-naive and previous treatment failure patients. A novel response-guided therapy algorithm which individualized treatment duration based on on-treatment HCV RNA responses allowed early responders to receive shorter durations of therapy.

    The safety of boceprevir has been evaluated in over 1500 patients who received boceprevir for up to 44 weeks in combination with peginterferon and ribavirin. The safety profile of boceprevir largely reflects the known safety profile of peginterferon and ribavirin standard of care. Incremental hematological side effects with boceprevir are manageable and are not treatment-limiting.

    The proposed indication for boceprevir is for the treatment of chronic hepatitis C genotype 1 infection in combination with peginterferon and ribavirin in adult patients with compensated liver disease who are previously untreated or who have previously failed therapy. Boceprevir is administered at a dosage of 800 milligrams three times a day with food. Treatment duration is individualized using a response-guided therapy algorithm.

    Merck has several consultants in attendance today, and I'd like to acknowledge them. We have Dr. Bruce Bacon, Professor of Medicine at the St. Louis University School of Medicine; Dr. Gary Koch, Professor of Biostatistics at the University of North Carolina at Chapel Hill; Dr. Fred Poordad, Associate Professor of Medicine at the David Geffen School of Medicine, UCLA; and, Dr. Jerry Spivak,
    Professor of Medicine and Oncology at the Johns Hopkins University School of Medicine.

    I will now ask Dr. Jan Albrecht, who has been involved in hepatitis C research for many years, to present a detailed discussion of the efficacy data that support the use of boceprevir for the treatment of patients with chronic hepatitis C infection.


    Sponsor Presentation – Jan Albrecht
    DR. ALBRECHT: Good morning. On behalf of Merck, I'm pleased to review the efficacy results of the Phase 3 studies supporting the use of boceprevir in combination with peginterferon plus ribavirin for the treatment of chronic hepatitis C in adult patients. During the presentation, for purposes of brevity, peginterferon plus ribavirin will be referred to as peg/ribavirin and capital PR will be used in the slides.
    During the past decade, there has been significant progress in successfully treating patients with chronic hepatitis C. Response rates have increased from approximately 10 percent in the
    1990s to approximately 40 percent with the current standard of care, peg-interferon plus ribavirin. The addition of boceprevir to peg-interferon plus ribavirin is another significant step forward.
    The efficacy review of boceprevir in combination with peg/ribavirin will include a description of the key concepts from the Phase 2 program that supported the design of the Phase 3 studies, a description of the study designs, the key efficacy results from the pivotal studies, and will conclude with recommendations for treatment.
    Phase 2 outcomes were used to design the Phase 3 studies. In Phase 2, we evaluated boceprevir in combination with peginterferon plus ribavirin in previously untreated patients, treatment-naive, for treatment durations of 28 and 48 weeks. Both 28 and 48 weeks of treatment were significantly more effective than peg/ribavirin for 48 weeks alone.
    Early anti-HCV response was shown to predict treatment duration. Patients who became HCV RNA undetectable by treatment week 8 had a similar
    response rate whether they were treated for 28 or 48 weeks of treatment, while patients who responded later required 48 weeks of treatment. This data provided the basis for a treatment strategy of response-guided therapy in which duration of treatment is based on time to undetectable HCV RNA. We also confirmed that 800 milligrams three times a day is on or near the plateau of both the dose exposure curve and the exposure response curve.

    The week 4 peg/ribavirin lead-in was shown to provide an assessment of interferon responsiveness and to be an important prognostic factor for SVR. A peg/ribavirin four-week lead-in prior to boceprevir treatment was used for the Phase 3 studies.
    Two boceprevir treatment strategies were assessed in the Phase 3 studies. Boceprevir added to the peg/ribavirin 48-week regimen administered as four weeks of peg/ribavirin lead-in and 44 weeks of triple therapy. Response-guided therapy, in which treatment duration is based on time to undetectable HCV RNA; early responders, defined as
    those patients having undetectable HCV RNA by treatment week 8 who receive a shorter duration of therapy; late responders, defined as those patients whose HCV RNA first becomes undetectable after week 8.
    Two pivotal Phase 3 studies were conducted in patients with chronic hepatitis C and genotype 1, one study in previously untreated patients, treatment-naive, and the other in patients who had previously failed treatment with peginterferon alfa-2a or alfa-2b plus ribavirin treatment failures.

    The study in treatment-naive patients, SPRINT-2, was a multicenter, double-blind, randomized Phase 3 study comparing peginterferon alfa-2b plus ribavirin to two boceprevir-containing regimens in combination with peginterferon plus ribavirin. Patients were adult with chronic hepatitis C genotype 1 with compensated liver disease. Peginterferon alfa-2b was administered according to the U.S. label. Ribavirin was given orally using weight-based doses of 600 to
    1400 milligrams per day in a divided dose twice daily.
    The study included three treatment arms, a peg/ribavirin control and two boceprevir-containing regimens. All patients received peg/ribavirin for four weeks in a lead-in phase. At the end of the lead-in, patients in the boceprevir control arm had a boceprevir placebo added to their regimen and received peg/ribavirin plus boceprevir placebo for an additional 44 weeks.
    In a parallel boceprevir-containing arm, patients had boceprevir added to their peg/ribavirin and were treated for an additional 44 weeks. The third arm evaluated a response-guided therapy regimen based on detectability of HCV RNA at treatment week 8. At the end of the lead-in, all patients had boceprevir added to their treatment and received 24 weeks of boceprevir plus peg/ribavirin.

    Patients with undetectable HCV RNA at treatment week 8 who remained undetectable through treatment week 24, early responders, completed
    treatment at treatment week 28. Patients with detectable HCV RNA at treatment week 8 who became undetectable by treatment week 24, late responders, had their boceprevir replaced by placebo and received an additional 20 weeks of treatment, for a total of 48 weeks of treatment.
    Patients in all treatment groups who remained HCV RNA detectable at treatment week 24 were discontinued from treatment, the futility rule for the study. Patients were followed after the end of treatment for 24 weeks, at which time their response to treatment was determined, SCR, HCV RNA undetectable versus non-responder HCV RNA detectable.
    SPRINT-2 enrolled two cohorts of patients. Patients in each cohort were randomized equally to the three treatment groups. Non-black and black patients were enrolled separately because of the known lower SVR rates in black patients.
    The goal was to assure that adequate numbers of black patients were treated to assess the impact of boceprevir on this important subgroup; 938 non-black and 159 black patients were enrolled. Of the black patients, 85 percent identified themselves as African-American.
    Patients were stratified by baseline HCV RNA level and genotype -- and HCV-1 subtype 1a or 1b. Base line liver histology was assessed by a central pathologist using the METAVIR score. The METAVIR score is a semi-quantitative score that assesses the fibrosis component of liver histology. The fibrosis component of the score goes from zero to 4, no fibrosis to cirrhosis. During the presentation, fibrosis zero, 1, 2 and 3 will also be referred to as no fibrosis and F4 as cirrhosis.
    HCV RNA was measured by polymerase chain reaction using the Roche TaqMan version 2 with a lower limit of detection of 9.3 international units per milliliter. Lower limit of detection was used for all decision points during the study and for determination of SVR.
    The study's primary efficacy endpoint was sustained virologic response rate, SVR, defined as a patient having an undetectable HCV RNA at
    24 weeks post-treatment. The population for the primary endpoint was all randomized patients receiving one dose of any study drug, the full analysis set, or FAS.

    The primary comparisons for the study were designed to demonstrate the superiority of each of the boceprevir-containing regimens compared to the peg/ribavirin control. SVR for the boceprevir/peg/ribavirin 48-week treatment group was compared to the peg/ribavirin control. And if that comparison was statistically significant at the alfa-equal-.05 level, then the response-guided therapy group was to be compared to the peg/ribavirin control.

    The key secondary endpoint for the study was SVR in patients who received one dose of boceprevir or placebo, the modified intent to treat analysis, mITT. Patients who discontinued treatment during the peg/ribavirin lead-in were excluded from the analysis, allowing a more precise estimate of the effect of boceprevir.

    The key demographic and disease
    characteristics of the patients included in the study are shown by cohort, non-black versus black. The study was conducted in North America and Europe, and the patient characteristics, in general, are reflective of these regions. Approximately two-thirds of patients were male with a mean age of around 50. Approximately three-quarters of the patients were from North America, mainly the United States.
    There were nearly twice as many patients with HCV subtype 1a compared to 1b, reflecting the distribution between the U.S. and Europe. Most patients had high baseline viral loads, approximately 9 percent had advance liver fibrosis, and, overall, the patient characteristics were similar across the three treatment groups.
    The disposition of patients was included in the briefing book and so will not be discussed in detail. Patients were randomized and treated in the study with 363 in the peg/ribavirin control, 368 in the boceprevir response guided-therapy arm, and 366 in the boceprevir/peg/ribavirin 48-week arm.

    The primary efficacy outcome for the study is sustained virologic response rate in all patients, blacks and non-blacks combined. Subsequently, the data will be shown for the non-black and black patient cohorts separately, but for most analysis, the results will be presented for the combined population.

    On the Y-axis is the percent of patients achieving SVR and on the X-axis the three treatment groups. The peg/ribavirin control is in white, response-guided therapy with boceprevir in yellow, and the boceprevir/peg/ribavirin 48-week regimen in orange. Ninety-five percent confidence intervals are shown for each of the SVR relapse rates. This format will be used for all results in the presentation.
    In this overall analysis, the SVR rates are significantly higher in patients receiving a boceprevir-containing regimen compared to peginterferon plus ribavirin, 63 percent in the response-guided therapy arm and 66 percent in the boceprevir 48-week arm compared with 38 percent in the peg/ribavirin control, with a p-value of less than 0.0001 for each of the boceprevir-containing regimens versus control.
    The high SVR rates in patients receiving a boceprevir-containing regimen are achieved through both a higher response rate during therapy and a lower rate of relapse following the end of treatment. Relapses in the boceprevir-containing regimens are approximately one-half of those in the peg/ribavirin control.

    The superiority of the boceprevir-containing regimens compared to the peg/ribavirin control, overall and across patient subgroups, is shown in this forest plot. Odds ratios and corresponding 95 percent confidence intervals are shown for response-guided therapy and boceprevir/peg/ribavirin 48 versus the peg/ribavirin control for the overall SVR, as well as for each subgroup. Dots to the right of the vertical line represent a higher SVR for boceprevir versus the left of the vertical line for the peg/ribavirin control.

    As shown, for the two boceprevir treatment regimens, response-guided therapy on the left and boceprevir/peg/ribavirin 48 weeks on the right, the boceprevir regimens have higher SVR rates compared to the peg/ribavirin control in all key subgroups, including race, baseline viral load, gender, age, fibrosis score, HCV 1 subtype, and regions where the studies were conducted. For both the response-guided therapy group and the boceprevir/peg/ribavirin 48-week group, these data demonstrate the strength of the overall treatment effects with boceprevir, as well as the consistency of treatment effect across subgroups.

    When the two patient cohorts enrolled in the study are analyzed separately, the SVR rates for each of the two groups, non-black and black, are significantly higher in patients receiving a boceprevir-containing regimen compared to the peg/ribavirin control. On the left side of the slide, the SVR rates in non-black patients are 67 and 68 percent for response-guided therapy and boceprevir/peg/ribavirin 48-weeks, respectively, compared to 40 percent for the peg/ribavirin control.
    Focusing on the response rates in the black patients, SVR rates in black patients, shown on the right, are 42 percent for response-guided therapy and 53 percent for boceprevir/peg/ribavirin 48-week compared with 23 percent for the peg/ribavirin control. For both non-black and black patients, the increase in the SVR for the boceprevir-containing regimens is statistically significant compared to the peg/ribavirin control.
    Consistent with historic response data, the overall SVR rates for black patients are lower than in non-black patients receiving the same treatment regimens. However, the advantage of adding boceprevir to peginterferon plus ribavirin in black patients is clearly evident.

    The response rate in the black patients is 11 percent lower for boceprevir response-guided therapy compared to the boceprevir/peg/ribavirin 48-week group. In an effort to understand the
    difference in these response rates, we conducted additional analysis.
    The first analysis is the mITT analysis in which patients who discontinued treatment during the peg/ribavirin lead-in and who never received boceprevir are excluded. The focus for this slide is the SVR rates for black patients, shown on the right. The data for non-black patients are also shown on the left for completeness. The FAS response rates are shown in the body of the response bars, with the mITT response rates on the top. The difference in response rate is represented by the crosshatch
    In the mITT analysis, SVRs are 47 percent for response-guided therapy and 53 percent for boceprevir 48 weeks versus 26 percent for the peg/ribavirin control. The 5 percent increase in the response rate in the boceprevir response-guided therapy group is a result of five patients in the response-guided therapy arm discontinuing treatment during lead-in, prior to receiving boceprevir, versus no patients in the boceprevir/peg/ribavirin
    48-week arm. This analysis explains about one-half of the difference in SVR between the boceprevir response-guided therapy group and the boceprevir 48-week groups for black patients.
    The presence of liver cirrhosis can have an effect on response to treatment. Therefore, we assess the SVR by the presence of no cirrhosis versus cirrhosis. Non-black patients are shown on the left side of the slide and black on the right side.
    For each patient group, no cirrhosis, F0 to 3, is shown on the left side and cirrhosis, F4, on the right side. In black patients without cirrhosis, the SVRs are similar for the two boceprevir-containing regimens, 50 percent for response-guided therapy and 54 percent for boceprevir 48 weeks.
    In black patients with cirrhosis, F4, the numbers of patients are too small to make an assessment of the response. For non-black patients, the SVRs for both of the fibrosis categories and the treatment groups within the categories are nearly identical. For black patients without cirrhosis, the recommended treatment duration is response-guided therapy.
    The goal of response-guided therapy is to significantly shorten the duration of treatment in patients who rapidly achieve an undetectable HCV RNA. The response-guided therapy regimen was included in the study to confirm that there is a group of patients, early responders, who can be treated for 28 weeks and have a response equal to that of patients receiving boceprevir added to the standard 48-week regimen of peginterferon/ribavirin.

    As a reminder of the study design, this slide shows the treatment regimens for early responders in the response-guided therapy and the boceprevir 48-week treatment groups.
    In order to compare the efficacy of response-guided therapy for early responders to a like group in the boceprevir/peg/ribavirin 48-week arm, we conducted an analysis in patients in the boceprevir response-guided therapy group who were
    early responders, HCV RNA undetectable at treatment week 8 through treatment week 24, and stopped treatment at treatment week 28, the yellow bar on the treatment diagram on this slide, with patients in a parallel boceprevir 48-week group, who were also early responders, orange bar on the treatment diagram on this slide, and continued on treatment for 48 weeks.
    When the non-black and black cohorts of patients are combined, overall, 44 percent of patients meet the criteria for early response, treatment week 8 through treatment week 24 undetectable.
    In these early responders, nearly all patients, 96 percent, in both the response-guided therapy and boceprevir/peg/ribavirin 48-week groups achieved SVR with the boceprevir response-guided therapy shorter, 28-week duration of treatment, being equally as effective as the boceprevir/peg/ribavirin 48-week treatment group.

    Overall, the response of these analyses clearly demonstrate that early responders,
    including black patients, can be treated for a shorter duration of 28 weeks and achieve an SVR equal to that of patients who received boceprevir plus ribavirin for 48 weeks.
    Before leaving the discussion of SPRINT-2, it is important to review the data for the late responders, the other aspect of response-guided therapy. As you will recall, in the response-guided therapy group, these are patients who first have an HCV RNA that becomes undetectable after treatment week 8 and remains undetectable at treatment week 24, the yellow bar on the treatment diagram in this slide.
    We compared these patients to patients in the parallel boceprevir/peg/ribavirin 48-week group who were HCV RNA detectable at treatment week 8 but who became undetectable at 24, orange bar, and continued on treatment for 48 weeks.
    In the late responders, using the assigned per protocol analysis, SVRs for the response-guided therapy boceprevir regimen and the boceprevir/peg/ribavirin 48-week groups are
    similar, 72 and 75 percent. However, in the FDA briefing book, the SVR rates, based on a modified FDA analysis, are 66 and 75 percent for response-guided therapy and boceprevir/peg/ribavirin 48-weeks, respectively.

    The FDA analysis excludes 14 patients on the basis that these patients may have had a false positive HCV RNA during the initial treatment period from treatment week 8 to treatment week 24 and should have been assigned to the 28-week response-guided therapy regimen.

    To conclude the discussion on the SPRINT-2 study, addition of boceprevir to peg/ribavirin standard of care results in a statistically significant increase in efficacy in treatment-naive patients. Using boceprevir response-guided therapy, 44 percent of patients received only 28 weeks of treatment and achieve an SVR rate of 96 percent.
    Boceprevir in combination with peginterferon plus ribavirin significantly improves efficacy in the difficult to treat black patients. And for
    these patients, response-guided therapy is the recommended regimen. Boceprevir in combination with peginterferon with ribavirin also improves efficacy in the difficult to treat cirrhotic patients. However, patient with cirrhosis may need 44 weeks of boceprevir treatment.
    A second pivotal study in the boceprevir Phase 3 program is RESPOND-2, evaluating boceprevir in previous peginterferon plus ribavirin treatment failures. The study was a multicenter, double-blind, randomized Phase 3 study in adult patients with chronic hepatitis C genotype 1 and compensated liver disease who had previously failed treatment with peginterferon alfa-2a or alfa-2b plus ribavirin.
    The objective of this study was to compare re-treatment with peginterferon plus ribavirin to each of two boceprevir-containing regimens in combination with peginterferon plus ribavirin. Peginterferon alfa-2b and ribavirin and boceprevir were administered using the same doses and regimens as in the SPRINT-2 study.

    The design for this study included three treatment arms and is the same as for the previous treatment-naive study, with two exceptions: the duration of the boceprevir treatment, 32 weeks, in the boceprevir response-guided therapy arm and the timing of the futility rule, treatment week 12.
    In the boceprevir response-guided therapy arm, patients received 32 weeks of boceprevir plus peg/ribavirin. Duration of therapy, 36 or 48 weeks, was determined by the detectability of HCV RNA at treatment week 8.
    Patients with undetectable HCV RNA at treatment week 8 completed treatment at treatment week 36. These were the early responders for the study. Patients with detectable HCV RNA at treatment week 8 who became undetectable by treatment week 12, the late responders for this study, had their boceprevir replaced by placebo at treatment week 36 and received an additional 12 weeks of treatment, for a total of 48 weeks of treatment.
    Primary and secondary endpoints and
    assessment of HCV RNA for this study are the same as for the treatment-naive study; 403 patients were randomized 1:2:2 to peginterferon alfa-2b plus ribavirin or the boceprevir treatment groups, respectively. Patients were stratified by response to prior peginterferon/alfa-ribavirin treatment and HCV-1 subtype 1a versus 1b.
    The study enrolled patients who are treatment failures to peginterferon plus ribavirin. Classification of non-responders is based on the patients' historic response to previous peginterferon plus ribavirin treatment. Subcategories of non-responders are based on the degree of previous interferon response.
    The study included relapsers and non-responders. Relapsers are defined as patients who are HCV RNA undetectable at the end of treatment, becoming positive when therapy is discontinued. Non-responders are defined as patients who did not become HCV RNA undetectable during treatment.
    Among the non-responders, the study included patients with a greater than or equal to 2 log decrease in HCV RNA by treatment week 12, partial responders. The study excluded patients with a less than 2 log decline in HCV RNA at treatment week 12, null responders.
    The decision to exclude null responders was based on a concern that for patients with a poor interferon response, that the addition of a direct-acting antiviral would result in what would be essentially monotherapy of the direct-acting antiviral and allow resistance to emerge. However, the data from the Phase 3 study support that boceprevir can be used to treat patients who have a very poor response to interferon.
    The demographics of the study are similar to the treatment-naive study in that they reflect the current demographics of HCV in the United States and Western Europe where the study was conducted. The major difference in the demographics between studies is the higher number of patients with advanced fibrosis or cirrhosis in the treatment failure study, approximately 19 percent compared to approximately 9 percent in the treatment-naive study. Overall, the baseline characteristics are similar across the three treatment groups.

    The disposition of patients was included in the briefing book and so will not be discussed in detail. Patients randomized and treated in the study include 80 patients in the peginterferon plus ribavirin control, 162 patients in the boceprevir response-guided therapy arm, and 161 patients in the boceprevir/peg/ribavirin 48-week treatment group.
    As in the treatment-naive study, the primary efficacy endpoint for the study is sustained virologic response, SVR in the FAS population. SVR rates in patients receiving a boceprevir-containing regimen are approximately three times higher than those in patients re-treated with peginterferon plus ribavirin.
    The differences are statistically significant; 59 percent of patients in the response-guided therapy group and 66 percent of patients in the boceprevir/peg/ribavirin 48-week group compared to 21 percent of patients in the peg/ribavirin control achieved SVR. There is a numerical difference of 7 percent between the response-guided therapy and boceprevir 48-week groups, which I will discuss later.

    As in the treatment-naive study, the high SVR rates in patients receiving a boceprevir-containing regimen are achieved through both a higher response rate during therapy and a lower relapse rate. The boceprevir-containing regimens have superior SVRs compared to peginterferon ribavirin across all patient subgroups, as shown in this forest plot.
    As a reminder, dots to the right of the vertical line represent a higher SVR for boceprevir versus the left of the vertical line for peg/ribavirin control. As in the treatment-naive study, these data demonstrate the strength of the overall treatment effect with boceprevir, as well as consistency of treatment effect across subgroups.
    As previously discussed, there was a 7 percent in SVR between the response-guided therapy and boceprevir/peg/ribavirin 48-week treatment groups. Additional analysis suggests that this difference is largely related to a difference in SVR in patients with cirrhosis.
    As in the treatment-naive study, SVR was analyzed by presence of baseline cirrhosis versus no cirrhosis. SVR for patients with cirrhosis is shown on the left side of the slide and for patients who have no cirrhosis on the right side.

    The majority of the patients in the study, approximately 85 percent, did not have cirrhosis. For patients with no cirrhosis, the SVR rates are nearly identical for the boceprevir response-guided therapy and boceprevir 48-week groups, 64 and 66 percent, respectively, and substantially higher than the peg/ribavirin control, 24 percent.
    For the approximately 15 percent of patients with cirrhosis, both boceprevir-containing regimens have substantially higher SVR rates compared to the peg/ribavirin control, in which no patient achieved an SVR. However, the boceprevir/peg/ribavirin 48-week SVR rate is substantially higher than the
    response-guided therapy group, 77 versus 35 percent.
    Boceprevir improves efficacy in the difficult to treat patients with cirrhosis. However, these patients may require a longer duration of therapy, boceprevir plus peginterferon ribavirin for 48 weeks.
    As previously noted, the majority, approximately 85 percent of patients in the study did not have cirrhosis. Therefore, the remainder of the analysis will exclude these patients to allow a better comparison of the response-guided therapy and boceprevir/peg/ribavirin 48-week groups.
    This slide shows the patients' historic response to peginterferon plus ribavirin. Previous non-responders are on the left and relapsers are on the right. Both boceprevir-containing regimens have multiple-fold increases in SVR compared to the peg/ribavirin control.
    For both previous non-responders and relapsers, the SVR rates within the historic response categories are similar for response-guided therapy and boceprevir 48-weeks, 47 to 51 percent for non-responders and nearly identical, 74 and 75 percent, for relapsers.
    These data demonstrate that response-guided therapy and boceprevir/peg/ribavirin for 48 weeks are approximately equally effective in patients without cirrhosis.

    As for the treatment-naive study, we compared the response rates for patients in the boceprevir response-guided therapy group who were early responders, defined as HCV RNA undetectable at treatment week 8, who received 36 weeks of treatment, yellow bar, to a parallel group from the boceprevir/peg/ribavirin 48-week group who were HCV RNA undetectable at treatment week -- excuse me -- who were HCV RNA undetectable at treatment week 8 and received 48 weeks of treatment, orange bar.
    Overall, as before, 44 percent of the patients met the criteria for early response. As in the treatment-naive study, the early responders
    in both boceprevir-containing regimens have very high SVRs, 89 and 97 percent for boceprevir response-guided therapy and boceprevir/peg/ribavirin 48 weeks, respectively. When the patients with cirrhosis are excluded from the analysis, the SVRs for the two regimens are also extremely high, 91 and 96 percent, supporting that response-guided therapy is a viable treatment strategy for early responders.
    As with the treatment-naive study, we compared the late responders from the response-guided therapy group, patients who were HCV RNA detectable at treatment week 8 and become undetectable at treatment week 12, yellow bar on the treatment diagram on this slide. We compared these with a parallel boceprevir/peg/ribavirin 48-week group who were HCV RNA undetectable at treatment week 8 but became -- excuse me -- detectable at treatment week 8 but became undetectable at treatment week 12, orange bar, and continued on treatment for 48 weeks.
    The similarity of response rates between
    response-guided therapy and boceprevir/peg/ribavirin 48 weeks in early responders is also observed with late responders, 80 percent of patients in the response-guided therapy arm and 73 percent of patients in the boceprevir/peg/ribavirin 48-week arm.
    When the patients with no cirrhosis are also considered, the response rates for the two boceprevir regimens are similar, 85 and 78 percent for response-guided therapy and boceprevir 48 weeks, respectively, again, supporting response-guided therapy as a viable treatment strategy.
    Merck is seeking a label indication for previous treatment failures that encompasses all categories of non-responders, including null responders. Therefore, the issue arises that we did not enroll historic null responders in the pivotal Phase 3 RESPOND-2 study.
    The next few slides will summarize the data from the RESPOND-2 and SPRINT-2 studies that demonstrate the efficacy of boceprevir for all categories of non-responders and the new data from
    the PROVIDE rollover study that is treating prospectively defined null responders. As a reminder, historical null responders are defined by their very poor response to interferon; that is, less than 2 log decline in HCV RNA at treatment week 12. Although we excluded these patients from the RESPOND-2 study, there are patients in both the RESPOND-2 and the SPRINT-2 studies with very poor interferon response that were included and treated.
    The inclusion of patients with poor interferon response in the boceprevir studies can be documented in two ways. Firstly, 20 percent of the patients in the peginterferon ribavirin control in both SPRINT-2 and RESPOND-2 failed to achieve a 2 log decline in HCV RNA at treatment week 12, null responders. We would expect to find a similar number of patients in the boceprevir groups since these were randomized studies. These control patients had little or no response.
    Second, both treatment week 4, less than 1 log, as outlined in the briefing document, and treatment week 12, less than 2 log, historical non-
    responders, defined patients who are unlikely to respond to peginterferon plus ribavirin treatment.


    Lastly, we have end of treatment results from the PROVIDE study.

    This is a study which includes patients from the peg/ribavirin control groups in the Phase 3 trials who, one, failed therapy with peginterferon plus ribavirin; two, met the week 12 definition of null response; and, three, who received treatment with boceprevir. This data has been shared with the FDA but not fully reviewed.
    This slide shows the SVR rate in the patients in the peg/ribavirin control arm of the RESPOND-2 study who met the definition of null responder at treatment week 12. No patient achieved an SVR. Likewise, this slide shows the SVR rate in the patients in the control arm who met the alternative less than 1 log decline at treatment week 4, definition of poor interferon response. Just as with the 12-week definition, no patient achieved an SVR.
    These data support that both definitions
    identify patients unlikely to respond to peg/ribavirin plus ribavirin.
    In contrast, in the boceprevir-containing treatment regimen using the less than 1 log week 4 definition, 33 and 34 percent of patients in the boceprevir response-guided therapy and boceprevir/peg/ribavirin 48-week groups achieved SVR.
    Overall, the data shows that, one, the treatment week 4 definition of less than 1 log decline at treatment week 4 defines patients who have a poor response to interferon; and, two, clearly demonstrates that boceprevir in combination with peginterferon ribavirin can achieve SVR in these difficult to treat patients.
    The findings in the SPRINT-2 were similar. In the peg/ribavirin control, SVR was 0 percent and 4 percent for the treatment week 12 and treatment week 4 definitions, respectively. These studies again demonstrate that patients who have poor interferon response by either definition rarely achieve an SVR.

    As in the RESPOND-2 study, the addition of boceprevir resulted in a substantial number of patients achieving an SVR in the boceprevir response-guided therapy and boceprevir/peg/ribavirin week regimens, 28 and 38 percent, respectively.
    As I had noted previously, PROVIDE is a rollover study that provides treatment with boceprevir to patients who are peg/ribavirin controls in one of the clinical studies and who failed treatment with peginterferon plus ribavirin.

    We have end of treatment HCV RNA results for patients that were in the SPRINT-2 and RESPOND-2 studies who, one, failed to respond as a peginterferon ribavirin control; two, met the definition of null response, treatment week 12, less than 2 log decline; and, three, who received treatment with boceprevir for 44 weeks.
    There are 38 null responder patients who have completed treatment, of whom 15 of the 38, 39 percent, are undetectable at end of treatment. On the left of the slide are the weeks of
    boceprevir they've received, 44 weeks, end of treatment. On the right side of the slide, those are currently undetectable, have undetectable HCV RNA, 15 of the 38, or 39 percent. While SVR data is not available, this end of treatment response rate is consistent with the end of therapy data from the Phase 3 trials in patients with less than 1 log decline at treatment week 4.

    To summarize the data, these data demonstrate that in patients with poor interferon response, that treatment week 4 definition is a good surrogate for poor interferon response and a useful alternative to the historical treatment week 12 definition. They demonstrate a broad range of treatment failure, and treatment-naive patients were enrolled in the Phase 3 boceprevir program, including hard to treat patients comparable to the conventional null definition.
    We also see from these data that boceprevir added to peginterferon plus ribavirin therapy resulted in an increase of approximately 30 percent in SVR in those difficult to treat patients. Thus,
    the data support our proposed indication for peginterferon plus ribavirin for use in the treatment of all previous peginterferon plus ribavirin treatment failures.

    To conclude the discussion of the RESPOND-2 study, adding boceprevir to peginterferon plus ribavirin for the treatment failure population resulted in a statistically significant approximately threefold increase in SVR. Boceprevir response-guided therapy allowed a shorter duration of treatment in 44 percent of patients, the early responders, who achieved an 89 percent SVR.

    Regardless of a patient's historic response classification, the addition of boceprevir substantially increase SVR, with the highest responses being observed in patients with good interferon response, such as relapsers. However, there was also a robust response in the more poorly responsive non-responders. This data supports that response-guided therapy is the optimal regimen in these populations.

    Boceprevir also improves efficacy in the difficult to treat cirrhotic patients. However, patients with cirrhosis may require 44 weeks of boceprevir treatment.
    In conclusion, the results of the boceprevir Phase 3 studies support that the addition of boceprevir to peg/ribavirin standard of care results in significant increases in SVR in both naive and treatment failure patients.
    Boceprevir in combination with peginterferon ribavirin increases SVR rates in all subgroups of patients, including patients with a poor interferon response. And most importantly, boceprevir response-guided therapy allows the early responder to receive a shorter duration of therapy with efficacy equal to adding boceprevir to the standard peginterferon plus ribavirin 48-week regimen.

    Thus, boceprevir response-guided therapy is the recommended strategy for both treatment-naive and previous treatment failure patients. However, patients with cirrhosis may need 44 weeks of boceprevir in combination with peg/ribavirin treatment.

    I would now like to introduce Dr. Clifford Brass, who will review the resistance profile and safety of boceprevir.
    Sponsor Presentation – Clifford Brass
    DR. BRASS: Thank you, Dr. Albrecht.
    This portion of the presentation will show that the addition of boceprevir to peg/riba standard of care has a favorable benefit-risk ratio, as reflected in the profiles of resistance, adverse events, and clinical pharmacology. The use of the response-guided therapy paradigm further increases the benefit-risk ratio of boceprevir combination therapy and supports the proposed indication.
    Replication of HCV viruses is dependent on an error-prone, RNA-dependent RNA polymerase. This results in a population of genetically distinct HCV particles within the population of viruses in an infected patient.
    Boceprevir is a direct-acting antiviral agent that targets the HCV NS3 protease. The loss of susceptibility to HCV protease inhibitors is associated with the selection of variants with amino acid changes in the NS3 coding region.
    Variants that cause a decreased susceptibility to boceprevir are termed resistance-associated amino acid variants or RAVs. RAVs are identified in vitro and in vivo and map to the NS3 region of the HDB genome. RAVs were characterized in the pivotal Phase 3 trials at baseline, at treatment week 8, and at the time of virologic failure. Finally, RAVs are followed post-therapy to evaluate their persistence.
    Putative RAVs were first identified as amino acid changes appearing in replicons during in vitro resistance selection experiments or in viruses from patients treated with boceprevir. Each of these variants was engineered into prototypical genotype 1a or 1b proteases and tested in vitro for a decreased susceptibility to boceprevir.
    Variants that cause a decreased susceptibility to boceprevir were observed at 10 loci in the NS3 and are listed on this slide. The
    full change in susceptibility to boceprevir differed for each variant and ranged from twofold to 300-fold in enzymatic assays in vitro.
    Samples for resistance testing were taken frequently from patients enrolled in boceprevir clinical studies. Population sequence data was available for 96 percent of patients at baseline and post-baseline sequence data was available for 86 percent of patients that failed to achieve an SVR.

    Viruses at the above time points were analyzed using a population sequencing method that overall detects minor variants existing in the population at a frequency of 20 percent or higher. For technical reasons, samples were sequenced only when the viral load was greater than 1,000 international units per ml. In the boceprevir pivotal studies, the sequence was obtained on the NS3 region.
    The following slides show the resistance data for patients that received boceprevir plus peg/riba in the SPRINT-2 and RESPOND-2 studies.

    In SPRINT-2 and RESPOND-2, RAVs were associated with virologic failure and were detected in 53 percent of all non-SVR patients. However, as the majority of patients treated with boceprevir achieved an SVR, this number represents 15 percent of all patients treated with boceprevir. RAVs that were identified in greater than 25 percent of patients with RAVs at virologic failure differed by genotype. The most common RAVs associated with genotype 1a and 1b are shown on this slide.
    One of the benefits of the lead-in strategy is to allow the assessment of the peg/riba response based on HCV RNA declines at in treatment week 4 prior to receiving boceprevir, as discussed in detail in the briefing book. This allows a relationship between interferon responsiveness and the development of RAVs to be determined.
    Subjects with a greater than or equal to 1 log reduction in HCV RNA at treatment week 4 are defined as interferon responsive, while subjects with a less than 1 log reduction in HCV RNA at the same time point are defined as poorly interferon responsive.

    This slide shows the percent of RAVs detected in non-SVR patients by interferon responsiveness. RAVs were detected in 69 percent of poorly interferon responsive patients compared to 31 percent of interferon responsive patients. This indicates that interferon responsiveness can influence the selection of RAVs in non-SVR subjects, most likely due to the ability of the peg/riba therapy to suppress viral replication.
    RAVs were detected in 7 percent of patients with samples sequenced at baseline by population sequencing. SVR rates in these patients were similar to the majority of patients that did not have RAVs detected at baseline. Therefore, detection of RAVs at baseline alone was not predictive of treatment outcome on boceprevir plus peg/riba therapy.
    In a new analysis not presented in the briefing book, we examined the persistence of RAVs in the non-SVR patients in SPRINT-2 and RESPOND-2. This slide shows the detectability of the most
    common RAVs declined during follow-up. During the 6 to 14-month post-therapy period, the majority of patients had RAVs that became undetectable by population sequencing. Although different RAVs declined at different rates, shown here for the six most common RAVs, they all declined over this period of time.
    To summarize, overall, the presence of RAVs at baseline was not predictive of outcome. Fifty-three percent of non-SVR subjects had viruses with RAVs detected post-baseline. In treatment failures, the most common RAVs differed by genotype and are listed here.

    RAVs detected in non-SVR patients declined over time. The majority of patients did not have detectable RAVs 6 to 14 months post-therapy, but the percent decrease differed based on the RAV selected and ranged from 68 percent to 94 percent. Studies to evaluate whether RAVs returned to pre-treatment levels are ongoing, but the clinical significance of RAVs is unknown.
    The addition of boceprevir to the peg/riba
    standard of care was generally safe and well tolerated, with the profile of adverse events consistent with backbone therapy. The pooled safety database showed some incremental increases in a small number of events, which did not generally lead to discontinuation of boceprevir.
    The safety database includes the pivotal Phase 3 studies, SPRINT-2 and RESPOND-2, and SPRINT-1, an open-label Phase 2 study in treatment-naive patients. SPRINT-1 was the first study to use the proposed regimen of peg/riba plus boceprevir, 800 milligrams TID.
    In all three of these key studies, boceprevir was dosed at 800 milligrams TID for 20 weeks or more. For the integrated analysis of safety, all boceprevir-containing regimens were merged, including the response-guided therapy arms and the 48-week peg/riba/boceprevir arms. Overall, 2,095 patients are in the key studies, including 1,548 patients in the boceprevir arms, which was threefold greater than the 547 peg/riba control patients.
    The majority of patients, 78 percent, received more than 24 weeks of boceprevir, and total boceprevir exposure in these three key studies was 840 patient years.

    The following analysis includes all patients who received at least one dose of any study drug, whether or not they discontinued during the lead-in period, and the full study period of 72 weeks is described.
    The safety profile of boceprevir combination therapy is largely a reflection of the peginterferon and ribavirin backbone, with some increases in events that are generally predictable and manageable. Nearly every subject in the key studies reported an adverse event.
    Although the actual number of events varied widely due to the threefold greater number of patients in the boceprevir arms, the percentages of subjects reporting SAEs in treatment discontinuations due to an AE were similar in the peg/riba control and boceprevir-containing treatment arms.

    Of the 2,095 subjects treated, eight died during the course of the key studies. Four deaths, or 1 percent, occurred in the peg/riba control arms and four deaths, which is less than 1 percent, occurred in the boceprevir-containing treatment arms. Dose reduction, primarily ribavirin dose reduction for anemia, was greater in the boceprevir arms.
    The most common adverse events reported in the boceprevir/peg/riba arms are largely reflective of the known events associated with the peg/riba backbone. Most of these adverse events, highlighted in orange, occur commonly with peg/riba therapy due to the flu-like symptoms or the gastrointestinal side effects associated with peg/riba therapy. These events were reported in a similar proportion of patients in the boceprevir/peg/riba and peg/riba control arms.

    Only two adverse events, anemia and dysgeusia, occur 20 percent more often in the boceprevir-containing regimens, and these will be discussed as events of special interest. Nausea,
    diarrhea and neutropenia are the only other common events listed here, which are 5 percent greater when boceprevir is added to the peg/riba standard of care.

    Events of special interest include events increased on boceprevir or expected to be increased with some ketoamide protease inhibitors.
    Three broad areas are examined as events of special interest: hematologic events, dysgeusia and rash. The increased hematologic events occurring with boceprevir use, primarily an increase in anemia, appear to be class effects associated with protease inhibitors of the ketoamide class.
    In addition, the incidence of dysgeusia, an altered sense of taste, was specifically increased with the addition of boceprevir. Skin rash, an event of special interest based on its association with ribavirin treatment and its increased occurrence with another drug in the same ketoamide class was not increased in the pooled boceprevir arms.

    An approximately 3 gram per deciliter drop in hemoglobin concentration is a well recognized side effect of treatment with peginterferon and ribavirin. Interferon is a bone marrow suppressant, and ribavirin causes a dose dependent hemolysis, which is the primary driver of this drop in hemoglobin.

    Boceprevir as a single agent did not cause a drop in hemoglobin in either animal studies or in a Phase 1 monotherapy study. When healthy male adults were treated with boceprevir for up to 57 days, no effects were demonstrated on red cell production, survival or destruction, or on markers of anemia. However, when boceprevir is added to peg/riba therapy, it causes an incremental drop in hemoglobin of approximately 1 gram per deciliter.

    The management of anemia for patients receiving boceprevir combination therapy is identical to that for anemic patients receiving peginterferon plus ribavirin standard of care. The incidence of discontinuation from therapy because of anemia is similarly low, at approximately 1
    percent in both the peg/riba control and boceprevir experimental arms. Anemia is transient as hemoglobin returns to baseline post-therapy in control and experimental arms.

    This slide shows the mean hemoglobin concentration on the Y-axis during 48 weeks of treatment and 24 weeks of follow-up, X-axis, for patients in the control, white line, and boceprevir experimental arms, yellow line, in the key safety studies. The pattern of mean hemoglobin concentration over time, with a return to baseline on follow-up, was similar in the boceprevir/peg/riba arms and the peg/riba control arms.
    Boceprevir exacerbates the hemoglobin drop caused by peginterferon ribavirin therapy, as reflected in the nadir hemoglobin levels on treatment of patients in the control arms, white bars, and boceprevir/peg/riba arms, yellow bars, which are displayed here according to the WHO toxicity grade. Three percent of patients in the boceprevir combination arms had WHO grade 3 or 4
    anemia and hemoglobin of 8 grams per deciliter or less compared to 1 percent in the pegloticase-riba control.

    Anemia was managed in the key trials with the same tools used to treat anemia with peg/riba therapy in clinical practice today, ribavirin dose reduction and/or treatment with erythropoietin, also referred to as EPO. Guidelines for the use of EPO were provided in the study protocols and were based on product inserts for erythropoietin use in other disease states. The decision to use EPO and implement these guidelines was based solely on investigator judgment.
    Among anemic patients, both ribavirin dose reduction and erythropoietin use was common. When aggregated, more patients receiving boceprevir, 25 percent, had a ribavirin dose reduction compared to control at 12 percent. Similarly, 38 percent of patients in the experimental arms were treated with erythropoietin to ameliorate their anemia compared to 19 percent in the control arms.

    Erythropoietin utilization was greater than
    noted in clinical practice or previous trials. This may relate to the fact that it was provided by the sponsor upon physician request. Overall, the management strategies for anemia were similar between treatment and control arms, reflecting the standard management of anemia in subjects receiving combination therapy with peg/riba, as described in the VA Guidelines for Health Care Providers.

    There were no significant increases in events that are associated with EPO use in other chronic diseases, such as hypertension, thrombosis or cancer. However, there was00:15:39 one case reported as antibody-mediated pure red cell aplasia. A recent update now reveals that this patient fully recovered with immunosuppressive therapy.
    Although anemia can be a safety issue, paradoxically, it also seems to be a marker of increased virologic response with two-drug therapy, as previously demonstrated in the IDEAL trial, a 3,000-patient study of treatment-naive patients.
    This slide shows the SVR rate in all three
    arms of SPRINT-2 and RESPOND-2 displayed for patients who became anemic on therapy, stippled bars, and patients in the same arm who did not become anemic, solid bars. For all three arms in both studies, patients developing anemia on boceprevir combination therapy or control treatment had higher SVR rates than patients who did not develop anemia.
    Based on exploratory analyses, these high SVR rates in anemic patients were maintained independent of the interventions used to manage anemia. In both SPRINT-2 and RESPOND-2, similar high levels of SVR were achieved in anemic patients managed with ribavirin dose reduction alone or EPO treatment with or without ribavirin dose reduction.
    Anemia is captured in these studies either by a laboratory report of a hemoglobin less than 10 grams per deciliter or was reported by the investigator as an adverse event. The rates of anemia reported as an adverse event were 29 percent in the control arms and 49 percent in the boceprevir arms.

    Anemia was generally manageable. Approximately one-quarter of all boceprevir-treated patients required dose modification for anemia, but only 3 percent were transfused, and only 1 percent reported SAEs of anemia. Ribavirin dose reduction and/or erythropoietin use were effective in maintaining subjects on therapy, as only 1 percent required treatment discontinuation for anemia.

    There were also increases in neutropenia on boceprevir. Peginterferon plus ribavirin leads to a decrease in neutrophil counts during treatment. An incremental effect of boceprevir in neutrophils resulted in increased rates of grade 3 and 4 neutropenia and absolute neutrophil count of less than 750 in 500 million cells per liter, respectively, when compared to peg/riba control.
    Neutropenia, as reported by the investigator, was managed in the key trials primarily by peginterferon dose reduction. Granulocyte colony stimulating factor, GCSF, was also occasionally used, but the drug was not provided by the sponsor. GCSF use was low but
    higher in the boceprevir/peg/riba arms compared to the peg/riba control at 9 percent and 6 percent, respectively.
    Eight percent of patients in the control arms and 13 percent of patients in the boceprevir arms dose reduced the study drug for neutropenia. Zero and 1 percent of patients in the control and experimental arms, respectively, discontinued treatment for neutropenia.
    As discussed in the briefing book, there were five cases of grade 3 and 4 neutropenia temporally related to severe infections or cases that the investigators judged to be life-threatening neutropenia. All resolved with the antibiotics and/or cessation of boceprevir/peg/riba therapy.
    There were also incremental increases in thrombocytopenia associated with the addition of boceprevir at the standard of care. Interferon causes a decrease in circulating platelets, sometimes leading to thrombocytopenia.
    When boceprevir was added to the
    peginterferon plus ribavirin backbone, there was an increase in subjects with thrombocytopenia during therapy. Grade 3, less than 50 x 10 to the 9 cells per liter was increased from 1 percent in the control to 3 percent with boceprevir combination therapy. No severe events were reported in association with thrombocytopenia.
    Dysgeusia, although clearly increased with the addition of boceprevir to the peg/riba backbone, was mild, transient, and rarely required medical intervention. Some of the verbatim terms coded to this event included metallic taste in mouth, earthy aftertaste, and bitter taste.
    While 37 percent of patients in the key studies experienced dysgeusia, less than 1 percent dose reduced or discontinued drug due to dysgeusia. Dysgeusia was typically mild. Only 1 percent were reported to be severe. None were serious. Dysgeusia events tended to begin early in the course of treatment and were associated with gastrointestinal symptoms, such as nausea, vomiting, and diarrhea.

    Rash, mainly related to ribavirin use, is fairly common with peginterferon and ribavirin therapy. Overall, the rate of rash was similar in the control arm and the boceprevir regimens, 27 percent and 30 percent respectively. Rash was typically described as being consistent with ribavirin rash without mucosal involvement. There were no cases of Stevens-Johnson syndrome, toxic epidermal necrolysis addressed.
    Treatment-related rash AEs, such as pruritis, occurred in similar proportions of subjects in the boceprevir-containing and peg/riba control arms. Rashes were generally mild to moderate in severity and were usually managed with oral antihistamines and topical corticosteroids.
    One percent or less of subjects required dose reductions or study drug discontinuation due to rash, and only one patient in the boceprevir/peg/riba arms reported rash as an SAE. This was an erythematous rash that did not require hospitalization and resolved on full doses of study drugs.

    Safety was examined in subgroups, including age, gender, race, BMI, and co-morbid conditions, such as cirrhosis and hypertension. The degree of anemia was increased in the elderly, defined as those greater than 65 years of age, as well as women and cirrhotics. All three groups had lower baseline hemoglobin.

    Cirrhotic subjects also experienced relatively more thrombocytopenia compared to non-cirrhotic subjects. All these observations have been made previously for peg/riba standard of care treatment. There were no clear differences in adverse events reported by race.
    Response-guided therapy by decreasing treatment duration of all three study drugs offers several safety advantages. Dr. Albrecht has already shown that the increase in efficacy over control for boceprevir RGT arm was similar to that of the 48-week boceprevir/peg/riba arm.
    In particular, SVR rates were very high in all patients with an early response to therapy. The RGT arms were designed to limit drug exposure,
    particularly in these early responders. Response-guided therapy is anticipated to reduce both discontinuations due to adverse events and the duration of adverse events.

    The safety advantages of RGT can be evaluated by comparing safety in early responders in the RGT arm to the corresponding early responders in the peg/riba/boceprevir 48-week arm. Because of this 20-week difference between short therapy and long therapy, this safety difference is most obvious in the naive study, SPRINT-2, as shown here. Mean boceprevir/peg/riba exposure was reduced 37 percent and mean EPO exposure was reduced 46 percent in the early responders in the RGT arm.
    Although there were similar rates of SAEs, there were fewer study drug discontinuations for adverse events, 9 percent compared to 17 percent in the early responders in the RGT arm compared to the corresponding early responders in the boceprevir/peg/riba 48-week arm. And the rate of severe anemia was reduced from 5 percent to
    1 percent with the shorter therapy.

    Additionally, the length of an adverse event when it occurred was decreased for early responders in the RGT arm. For example, in patients who developed moderate to severe depression, a side effect associated with interferon therapy, the mean length of depression was 16 weeks for patients in the RGT arm and 30 weeks for the corresponding early responders in the boceprevir/peg/riba 48-week arm.
    As previously shown for hemoglobin, all hematologic parameters returned to normal after discontinuing therapy. The advantage of the shorter duration of therapy in the early responders in the RGT arm compared to those early responders in the boceprevir/peg/riba 48-week arm is shown here.
    Hemoglobin levels normalize after therapy is completed at treatment week 28 for patients in the RGT arm, yellow line, compared to the continued lower hemoglobin of those still on therapy in the 48-week regimen, orange line.

    Similarly, neutrophils counts normalize quickly for early responders in the RGT arm after therapy is completed at treatment week 28, yellow line, compared to the continued lower neutrophil counts of the corresponding early responders still on therapy in the 48-week treatment regimen, orange line.

    The Phase 3 trials confirmed the use of boceprevir 800 milligrams TID as an efficacious and safe dose to be used in addition to the current standard of care, peginterferon and ribavirin. The clinical pharmacology of boceprevir supports this regimen with no dose adjustments required for special populations.

    Absorption of boceprevir is rapid, with less than dose proportional exposure and no accumulation. Food increases exposure to boceprevir by 40 to 60 percent, regardless of meal type. Boceprevir is extensively metabolized by two distinct pathways, via aldoketorereductase mediated reduction and via CYP3A4 mediated oxidation.
    The half-life is approximately three and a
    half hours, with the majority of boceprevir excreted hepatically and focally. Boceprevir is a strong, reversible inhibitor of CYP3A4 and is not a CYP P450 isoenzyme inducer.
    Merck has conducted several investigations to characterize the drug-drug interactions with boceprevir. Since the biotransformation and clearance of boceprevir involves two different enzymatic pathways, boceprevir is less likely to be a victim of drug-drug interactions with concomitant medications that affect either of these pathways. Investigations of drug-drug interactions with boceprevir do not show clinically relevant changes in boceprevir exposure that would require a dose or schedule adjustment.
    In clinical trials, co-administration of aldoketorereductase inhibitors, such as ibuprofen or diflunisal, did not increase exposure to boceprevir. The published literature on known inhibitors does not show evidence of saturation of aldoketorereductase isoforms to an extent that has observable clinical effects.

    Co-administration of boceprevir with strong CYP3A4 and PGB inhibitors, such as ketoconazole or CYP3A4 inducers, such as efavirenz, while showing some variability, did not change the exposure to boceprevir to a clinically concerning extent. In addition, boceprevir will be administered with standard of care and no interaction has been observed between boceprevir and peginterferon alfa-2b or ribavirin.
    Since boceprevir is a strong inhibitor of CYP3A4, drugs metabolized primarily by CPY3A4, such as midazolam, may have increased exposure when administered with boceprevir. This could increase or prolong therapeutic and adverse effects, and dose adjustment and/or clinical monitoring of the co-administered drug may be warranted.
    Sub-analysis of safety in the Phase 3 trials when boceprevir and standard of care was administered concomitantly in the presence or absence of specific classes of medications, such as CYP3A4 substrates, including oral benzodiazepines or CYP3A4 inhibitors, such as the azoles, suggest
    that, in general, the safety profile was similar. However, a number of additional studies examining drug-drug interactions for drugs metabolized by CYP3A4, including methadone and oral contraceptives, are being prepared.
    No clinically meaningful change in exposure to boceprevir was observed in subjects with hepatic impairment or renal insufficiency or when analyzed for specific demographic factors, such as gender or race. Boceprevir was not associated with clinically relevant effects on cardiac conductions as assessed in a multiple dose thorough QT study.
    To summarize, the addition of boceprevir to peg/riba standard of care is safe and generally well tolerated. Treatment duration of boceprevir/peg/riba therapy is not limited by toxicity and safety data available for 44 weeks of boceprevir administration. The safety profile of boceprevir combination therapy largely reflects the known profile of peginterferon and ribavirin with incremental hematologic effects and dysgeusia, which are manageable.

    No dose adjustment of boceprevir is required when co-administered with other medications or in the treatment of special populations, including those with hepatic impairment. Response-guided therapy leads to decreased exposure to all three study drugs for nearly half the patients.
    I would now like to introduce Dr. Keith Gottesdiener, who will review for you the risk-benefit of boceprevir combination therapy.


    Sponsor Presentation – Keith Gottesdiener
    DR. GOTTESDIENER: Thank you.

    Boceprevir added to standard of care has a positive benefit-risk ratio, with increased SVR rates seen in every population treated and no new safety concerns that limit the duration of therapy.
    The current standard of care for chronic genotype 1 HCV infection is associated with low rates of SVR, particularly in certain patient populations. There is a need for therapies that increase rates of SVR without adding new safety concerns. Boceprevir, a first generation HCV protease inhibitor, meets this need.
    In pivotal studies of treatment-naive and treatment failure patients, boceprevir increased SVR rates nearly twofold and threefold, respectively, compared to standard of care and also decreased relapse rates.
    Improved SVR rates were demonstrated in all study populations, with substantial improvement in the most difficult to treat groups, such as black patients and those with advanced hepatic fibrosis or cirrhosis. Response-guided therapy also offered a shorter duration of therapy for nearly half of patients.
    I've shown in the briefing book the durability of SVR has been documented in follow-up studies that have been conducted for more than two years. No cases of late virological relapse were identified.
    Consistent with ICH guidelines for the evaluation of drug safety, the clinical development program for boceprevir provides a safety database that is extensive both in terms of size and duration. Overall, the addition of boceprevir to
    the peginterferon and ribavirin standard of care has been generally well tolerated, with low discontinuation rates due to adverse experiences. No specific safety issues have been identified that would preclude the use of boceprevir for the proposed indication.
    The addition of boceprevir to interferon and ribavirin results in a safety profile that largely reflects that of standard of care therapy alone. Incremental adverse experiences are generally monitorable, manageable, and reversible. And safety has been demonstrated with boceprevir treatment for up to 44 weeks, and there are no specific toxicities that limit duration of therapy.

    Anemia is a known adverse event associated with standard of care treatment of HCV. Boceprevir added to standard of care causes an additional decrease in hemoglobin of about 1 gram per deciliter, but this can be managed with the current tools used to manage anemia with standard of care therapy; in other words, ribavirin dose reduction and/or erythropoietin. Both strategies are
    associated with increased rates of sustained virological response. Of note, there were smaller effects on other hematological parameters, such as neutrophils.
    Boceprevir affects the plasma concentration of drugs which are metabolized by CYP3A4; and so concomitant administration of drugs that are CYP3A4 substrates and have a narrow therapeutic index should be avoided. Additional studies we'll be initiating shortly to complete the profile for potential drug-drug interactions. Boceprevir is contraindicated in pregnancy due to its required use with ribavirin, a known teratogen.
    HCV-resistant associated variants, or RAVs, occur in many non-SVR patients. Long-term, ongoing follow-up studies show that RAVs diminish over time, but the clinical implications of RAVs are unknown.
    Some of these risks persist for as long as therapy is continued, so an early futility rule would be a benefit. This might be particularly important for treatment-naive patients, where, in
    SPRINT-2, the FDA and the sponsor agreed on a futility rule at treatment week 24, at which time patients with detectable HCV RNA were discontinued.
    With the help and the support of the FDA, the sponsor has just completed an extensive post hoc evaluation of alternative early futility rules.
    Now, we will briefly present our proposal for which the FDA is in general conceptual agreement, although the details are still under discussion. This slide is not in your briefing booklet, but the proposal can be explored in the Q&A, if desired.
    The goals of these evaluations were, one, to preserve SVR rates for those patients who might achieve SVR with further therapy; in other words, not to discontinue patients who might achieve an SVR.
    Number two, stop therapy early in patients unlikely to ever achieve SVR with continued therapy; in other words, to prevent the development of RAVs and adverse experiences in patients who have little, if any, chance of achieving SVR. And three, to harmonize futility rules across populations to reduce the complexity of the treatment regimens.\

    The proposed futility rule would discontinue treatment if patients had HCV RNA levels greater than 100 units at treatment week 12 or detectable HCV RNA at treatment week 24. In order to minimize confusion and based on the public health benefit of having a consistent rule for both populations, this rule will be similar for both treatment-naive and treatment failure patients.
    This rule meets the goals above and would identify a substantial number of patients who would meet early futility, thereby minimizing patients who would receive continued therapy without benefit but would also preserve SVR rates.

    An early futility rule is one approach to maximizing the benefit-risk ratio for boceprevir. The use of a response-guided therapy paradigm is a complementary approach that further increases the benefit-risk assessment of boceprevir combination therapy.

    Early responders, representing approximately half of all patients, attained similar and very high SVR rates when treated with a short RGT regimen versus 48 weeks of therapy in the trials for both treatment-naive and treatment failure patients. Additionally, with RGT, the risks of therapy are less due to shorter exposure to all three drugs, boceprevir, peginterferon, and ribavirin. However, based on limited data available from these trials, patients with cirrhosis should receive 44 weeks of boceprevir.

    After very recent scientific discussions with the FDA, the sponsor agrees it would be appropriate to recommend a duration of boceprevir therapy that is greater than 24 weeks in treatment-naive late responders. This is reflected on this slide, which also is not included in your briefing booklet.

    Therefore, the proposed dosage administration should be, number one, all patients will initiate treatment with four weeks of lead-in with peginterferon/ribavirin, after which
    boceprevir at 800 milligrams TID will be added to the regimen; number two, response-guided therapy would be the optimal regimen for most patients; number three, after the lead-in, early responders treatment-naive patients would receive 24 weeks of additional boceprevir/peg/riba, while early responders treatment failure patients would receive 32 weeks of additional boceprevir/peg/riba.
    After the lead-in, late responders, in this case, both naive and prior treatment failure patients, would receive 32 additional weeks of boceprevir/peg/riba, followed by 12 weeks of peg/riba alone.
    As mentioned previously, cirrhotics, defined either clinically or histologically, would receive a longer duration of therapy, lead-in plus 44 weeks of therapy, triple therapy.
    Should the advisory committee and the FDA agree that boceprevir can be used to treat historical null responders? Null responders would also receive longer duration of therapy, lead-in plus 44 weeks of triple therapy.

    Overall, the benefits of adding boceprevir to the peg/riba standard of care provide a favorable benefit-risk assessment and support the proposed indication for boceprevir. Boceprevir is indicated, in combination with pegylated interferon and ribavirin, for the treatment of HCV genotype 1 infection in previously untreated and previous treatment failure adult patients with compensated liver disease.
    Genotype 1 is the most common form of HCV infection in the U.S. and in Western Europe. The addition of boceprevir to peginterferon/ribavirin standard of care represents the next major advance in the treatment of the disease.

    Thank you very much for your kind attention. And I'll invite Dr. Jan Albrecht back to the podium, who will respond to clarifying questions.


    Clarifying Questions from Committee to Sponsor
    DR. CARGILL: Thank you very much.
    Clarifying questions from the committee for the sponsor? Dr. Ellenberg?
    DR. ELLENBERG: Could you clarify for me the
    follow-up procedures for subjects, the follow-up for safety, including people who were taken off after the lead-in, including people who were taken off after 12 weeks, including people who were taken off because of adverse events, whatever?
    I would like to know what the protocol was for following for safety and whether all these subjects continue under follow-up for safety.

    DR. ALBRECHT: The protocol specified that patients who discontinued early or were taken off for any reason would enter a six-month follow-up period. So if a patient discontinued therapy at 12 weeks, for example, they would go then into the 24-week or six-month follow-up period.
    In addition, the study that we mentioned in the briefing book, the long-term follow-up study, all patients, whether responders, non-responders, who completed the six-month follow-up period were invited to join a three-year long-term follow-up study in which they're being followed both for safety, severe adverse events, and they're also being followed for HCV RNA, either loss of response
    or presence of HCV RNA.

    DR. ELLENBERG: So if they were discontinued, they have a six-month follow-up period. And if they didn't discontinue and completed the study, then how long were they -- was follow-up also discontinued after six months?
    DR. ALBRECHT: The same criteria, that the patient completed the study, they had a six-month follow-up period, at which time they were invited to join the three-year follow-up study.
    Not all patients accepted to join the three-year follow-up study, I should comment.

    DR. CARGILL: Dr. Knodell?
    DR. KNODELL: I was just wondering. Maybe in Europe they have breakfast at 6:00 and dinner at 2:00 or lunch at 2:00 and dinner at 10:00, but in the United States, that would be unusual timing for meals. And since you think that food is important, how important is this QA hour dosing in terms of efficacy? Because that might be hard to enforce in the United States.
    DR. ALBRECHT: The food that the patient was
    asked to take did not need to be an entire meal. It could be just simply a cracker or a small snack.
    We suggested that they dose in a 7 to 9-hour window, so it was essentially three times a day dosing. And we also did look at the effect of adherence to the boceprevir dosing regimen on the SVR rate, and what we found was that in a group of patients who were generally adherent to their dosing overall, their dosing schedule, that the adherence to the strict 7 to 9-hour window did not have a substantial effect on the SVR. However, we're not recommending that the drug be dosed other than the 7 to 9-hour window that was used in the clinical trials, because that's what we studied.

    DR. CARGILL: Dr. Clay?
    DR. CLAY: I had another question, but I want a follow-up to his question, because I think what he's getting at is forgiveness in a dosing regimen of TID or every eight hours. And your response was that if they were late with their medicines, it didn't really make a difference.
    Did you capture in the clinical trials what

    your definition of late constituted?
    DR. ALBRECHT: Yes. The data that I just referred to, we used an electronic dosing diary that the patient completed every day, and we collected the fact that they not only took their drug, but when they took it. And when we looked at patients who were generally adherent in taking their drug, meaning that they took their three drugs on the daily schedule they were supposed to, we saw that in the patients that were, say, 80 percent adherent to the seven --
    DR. CLAY: No, I understand that, but you're talking about total daily. I'm talking about hours; was there a window of hours?
    Is that what you were getting to?
    DR. ALBRECHT: Yes.
    DR. CLAY: Okay. I'm sorry, apologize for interrupting.
    DR. ALBRECHT: We actually tracked whether they took their dose according to the hours, the 7 to 9-hour window. And what I'm saying, in those people that were generally adherent taking their
    drug, they were also adherent to the 7 to 9-hour window. But there were some cases where patients didn't adhere to it, and it doesn't appear to make a difference.
    DR. CLAY: And generally refers to what percentage? I mean, we're HIV people. We're used to talking adherence rates here. So that's why I'm asking.

    DR. ALBRECHT: Could I have slide 2634, please?
    DR. CLAY: I knew you'd have it.
    DR. ALBRECHT: This is the adherence to the boceprevir dosing schedule. If you'll look at the left panel of the screen, it's the proportion of time they were adherent to the 7 or 9 hours. And this is the SVR in the patients that we were tracking. And as you can see, the SVRs tend to be very similar, someplace between 60 and 70 percent.
    DR. CLAY: Thank you. I apologize, but my original question was you mentioned the drug is broken down and you give the impression it's equally between the CYP3A4 pathway and the AKR
    pathway. And I see from a recent publication from Merck that it's AKR1C2 and AKR1C3. And I'm curious if you have looked at the expression difference in black patients versus non-black patients in your trial to see if that factored into your response rates, because you get two active metabolites. You get M28 and M31 and the activity level of those.
    I'm just wondering if you've looked at that.

    DR. ALBRECHT: Could I ask Dr. Claudia Kasserra from clin-pharm to answer the question, please?
    DR. KASSERRA: We have not conducted an analysis of the different AKR isoforms in different subpopulations.
    DR. CLAY: Because it's in the same family as NADPH, which may have some role in the anemia which you talked about. You saw good SVRs in people with anemia. So I'm just -- it's all tied together. It's all one person. Thank you.
    DR. CARGILL: I'm going to pick up, as the chair, with some additional questions. Beginning with your slides in 89, looking at some of the
    hematological profile consequences, do you have those also broken out by race, as these cohorts are lumped for this data?
    [Pause.]
    DR. CARGILL: If you'd like, we can come back to that later since it looks like --
    DR. ALBRECHT: Could we come back to that one in just a moment?
    DR. CARGILL: That's fine. But I do want to leave that question standing on the table.
    DR. ALBRECHT: We have it. I'm sorry. It took us a minute to get the slide up.
    DR. CARGILL: That's fine. You have it?
    DR. ALBRECHT: Slide 860, please. This is the same slide that we showed you for the overall hemoglobin. This is broken out on the left panel. This is the WHO grading. On the left panel is the non-black and on the right panel is the black. As you can see, there are much smaller numbers among the black patients, but there essentially is really no difference between the black and the non-black with the hemoglobin nadirs that we saw.
    DR. CARGILL: Thank you.

    Dr. Strader?
    DR. STRADER: Thank you.
    I have two questions. The first one is that your definition of early response or what prompted you to use response-guided therapy varied, whether the patients were naive or had prior non-response.
    In the naive, it was negative HCV RNA from weeks 8 through 24, and in the patients who did not have a -- who would had previous treatment, it was 8 through 12.
    Can you clarify that that's correct, first of all, and why you chose to make those different from the patients who were naive versus patients who had a prior treatment to non-response?
    DR. ALBRECHT: Yes, that is correct, and the reason being the futility rule. The futility rule was treatment week 24 in the treatment naïve, and the futility rule in the previously treated patients was treatment week 12. So if a patient became HCV RNA negative at treatment week 8, they were required to remain that way all the way to the
    futility rule, when we allowed them to either stop or go on.
    In the previously untreated patients, the futility rule was at treatment week 12. And in the study, a patient could come out at treatment week 12. So the patient was required to be HCV RNA negative at treatment week 12 and then was rechecked, if he passed that criteria, at treatment week 36, when the decision was made whether they go to 48 weeks or 36 weeks.
    DR. STRADER: And the futility rule at treatment week 12 for patients who had a prior response -- I'm just curious, because you make an argument about including null responders, because you claim that there's no real -- that people who do not achieve HCV RNA negativity in the first four weeks, that there's not a big difference between those.
    So I'm curious as to why people who had prior response had to stop at treatment week 12 as opposed to 24.

    DR. ALBRECHT: The futility rule for the
    previously treated patients was based on a large study that was published by Poniard about two or three years ago in which we re-treated patients who had previously failed therapy with peginterferon plus ribavirin. And when we ran that study, what we determined was that if a patient had previously been treated did not respond by treatment week 12, it was highly unlikely that they would respond. And we selected the treatment week 12 futility rule based on that study. And, actually, that is a labeled indication.

    DR. STRADER: Okay. In your study looking at blacks versus non-blacks, the numbers are -- it seems to be in each treatment group probably fivefold as many non-blacks as blacks. I assume you did some sort of calculation to determine what the number of blacks in each group would need to be in order to determine whether your differences in response were significant or not; is that correct?

    DR. ALBRECHT: That's correct. We enrolled the two cohorts separately because we wanted to assure ourselves that we would have enough blacks
    in the study to be able to look at it and hopefully achieve a statistically significant difference.
    This was a multinational study conducted in other countries in addition to the United States, and, of course, when we go overseas, and particularly Western Europe, we don't accrue very many black patients. So we enrolled the non-black cohort and the black cohort separately to assure that there would be an adequate number of patients. And as you saw, the deltas were statistically significant between the controls and the treated groups.
    DR. STRADER: So that the 50 patients in each group or thereabouts was considered to be enough to be able to determine a difference.
    DR. ALBRECHT: We had been able to determine, based on our Phase 2 study, that we could have an expectation that the differences in response rates would be quite large, and what was how we decided on our sample sizes.
    DR. CARGILL: We will now take a 15-minute break, but let me just first make sure that I
    assure the panel members who had raised questions that we're not going to forget. You are definitely going to come back to them.
    In addition, we'll also have Dr. Van Dyke introduce himself, since he's been able to join us.
    So, please, before you leave, especially panel members, please remember that there should be no discussion of the meeting topic during the break amongst yourselves or any members of the audience. We will resume promptly at 10:30. Thank you.
    (Whereupon, a recess was taken.)
    DR. CARGILL: We are going to resume the meeting of the advisory panel. However, first, we would like to have Dr. Van Dyke introduce himself, as he was not here at the beginning of our meeting.
    DR. VAN DYKE: Good morning. I'm Dr. Russell Van Dyke from the Department of Pediatrics, Tulane University in New Orleans. Unfortunately, New Orleans doesn't have a whole lot of flights and mine was canceled last night. So I had to catch an early one this morning. So I apologize for being late.

    DR. CARGILL: Welcome.
    Again, just to repeat that we will resume the questions this afternoon. We'll pick up so that the advisory panel members who did not have a chance to ask a question or have additional questions, we will continue at that time.
    We will not proceed with our presentation from the FDA. I would like to remind our public observers at this meeting that while this meeting is open for public observation, public attendees may not participate except at the specific request of the panel.
    FDA Presentation – Poonam Mishra
    DR. MISHRA: Good morning. Today, Dr. Jeffrey Florian and I will be presenting the FDA analysis for boceprevir on behalf of the review team. This presentation will include the key safety and efficacy results and issues. I will try to focus only on the analysis where FDA has noted some differences from those of applicant and will not include the topics which have already been covered in the applicant's presentation earlier
    today.

    I will first start with the safety of boceprevir, focusing on anemia and other bone marrow suppressive effects. Then I will go over some of the evidence data currently available for boceprevir.
    I will briefly go over the key terms and the trial designs to rearticulate the difference in the duration of therapy and the response-guided therapy arms in the two Phase 3 trials. I will discuss the primary efficacy results and pertinent subgroup analysis. Then Dr. Florian will discuss the issue of proposed indication in non-responders and optimal duration of therapy in treatment-naive late responders.
    The most notable safety concerns with boceprevir are hematologic adverse events, mainly anemia, and to a lesser extent, neutropenia and thrombocytopenia.

    The two Phase 3 trials, which have been analyzed for hematologic adverse events include SPRINT-2, done in previously untreated subjects. I
    will also be referring to these subjects as treatment naive during my presentation today.

    RESPOND-2 was done in subjects who had failed the previous course of pegylated interferon and ribavirin therapy. I will also be referring to these subjects as treatment-experienced during my presentation.
    Please note that the sponsor also pulled the data from the Phase 2 open label trial, SPRINT-1 in treatment-naive subjects; hence, the analysis presented here may differ from those of the applicant's presentation earlier.

    Anemia was reported as an adverse event in 52 percent of boceprevir-treated subjects compared with 30 percent of controls; that is, treated with pegylated interferon and ribavirin therapy, also referred to as PR. In these trials, investigators were advised to intervene when hemoglobin concentrations fell to 10 grams per deciliter or lower. The definition of an anemia adverse event included any laboratory value resulting in an intervention; hence, anemia adverse event reporting
    was linked to the appearance of an intervention for anemia management.
    This table shows an analysis of number and proportion of subjects who reached hemoglobin levels of less than 10 grams per deciliter and less than 8.5 grams per deciliter in the Phase 3 trials.
    A higher proportion of boceprevir-treated subjects compared to those subjects who received PR experienced hemoglobin levels of less than 10 grams per deciliter or less than 8.5 grams per deciliter in the Phase 3 trials, as shown in this table.

    Anemia was reported as a serious adverse event in 12 boceprevir-treated subjects compared to one subject in the PR control arm. Anemia resulting in drug discontinuation, dose reduction, or dose interruption was also higher in the boceprevir-containing arm compared to the PR control arm, shown in the rows below.
    Before I discuss any further, I would like to point out some of the challenges in the assessment of anemia in these trials. Assessment of anemia was confounded by baseline hemoglobin
    values. Subjects with lower baseline hemoglobin had higher rates of nadir hemoglobin less than or equal to 10 grams per deciliter. These subjects also had more anemia adverse events reported.
    Subjects with lower baseline hemoglobin had higher rates of interventions for anemia management. Subjects with lower baseline hemoglobin also had smaller magnitude of hemoglobin decline, and this I will discuss a little further.

    Boceprevir-treated subjects experienced additional decline in hemoglobin. Mean hemoglobin decline beyond that seen with the PR arm was approximately 1 gram per deciliter. In some subjects, additional hemoglobin decline was greater than 1 gram per deciliter. However, assessment of risk by absolute maximum hemoglobin decline is difficult to interpret because of differential post-baseline interventions for anemia management in these trials.

    This graph shows the mean decline in hemoglobin by treatment arm and visit. The X-axis denotes treatment visit and the Y-axis represents
    hemoglobin decline. If you focus your attention on the left side of the graph, you will see that all arms had an initial decline in hemoglobin during the first four weeks of lead-in therapy with pegylated interferon and ribavirin, and the magnitude of decline is similar in all arms.
    Boceprevir was added on at treatment week 4 and, as you can see, there was an additional decline in hemoglobin in boceprevir-containing arms, shown below the graph, whereas the red line shows the control arm.
    You can see that in the treatment-naive trial, boceprevir was stopped in the RGT arm at week 28 and hemoglobin values returned to the same mean level as that with the PR alone. Similarly, when boceprevir was stopped in the RGT arm of the treatment-experienced trial, hemoglobin returned back to the PR level. Hemoglobin values returned to baseline values after all the study drugs were discontinued.
    This graph shows the mean hemoglobin decline in subjects with baseline hemoglobin less than or
    equal to 14 grams per deciliter compared to those with baseline hemoglobin greater than or equal to 16 grams per deciliter.
    As shown in the graph on the left side, subjects with lower baseline hemoglobin values, that is, less than or equal to 14 grams per deciliter, the magnitude of additional boceprevir-associated hemoglobin decline was lower than those in subjects with baseline hemoglobin greater than or equal to 16 grams per deciliter, as shown in the graph on the right. This could be partly due to the early interventions for anemia in subjects with lower baseline hemoglobin.
    These figures show the exposure-response relationship for anemia. An upward trend of increasing incidence of anemia was observed with increasing boceprevir AUC in the PK population. However, this trend was not significant. A significant relationship between incidence of anemia and ribavirin AUC was observed in the PK population receiving triple therapy.

    In Phase 3 trials, as shown in our
    background document, a shallow and non-significant relationship was identified between boceprevir exposure and SVR. A non-significant, but upward trending relationship between ribavirin steady-state AUC and SVR was also observed. These data also suggest that improvement in SVR rates in subjects who develop anemia compared to those who did not develop anemia during treatment may be related to higher ribavirin exposures.
    Use of erythropoietins as stimulating agents, also referred to as ESAs, was allowed for the management of anemia in these trials at the investigator's discretion. This slide shows that the proportion of subjects receiving ESA was higher in the longer duration boceprevir treatment arms compared to response-guided therapy arms. This pattern continued with varying ESA treatment duration as shown in the table. Overall, ESA use was higher in the boceprevir treatment arms than the PR control arm.
    ESAs are not FDA approved for the treatment of anemia in patients with chronic hepatitis C.
    ESA use in itself may potentially pose an additional safety risk, the extent of which has not yet been fully described. ESA use has been associated with increased incidence in thromboembolic events. There were a few thromboembolic events reported in these trials, including a case of arterial thrombosis in one subject. A case of pure red cell aplasia, which is a rare erythropoietin side effect, was also reported in one subject during the follow-up period.

    Some adverse events which may represent clinical manifestations of anemia, such as dyspnea, exertional dyspnea, dizziness, and syncope were reported more commonly in the boceprevir-containing arms compared to control. Other adverse events of interest, such as myocardial infarction, myocardial ischemia, cerebral accident or ischemia, were reported too infrequently in these trials to assess differences in proportions between the arms. Trials were not designed to specifically assess incidence of symptoms and events associated with anemia.

    This table shows that dyspnea, dizziness and syncope were numerically more frequent in boceprevir-containing arms compared to control arm. Analysis of anemia events was also done by treatment arm and gender. Overall, it appears that females had more anemia events reported. However, it should be noted that subjects who had lower baseline hemoglobin levels, particularly females, were more likely to receive an intervention for anemia and have that intervention earlier, and, thus, would be reported more frequently as having had an anemia event. When you look at the lower row, this is a subset of subjects with similar baseline hemoglobin. This difference is not seen.
    In the Phase 3 trials, grade 3 and grade 4 neutropenia was reported more frequently in boceprevir-containing arms compared to control. Neutropenia was reported as a serious adverse event in three subjects in boceprevir-containing arms compared to none of the subjects in the control arm.

    Neutropenia resulted in study drug
    discontinuation in eight subjects in boceprevir-containing arms compared to none in the control arm. As presented by the applicant earlier, GCSF use was allowed in the Phase 3 trials.
    Three subjects, all in boceprevir-containing arms, experienced severe infections within two weeks of grade 3 and 4 neutropenia. These include a case of epiglottitis, which was life-threatening and required tracheostomy; an upper respiratory tract infection resulting in hospitalization; a case of severe salmonella gastroenteritis.
    Additionally, two cases of life-threatening neutropenia, both in boceprevir-treated subjects, were reported in Phase 2 open label trials. One subject developed multi-organ system failure due to sepsis and the other experienced a fever of 104.5 degrees Fahrenheit. A specific infection was not reported in these cases.
    In Phase 3 trials, as shown in the table, a higher proportion of subjects in boceprevir-containing arms than the PR arms experienced grade 3 or 4 lowest platelet count on treatment.

    Thrombocytopenia was reported as a serious adverse event in three subjects in boceprevir-containing arms compared to none in PR arms. Thrombocytopenia resulted in study drug discontinuation in four subjects in boceprevir-containing arms and in none of the subjects in the PR control arms. No cases of significant bleeding were reported in Phase 3 trials. However, one of the subjects received multiple platelet transfusions because of severe thrombocytopenia.
    In conclusion, the most notable safety concern is the additional decrease in hemoglobin above and beyond that observed with pegylated interferon and ribavirin therapy alone. Boceprevir-treated subjects experienced more anemia, neutropenia, and thrombocytopenia. These appear to be part of an overall bone marrow suppressive effect of boceprevir.
    Anemia appeared to be managed effectively during the clinical trials and was reversible after all the study drugs were discontinued. A few serious or life-threatening infections were
    reported. Close monitoring of laboratory parameters is recommended in clinical practice if drug receives marketing approval.
    Now, I will briefly summarize the clinical resistance issues.

    This table summarizes a pooled analysis of substitutions that emerged in boceprevir-treated subjects who failed to achieve SVR in either of the two Phase 3 trials. Slightly more than half of boceprevir treatment failures had at least one of these specific boceprevir treatment emergent substitutions detected. Patterns of treatment-emergent substitutions were different in genotype 1a and genotype 1b subjects, as shown in this table.
    A relatively small proportion of study subjects had one or more of the major boceprevir treatment emergent substitutions detected as baseline polymorphism. These substitutions were detected using population-based nucleotide sequence analysis. So the viral populations carrying these substitutions are highly evident in these 40
    subjects.

    A pooled analysis of SVR rates from these subjects was conducted to explore the effect of having detectable baseline boceprevir resistance-associated substitutions on treatment efficacy. As shown in this table, this analysis is limited by the small number of subjects with these substitutions at baseline, but it appears that these detectable substitutions reduced boceprevir efficacy among subjects with less than 2 log decline in response to PR therapy compared to those without baseline substitutions; a possible effect on boceprevir efficacy among subjects with a relatively poor virologic response to pegylated interferon and ribavirin therapy based on lead-in period.
    Certain boceprevir treatment emergent substitutions may persist at a high level for years in some patients who fail a boceprevir-based treatment regimen. Based on an analysis of long-term follow-up isolates from subjects who failed to achieve SVR in Phase 2 boceprevir trials,

    approximately 25 percent of subjects with one or more boceprevir treatment emergent substitutions still had at least one such substitution detected after 2.5 years of follow-up.
    The substitutions that remained detectable were mostly T54S and R155K. Note that these data are based on population nucleotide sequence analysis, which typically does not detect variants that comprise less than 25 percent of the total viral population.
    Now, let's look at some of the efficacy results and issues. Some of the terms commonly used to describe treatment response in patient populations were earlier described by Dr. Birnkrant. It should be noted that applicant's non-responders include partial responders, but does not include null responders, as defined above.
    I'll go over the primary efficacy results in the treatment-naive population.
    This slide shows the trial design for the treatment-naive population, which has been already presented by the applicant. As you know, this
    trial had three arms. Arm 1 was the control arm with pegylated interferon and ribavirin therapy for 48 weeks. Arm 3 was a triple therapy arm with pegylated interferon and ribavirin for an additional 44 weeks after four weeks of lead-in. All three arms had four weeks of lead-in therapy with PR. Arm 2 was the response-guided therapy arm, and subjects in this arm received a different duration of therapy depending on virologic response at weeks 8 through 24. I will discuss this later.
    Two separate population cohorts were enrolled in the treatment-naive trial; cohort 1, non-black subjects, and cohort 2, black subjects. However, for the primary endpoint analysis, cohorts 1 and 2 were combined. Overall SVR rates were 63 to 66 percent in boceprevir-containing arms compared to 38 percent in controls. Relapse rates were lower in the boceprevir-treated subjects compared to controls.
    SVR rates were lower in cohort 2, black subjects, than in cohort 1, non-black subjects, for both boceprevir treatment groups, arm 2 and arm 3.

    But this was also true with the PR control arm. Within each cohort, SVR was higher in both boceprevir treatment arms than in the PR control arm. In cohort 2, black subjects, there was an 11 percent numerical difference in SVR between the RGT boceprevir arm and the 48-week triple therapy boceprevir arm. This is of some concern and could be of clinical significance.

    Now, I will walk you through the response-guided therapy arm in the treatment-naive trial. As I said earlier, all subjects received four weeks of lead-in therapy with pegylated interferon and ribavirin, which is also referred to as standard of care therapy.
    All subjects received 24 weeks of boceprevir in combination with PR after the first four weeks of lead-in period. For subjects with undetectable HCV RNA at treatment week 8 through treatment week 24, all three drugs were stopped at week 28, and these are also referred to as early responders.
    While those with detectable HCV RNA at week 8, but undetectable at week 24, boceprevir was
    stopped and subjects received an additional 20 weeks of PR therapy. These subjects are also referred to as late responders. For subjects in each of the treatment arms, all treatment was discontinued for futility if HCV RNA was detectable at week 24.

    This analysis shows the comparison of early responders in the RGT arm to those subjects in arm 3 who had a similar virological response. Early responders were compared to the subjects in arm 3 who were HCV RNA undetectable at treatment week 8 up to treatment week 24 and received treatment duration greater than 31 weeks. SVR was similar in both arms, in the RGT arm, as well as triple therapy arm 3, 97 percent versus 96 percent.
    In late responders, subjects in the RGT arm who were assigned to longer therapy were compared to subjects in arm 3 who had HCV RNA detectable at treatment week 8 or at any subsequent visit up to treatment week 24.
    SVR was numerically higher, approximately 9 percent, in the triple therapy arm 3, than the

    RGT arm 2 and these late responder subjects. This difference was not statistically significant, but the trial was not designed to detect differences in this subgroup.
    Our analysis differs from that presented by the applicant, and then I will explain it in the next slide. Arm 2 was the RGT arm. Subjects who were undetectable from treatment week 8 through treatment week 24 are referred to as early responders. However, subjects who were detectable at treatment week 8, but undetectable at treatment week 24 are referred to as late responders.

    There were 14 subjects who should have received 28 weeks of treatment duration, HCV RNA undetectable through treatment week 8 through treatment week 24, but received longer duration of therapy and, hence, were excluded from the FDA analysis.
    These 14 subjects in arm 2 were really early responders but received the wrong duration of therapy because of detectable HCV RNA results that were not confirmed with a second analysis. One
    subject should have received 48 weeks of treatment based on HCV RNA detectable at treatment week 24, but received a shorter duration of therapy of 28 weeks treatment. Hence, this subject was also excluded from our analysis.
    There were some subjects classified as other who discontinued treatment prior to week 28 assignment due to reasons for treatment failure or discontinuations due to adverse events, and these are not included in the analysis, as well.
    On the right side of the slide you will see the triple therapy arm, which is boceprevir and PR for 44 weeks after lead-in therapy. And the subjects who received more than 31 weeks of treatment and who had any positive from treatment week 8 through treatment week 24 were compared with late responders in the RGT arm. So the subjects -- the early responders in the RGT arm were compared with the similar responders in arm 3, and those late responders in the RGT arm were compared with the similar responders in arm 3.

    So just to reiterate, this 9 percent
    difference between RGT arm and arm 3 and late responders, cohort 1 is very similar to overall. Cohort 2 is similar, although the differences are bigger. But you should note that these numbers are very small, and this is not statistically significant.
    Now, I'll discuss efficacy results in the treatment-experienced trial. This is a trial designed for the treatment-experienced trial, and, as you may recall, arm 1 and arm 3 are similar to the treatment naive trial. The only difference here is in the RGT arm, which is the response-guided therapy arm.

    Subjects received different treatment durations depending on virologic response at weeks 8 and 12, and I will discuss that later. In all treatment arms, subjects with detectable HCV RNA at week 12 discontinued all therapy for futility and were considered treatment failures.
    This slide shows the primary efficacy results for treatment-experienced population. SVR was higher and relapse rates were lower in both
    boceprevir-containing arms compared to the PR control arm. However, SVR was numerically 7 percent higher in arm 3 than in the RGT arm in this population. The 7 percent difference in SVR between the two arms was due to differences observed by subjects in each arm who were receiving identical therapy prior to week 36, and this may be due to the differences in responses in the subgroup of subjects with cirrhosis, as pointed out by applicant earlier. Within the boceprevir treatment arms, subjects who were previous relapsers had higher response rates than those who were previous partial responders.

    This slide shows you the response-guided therapy for the treatment-experienced trial. All subjects received four weeks of standard of care therapy, followed by eight weeks of boceprevir and standard of care therapy.
    Virologic response was checked at treatment week 8 and treatment week 12. Subjects with an undetectable HCV RNA at week 8 completed all therapy at week 36, and these are referred to as
    early responders, while those with detectable HCV RNA at week 8, but undetectable HCV RNA at week 12 received triple therapy through week 36, followed by an additional 12 weeks of PR alone. These are referred to as late responders.
    In all treatment arms, subjects with detectable HVC RNA at week 12 discontinued all therapy for futility and were considered treatment failures.

    This slide shows you the response rates compared for early responders and late responders in arm 2 versus arm 3. Early responders had higher response rates of greater than 90 percent in both boceprevir-containing arms. Response rates were also similar in late responders in the RGT and triple therapy arms.

    Primary efficacy results for advanced fibrosis are patients with cirrhosis at baseline that are presented in this table here. In the treatment-naive trial, subjects both had higher METAVIR fibrosis scores or cirrhosis had lower SVR than those with lower scores or no cirrhosis at
    A Matter of Record baseline.

    Patients with cirrhosis or advanced fibrosis did better with longer triple therapy arm than shorter response-guided therapy arm. This was seen in both treatment-naive and treatment-experienced subjects. We think the difference in SVR between RGT and the longer triple therapy arm is mainly driven by those with cirrhosis.
    To conclude, overall, in the treatment-naive subjects, SVR was 63 to 66 percent in boceprevir-containing arms compared to 38 percent in the control arm. Overall, in the treatment failure subjects, SVR was 59 percent to 66 percent in boceprevir-containing arms versus 21 percent in the control arm.
    The treatment difference was substantially significant and robust for each trial based on the primary efficacy endpoint. Relapse rates were also lower in boceprevir-treated subjects. Efficacy of boceprevir was demonstrative in subjects regardless of race. Response-guided therapy approach in early responders provides a potential shorter duration of
    therapy.

    This concludes my portion of the presentation, and Dr. Florian will continue over the next part of the presentation.
    FDA Presentation – Jeffrey Florian

    DR. FLORIAN: Thank you, Dr. Mishra.

    For my part of this presentation, I am going to cover two key questions. The first question relates to the evidence of effectiveness for prior null responders.
    The applicant provided us information in treatment-experienced subjects who were prior relapsers and prior partial responders. However, the applicant did not explicitly study the prior null responder population. So question 3 to the committee is to weigh on the evidence for use of boceprevir in combination with PR in prior null responders.
    So let me walk you through the rationale to illustrate that the treatment-naive population can inform about drug effect in null responders. Shown here on the left in the pie chart are treatment
    outcomes on PR from treatment-naive P05216. The distribution of end-of-study response rates are expected based on our experience with PR treatment.

    What we have learned in this review is that week 4 decline is an indicator of end-of-study responses. As expected and shown on the right are week 4 viral log declines grouped according to their response. As expected, the responders had the highest decline by week 4, shown at the top, and null responders had the lowest, at the bottom.
    If we select a less than 1 log decline at week 4 to identify null responders, a large portion of the subset will also include subjects who may have been partial responders, relapsers, or responders. Alternatively, we can select the population comprised predominantly of null responders using a less than 0.5 log decline.
    Now, because the lead-in with PR was included in both the control and boceprevir arms, we can look at the week 4 response in the boceprevir treatment arms and derive information on drug effect in prior null responders.

    First, let's focus on the PR-treated subjects who had a less than 1 log decline at week 4, and we see that 69 percent of these subjects were null responders and that 4 percent of the subjects in the PR group achieved SVR.
    For boceprevir-treated subjects with a week 4 response during PR lead-in, less than 1 log, we see that the SVR in these subjects was 28 percent in the RGT arm and 38 percent in subjects treated with boceprevir for 44 weeks. This analysis shows an improvement in SVR with the addition of boceprevir, but may overestimate SVR as this analysis includes a proportion of subjects who are not null responders.
    If we perform the same comparison for less than 0.5 log decline at week 4, we see that 88 percent of these subjects, or 22 out of 25, were null responders. SVR for subjects treated with PR in this group was 0 percent, while subjects receiving boceprevir who had PR lead-in response less than 0.5 had an SVR of 28 to 30 percent.

    Again, an improvement over PR was observed
    with the addition of boceprevir, but a less than 0.5 week 4 response selects a population that is primarily null responders and provides a more conservative estimate of SVR in this population.
    Summarizing these observations, data from subjects with less than 0.5 or less than 1 log decline at week 4 provides information about boceprevir's effectiveness in prior null responders. Also, treatment with boceprevir improves SVR in these subjects compared to treatment with PR.
    Now, it is anticipated that these subjects would be the most difficult to treat and as such, they may benefit from a full 44 weeks of boceprevir treatment regardless of initial response.
    So returning to the original table for treatment-experienced subjects, one option, one possible option for prior null responders would be what is shown on the far left, four-week PR lead-in, followed by 44 weeks of boceprevir plus PR.

    I'd now like to discuss the second question that relates to treatment duration in treatment-
    naive subjects.

    For this question, we would like you to weigh in on treatment duration for the treatment-naive late responders, as we believe this population may require a longer duration of boceprevir and was originally studied and proposed by the sponsor.

    Shown here are the SVR results for the boceprevir treatment arms from P05216, according to the FDA analysis discussed earlier, for early and late responders. As has been presented previously, there was no difference in SVR between early responders receiving treatment for 28 weeks or 48 weeks. However, shown at the bottom of this table, a 9 percent numerically lower SVR rate was observed for late responders in the RGT arm compared to subjects receiving boceprevir for 44 weeks. Based on this observation, we explored what was happening in these late responders once boceprevir treatment was stopped.

    The plots here show percent of treatment-naive subjects with undetectable viral load on the Y-axis at each visit over the course of the trial
    on the X-axis. On the left-hand side, we have treatment-naive early responders, and we see that both early responder treatment arms had similar results, responses, further supporting the selection of 28 weeks of treatment in treatment-naive early responders.
    In contrast, on the left side, which is treatment-naive late responders, the percent of subjects with undetectable viral load separates immediately after removal of boceprevir in the RGT arm at week 28 due to viral breakthrough during treatment with PR. This demonstrates that 24 weeks of boceprevir treatment may be suboptimal in treatment-naive late responders.
    Based on this observation, one treatment option for treatment-naive late responders would be four weeks PR lead-in followed by 44 weeks of boceprevir with PR, and the evidence for this regimen was from the treatment-naive late responder treatment arms where this arm performed better for the RGT in late responders.

    What I'm going to show next is that we may
    have another treatment option to consider by bridging data from the treatment-experienced patients. Before I introduce the information on bridging, let me walk you through the concept that we have learned during this review to help us with this decision.
    Now, when a treatment-naive subject is first treated with PR, these subjects may have one of four treatment outcomes: responder, relapse, partial response, null responders. Those subjects who don't achieve a response would be considered the treatment-experienced subjects.
    Now, when these subjects would be re-treated with PR, what do we expect their response to be and how does this tie into a patient's response when they were first treated with PR?

    Design of these studies has allowed us to make comparison between the treatment-naive subjects from P05216 receiving PR and treatment-experienced subjects from P05101, whose prior PR outcome is already known using their week 4 response. The week 4 response data has also
    allowed us to extend these observations to early and late responder subjects from the boceprevir treatment arms as all treatments included four-week PR lead-in.

    For the first point, it was observed that the week 4 response was similar between treatment-naive subjects treated with PR who relapsed during treatment, shown as the first line in this table, and the treatment-experienced subjects who were prior relapsers. We see the mean response was similar between these two groups.
    Likewise, week 4 response was similar between PR treatment in treatment-naive subjects who had a partial response, shown as the third line in blue, and treatment-experienced subjects who were partial responders, shown as the final line.

    Now, only the median values are shown here, but the intra-quartile ranges were also similar for these groups. This observation tells us that the week 4 responses is similar in subjects whether receiving PR for the first time or having received it previously.

    For the second point, let's look at the week 4 responses in treatment-naive subjects treated with PR who failed treatment, shown as the top three lines, going as relapsers, partial responders, or null responders. If we then look at the week 4 responses for the late responders from either the treatment-naive or treatment-experienced studies, we observe that these subjects have similar week 4 responses to those subjects who went on to fail PR treatment.
    So what we've seen is evidence that the PR response is similar for subjects receiving PR treatment for the first or subsequent times and that the late responders are those subjects who may have failed treatment with PR alone. Together, these observations allow us to make a comparison between late responders in the treatment-naive and treatment-experienced trials.

    The plots here, again, show percent of subjects with undetectable viral load along the Y-axis at each visit over the course of the trial along the X-axis. Now, on the far right we have
    the figure for the treatment-naive late responders again, and on the far left, treatment-experienced late responders.
    What we see is that the late responders in the treatment-experienced trial, which only enrolled prior relapsers and prior partial responders, had similar trends for both boceprevir arms. No separation between the treatment arms was observed once boceprevir was stopped at week 36.
    Also, at the bottom, we observed SVR for late responders from this trial were similar for both arms. It supports 32 weeks of boceprevir in the treatment-experienced late responder population, where null patients were excluded and could be applied to treatment-naive late responders.
    So going back to the tables we have shown, based on these observations and presented as option 2, one treatment option would be four weeks of PR, followed by 32 weeks of boceprevir plus PR, followed by 12 weeks of PR. And the evidence for this treatment regimen would come from the
    treatment-experienced trial and the bridging analysis that we've shown on the previous slides.
    So bringing everything together for the treatment-naive and treatment-experienced patients, treatment options for these patients may look as follows.
    Under the treatment-naive late responder group, both options are presented and will be reduced to one based on feedback from the committee.

    Now, it should be noted that both of the currently proposed treatment options may result in suboptimal treatment duration in a subset of patients based on the currently proposed set of treatment-experienced regimens. What we mean by this is that treatment-naive late responders that are similar to the prior partial responders and prior relapsers may receive unnecessary additional boceprevir treatment if the first option, the PR4, 44 weeks boceprevir plus PR, is used.
    Likewise, treatment-naive late responders, similar to prior null responders, may receive a
    suboptimal treatment duration with treatment option two. However, more granularity in the late responder treatment may overcomplicate treatment for these patients without substantially contributing to subject benefit-risk ratio.

    Clarifying Questions from Committee to FDA
    DR. CARGILL: Clarifying questions from the committee for the FDA? And then we will proceed.

    Ms. Dee?
    MS. DEE: I'm wondering, on the safety, our briefing document talks about the open label use of EPO and the AE reporting system that was used by the sponsor. And it says that interventions actually led to AE reporting and that 21 percent of patients that actually had a less than or equal to 10 GDO decrease were not reported as adverse events and that the discretion of the investigator actually led to -- using the discretion of the investigator instead of hard and fast rules led to differences in reporting and inherent miscalculations and claims that it makes this analysis limited and that only simple
    characterizations are possible.
    I didn't hear that in the FDA presentation.
    Is that still the agency's position? I'm trying to understand if, in fact, these data are reliable when it comes to the number of cases of anemia.
    DR. COOPER: Those issues mainly affected our ability to interpret subgroup analyses, because there were differences across treatment arms within the subgroups with regard to whether interventions occurred and according to baseline hemoglobin.
    So the main issue was with regard to confounding you to baseline hemoglobin and differences in interventions impacted our ability to interpret subgroup analyses. So that was the main limitation incurred by those findings.

    MR. TRAN: Could you state your name into the record, please?
    DR. COOPER: Chuck Cooper from FDA.
    MS. DEE: So in other words, can we expect more cases of anemia and more serious cases of anemia if people aren't allowed to use EPO for
    whatever reason, if this is approved?

    DR. COOPER: I'm not sure we can answer that question from this data. There was no formal comparison made with regard to patients who received ESA versus those who didn't receive ESA as part of the study design.
    MS. DEE: So are the data confounded and biased because of this EPO use and this discretion of the investigators instead of hard and fast rules? Help me understand that.
    DR. COOPER: I think when you're looking overall at what happened, I think you can believe that what happened to the patients is accurately represented according to the interventions that actually occurred. But I think when you start to look at subgroups, such as by race or gender or sex or other subgroup analyses, then these confounders actually -- as you saw with the example of females versus males -- can actually have a significant impact on your ability to interpret the results of the subgroup.
    So that was the main point there, that
    differences in interventions and differences in baseline that exist between each treatment group, within whatever subgroup you're looking at, can have an impact and make the ultimate subgroup analyses difficult to interpret and potentially misleading.
    So it really has to do with subgroup analyses, and that's the reason why we actually didn't present any subgroup analysis, because we didn't have the time to try and do corrections for these different issues. But we fully intend, after we have more time, to explore subgroup analyses, trying to control for these different things.
    MS. DEE: Okay. And one more quick question.
    The data in blacks, what I read when I read the briefing paper was that the agency didn't seem to be confident that response-guided therapy versus 44 weeks of boceprevir and PR was -- in other words, response-guided therapy wasn't sufficient, and that's not what I heard today.

    So what's the issue there?

    DR. MISHRA: So further analyses were done, and, at this point, we think that most of those differences in the responses are driven by subjects with cirrhosis. Applicant has presented that data, and we have confirmed with our own analysis, and we do believe that those differences are mainly driven by subjects with cirrhosis.
    MS. DEE: And the numbers of blacks and people with cirrhosis were enough to make you feel confident that that's correct.
    DR. MISHRA: We have to keep in mind that these subset numbers are very small.

    DR. CARGILL: Dr. Strader?
    DR. STRADER: I have two questions, but first I want to follow-up on the question that she just asked.
    Are you saying that there are sufficient number of black patients in the study with cirrhosis that we can say that we would recommend a method of treatment for them?
    DR. MISHRA: No. I'm not saying that the numbers are sufficient. They are small numbers,
    but still we have seen the patients with cirrhosis or in advanced fibrosis stage at 4. Those with triple therapy did better than response-guided therapy. So our recommendations at this time are that maybe subjects with cirrhosis at baseline should be treated with triple therapy, boceprevir, for 44 weeks.
    DR. STRADER: Including black patients with cirrhosis is what I'm asking.
    DR. MISHRA: Yes. Yes. All patients with cirrhosis.
    DR. STRADER: Okay. And my original question, I'm confused a little bit about the explanation for why null responders should be treated. It appears in the briefing document, in the beginning, they were excluded from the trial because there was insufficient data in the Phase 2 trials to suggest that these patients might respond.
    Now you're saying that based on what happens in the group of naive patients, that there is going to be a subset of people who would be null
    responders; that somehow we should extrapolate that data from the group of naive patients into the group of mull responders.
    There appears to be a difference between people who relapse and people who have partial response to all three drugs. So I'm not quite sure why you think that that doesn't extrapolate to null responders. Someone who did not respond to a prior course of interferon and ribavirin would not be different than someone who had a partial response or a relapse.
    DR. MISHRA: I will have our pharmacometrics team to answer that question.
    DR. FLORIAN: So if I understand the question, it's not so much is this working in the group, but is it asking why the recommendation is at 44 instead of 32?
    DR. STRADER: No. It's asking why you're saying --
    DR. FLORIAN: It's asking why a recommendation --
    DR. STRADER: -- null responders. They were
    specifically excluded, but you're trying to give us a justification for why they can be treated now, and I'm trying to understand your justification for that.
    DR. FLORIAN: You're correct that they were originally excluded, but during the course of analysis, looking at that treatment-naive group, we believe that we can use information from that population to show -- select a subset of patients who if, instead of receiving boceprevir with PR, if they had received just PR, would have resulted in a null response, and then perform that comparison between what those response rates -- just get an idea on what an SVR rate might be in this population had it actually been studied.
    DR. STRADER: So you don't believe that a prior null responder, someone who did not respond to pegylated interferon and ribavirin in the very beginning, is different than a prior relapser or a prior partial responder is what you're saying.
    DR. FLORIAN: No. We are saying they're different.
    DR. STRADER: In what way? You're saying you can use data from people who have not been treated and compare the two.
    DR. FLORIAN: So in the prior -- when you're looking at the treatment of experienced, we are saying they should be treated different, those two groups.
    DR. JADHAV: Pravin Jadhav, team leader, pharmacometrics. Let me offer more help here.
    I think the question you're asking is, are prior relapsers and partial responders, do we think they're different. And there are data to show that the applicant's analysis, as well as our analysis, has shown that SVR in that combined population is about 66 percent.
    What we have done is to get us an idea about what's that drug effect in null responders, we looked at less than 1.0 log or .5 log, and that treatment effect turns out to be 30 percent.
    So these analyses show that the response to these triple therapies in null responders compared to the partial responders and relapsers is lower.

    The second question you were asking at the beginning was we didn't exclusively study the treatment-experienced trial, for sure, but, again, as Dr. Florian mentioned in his presentation, we have learned during this review that even naive populations can inform us about the drug effect in the experienced. It's unlike HIV, where your response, after you fail the fourth round, changes because of mutation. That's a very distinct difference we had to make here in HCV. After the fourth round of therapy, your response to peg/ribavirin on which you fail is still the same.
    So the patient process still remains the same, except in the second part, we know who they are. But week 4, where it comes in is it tells us about interferon responsiveness. And we can choose who are those -- as Dr. Florian mentioned in his presentation, who are those future null responders, and because you believe the future null responders and the prior are the same, we are able to link and get an estimate of drug effect that may not be accurate, but what this data shows is better than
    just treating them with peg/ribavirin.

    Is that clear?
    DR. STRADER: No, but --
    [Laughter.]
    DR. JADHAV: All right. I tried.
    DR. VAN DYKE: Can I just follow-up on that? Because you pretty much had sold me when you presented the slides, but now I'm thinking more.
    It seems to me that the non-responders who have already non-responded are different than the presume non-responders that haven't failed yet, but you're thinking they will. So it's kind of hard to compare them directly, because you don't know the response of the non-responders who haven't been treated yet, the future non-responders. It seems to me that they're likely to be different than the non-responders who have really non-responded.
    DR. JADHAV: Can I have slide number 69 from our backup slides, 69?
    So what is shown here is the treatment outcome in the pegylated interferon/ribavirin arm for relapsers after they ended their study; so knew
    this end-of-treatment status and we looked back at what happened at week 4.
    So at week 4, we didn't know who these patients were going to be, but we went and looked at their end-of-treatment status, and we found there were relapsers and partial responders and plotted out the box plots for their week 4 response.

    Next. On the other hand, in the treatment-experienced trial, we knew who they were. At the beginning it says they were the prior relapsers and the partial responders.
    If you look at the distribution of responses, whether they were future, that is the left side, or the known, it is still the same. That's why these are the same patient population to which, on an average, the treatment PR responsiveness -- to be very clear, the PR responsiveness hasn't changed.
    So it is not on this slide, but if I bring in the third one, which is the null responder, which will be the lower end of this curve, whether
    it's the first round or the second round, they will still be the same. That's why we think the first round of treatment in these patients can inform us what will happen to the drug effect on the second round, because we have the data from treatment-naive population who were the null responders at the end of the study.

    DR. CARGILL: Dr. Ghany?
    DR. GHANY: Thank you. Maybe if I could just continue on this and then I'll get to my original question.
    So what you're trying to tell us is that a week 4 response is essentially the same as a week 12 response in terms of that the week 4 antiviral decline is predictive of the week 12.
    But what about development of virological breakthrough and resistance in those patients? Because, one, you're dealing with a treatment-naive group and the other with an experienced group. So I don't think there's data to answer that question.
    Is that correct?
    DR. JADHAV: We are not saying that week 4
    response is predictive of week 12. And, in fact, the resistance doesn't come into this argument, because we're only talking about the PR responsiveness to link these patients.
    DR. GHANY: Yes. But you're saying that a week 4 peg/ribavirin response equates a week 12 null response.
    DR. JADHAV: Yes.
    DR. GHANY: Isn't that what you're saying? That's the argument that you're trying to make.

    DR. JADHAV: Specifically, if you talk about null response, because it's defined at week 12, yes. What I meant was week 4 response is informing us about what will happen at the end of the study. Well, for null responders, the end of the study is week 12, because they are done there.
    But what I'm saying is you can grade them -- Dr. Florian has one of the slides where he showed greater week 4 response for the responders being the highest and the null responders being the lowest. It's telling us what will happen to them at the end of the study. Week 4 itself, you can
    kind of tell what will happen. That's what I mean.

    DR. GHANY: So I understand that. But then my question is, do we know about the development of virological breakthrough in these patients during therapy?

    So my question is, would a null responder have a higher rate of virological breakthrough than a naive patient who fails to achieve a week 4 interferon response?
    DR. MURRAY: Well, we think that that has been studied in the naive population, the people who had -- if you pick less than a half a log, 80 percent of those people would be null responders if just given peg-interferon. And those people, when they added boceprevir in the naive trial after four weeks, still had an SVR rate of 30 percent, around about.
    Now, do we know exact magnitude of a null responder? No, because they weren't studied. But I guess what we're saying -- that this is a question for the committee to decide. We're just offering some evidence that might give you some
    flexibility to consider whether to include them or not. But what we're saying is, is it likely that boceprevir offers an added benefit, and we are saying, yes, by the data in treatment-naive , it's almost inconceivable that they wouldn't have a benefit of boceprevir.
    When you look at the four-week response, if it's less than half a log, about 80 percent, 90 percent of those would be null responders. And in the ninth trial, they did add boceprevir, and they still had some sort of response, which was zero, and in the PR arm, about 30 percent. It's a pretty big number treatment difference to be able to explain away as having no effect at all in that population, who would be null responders. But we don't know exactly what it is.
    Are they exactly the same? No. Four-week would be a surrogate for 12-week.

    DR. CARGILL: Dr. McGovern?
    DR. MCGOVERN: The sponsor said that the reason that they did not study the null responders was because they were concerned about essential
    monotherapy. I'd like to follow-up on Dr. Ghany's question in that I would like to know -- I don't think there's a question that the addition of boceprevir changes virtually a null response to about a third of those patients getting a response.
    I think what I'd like to know more about are the people who don't get the SVR. Do we have any information about teasing out virologic resistance based on the four-week virologic response to peg/riba of less than a log? Are the numbers of RAVs enriched in patients who have less than a 1 log drop? That would be helpful to know. In term so a risk of exposing null responders who have not been studied in this trial, that would be very helpful.
    DR. MISHRA: I'll have Dr. Patrick Harrington from our virology team to answer that question.
    DR. HARRINGTON: Yes. The relationship between interferon responsiveness and the ultimate detection of resistance variance upon treatment failure is quite clear. Those who respond more
    A Matter of Record
    (301) 890-4188
    181
    poorly to the peg-interferon/ribavirin lead-in phase are more likely to have the detection of resistance-associated substitutions upon treatment failure compared to those who respond more favorably during the lead-in phase.
    So it's about 70 percent of patients who have a less than 1 log decline at week 4 who then go on and fail to achieve SVR have a detectable resistance-associated variant at the end of treatment compared to somewhere about 35 percent of those who fail treatment but have a greater response during that lead-in period.
    DR. MCGOVERN: Just a follow-up, too. In terms of this issue of using week 4 as predictive of what happens at week 12, there were documents that were given to us by the sponsor that had a table that I felt, when I looked at the table, that if you had more than a 1 log drop, that issue of predicting more than a 2 log drop at week 12 was much tighter than if you had less than a 1 log drop as to whether you would have less than a 2 log drop at week 12.

    So I guess I'm still wondering. Are we really truly convinced that what happens at week 4 defines a null responder? And we've looked at this for years before this. This isn't the first time this has come up.
    DR. FLORIAN: Just on that point, with the 1 log drop, that was sort of the motivation on why we looked at an even lower decline. It was recognizing that the 1 log drop, there were other patients within that group, except for null responders. When you go down to like a half a log drop, what we saw is that you were getting closer and closer to a homogenous patient set that are null responders.
    Certainly, I think one thing that might be asked, if the data is available from IDEAL trial, putting together that similar table, but for a half a log, and possibly after lunch, that's something that could be looked at.

    DR. CARGILL: Dr. Knodell?
    DR. KNODELL: Knodell. These resistant mutants are what are responsible for most people
    not responding and getting an SVR -- perhaps this should go to the sponsor. But how long a time period was needed before you could access or admit a non-responding patient to this trial? And has anybody looked at whether the duration between prior treatment and access to this trial or admission to this trial had anything to do on these early viral responses, whether that time period is -- because we know these mutants last or are around for a long time. And if there is a difference in time period between a previous treatment and accession into one of the boceprevir trials, that may well affect your early virological response.
    What was the interval that was required between treatment, prior treatment, and accession into one of these trials?
    DR. SINGER: I think we'd ask the sponsor to answer that question.
    DR. ALBRECHT: For the treatment-experienced trial, a patient must have finished his last course of peg/ribavirin at least six months. However,
    there were patients that were in the trial that had been off drug for several years, because, as you'll remember, these are people that were treated and for reasons, probably because they knew they wouldn't re-respond, they weren't re-treated again.
    DR. KNODELL: Do you know whether there was any difference in virological response as to whether these people had been previously treated six months ago or three years ago when the mutants would probably have disappeared?
    DR. ALBRECHT: Well, basically, with peg-interferon and ribavirin, we do not see resistance. So these are patients that only receive pegylated interferon and ribavirin, so they really would not have had RAVs when they came into our trial.

    DR. CARGILL: Dr. Clay?
    DR. CLAY: Sorry, I'm not going to ask about the non-responder thing.
    Was there any sort of analysis done looking at the timing with respect to the onset of anemia, the onset of thrombocytopenia or neutropenia? I guess what I'm thinking is, was there anything in
    the way in which it presented that would give the impression of a cascade of events occurring?
    DR. MISHRA: As I presented earlier, there was an initial decline during the first four weeks of therapy, which we contribute to ribavirin. And then after boceprevir was added at treatment week 4, you see an additional decline in hemoglobin which was approximately 1 gram, but which was higher in some subjects and lower in some subjects based on their baseline hemoglobin. But we haven't looked at neutropenia or thrombocytopenia as far as timing of onset of events.

    DR. CLAY: And I apologize because I don't have it all memorized.
    Would they discontinue or stop therapy or treat the anemia? I mean, was the potential there for someone to continue with the drop in hemoglobin that could have then gone on to develop either thrombocytopenia or neutropenia?
    DR. MISHRA: I don't think I understand the question. Can you please state it again?

    DR. CLAY: I'll try. Okay. In the
    individuals who either had thrombocytopenia or neutropenia, did any of them have anemia or not defined in as anemia in the study, but a drop in hemoglobin prior to that occurring?
    DR. MISHRA: Overall, they think that this is an overall bone marrow suppressive effect. So subjects who had anemia also had neutropenia and thrombocytopenia, but thrombocytopenia and neutropenia were to a lesser extent compared to anemia. But it seems that this was an overall bone marrow suppressive effect.
    DR. SINGER: And maybe the sponsor could address that question a little further as far as timing of neutropenia and thrombocytopenia. There were some patients that were pancytopenic in the trials.
    DR. KORN: Scott Korn, K-o-r-n.
    Slide 103, please, from the core deck. We think this may help. We showed this earlier.
    You can see that the predominant amount of the decline in the neutrophil count occurs very early within the first four weeks when there's no
    boceprevir on board, and then there's a small incremental increase. This really mirrors the pattern that we saw in the anemia slide, with the decrease in hemoglobin at the same time.
    So we agree with the FDA. These are occurring concurrently and not really sequentially, if that answers your question.
    DR. CLAY: But it's still not broken down by individual patient, in other words. That's trends over the entire study, correct?
    DR. KORN: That's correct.
    DR. CLAY: Okay. Thank you.

    DR. CARGILL: Dr. Ellenberg?
    DR. ELLENBERG: Could you put up the FDA's slide 52, again, on the treatment-naive late responders?
    So I'm a little confused by this slide. I'm looking at the bottom, and it looks like there's sort of a small number in each group, and there's 27 out of 34 versus 29 out of 40. And you're pointing at week 24 or week 28 in one place and week 36 in another.
    I'm just trying to understand how much we should really believe that this little separation over there to the right is something real or whether this is easily -- could easily just be random noise. I just want to make sure I'm not -- because I was surprised at the attention to that, and I want to make sure I'm not missing something about why you think that's real. It might be real, but it looks to me like it could also just be noise.

    DR. MURRAY: You're right. There are differences, small differences if you look at certain subgroups, and sometimes the differences go one way or the other in favor of either the triple or the RGT.
    But I guess when we compared arm 2 to arm 3, it did seem like that there was a pattern, blacks, treatment-experienced, late responders, if your HCV RNA at four weeks was less than a log. There seemed to be a pattern for the poor responders from the outset to be doing a little worse on arm 2. So we were particularly looking for breakthroughs that
    would occur after boceprevir stopped. It looked like that there might be a small difference in the naive.
    Whether any of this is real, I guess that's the question that the committee is going to have to deal with. As I said, there are so many different factors that impact SVR, and we're looking at so many subgroups, but what we're trying to figure out is, I guess, maximizing SVR and decreasing toxicity. So you're right.
    Then the table at the bottom goes with the figure at the left, and so that's the treatment-experienced, and a subset of those, the late responders who had received therapy for a certain amount of time, as well.
    But it did seem to be -- the treatment-naive, there did seem to be a slightly different pattern than the experienced based on small numbers. It's difficult. I don't know what to tell you.
    DR. ELLENBERG: They come apart and they go together, and it's --
    DR. MURRAY: Well, they really don't go together on the naive. I think the difference that you see between the blue and the red lines occurs sometime right after the boceprevir is stopped at 28 weeks. And then they look like they're coming together, but really that's end of therapy. And then when you stop therapy week 4, they both declined kind of in a parallel fashion. The difference narrows a bit.
    DR. HARRINGTON: I just want to add on to Dr. Murray's explanation a little bit further. If you look at the data for the patients in the right panel -- and we just use a strict definition of virologic breakthrough -- after week 28, there are 12 subjects in the RGT arm who experience a strict definition of virologic breakthrough compared to two subjects in the triple therapy arm.
    So that difference between the red and the blue lines does appear to be associated with bona fide virologic breakthrough after you take boceprevir away from that regimen.
    DR. CARGILL: Thank you.

    Ms. Dee?
    MS. DEE: This may just require yes and no answers. So what I'm hearing is the agency is not recommending that null responders be included. You're asking us that question; correct?
    DR. MURRAY: That's correct. That's a question to the committee. We were just trying to present an argument that you might want to consider, and we're permissive to some flexibility, if you think it's sound.
    MS. DEE: So in other words, it's up to us -- or not up to us, it's up to you ultimately, but it's a question for us.
    So in the lead-in, the four-week lead-in predicts outcome, ultimate SVR outcome, but that doesn't mean -- it doesn't say anything about null responders, per se, really about their SVR outcome as far as these trials are concerned.
    DR. MURRAY: Lynda, state the last part of your question again, please.
    MS. DEE: So in other words, we've got null responders in this naive trial. We know they're in

    there, okay? And we know that some people that do worse during the four-week lead-in period will probably -- may be nonresponsive.
    Does that mean that we know that boceprevir and PR for 44 weeks plus the lead-in is going to be sufficient for those patients?
    DR. MURRAY: Well, that's the longest duration that was studied in arm 3. I think that's the length that I had problems with, is that the null responders -- duration of null responders was not really studied. So if you look at a half log and use that as a cutoff for null responders, most null responders had greater than a half log. The median was greater than a half log.
    So I think when we use a half log cutoff, it's kind of the worst case scenario. Those are the worst of the worst of the null responders, and it looked like they would still have a treatment effect when you added boceprevir.
    How long do you treat them? That's up in the air. So I think the default for us was the maximum amount that was studied as far as the
    safety database, and that would be a 44-week after the four-week lead-in.
    Could you go shorter? Maybe, but I wouldn't know how to really tease that out.
    MS. DEE: Right. And what we've also heard is people that are null responders that don't have SVR will definitely have boceprevir resistance, resistant mutants. They're detectable in these patients, correct?
    DR. MURRAY: Again, if the null responders -- again, rephrase your question.
    MS. DEE: So in other words -- I can read it to you from here. In other words, these people that don't respond will have resistance, boceprevir resistance at failure, detectable mutants at failure. Right?
    DR. MURRAY: I think the answer would be most. But, Patrick?
    DR. HARRINGTON: For somebody who is poorly responsive to peginterferon and ribavirin, based on the lead-in phase, you can assume that the majority of treatment failures will have a treatment
    emergent resistance-associated substitution.

    MS. DEE: Right. So the danger is resistance for these patients rather than an SVR, correct?

    DR. HARRINGTON: Right. Right.
    MS. DEE: And this might be for the sponsor, but on slides 61 and 62, they talk about -- they showed this difference in the null responding patients. And I'm wondering. It looks like to me that there are less than 40 patients in the RESPOND and SPRINT studies.
    Do you think that's relevant at all?
    It's actually the sponsor's slide.
    DR. GOTTESDIENER: Would you like us to respond to that question?
    DR. CARGILL: Yes, please.
    DR. GOTTESDIENER: So before we get into this particular question, maybe I could help just clarify one or two points that have come up that really seem to be causing a lot of confusion for the group as a whole.
    The sponsor has actually put forward three

    arguments why we think, in fact, we can understand what the treatment regimen and the effect of boceprevir would be for null responders, one of which we obviously share with the FDA. The second one is the question that was being addressed today, and the third one is even some additional data, as well.
    Let me start with the question that was just asked, and if the chairman will allow me just one or two more minutes, I'll see if I can get back to the FDA question and see if I can just help to clarify that a little bit, because it's hard to listen to some of this. We know now complex it is for people who have been thinking about this for some period of time.
    If I could show the PROVIDE data from the main presentation. If you could put up slide 63, please.

    So this will address the final question. This is data that is unequivocal data from historic null responders. So in this particular population, you don't have to believe anything about treatment
    week 4, who is poorly interferon responsive, don't really have to make any assumptions at all.
    These were people who were in the boceprevir trials, Phase 3 trials, who were in the control arms failed and who were null responders by meeting the definition at treatment week 12, less than 2 logs definition. And if we rolled this people over into a boceprevir trial, where they receive 44 weeks of boceprevir in that particular trial, what you can see is -- and we don't have SVR data yet, but what you can see is these people haven't quite gotten to SVR yet. They will be in about 12 weeks.
    But what you can see is there are 38 of those patients who were in that trial who have received a boceprevir regimen for 44 weeks on top of the lead-in, and in that population, 39 percent of them have achieved end-of-therapy undetectable values.
    Now, there are two ways to take a look at that data. One can say to themselves, "Well, this compares very favorably in the poorly interferon responsive patients who were actually in the trial
    where the end of therapy data was 44 percent." The other way one can look at that and try to estimate what the SVR rates are going to be is by applying the relapse rates from the previous trials, and those relapse rates were about 10 to 15 percent. So if one takes 10 to 15 percent of the 39 percent, the final SVR rates are very strongly predicted to be somewhere around 30 to 35 percent.
    So this data, historical null patients, really supports that boceprevir can have a 30 to 40 percent effect overall.
    Now, we find this data reassuring if we don't have SVR data because it agrees with the same conclusions that the FDA and we do if we look at these patients a different way.
    I think what the FDA has been saying, and we strongly endorse this, is that the definition of poorly interferon responsive, null responders, the less than 2 logs at treatment, is just one definition to define these patients. It's a very useful definition.
    What we've done is we've said, if you take
    an alternate definition of these patients, which is very easy to assess in our trials, because, remember, we did a four-week lead-in in all of our trials, so that data is available to everybody.
    Let's see how we do.

    DR. CARGILL: Excuse me. I'm going to say that I think that will help us right here, so we can keep moving our questions forward.
    DR. GOTTESDIENER: Okay. Thank you very much.
    DR. CARGILL: Dr. Ghany?
    DR. GHANY: Yes. I had two questions. The first has to do with antiviral resistance. I wonder if either the sponsor or the FDA could tell us about the timing of onset of virological breakthrough. Was it early? Was it late? And what was the clinical significance? Were there any ALT flares associated with this? And how was it managed? What was the algorithm for managing patients who had virological breakthrough?
    My second question is actually to Dr. Florian. In the bridging analysis that you
    performed to look at late responders, based on your analysis, you presented two treatment paradigms, one to treat with a lead-in phase followed by 44 weeks of triple therapy or to use a lead-in phase followed by 36 weeks and then 12 weeks of consolidation with peg plus ribavirin.
    The latter part, I'm being a somewhat vociferous bearing.
    Was there any difference in SVR rates in that analysis? That would help me at least in addressing the question the FDA would like us to.
    DR. HARRINGTON: I'll first answer the question about resistance. I can't comment on the ALT flares. Maybe Drs. Mishra or Singer could comment on that.
    But in terms of virologic breakthrough, most of the breakthrough -- about two-thirds of the breakthrough that occurred in the treatment-naive trial occurred between weeks 12 and 28. There were some breakthroughs a little bit earlier on, but it's sort of varied. There wasn't a clear cut point of one breakthrough absolutely had to occur,
    but it appears that most of it was right in the middle of the treatment regimen.

    DR. SINGER: And maybe the sponsor can address the second part of that question as far as what happened clinically in patients that had breakthrough.
    DR. ALBRECHT: We did not see any ALT flares when the patient broke through. And the trial monitored the patients closely, so we stopped therapy immediately if there was a breakthrough. And I would remind you that a breakthrough was defined as a patient becoming HCV RNA undetectable and then coming back up to 1,000 international units.
    We also looked at incomplete virologic response, which I think you're referring to as breakthrough, where they came down significantly and then rebounded back up. And we didn't see any ALT flares in either of those situations.
    DR. CARGILL: Thank you.

    Dr. Roland?
    DR. ROLAND: I want to get back to the
    question of differences or potential differences in the black cohort.
    There was a lot that was in the briefing materials that I think set us up to be very concerned about that, and I was initially reassured when I saw the sponsor's data on fibrosis and then to hear your response. But when I look at the slides, there are so few patients and there are virtually no black patients, only five who are METAVIR score F4.
    So I'm just wondering what your level of confidence is in attributing the difference in the black patients in the two treatment arms to differences in fibrosis and if there's anything in the literature that might help us to either support that hypothesis or be concerned about that hypothesis.
    DR. SINGER: Certainly, we know from many studies with standard of care, peg/ribavirin, that black patients do not respond as well as far as SVR to standard of care. Of course, these differences we were pointing out were not statistically
    significant. And when we first looked at them, we noticed that there was a numerical difference, and it raised a little bit of a flag, and we realize that we're talking about subsets of subsets, very small numbers of patients.
    So I think in the end, we can't really say what is the optimal duration for those subsets of patients.
    Does that answer your question?
    DR. ROLAND: I guess it doesn't directly answer the question about how confident you feel in your hypothesis that the potential difference -- I mean, the study is not powered to look at a difference between the two treatment arms, and that's, obviously, the fundamental challenge that we have in trying to decide what is the appropriate duration of therapy for anyone.
    But the specific question about given the incredibly small numbers, how confident you are that you can explain the difference in blacks based on fibrosis, because that would make it easier for us, right? Then we just have to look at fibrosis,
    we don't have to look at black race.
    DR. SINGER: Right. See, we can only be as confident as the numbers we have. They're not large. Perhaps more significant is looking at early responders versus late responders or the treatment week for response as predictive.
    DR. MURRAY: And I guess I want to get away from our confidence in this data, because this is really the questions that we're asking you. So it was the sponsor's argument that the cirrhosis makes up the difference between the blacks triple in the RGT arm 2 and 3. We think it's interesting. We note that there are small numbers. But at the end of the day, you have to decide, I guess, risk-benefit for giving more boceprevir for all black patients for a little bit higher SVR versus the toxicity or if you feel comfortable allowing some sort of RGT recommendation for black patients if you can tease out perhaps the groups, the subgroups, that's the poorest responders.
    So it goes back to my opening remarks. It's all kind of a risk-benefit based on a lot of
    uncertainty in the data, and that's why we're asking you the hard questions. We're trying to present all arguments but don't necessarily look for our confidence in it at this point, because we want the committee's true opinion on these matters.

    DR. CARGILL: Thank you. And our last question, if you can be succinct, Dr. Giordano.

    DR. GIORDANO: For the FDA. Your analysis excluded 14 patients, and I'd just like clarification. Were those 14 people excluded -- I think you said there was unconfirmed virologic suppression. Was that because it was never repeated or because it was repeated and found to be, in fact, in error or detectable?
    DR. MISHRA: Dr. Wen Zeng from our statistics group can answer that question.
    DR. ZENG: Wen Zeng, staff reviewer. The 14 subjects have arisen from the week 8 to week 24 or in each, the worst value is inactive, undetectable.
    But I think that during the trial, they assigned to the longer duration of treatment. So we think it's early responder -- when we do a
    comparison, for the early comparison, we include this subject (indiscernible). That's the reason.
    Is that clear?
    DR. GIORDANO: I'm sorry. I did not understand your answer.
    DR. ZENG: Okay. The question is that these 14 subjects, from the treatment week 8 to week 24, every visit window, they have the viral load undetectable, is not undetectable. So it should be early responder. But during the trial, assigned to longer duration treatment. They should be assigned to the short duration of treatment. It's not just the one-sample or two-sample repeated measures, just different windows.
    DR. GIORDANO: So the company can clarify, but my understanding is the 14 patients, at the time they were assigned a longer duration, they were thought to have at least one positive value between week 8 and 24. But on re-testing of the samples, I guess it was determined that they were probably really undetectable if you re-tested the samples, and so might have considered that they
    were false positive.

    Now, whether to include or exclude those 14 is a judgment call, it is our judgment that they probably should be excluded.
    DR. GOTTESDIENER: Correct. And that is correct. In fact, these 14 patients were negative at treatment week 8 and were negative at treatment week 24. The protocol said that if they were positive anytime between treatment week 8 and treatment week 24, they should be assigned to the late responder category.
    So, of course, we assigned them per protocol and we did an analysis as they were assigned. But the FDA has pointed out that, in fact, one could look at the same data, as was mentioned by Dr. Murray. It's a judgment call.
    In the post-hoc analysis the FDA did, they said you could assign them to early response because they were negative at treatment week 8 and treatment week 24, and they were re-tested -- they're a single value that was above the limit of detection -- was rested and was negative
    on subsequent measures. They could just as reasonably have been assigned to the early response category, and so the FDA took them out.
    That is the place where we differ, at the per protocol analysis 72 versus 75. The FDA analysis, also a reasonable analysis, is 66 versus 75, and that raises the question of how to treat the treatment-naive late responders.

    But we do agree with the FDA, as they showed on slide 52, that if you actually look at the late – sorry -- if you look at the late responders in the treatment failure patients, there they seem to be about the same between the 32 and the 44 weeks of boceprevir. And so the company feels that 32 weeks would be more than sufficient for the late responder treatment-naive patients.

    DR. CARGILL: All right. Thank you.

    We'll now break for lunch. I can see people shifting in their chairs, so I assume your stomachs have been talking to you during the last part of this meeting. We won't ask you to repeat verbatim what we've said.
    We will reconvene in this room again in one hour from now, which is at 1:00 p.m. Please take any personal belongings you want with you at this time, as the ballroom will be secured by FDA staff during the lunch break.
    Panel members, please remember that there should be no discussion of the meeting during the lunch amongst yourselves or with any member of the audience.
    Thank you, and we'll see you back at 1:00.
    (Whereupon, at 12:06 p.m., a lunch recess was taken.)
    A F T E R N O O N S E S S I O N
    (1:01 p.m.)

    Open Public Hearing

    DR. CARGILL: Good afternoon.
    Both the Food and Drug Administration and the public believe in a transparent process for information-gathering and decision-making. To ensure such transparency at the open public hearing session of the advisory committee meeting, FDA believes that it is important to understand the context of an individual's presentation.
    For this reason, FDA encourages you, the open public hearing speaker, at the beginning of your written or oral statement, to advise the committee of any financial relationship that you may have with the sponsor, its product, and, if known, its direct competitors.
    For example, this financial information may include the sponsor's payment of your travel, lodging, or other expenses in connection with your attendance at this meeting. Likewise, FDA encourages you, at the beginning of your statement,
    to advise the committee if you do not have any such financial relationships. If you choose not to address this issue of financial relationships at the beginning of your statement, it will not preclude you from speaking.
    The FDA and this committee place great importance in the open public hearing process. The insights and comments provided can help the agency and this committee in their consideration of the issues before them.
    That said, in many instances and for many topics, there will be a variety of opinions. One of our goals today is for this open public hearing to be conducted in a fair and open way, where every participant is listened to carefully and treated with dignity, courtesy and respect. Therefore, please speak only when recognized by the chair, and thank you for your cooperation.
    Our first speaker will be Martha Saly, Director, National Virus Hepatitis Roundtable.
    MS. SALY: So my name is Martha Saly and I'm the Director of the National Viral Hepatitis

    Roundtable, and we are a coalition of more than 175 public/private organizations. And some of those are patient groups, medical associations, government agencies, and pharmaceutical companies.
    So I do want to mention that Merck is an associate member of NVHR and has provided funding to NVHR at various times in the past and present. They have not paid for my travel here today, though.
    We are dedicated to reducing the infection, morbidity and mortality from viral hepatitis in the United States. So although NVHR has really been eagerly awaiting the approval of a new class of hepatitis C drugs, I'm really here today to speak personally as a former hepatitis C patient and as an advocate for people who have not been as fortunate as I.
    It strikes me a little bit funny to be able to say I was treated for hepatitis C in the last century. But when I think about what a long road it's been for many of my aging baby-boomer friends who are still suffering from hepatitis C, it seems
    totally appropriate to say it this way.
    I was diagnosed in late 1998. At that time, there was much anticipation of pegylated interferon, but my doctors decided I could not wait for that. So I embarked on the 48-week journey of three times weekly shots. I was really happy to see that Dr. Poordad is here today because he was my first physician.
    I was told that I had about a 30 percent chance of a sustained response, and the word "cure" wasn't even being used at that time. But I decided to chance it. Had I not, I have no doubt that I'd be dead or on a transplant list today.
    As I said, I was fortunate. Some would say more than fortunate. Against the odds, I was cured. I'm here today because I believe that something quite historic is really happening here today in this room, and I really want to be a part of it.
    For more than 10 years, I've dedicated my personal and professional life to addressing this epidemic. In this role, I've spoken with too many
    patients to count who are ill, unable to access the care they need, or who were failed by earlier treatments and who are waiting for something that will give them better odds.
    Some just want to tell me their story or tell me how my story inspired them to be able to take the next step. I tell these patients that treatment is not easy and that it has to be a personal choice but one that might save their lives.

    It is for these patients that this day is of paramount importance, because for the first time, I can actually tell them although treatment may still be extremely challenging, hope for a cure should outweigh their fear that treatment will fail them.
    Yes, I believe that approval of these drugs will be the biggest thing that's happened in the years I've been doing this work and certainly what many patients have been anxiously awaiting. However, I'm still concerned that treatment will continue to be a difficult and challenging goal.
    I'm hopeful that a shortened course of
    treatment for many patients will equate with a greater ability to tolerate the drugs and manage the side effects, but I fear that the issue of compliance with strict medication regimens might affect treatment outcomes and treatment success. And these drugs will help few if awareness of the hepatitis C epidemic is not increased. And if awareness does not go hand-in-hand with improved screening and affordable access to care for the people who need it, it won't do us much good.
    I do have every hope that Merck and Vertex tomorrow --
    [Microphone times out.]
    DR. CARGILL: Thank you. Yes. We are going to adhere strictly to time because we still have quite a bit of work yet to do. Thank you very much.
    Tracy Swan, Treatment Action Group.
    MS. SWAN: I'm going to skip the introduction for time, but I do want to say I'm speaking on behalf of Treatment Action Group, a nonprofit think tank where I work. I do not have
    any financial relationships with the sponsor or their competitors nor has anyone supported my travel other than my employer.
    So I just want to very quickly, obviously, talk about some key issues that are involved with the development program for this drug and frame them in the context of my pleasure at being here for such an exciting and historic occasion.
    So as we know, hepatitis C is curable, and we've already heard quite a bit about the benefits of a cure. We've also heard about the grim scenario that patients in the United States face who don't have adequate treatment options.
    So if you look at this list, last week there were more than 16,000 people waiting for a liver transplant. And the Phase 3 program for this drug was started in June three years ago, but there have been no early access trials in any extremely ill populations.
    I have no doubt that doctors will use these drugs for people without the information that would really support a more well informed treatment
    program, and I want to use this as an opportunity to please ask sponsors to consider early access trials for their late stage drugs.
    I'd also like to highlight the low enrollment of populations among whom hepatitis C is highly prevalent and also among whom need better treatment, beginning with African-Americans, Latinos and Latinas, treatment-naive, people with cirrhosis, null responders. We don't know if people on methadone or buprenorphine were included in these trials, and we have very little information about the numbers of elderly people who are participants in these studies.
    Particularly upsetting for me is a lack of interaction data. There are some high stakes here. There's data in efavirenz and tenofovir, two antiretroviral agents, but the stakes are pretty high. If there as an uncharacterized drug-drug interaction, people could have HIV and/or hepatitis C treatment failure. And co-infected patients have faster progression to cirrhosis and liver failure and really need better treatment, and
    they've been waiting a long time.
    Methadone. People are at risk for a relapse to active drug use if they're under-medicated or experiencing withdrawal and hepatitis C treatment failure. Transplant recipients risk graft rejection if they are uncharacterized drug-drug interactions with immunosuppressants, as well as antidepressants, the risk for suicide. We know the neuropsychiatric side effect profile of interferon.

    I'm surprised that the drug has gone this far without that information, and I hope that in the future this doesn't happen. I also want to make a plea for simplicity. It's very confusing, this data. The trial design is confusing. Interpreting it is confusing.
    How on earth is this going to get incorporated into clinical practice, especially for non-specialist providers? We don't even have a treatment guidelines panel to really take this in hand and make it easy for people to optimize treatment.

    I am glad that the sponsors and the agency
    has thought about the week 24 stopping rule and have thought about ways to modify it. And I'm also glad to see the percentage of people who had an SVR and used EPO, not just the people in the trials who used EPO.
    I'd like to close with my list of post-approval trials, which is up there. Thank you.
    DR. CARGILL: Thank you.
    Lorren Sandt.
    MS. SANDT: Good afternoon. My name is Lorren Sandt. I'm the Executive Director of the Caring Ambassadors program in Oregon City, Oregon. I'm also the chair of the National Viral Hepatitis Roundtable that Martha talked to you about earlier. We're a nonprofit patient advocacy group. My company -- my organization does take money from Merck and Vertex. However, they did not pay for my trip to be here today.
    I'm here on behalf of my three family members who have hepatitis C, and I'm also here on behalf of the hundreds of thousands of people who are treatment-experienced who are looking forward
    to new treatments, and also the millions that are undiagnosed and aren't even aware that we're here today and what we're doing on their behalf, all of us working so hard to make sure they have care when they are diagnosed.
    But I do have some concerns. As an advocate, I have spoken with hundreds of patients, and I really have grave concerns about the difficulty of this regimen and the complications within it. I heard this morning the questions that were being asked, and all I kept thinking was, "Boy, if we can't figure it out in this room, how is a doctor in rural Oregon going to figure that out?"

    I'm really worried about patients. Boceprevir is not a simple protocol. We know adherence has an impact on treatment outcomes. We heard today we don't know whether it matters whether you take it at seven or nine hours. Patients need to know. So when the FDA comes up with their label, we would really appreciate some really strong guidance on that. There's a lot of
    studies out there that show medication adherence is very low just in general. It's at about 50 percent. And if you add depression or if you add any other co-morbidities, like diabetes or HIV, which many hepatitis patients have, the regimen is going to be increasingly difficult. And whenever you have a long, complex regimen, it really changes the adherence picture. So I'd like to see for the patients not to get resistance and to be best managed, to really have some strong guidance on adherence.
    Clinicians must incorporate supportive services and they need to learn what those supportive services are. We need some guidance on that. In HIV and TB, we have directly observed therapy. That does not exist. There's no one who's going to go home with a hepatitis patient and make sure that they took their drugs. Those case managers don't exist in hepatitis.
    Alternative therapies. One of the most widely used alternative therapies in this country is milk thistle, and it's certainly used within
    hepatitis C patients. And I'm really concerned about the lack of data since milk thistle and some protease inhibitors share the same metabolic pathways. Patients and providers need this information, so I'd really like to see that study done.

    As has been mentioned previously, special populations are really a concern for all of us. The FDA recently came out with their strategic guidelines and said that addressing unmet public health needs of special populations is a priority. And what I saw here today, we need to do a lot more work on those special populations.
    One that was not talked about today that I'd like to bring your attention to is the co-infection with hepatitis B. It's estimated that up to 30 percent of people with chronic hepatitis B have hepatitis C, and 8 percent of the chronic hepatitis C patients in the VA have hepatitis B. What is going to happen when we give them a protease inhibitor? What's going to happen to their viral titer.
    EPO. How are cash poor Medicaid states going to pay for EPO on top of the cost of all of these other drugs? It's not going to be an option for many people who are on not first tier insurance programs? So I want to make sure that those people have the best chance of getting cured, as well.
    So I thank you for your time. Really looking forward to all of these --
    [Microphone times out.]

    DR. CARGILL: Thank you for sharing your remarks.
    Michael Ninburg?
    MR. NINBURG: Good afternoon. My name is Michael Ninburg. I am the Executive Director of the Hepatitis Education Project, a nonprofit organization based in Seattle, Washington. Our organization has received funding support from both Merck and Vertex.
    I was a hepatitis patient myself for many years until May of last year when I learned that I was cured after a course of 48 weeks of treatment. I had been diagnosed about 20 years ago and
    initially sought treatment about 10 years ago and was given the prognosis of about a 50 percent likelihood of cure. I was also told that newer, better drugs would likely be available within five years.
    Well, as we know, that timeline proved to be optimistic. In the meantime, my liver disease did progress. I eventually did progress to borderline cirrhosis and was told that it would be a good idea to enter treatment. I was fortunate enough to get into a clinical trial about two years ago, and, as I mentioned, I am now cured.
    I'm here today to advocate on behalf of approval for boceprevir. That said, I do have some concerns, all of which have been addressed here today, but I will point out a few. Among them, further study is certainly warranted among drug-drug interactions.
    The effect of boceprevir on methadone and buprenorphine, in particular, is one of the more important areas for study, especially given that most new hepatitis C infections in the U.S. are
    related to injection drug use and many of those persons who are newly infected will eventually go on to opiate substitution therapy.
    Similarly, as has been pointed out already, drug-drug interaction studies are very important for antidepressants. This is crucial because we know that other protease inhibitors do reduce exposure to some SSRIs. And because depression is such a common side effect for treatment, more and more patients are being given antidepressants prophylactically before initiating treatment.
    Proper management of side effects will also be crucial to ensure that patients have the best chance of being cured, and I do look forward to seeing the results of the studies underway looking at the EPO versus dose reduction to address anemia.
    I would finally like to reiterate that adding boceprevir to peg/riba with a lead-in and response-guided therapy is going to make a complex treatment regimen even more complex. This will be equally true for patient and provider. Adding to the complexity is a fact that providers who treat
    hepatitis C are generally inexperienced with drugs that can cause viral resistance, and this has been pointed out, as well, today.
    Adherence will be crucial to successful treatment, as will effective mitigation of side effects. For these reasons, labeling recommendations should be as specific as possible.
    This is truly a watershed moment in the treatment of hepatitis C. The drugs you are reviewing today and tomorrow will increase the cure rate for patients and, equally importantly, reduce the length of treatment for many patients. This is the first real breakthrough in treatment in the past 10 years and is the next step toward a universally effective cure for hepatitis C.
    Thank you.

    DR. CARGILL: Thank you.
    Jules Levin?

    MR. LEVIN: I believe most of you know just who I am, and I have a few comments. The first thing I want to say is I read all the briefing documents and listened to the FDA presentation, as
    well as the Merck presentation, so I think I'm fairly familiar with the data.
    I want to say that I do support, for a change, the FDA position on the 32 weeks. I think that we need longer therapy for certain patient populations, the harder to respond patients. So I'm in full agreement on their analysis on that. And I want the committee to know that I support that, and I think it's a good thing for patients. A little extra therapy is better than too little therapy for patients who are hard to treat.
    I want to talk about the null responder situation a little bit, and what I'd like to say is I think that it's a fluid situation, it's a difficult situation. And I think that what might be helpful here is perhaps for the FDA to put some language in the label about null responders, maybe without taking the position that supporting treatment or not supporting treatment for null responders. But maybe some language about what we know, what we don't know, the strength of the data, the weakness of the data, because what does concern
    me is some patients who are null responders who maybe can't wait too long for new drugs to come along.
    So I don't want to deny them the opportunity. At the same time, I don't want to treat somebody unnecessarily that could wait. So maybe some language would help, if you could do that. Maybe you could comment on whether you can just put some language in there that would help people interpret things.
    I want to talk about co-infection and DDIs, drug interaction studies. I'm a little disappointed perhaps. Maybe Merck and the FDA could fill us in a little bit about what has been done in co-infection, what has -- maybe Merck could explain a little more about what they've been doing in co-infection and what their plans are or what they have done in drug interaction data, because we're going to hear Vertex's data tomorrow; we know what that is in drug interactions and co-infection studies.
    What does concern me is that there are
    patients with co-infection that really can't wait too long for new drugs to come along, and if they're not on Reyataz or telaprevir, what are they to do? So I'm concerned about that.
    Then lastly, the last point I want to make, I would love to hear some discussion from Merck and the FDA and Vertex tomorrow. We know how complicated this is becoming now in terms of therapy and how difficult this is going to be for patients and clinicians to understand how to do it and what to do.
    I really would love to see better education programs from the companies and the FDA to help us on this for clinicians and patients about understanding resistance. I really don't want to hear companies saying, "We won't have resistance. It's going to disappear, and you can re-treat," because we really don't have an answer to that question. That could happen but we don't know.
    So what I'd like to hear is some real education for clinicians and patients not just about resistance but about adherence, about all the
    stuff that we need to know about coming in at week 4 or week 8 to do your viral load testing and why that's important. There are many, many issues that I could list here, but what we need is better education programs from the companies and the FDA to support this and encourage this and perhaps to demand it from all the companies as we move forward here.
    Lastly, I just want to comment on early access.
    [Microphone times out.]

    DR. CARGILL: I'm sorry. We have to keep moving forward. Thank you.
    Murray Penner?

    MR. PENNER: Good afternoon. My name is Murray Penner, and I'm the executive director at the National Alliance of State and Territorial AIDS Directors, or NASTAD. Neither NASTAD nor me have a financial relationship with Merck. NASTAD does receive money from Vertex, but they did not pay for my travel here.
    NASTAD is a nonprofit organization that
    represents the nation's chief state health agency staff who have programmatic responsibility for administering HIV and viral hepatitis education, prevention, care, supportive service programs funded by state and federal governments.
    I'm pleased to have the opportunity to speak briefly about your consideration of this new drug application for boceprevir. While not necessarily in the purview of the Antiviral Drugs Advisory Committee, I must point out some very important considerations regarding this new treatment and the public health infrastructure that will impact access and uptake of this drug.
    Given the lack of public health infrastructure for viral hepatitis and the many barriers to care for persons living with chronic hepatitis C, the cost of this new treatment will significantly impact affordability and comprehensive coverage of patients.
    NASTAD encourages Merck and the FDA, as well as other government agencies, to work together to ensure that uninsured and underinsured patients who
    need treatment are able to afford the drug, as well as the entire standard of care, given that this new treatment will be added to and not replace the current hepatitis C standard of care. Further, coverage must include ancillary medical costs, such as liver and viral tests, liver biopsies, and management of treatment side effects.
    As you know, this treatment will not eradicate hepatitis C. A cure must rely on infrastructures, such as funding for screening and testing, staff capacity in medical settings, and educated health providers in order to identify those who need treatment in the first place.
    In order for this new treatment to work, new diagnoses are required. Unfortunately, as you know, 75 percent of people living with chronic hepatitis C, or up to three million people, do not know they are living with infection and must be identified.
    The IOM recently reported a lack of knowledge and capacity among providers to identify infection and deliver expert care, as well as a
    lack of knowledge among the public, most importantly, among populations at high risk of infection. Even with this new treatment, low provider awareness will continue to lead to lower than anticipated utilization of this drug and misdiagnoses.

    In addition, there are limited specialists, such as infectious disease doctors, hepatologists, and gastroenterologists available to treat hepatitis C. Further, low public awareness will continue to lead to misinformation and missed opportunities for prevention and care.
    Targeted and difficult to reach populations must have access to this new drug, as well as appropriate care. Treatment will need to be given in tandem with other important public health and social services being provided in public health settings.

    If we do not capture the 75 percent of persons living unknowingly with chronic hepatitis C, the costs associated with liver cancer caused by chronic hepatitis and the care and
    treatment of persons living with chronic hepatitis are very high. In the absence of significant increases in screening and care above current medical management, an actuarial study from Milliman, Incorporated projects medical costs for patients with chronic hepatitis C will go from 30 billion to over 85 billion over the next 15 years.
    Finally, further post-marketing studies of this drug must be conducted in order to further define potential risks, other drug interactions, and impact on certain populations, including blacks, where there is evidence of lower sustained virologic response.
    Thank you for the opportunity to provide comments this afternoon.
    DR. CARGILL: Thank you.

    Paul Brayshaw?
    MR. BRAYSHAW: Hi. My name is Paul Brayshaw. I'm here on behalf of people with bleeding disorders, and I have no conflicts to report.
    We're a group of people who are made up of patients and supporters who were affected by the hepatitis C crisis. We organize and advocate for urgent development and access to new, far better HCV therapies. As stakeholders, we salute the efforts of the FDA for their early approval of telaprevir and boceprevir, and we appreciate the proposed guidance regarding the testing of the direct acting agents, and, more positively, guidance regarding drug combination testing.
    Our campaign has been recognized and is supported, in part, by the Hemophilia Federation of America, the National Hemophilia Foundation, and the Committee of 10,000. Among people with factor-dependent bleeding disorders over the age of 30, nearly the entire cohort was exposed to hepatitis C via contaminated factor. As a result, most of us now have long-term disease associated with these products and therapies which were used for the treatment of bleeding disorders.
    We support the earliest possible approval of the drugs, telaprevir and boceprevir. However,
    people with bleeding disorders have specific problems with hemostasis or the ability of the body to stop bleeding, which may be further exacerbated by hepatitis C and HIV disease, as well as HIV therapy. Standard of care therapy for hep C also depresses hemostasis, and while the next generation of hepatitis C therapy is becoming available, those could potentially have other complications, as well.
    People with bleeding disorders and hep C also have high levels of HIV co-infection which is associated with faster hep C progression and a decreased likelihood for hep C drug therapy success. The FDA should require advertising and/or educational materials on these drugs, as well as define difficult to treat populations and point out that treatment success rates for these groups may be substantially lower than those pertaining to the population as a whole.
    We also request the FDA require guidance and warn about treatment failures with the next generation of drugs, as well as the leading into
    resistance among the most endangered and difficult to treat populations.
    The medications for the treatment of bleeding disorders are extremely expensive. On average, they cost approximately 100 to $150,000 a year for insurance companies, and that has an impact on patients, as well. Employer-based health insurance or some other form of health insurance is critical to ensure access to these life-saving therapies, and we encourage the FDA, the pharmaceutical companies, as well as insurers to recognize this burden in the context of additional chronic illnesses. We are concerned that this affordability issue may affect access to therapy, as well as adherence.
    Our efforts have been collaborative, transparent, and with dedication, because our lives depend on the development and access to more efficacious therapy. And so we call on the FDA to protect us from advertising or educational materials with the promise of a high cure rate, as well as treatment success as treatment becomes
    available.
    Thank you.

    DR. CARGILL: Thank you.

    Dr. Belperio?

    DR. BELPERIO: Hi. I'm Pam Belperio. I'm the national public health clinical pharmacist for the Department of Veterans' Affairs' Office of Public Health, and I'm here to talk to you today about the VA experience with hepatitis C.
    This presentation will summarize the VA's experience with real world diagnosis and treatment of chronic hepatitis C over the last decade. We hope that this data will be useful to the committee in discussions about these promising new therapies, particularly with regard to populations for which more data may be helpful in assessing safety and efficacy.
    As seen in the chart on the left-hand side, chronic hep C has been a major clinical and public health issue in the VA over the last 20 years. Veterans have a hep C prevalence rate over three times that of the general U.S. population and even
    higher prevalence in certain high risk populations, such as homeless.
    In 2009, the VA had roughly over 210,000 veterans in care seropositive for hepatitis C and approximately 157,000 with chronic hep C, representing 5 percent of all individuals in the United States with chronic hep C. The VA has a sophisticated infrastructure for hep C policy, programs and products to achieve the highest possible outcomes. As a result, over 80 percent of veterans with chronic hep C have been linked to care.
    Despite this infrastructure, the reality is that most patients with hep C do not receive treatment. Whether the setting is in the VA, community practice or academic medical centers, the most common reasons for exclusion are psychiatric disorders and alcohol and substance abuse disorders.
    The reality is that many patients have significant co-morbidities that wouldn't qualify for clinical trials based on exclusions, and I'd
    like to focus on this for just a moment because these patients represent the majority of hep C patients in both the VA and non-VA settings and, in many ways, are those patients who can benefit most from these new therapies.
    Understandably, registrational trials have excluded these patients, and while it's appropriate, it's important to keep in mind that we do not have safety and efficacy data on hep C protease inhibitors for significant populations of hep C patients in need of treatment and that the benefit-risk profile may differ in these individuals from those enrolled in clinical trials.

    Yet, we often treat these patients regardless with little knowledge about what we can or should expect in terms of efficacy, safety and tolerability, as you can see here. As a result, we often see more adverse effects, higher discontinuation rates, and more dose modifications.
    As a consequence, we have observed SVR rates that are, as one would expect, lower than seen in registrational trials. And this should not be
    surprising to anybody in the audience, as real world populations outside of those studied in registrational trials show different levels of benefit. But we clinicians in the field and patients need guidance on what to expect in terms of efficacy and safety in these not so ideal, yet all too realistic populations so that reasonable, evidence-based discussions can take place between providers and patients with a practical understanding or at least some insight into the risks and benefits of treatment.

    I want to thank our colleagues at Merck, Vertex, and the FDA for the hard work in getting us to this point. These new therapies promise a paradigm shift. Yet, at the same time, it's important to recognize that efficacy in a clinical trial is not the same as real world effectiveness and that we're treating patients, not just a virus.
    We look forward to the discussion about what the next steps should be in studying populations who have yet to be enrolled in clinical trials.
    Thank you.

    Questions from Committee to Sponsor and FDA
    DR. CARGILL: Thank you. The open public hearing portion of this meeting has now concluded and we will no longer take comments from the audience. The committee will now turn its attention to address the task at hand, which is the careful consideration of the data before the committee, as well as the public comments.
    We're going to return, because we had to cut our questions short in both parts of our discussion, so we're going to return to questions, which will be for either the sponsor or the FDA or both.

    Dr. McGovern?
    DR. MCGOVERN: Thank you. I'd ask the sponsor to give us some slides about whether four-week responses to interferon predict 12-week responses. And this was supplied with the documents. So if we could have that slide, I'd just like to look at that data again to refresh my memory.
    DR. GOTTESDIENER: Yes. If we could see
    slide 550, please. Thank you. If you can show this slide.
    So this is a slide, and we have one subsequent slide to do the analysis that was requested earlier in the day to look at this same kind of concordance data at the 0.5 level at treatment week 4, which we would like to share with the committee, as well, since it was requested.
    What this slide attempts to do is to help people to understand how comparable are and how much overlap there is between the two populations, the two definitions for null responders, treatment week 12 and the treatment week 4; treatment week 12 less than 2 log and the treatment week 4 less than 1 log.

    What you can see is in the center is a 2by2 table, and this data is from the IDEAL trial, which had 3,000 patients. And we took this data, though similar data is available for both SPRINT-2 and RESPOND-2 for smaller numbers of patients, because this has a much larger cohort of patients who receive standard of care. So we could really see
    how these two definitions line up.
    In the top left, you see 533 patients who met both definitions, both the less than 2 log at treatment week 12 and the less than 1 log at treatment week 4. And remember, because this is PR, because this s standard of care, every patient could be fully defined as to being one or the other of those definitions, something that isn't always possible in the boceprevir-treated patients.

    If you look at the bottom right for the diagonal, you see the number 1926, which are patients who were greater than 2 logs in terms of treatment week 12, who are not null responders, and who also did not meet the treatment week 4 definition overall. And on the diagonals, you can see 172 patients and 146 patients who either met one of the definitions or met the other of the definitions.
    In practice, with this particular population, depending on which study you look at, somewhere between 65 and 85 percent of the patients are exactly the same patients. So in other words,
    whether you define them by a treatment week 4 definition or whether you define them by a treatment week 12 definition, they're exactly the same patients.
    If you could put up the slide that's done the analysis at treatment week 4, slide A1, please.
    This is the same study analyzed by the treatment week 4 definition of 0.5. Remember, the FDA proposed that definition to come up with a definition that was more stringent, that really used a treatment week 4 definition that totally identified null responders. And you can see, from the IDEAL trial, they've actually succeeded.
    If one looks across the row that says "less than 0.5 log," one can see that out of 253 patients who meet that definition, 234 of them, 90 percent, actually are the same patients who would have met the same definition for a treatment week 12 responder, a null responder. And there are only 19 patients, if one looks to the right, who really are discordant in terms of the treatment week 12 and the treatment week 4 definition.

    Before we go any farther, I want to remind everybody that the FDA has analyzed that particular definition in our trials -- they showed that in the slides -- and those patients had 28 to 30 percent response rates with boceprevir as opposed to zero in the control arms overall. So if one uses that definition, again, it's an alternate definition of poorly interferon responsive patients, one can see, in fact, they do identify exactly the same patients, at least 90 percent or so, and there's still 28 to 30 percent, increased percentage, with boceprevir.
    As the FDA did point out, it is a more stringent definition in the other way. If one looks down the column of less than 2 log, it only identifies about 30 percent of the true null responders. So this definition is a very stringent definition of null responder, the most poorly interferon responsive patients. And even so, it's still a very useful definition to help to understand that there's great concordance between the two definitions. And therefore, this data
    supports the 30 percentage points increase in the null responders.
    Thank you.
    DR. CARGILL: Thank you.

    Dr. Strader?
    DR. STRADER: I was just looking at those slides, so it brings up a similar question. I wanted to make a point initially that we've changed the definitions on a lot of things. So we're calling a null responder something different. You approach that in your slides there.
    These patients and the patients in your PROVIDE study are naive patients, correct?
    DR. GOTTESDIENER: The patients in PROVIDE are all patients who failed in the -- PROVIDE included many patients, but the ones we provided in the analysis are all patients from either study who failed on the control arm, met the definition of null responder, and then proceeded through 44 weeks of therapy.
    DR. STRADER: Okay. And there were 38 of those, but you say failed on either.
    DR. GOTTESDIENER: Could have failed in either from RESPOND-2 or from SPRINT-2 control arms.
    DR. STRADER: So RESPOND-2 was non-responders.
    DR. GOTTESDIENER: Correct.
    DR. STRADER: Right. So those would be partial responders and relapsers in that group that would have been in the control group that you would have put in your PROVIDE study.
    So do we know how many of those 38 patients you had came from the RESPOND control group as opposed to the SPRINT control group?
    DR. GOTTESDIENER: Yes. I'll have the slide up in a minute. But I do want to clarify one thing that actually is incorrect about what you said.
    While, by historical definitions, we only brought in partial responders and relapsers, we know from looking at the control group in RESPOND-2 that we didn't succeed in that particular goal. So in RESPOND-2, as we showed in the briefing booklet, 20 percent of the patients, despite that
    exclusion, demonstrated themselves as being null responders by hitting less than 2 logs at treatment week 12.
    So I want to be very careful. The historical definitions are useful tools, but in practice, despite all the efforts we made to keep patients out of that particular trial who were null responders, 20 percent of them came into the control group and demonstrated in the course of their control therapy.

    Now, if you actually want to look at the data you asked for -- show slide 566 please. This actually breaks the data down by the 38 patients by which study they came from. And what you can see if you look down the left-hand column is the 29 patients who came out of the SPRINT study and the nine patients who came out of the RESPOND-2 study, all of whom who met the definition of null responders in the control arm. And you can see the SVR rates are 34 percent and 56 percent.
    I would caution these are very small numbers. That's why we took the two groups
    together and we compiled it as 15 of 38, the last column, which really is the pooled data.
    I hope that helps to clarify a little bit.
    DR. CARGILL: Thank you.

    May I just caution the committee, because we have a number of people with questions, if your questions can be targeted, as well as the answers succinct, it would be helpful.

    Dr. Korman?
    DR. KORMAN: My question relates to the other side of the equation, not the benefit, the sustained virologic response, but the risk.
    Do we have any idea from the FDA or the sponsor as to the distribution of risk, side effect profile based on age, gender, fibrosis score, co-morbidity or race so that we try to get some sense of the risk stratification that a clinician might have to do in terms of interpreting which patients they're going to try to treat and which they may decide to delay?
    DR. MISHRA: The sponsor has presented some data, and we have also looked in the analysis,
    response-guided therapy versus long duration boceprevir, and we have seen that there is some difference, there is some benefit of having a shorter course of boceprevir in terms of anemia, in terms of serious adverse events and discontinuations.
    So we do feel that a shorter course of boceprevir therapy may provide a favorable benefit-risk assessment.
    DR. KORMAN: Within the groups where we've had these profound effects, the very unique events, is there any guidance in terms of which population or which subgroup? I understand that shorter is better, but is sicker worse?
    DR. SINGER: So if you look at the data comparing the treatment-naive and treatment-experienced populations, there were generally more serious adverse events, severe adverse events in the treatment-experienced population. It probably goes along with their stage of illness; although if you look at subgroups, there did not appear to be any difference in safety profile, for example, by
    age, by gender, by race.

    DR. CARGILL: Dr. Ghany?
    DR. GHANY: I guess this was just to try to address a question that I had that wasn't answered in the earlier session. It was to Dr. Florian.
    The question pertains to the bridging analysis that was performed and based on that analysis, the FDA proposed two treatment regimens for late responders, one using lead-in with 44 weeks of triple therapy versus lead-in with 36 weeks and then 12 weeks of consolidation with peg plus ribavirin.
    The specific question was what were the SVR rates for those two different regimens? Thank you.
    DR. FLORIAN: To that question, we have the SVR for the 32-week regimen. It comes from the treatment-experienced trial, but this is a group that only had prior relapsers and prior partial responders. So in a treatment-naive setting, it may be a little bit lower than what was seen there. We're not entirely sure what the difference in a treatment-naive late responder setting would be
    between the two regimens, but we do know that 32 weeks or extending it to 32 weeks would cover a majority of that numeric difference that was seen from the treatment-naive late responder arm.
    DR. GHANY: What were the actual rates?
    DR. FLORIAN: Oh, the actual numbers.
    DR. GHANY: Yes.
    DR. FLORIAN: So from the treatment-experienced arm, the RGT, so the one that looked at 32 weeks, 79 percent. For 44 weeks, it was 73 percent. And then for the treatment-naive late responder arm, where they looked at 44, it was 75 percent.
    DR. CARGILL: Thank you.

    Dr. Roland?
    DR. ROLAND: I have two quick questions. You used a last observation carried forward method and stated in the background materials that there was a very small number of people who were lost to follow-up after week 12.
    Can you address what the numbers were and whether there were any differences across arms in
    either of the studies? That's question number one.
    Question number two is, do you have a list prepared of the CYP3A4 substrates with narrow therapeutic index that we could take a look at?
    DR. ZENG: This is week 12 follow-up carried forward for subjects who missed week 24 follow-up. It's pre-specified in the protocol, and this is a pre-specified analysis.
    In terms of number of subjects, I can get a number to you later, but the number is very small, and we did sensitivity analysis, and probably the sponsor can give you the exact number. The numbers are very small. I can dig up for you.

    DR. ALBRECHT: Could you bring up slide 291?
    This slide gives you a comparison in the SPRINT-2 study of using the last observation carried forward. On the left side is the SVR for follow-up 24 and in the parentheses, missing equals failure when we do all of these studies. And the second row is SVR last observation carried forward.
    Arm 1 is the peg/ribavirin control. As you can see, the response rates are almost identical,
    37 and 38 percent. You can also see in arm 1 that they're 62 and 63; and, in the third arm, 65 and 66. So there were very, very few patients that were carried forward. There were four in arm 1, four in arm 2, and three in the arm 3 study.
    So we had a very good return on our patients coming back for their week 24 follow-up, and there was no difference essentially in the SVR rates.
    DR. CARGILL: Thank you.

    Dr. Connick?
    DR. CONNICK: I have two general questions. The first relates to conception. What were the instructions given to males and females about birth control in the study? How many pregnancies did occur and what were the outcomes?
    DR. ALBRECHT: Using the backbone therapy that we used, the peg/ribavirin, ribavirin is a teratogen, and we used the same criteria that is in the label of double-barrier contraception in the study. We required that patients use it prior to the start of the study and during the follow-up period. There were no pregnancies in patients that
    were being treated. There were pregnancies in partners, and those are reported in the ribavirin registry. There were no outcomes in partners to suggest that there was any ill effect.
    DR. ROLAND: Thanks.

    My second question is adherence. You said you did have adherence measures in the study. These are very difficult drugs to take. What was the level of adherence in the two studies?

    DR. ALBRECHT: We used an electronic diary to have the patient record their medication. They recorded all three medications and did so on a daily basis. The electronic diaries were taken to the site where they were uploaded, and that's the way we collected the data.
    Could I have slide 2632 please? This slide is illustrative of SPRINT-2, and it shows you that the level of adherence that was reported to us was basically very, very good. We found that in looking at SVR, that treatment duration was actually more important than dose. But as you can see on the left side, treatment duration greater
    than 80 percent, dose of ribavirin and boceprevir greater than 80 percent, most of our patients are in the more than 80 percent range.
    DR. ROLAND: Thanks.

    DR. CARGILL: Ms. Young?
    MS. YOUNG: I'd like to congratulate the sponsors for coming through with a breakthrough item, but I am concerned with some of the issues raised by the public groups in terms of complications of this regimen and wondering if you could give a little more description of what the response-guided therapy looked like in the trials. And then if the FDA could tell us what they're thinking of in terms of guidance, in terms of protocols for physicians to make sure that once these drugs do get into the system and there's demand for it, that they actually have a positive effect and are used appropriately.
    DR. GOTTESDIENER: If I could see the pathology slide, slide 118, please?
    This is the kind of slide -- obviously, there needs to be very good educational materials,
    as we've heard before and as we all acknowledge. This would be the kind of slide that would be available to practitioners, obviously, in a more patient-friendly or in a practitioner-friendly manner. But it lists all of the different groups and what the durations of therapy would be for treatment-naive and treatment failure, early responder and late responders.
    MS. YOUNG: As a follow-up to that, in terms of the use of EPO, is that going to be part of the instructions?
    DR. GOTTESDIENER: In our studies, we did give instructions for EPO, and I'm sure we'll have some very interesting discussions with the FDA about what will be included in the label. But we certainly think that that's an important part of the instructions to practitioners.
    DR. CARGILL: Thank you.

    Dr. Schechter?
    DR. SCHECHTER: A question more about the hematologic safety. You showed us a hemoglobin decline and neutrophil decline with timing. The
    same decline in terms of timing for the grade 3 and 4, are they similar?
    I also wanted to know what the directions you gave -- that is, how often were CPCs counted, that is, tested for? And given the complications that you've seen in terms of the severity of the anemia, whether those will be enhanced; that is, should people be tested more frequently during certain periods of time?

    DR. ALBRECHT: The peg/ribavirin label has actually a warning about anemia and the peg/ribavirin label states that at week 2 and week 4, the patient should be tested and more frequently if clinically indicated. This was also the way that we did it in the clinical trials. Now, in the clinical trials, patients were tested at every visit or as clinically indicated. So some patients, if they had steep declines in hemoglobin or neutrophils, were tested more frequently.
    With regard to your second question, we do have -- probably the best indication was in the slide that we showed you, 89, with regard to
    hemoglobin values during treatment and when we saw the nadirs. As we indicated, those hemoglobins dropped very rapidly in the first four weeks and then we saw the decline after that. So we were catching them as they came in on the various clinical visits. So a physician watching a patient would be calling them back if they were seeing a steep drop in hemoglobin.
    DR. SCHECHTER: So you're not thinking that one should treat -- one should be testing weekly in that period of the 8 to 12 or which -- one wonders if one could avoid the grade 4 particularly in the neutrophil and platelet.
    DR. ALBRECHT: The directions and the protocols were that if the hemoglobins or the neutrophils continued to drop, that the patient should be called back. So if the hemoglobin was dropping and there was a hemoglobin that was close to 11, one would expect a physician then to call that patient back the next week. And the physicians, at their discretion, could call the patient back as many times as they felt was
    appropriate.
    So those grade 4s generally -- as you could see, there were not that many grade 4s. Those hemoglobins were caught before they got down to a grade 3 or a grade 4.

    DR. SCHECHTER: In the patients who had more anemia, as in the blacks, particularly in the blacks, who often start at a lower hemoglobin level than Caucasians do, the question was did they have more ribavirin reduction and did that possibly account for the lower SVRs?
    DR. ALBRECHT: I'll start from the SVR perspective first, and that is, blacks do have lower SVRs with two-drug therapy. And using the base that we're using, the peg/ribavirin backbone, it's not surprising that with boceprevir they have lower SVRs.
    We know for some reasons why they have lower SVRs, IL28 being one of them. But if you look at the difference in SVR from the control with boceprevir in the blacks, the delta from the control is almost the same as we see in the non-blacks. So the SVRs are not so much related probably to their anemia as to other facts that are associated with the black race.
    I think Dr. Bacon will comment on the anemia and the way he manages anemia with these drugs.
    DR. BACON: Thank you, Jan.

    I'm Bruce Bacon. I'm at St. Louis University. That's the people who pay me. I am a consultant for a lot of pharmaceutical companies and my travel expenses were paid today by Merck.
    I think that the management of anemia, for anybody that has a little bit of experience with taking care of patients with hepatitis C on interferon and ribavirin will be comfortable handling the anemia, the leukopenia or neutropenia, and thrombocytopenia that we see with these regimens when they add boceprevir. I really don't think it will be that much different for most of us.
    It's comforting to know that there's good data that shows that you don't lose the ability to attain a sustained response by dose reduction of
    ribavirin, which is -- some people brought up the question about, oh, we're going to have to use all these erythropoietin that's so very expensive. Well, there's data that shows that you can do just as good a job with dose reduction of ribavirin and not lose that effect.
    So there's a lot of different strategies and ways in which these complications can be managed. But anybody that's done this for a little bit knows how to do that. I don't think it'll be very difficult.
    DR. CARGILL: Thank you.

    Dr. Friedman?
    DR. FRIEDMAN: Just to follow-up on a theme, my question had to do with the role of IL28B genotype testing, which I realize came out while you were doing your studies. But I wonder to what extent this impacted the differences in racial responses and, also, how you see that ultimately fitting into practice in patients treated with three-drug therapy, further complicating the management for the practitioner.
    DR. ALBRECHT: We found the IL28 genotype in the IDEAL study, and the IDEAL study had started before we initiated -- the IL28 was found after we had initiated our trials. So, basically, we had collected pharmacogenomic samples in the two trials, the SPRINT and the RESPOND-2 trial, and there were 60 percent of our patients who had agreed to give pharmacogenomic samples. So we only have a partial set of IL28 data. And in cooperation with Duke, we also looked at the IL28 in these studies.
    The distribution that we found in the two studies of CC, CT and TT were very similar to what we saw in the IDEAL study. So the proportion of patients that had these different alleles was quite similar.
    With regard to the response with regard to IL28 -- if I could see the slide with -- yes, 642, please. This is the data from SPRINT-2, again, as an illustration of SVR by IL28 genotype. And in the white are the controls, in the yellow, the response-guided therapy, and in the orange the
    boceprevir 48 weeks. Now, I will caution you this is a retrospective analysis. It's only part of the data, but it does illustrate what we've seen before.
    In the CC allele patients that received peg/ribavirin for 48 weeks, they do very well. And as you can see in the response-guided therapy in the boceprevir 48-week group, when you add boceprevir to the mix, they also do very well with the CT and the TT patients.
    I would call to your attention, however, that in the response in the 48-week group, with two-drug therapy, it's a 48-week therapy, and we know that the CC patients that are treated with boceprevir, that 70 to 80 percent of them are in the early responder group.
    So if you treat a CC patient with boceprevir, they are very, very likely to be an early responder, get 28 weeks of therapy, and be done with their treatment in 28 weeks as opposed to -- they do also respond well with 48 weeks of peg/ribavirin.
    DR. CARGILL: Okay. Thank you.

    Ms. Valbh?
    MS. VALBH: I have a couple of questions about the anemia.
    What percentage of patients were on EPO and GCSF at the same time? And then out of that, what percentage stopped therapy because of compliance?
    DR. ALBRECHT: We do not have the data on the combination of EPO and GCSF. But if you remember from the GCSF, there wasn't very much of it used. I think it was 9 percent. So it was relatively low.
    I can't comment on the compliance with regard to the use of GCSF and EPO because we didn't collect that.

    MS. VALBH: And, in addition, what percentage of patients -- it seems to me, with the anemia -- I'm a little confused on the magnitude of the hemoglobin drop, because if a patient came in with a high hemoglobin and they were at 18 grams per deciliter, for example, and they dropped to 14 grams per deciliter, it's unclear, because it
    was at the discretion of the investigator to report that as an adverse event. And if they intervened, then it was reported as an adverse event. So it's not very clear to me exactly how much the hemoglobin was dropping based on the addition of the triple therapy.
    Can you comment on that?

    DR. ALBRECHT: We looked at -- and I believe the FDA did the same thing. We looked at the hemoglobin in two ways. We looked at the hemoglobin reported as anemia, which was, as you said, the investigators' discretion or responsibility to report that as an adverse event. But we also showed you the data for -- the lab data by hemoglobin grade.
    So if you look at the concurrence between the two, they're actually quite close, because investigators that were reporting anemia, basically, less than 10 grams, was matching the laboratory data fairly closely.
    DR. CARGILL: Thank you.

    Dr. Giordano?
    DR. GIORDANO: Two, I think, quick questions. First, could the sponsor please describe the exclusion criteria related to depression or psychiatric illness? And second, on the IL28 question, given this very small number of black participants relative to the overall study populations, were there any analyses done to see if IL28 genotype explains the differential response in the black patients?
    DR. ALBRECHT: We'll start with the depression. We have in our protocols exclusions for patients with severe depression or a history of severe depression. So, for example, patients with a history of hospitalization for depression would have been excluded from the trial.
    The protocol also had guidelines to reduce the peginterferon if depression became moderate, to stop the interferon if the depression became severe, and there were guidelines around also tracking the patient from the concept of calling them if they have reported depression or having them come in more frequently. And if the patient
    had severe depression, all three drugs were stopped.
    DR. GIORDANO: How was depression assessed? Was there a quantitative assessment or was it purely subjective?
    DR. ALBRECHT: It was the investigator's assessment of the patient or the investigator's assessment of the patient in cooperation with a health care professional if he sent them to another physician to be evaluated.
    We don't have the analysis of IL28 in the black patients.
    DR. CARGILL: Thank you.
    If you could just state your name into the record, please.
    DR. PACANOWSKI: Mike Pacanowski, genomics group. Essentially, there were a very few number of black subjects in the treatment-naive trial that had IL28B genotype data available. So the total numbers of black subjects who were CC was 16, CT was 40, and 38, which is the low responder genotype, TT; there were 38 subjects.
    The response rates, even despite the low number of subjects, did track in the expected direction, being incrementally lower in subjects who had one or more copies of the T.
    DR. GIORDANO: Were they still lower than the non-black population, or when you adjusted for that, did it even out?
    DR. PACANOWSKI: It's very difficult to tell just given the very small sample sizes. So you're dealing with 5 to 10 subjects in each of the treatment arms. So it's very difficult to rely on the percentages and estimates from that.
    DR. CARGILL: Thank you. That's helpful.

    Dr. Knodell?
    DR. KNODELL: I have two questions for the sponsor. Slide 107 says that the sub-analysis shows that the safety profile with CYP3A4 substrates or inhibitors was similar when administered with boceprevir.
    Now, I do a lot of endoscopy. We use a lot of midazolam. A fivefold increase in AUC, it seems to me, would probably keep somebody on boceprevir
    asleep for a couple hours after an EGD.
    The second question is, you talked about futility, and all through the previous discussions and your data, futility was positivity at 24 weeks. And then somewhere during the presentation, the question of calling futility a 12-week 100 international unit level of RNA was raised. And I wondered if there was some support for that or why that was inserted.
    DR. ALBRECHT: I'll let Dr. Kasserra answer your first question, and then we'll go to the futility rule question.

    DR. KASSERRA: We did do sub-analyses within the Phase 3 of the CYP3A substrates that were administered with boceprevir. They did, in fact, include midazolam. What we did not collect was the dose that was taken or the timing of the dose as compared with the doses of boceprevir.
    We did not see any unusual or remarkable safety profile with any of the CYP substrates that were co-administered with boceprevir. With midazolam specifically, generally, patients did
    take, for example, one dose, and it was impossible to tell from the data we collected whether, for example, they slept two hours longer.
    DR. ALBRECHT: We'll turn now to the futility rule. As Dr. Gottesdiener mentioned, in cooperation with the FDA and the scientific consultants that we engage, we have been looking for an earlier futility rule, and the way we did this was as follows.
    In the SPRINT study we assessed our patients virtually every two weeks until week 12 and then every four weeks to week 28. So we conducted exploratory analysis using these HCV RNA results included by treatment week, log decline in HCV RNA, detectable/undetectable HCV RNA using lower limit of detection of a lower limit of quantitation, and absolute values of HCV RNA.
    The number of patients that had been identified with the treatment week 24 rule was used as our basis. We didn't want to miss anybody who would become an SVR, and we didn't want to leave anybody on therapy who could be taken off earlier.
    So we found a number of single time points that were very good negative predictors of response. But we did not find a single time point that would replace treatment week 24. So then we looked for a two-step rule so that a proportion of the patients would come off earlier and the rest of them would come off at week 24.
    Could I have slide 1712, please? And this was the stopping rule that we came up with, and this is the basis for how we picked it out. The stopping rule is that if the patient is greater than 100 international units, on the left side of the slide, they're discontinued at week 12; and if they remained detectable at week 24, then they're discontinued.
    So the number of patients that would be stopped by the early rule would be 63, which is 9 percent. The additional patients that would be stopped at treatment week 24 would be 50 patients, for a total of 113 patients stopped.
    So essentially what this is doing, it's allowing about 50 percent of the patients to be
    stopped at treatment week 12 and the remainder of the patients then to be identified and taken off therapy at treatment week 24. And this was the way that we selected the rule.
    Now, we looked at multiple iterations around using other criteria, but this was the best rule that we found. The other reason that this rule is somewhat easy for physicians to adopt is that they're used to using treatment week 12 with two-drug therapy to think about stopping therapy in these patients.

    DR. CARGILL: Dr. McGovern?
    DR. MCGOVERN: I have two separate questions on two different issues. The first question is, did you have adherence data for the lead-in phase on ribavirin and peg? This would help us understand also the point that you made that you tried very hard not to include null responders, and yet about 20 percent of them were null responders.
    The second issue is a totally different issue. In terms of safety, did you require EKGs at baseline? And in terms of the deaths, which we
    haven't heard characterized, do we have any information on whether anemia was very severe in those patients prior to death? Is there any information about cardiovascular disease?
    DR. ALBRECHT: Yes. EKGs were required at baseline, and they were instructed in the protocols, the investigators, to perform analysis sort of cardiovascular monitoring, including EKGs, as clinically indicated. So the investigator was able to do this at any time.
    With regard to the deaths, as noted, there were eight deaths in the study, four in each treatment group. Two of the deaths in each treatment group occurred in the follow-up period between 18 and 20 weeks later. Two deaths in each of the groups occurred -- I looked within one week of stopping treatment, because the other ones were very far out. We had one death on boceprevir that was due to cocaine toxicity and the other one that was due to suicide. In the peg/riba alone, we had one suicide and one cardiorespiratory arrest.
    With regard to the number of patients that
    A Matter of Record
    (301) 890-4188
    275
    dropped out in the lead-in, it was about 5 percent among the various treatment groups that dropped out. It was higher in the naive study than in the relapse study. It was about 3 percent in the relapse study and 5 percent in the naive study.
    We did not make any decision about the patient's treatment based on the treatment week 4 result. It was a result that was collected, and then retrospectively or as part of our analysis plan, we looked at the effect of treatment week 4 on SVR and other parameters.
    DR. MCGOVERN: Well, I just want to know, though, do you have any adherence data on ribavirin during the lead-in?
    DR. ALBRECHT: No, we don't have any data on that.
    DR. CARGILL: Thank you.

    Dr. Murata?
    DR. MURATA: I have a clarification request from the sponsor regarding efficacy, using the week 4 log decline in the lead-in phase.
    Clearly, the agency has presented data
    A Matter of Record
    (301) 890-4188
    276
    regarding their analysis on one and half a log viral load decline. Thus, my understanding of the analysis that's been presented early this morning involved week 4 data at less than a log, although in the afternoon, some of the sponsor's slides also included the IDEAL study retrospective analysis using half a log.
    Does the sponsor have any additional comments on whether or not one log or half a log may be an appropriate surrogate?
    DR. ALBRECHT: I'll let Dr. Gottesdiener further speak about this.
    DR. GOTTESDIENER: I don't think it's the sponsor's goal to propose a new definition of null responders for the general worldwide treating population and practitioners. I think the purpose of the FDA bringing forward their 0.5 log drop was to say if you want to address the question of are these people like null responders, this is a little bit more stringent definition. And so they said even with that more stringent definition than the sponsor applied, we applied less than 1 log, the
    expected effect of boceprevir is the same.
    So I don't think either we or the FDA would really be proposing a new definition of null responders as much as saying these people were essentially comparable to no responders, and, therefore, we can estimate and give a good estimate of what happens with the null responders, as well.
    I do want to ask Dr. Poordad if he'd comment on that, because he's one of the people who actually deals with these null responders really on a day-to-day basis.
    DR. POORDAD: Thank you.

    My name is Fred Poordad. My conflicts are that I do research studies with both Merck and Vertex. I'm an advisor and consultant to both, and I am paid to speak for both. My trip here was funded through Merck.
    I do treat a lot of patients and I've been involved in all of these clinical trials, so I do want to make a couple of points that I think are important. One is that while the definition of the 12-week null is an important and useful one, it's
    really the issue of interferon non-responsiveness that's more important.
    To me, as a clinician, if I can predict earlier who is not going to do well with interferon, it's an important thing. The negative predictive value of the week 12 2-log response is 97 percent. The negative predictive value of the four-week 1-log is 95 percent. In other words, I can predict who is not going to respond at week 4 of week 12 equally.
    The concordance is about 90 percent, but it's not trying to say that week 4 and week 12 are synonymous. They're very similar in that you're defining or you're identifying people who are not going to respond to interferon.
    Someone asked a very good question earlier. Should we even be treating null responders, people that don't respond well to interferon? And I think that's where the clinician really has to make that decision on a case-by-case basis. Granted, a 30 percent SPR is not great, and we know there are better therapies coming down the road. However,
    there are some patients who probably need the therapy earlier. And I think on those cases, the risk-benefit ratio will tell some clinicians that we should treat now.

    So I don't think we're trying to cut hairs here with half a log or a log. Please understand that with the PCR technology, there is a half-log variability, and that, in and of itself, could certainly explains some of the discordance that we see. So it's simply a matter of redefining what we've come to think of as 12-week 2-log being paramount. It's really interferon responsive, poorly responsive, or somewhere in between.
    DR. CARGILL: Thank you.

    Dr. Camardo?
    DR. CAMARDO: Thank you. I had actually a couple of questions about the null responders. The first one is -- well, first of all, it seems like that's going to be a lot of the population that is available -- I mean, has hepatitis C. So we can't really exclude them from treatment. And you commented that doctors would want to treat them and
    would have to make some decision.
    So my question is, would the futility rule apply that we have developed using all of the data from treatment-naive and treatment failures? There's a lot of data there, so we'd have to extrapolate, I suppose, to null responders.

    The second question is, is there data that could identify null responders who might have a higher risk to taking the medication or would have some way to identify a patient who may have a benefit to not waiting, because then we could at least put out some guidance about how to at least recognize patients that might be treated.
    I don't know exactly who can answer that question.
    DR. POORDAD: Let me make sure I understand the question. How do we identify these individuals early?
    DR. CAMARDO: Well, how would you -- you commented that there might be some null responders who you should treat if you're a physician, and there are some who may -- you can identify.
    ADR. POORDAD: Sure.

    DR. CAMARDO: Where would you start?
    DR. POORDAD: Only speaking for myself, if I know what their fibrosis stage is, that's very useful. A patient who's at stage 0 or stage 1 can certainly wait a few years until we have better therapies for this population. Someone who's a stage 3 or stage 4, who's clearly progressing or has already progressed, then you may feel that a 30 percent likelihood of response is worthwhile.
    Someone made the comment earlier that 70 percent of these individuals will develop resistant variants, and this is true, and we don't know what this means down the road; will they be eligible for some other therapy. So I don't think we should rush to treat all interferon non-responsive people, but for individuals where the risk-benefit ratio makes sense, I think we should.
    I'd like to also point out there are other studies ongoing where we're trying to find other identifiers early on. I think IL28 is one of them. There are other ones. And we'll certainly continue
    to do trials to try and pare down better predictability rules about who we should treat today and who can maybe wait a few more years.
    DR. CAMARDO: Thank you.
    But what about the futility -- would it apply or would a patient --
    DR. POORDAD: Yes. No. Absolutely. I think the futility rule, as you saw there, is very important. You want to have a backup futility as we currently do with peg and ribavirin. We have a 12-week stopping rule today. We also have a 24-week stopping rule. The reason I like the 12 and 24 is it's not a departure from what clinicians already know. If the patient doesn't meet the week 12 futility, they may meet the week 24, and you would not overdose people needlessly.
    So, yes, the futility rule would still apply to those individuals.
    DR. CAMARDO: Okay. Thank you.
    DR. CARGILL: Thank you.

    Dr. Clay?
    DR. CLAY: I was wondering if the sponsor,
    over the break, had any opportunity to look to see the impact on the AKR alleles. When we had talked about the metabolism of the drug split between AKR1C2 and 1C3, that gets you to their inactive metabolites, and then also the other 50 percent that go through the CYP enzyme pathway.
    DR. KASSERRA: I'm sorry. Could you clarify the question exactly for me?
    DR. CLAY: Yes. I guess when I asked the question earlier today about if you looked to see if there was any impact that this drug had on expression of the alleles, and just in case somebody had some information back in the company and maybe you got it over lunch.
    DR. KASSERRA: So we do know that, for example, in blacks, that the diastereomer ratio is very similar to what it is in whites, approximately 2 to 1. We looked at the Phase 3 data. We also know in the different studies that we have done that that diastereomer ratio is not affected by, for example, drug-drug interactions or within the hepatic impaired population.

    We have looked at the literature, but there's very little information on the literature in AKR, as I'm sure you're aware. It's a fairly new field. In addition, it's a pluripotent family of enzymes and very ubiquitously distributed. So trying to understand what impact or where the expression might differ could be a challenging type of study.
    DR. CLAY: You also included in your materials to us that you had not conducted any p-glycoprotein interactions to date. Is that still accurate from the time you submitted your data?

    DR. KASSERRA: That is correct. What I can tell you is that the data we submitted at the time of the filing was limited on P-gp. At the time, we had done a Caco-2 study where we had identified boceprevir as being an inhibitor of P-gp with an IC50 of 25 micromolar. That data was submitted with the filing.

    Since then, we've continued to conduct in vitro studies. These data have been submitted to the FDA. They may not, at this time, have been
    reviewed by the FDA. I am happy to share that information with you.
    We did conduct a further study in MDR. It's a P-gp over-expressing MDCK cells. And in that study, which is more specific and more sensitive than the Caco-2 cells, the IC54 digoxin transport and inhibition thereby by boceprevir had an IC50 greater than 300 micromolar. So we would not expect a direct inhibitory effect of boceprevir on P-gp that would affect other P-gp substrates.
    DR. CLAY: Thank you.
    DR. CARGILL: Thank you.

    Dr. Ellenberg?
    DR. ELLENBERG: Thanks.
    There are a lot of time points that are relevant here where very important measures are taken. And getting back to the issue of the missing data, I wanted to know how far off something would be before it was considered missing.
    So, for example, if you want a week 12 measure, probably not everybody had it exactly on
    week 12. Some might have had it on week 13 or week 11, and the same with week 24 and the same with the follow-up week 24, the key points.
    So I would like to know, how wide was the acceptable window for those measures and what did you do -- you talked about the last observation carried forward, I think, for one of them, but what did you do about the others?

    DR. BOPARAI: Navdeep Boparai, statistics. The windows were three-week windows, and it depended on the rigid schedule. Some were tighter two-week windows. The last observation carried forward was just done for the primary efficacy endpoint, and it was just the 12-week was carried forward. But the windows were just plus/minus three-week at the time point.
    MR. TRAN: Could you identify yourself, please?
    DR. BOPARAI: Navdeep Boparai, statistics.
    DR. ELLENBERG: So with regard to the other ones, they weren't counted. I'm sorry. The on-treatment week 12 was carried forward to?
    A Matter of Record
    (301) 890-4188
    287
    DR. BOPARAI: The follow-up week 12 was carried to follow-up week 24. No on-treatment data was carried forward.
    DR. ELLENBERG: So if they didn't have an on-treatment week 12 measurement within three weeks, what happened to that?
    DR. BOPARAI: They were counted as failures.
    DR. ELLENBERG: They were counted as failures.
    DR. BOPARAI: Correct.
    DR. CARGILL: Thank you.
    Dr. Connick?
    DR. CONNICK: This is a question both to the sponsor and to the FDA. I think the implications of the virus -- the resistance mutations are really critical in determining the risk-benefit. I'm curious how you plan to approach that question. I'm also curious if you have tried re-treating anybody with boceprevir who failed, to see if they had the same type of response the second time, one of the people with resistance mutations.
    DR. ALBRECHT: We'll talk about the
    re-treatment of patients that have failed boceprevir first. No, we have not tried to re-treat any of those patients, and we currently don't have any plans at the moment to do so.
    With regard to the development of the RAVs, I think that one of the really important things that we need to remember about these new drugs is that if a patient is cured, that's fine. If the patient fails therapy, based on the data that we showed you, it's highly likely that the patient will develop RAVs, which is why we're using the early futility rules, why Dr. Poordad talked about how he would assess a patient that was a null responder.

    So we take this very, very seriously, and I think that physicians will have to think about the options that he has with regard to patients such as null responders before they decide to treat. But, certainly, for many other patients that are highly interferon responsive, the likelihood of having a sustained virologic response is very high, particularly in people that are early responders,
    and we're going to use a much earlier futility rule.
    DR. CONNICK: How are you planning, though, to look at the clinical significance of it in the future?

    DR. ALBRECHT: Dr. Hazuda, do you want to answer that?

    We're going to continue to follow these patients, for example. We have them in three-year follow-up. We're looking to see whether they're RAVs decline over time, which we've already shown you that they decline over time. And we're looking to see, over time, whether they go back to wild type, at least by population sequencing, and we're also doing some more extensive work on looking at the RAVs in the follow-up.
    DR. CARGILL: Thank you.
    Well, we've reached the point in our agenda where we are going to take a 15-minute break. We will leave here now and return promptly at 2:45. And, again, as always, the committee is charged with not discussing this topic outside of this
    meeting room.
    (Whereupon, a recess was taken.)
    Discussion/Questions to the Committee
    DR. CARGILL: We are now going to begin the panel discussion portion of the meeting. Although this portion is open to public observers, public attendees may not participate except at the specific request of the panel.
    Yes, Dr. Roland?
    DR. ROLAND: I just have a process question. I'm wondering how many more people you had on your list with questions.
    DR. CARGILL: Because of the time that we need to go through the discussions, we are going to have to terminate the question portion of the session so that we can move forward and provide and answers and guidance as requested by our FDA colleagues.
    Here are the questions. The first question, and there are a total of five of them: Please comment on the safety of boceprevir in patients with chronic hepatitis C genotype 1, focusing
    mainly on the hematological effects of boceprevir in combination with pegylated interferon and ribavirin.
    So we'll now begin that discussion.
    DR. SCHECHTER: I think it's clear that there are increased hematologic effects, that is, adverse effects, with boceprevir. But the thing that I would -- my impression is that these have been well managed; that they're reversible, that very serious effects are fairly uncommon, that is, the grade 4, the grade 3 and 4. And the issues of the EPO use, and there does not seem to be any serious effects from the EPO use. But I think to the coming studies or the proposed studies -- I don't know whether there is actually an ongoing study of EPO versus dose reduction -- is going to be a very important one for the guidance of clinicians.
    The fortunate thing, also, is that the neutropenia and thrombocytopenia are much less and that the grade 5 complications in neutropenia and thrombocytopenia were pretty uncommon, grade 3 and
    4, I meant, and the fact that there didn't appear to be any deaths due to neutropenia or thrombocytopenia.

    DR. CARGILL: Thank you.

    Dr. Friedman?
    DR. FRIEDMAN: I'd like to concur with that view. As a hepatologist, I can say, at least for hepatologists, we've become very comfortable dealing with the anemia issues from peginterferon and ribavirin. So there will be a little more of it, but we can handle it.
    I really would like to see a study that compares EPO with dose reduction of ribavirin. I have shied away from using EPO because at least of theoretical concerns about thromboembolism and the case report of aplastic anemia. But with boceprevir coming out and the greater degree of anemia, there may be a greater tendency of physicians to want to go to EPO. So I think that issue ought to be settled.

    DR. CARGILL: Thank you. Dr. Clay?
    DR. CLAY: I think the data is really clear
    that this is a huge advancement in the treatment of hepatitis C and that in the appropriate situation, this drug is going to make huge differences.
    But as has been brought up by the members of the committee, as well as the public in attendance, education is going to be absolutely key. And while I would commend Dr. Friedman and the rest of the gastroenterologists and hepatologists here, I think the numbers of people that may begin seeking therapy for hepatitis C are going to increase to the point where they will seek their family physician, their primary care provider, and get treatment from them, and that education across the board, the Katie Couric effect for prostate cancer, if you will, really is going to have to be replicated in order for people to understand not only the importance of the therapy, but the importance of getting proper therapy and adhering to that proper therapy.

    I also think special attention needs to be paid to the fact that by people who become anemic or have a drop in hemoglobin responded really is
    encouraging the use of EPO, but yet that has not -- there's not approved labeling of use of EPO in this population. And so I would concur with Dr. Friedman that it would be wise to get some quickly assembled data as to some guidelines that could be used for the primary care physicians out there.
    DR. CARGILL: Thank you.

    Dr. McGovern?
    DR. MCGOVERN: In my clinical practice, it's been my usual standard to try to only decrease ribavirin after virologic suppression has been achieved. And so I'm hopeful that with this potent agent, that we're going to get to viral suppression much quicker and that the dose reduction of ribavirin will hopefully be something that is certainly going to be doable and will be safe in terms of bringing hemoglobins to a better level.
    One of the things I did notice in the trial was that about 70 patients did not have a change, either a dose reduction or the addition of EPO, even though their hemoglobins were less than 10.
    And I'm not sure if we're giving a crossed message that if you're anemic, you have better SVR rates. I think that's just a reflection of better drug exposure, and it doesn't mean that anemia is a good thing and we should be pushing it. But I do think that -- I'm hopeful that when I do back off on ribavirin, it's going to be because most of my patients are suppressed, which is really exciting.
    DR. CARGILL: Thank you.

    Dr. Ghany?
    DR. GHANY: Yes. Thank you.
    So my comment also has to pertain to the management of anemia. I guess what I'm going to say is going to partly echo what's already been said. But based on some of the data that were shown by the sponsor, I think on slide 92, that with dose reduction of ribavirin alone, albeit in a small number of patients, the SVR rate is equally as good as patients who are managed with just EPO alone or with a combination of dose reduction plus EPO.
    So given the concerns with EPO, including
    all the side effects, the cost and the fact that there's been no study that's demonstrated an effect of EPO in increasing the rate of SVR, I would like to suggest that the first approach to management of anemia should be dose reduction of ribavirin, and the use of erythropoietin should be reserved only if the ribavirin dose has to be reduced to a level where it would affect SVR, and I wonder if such data exists, or to preserve the use of boceprevir. Otherwise, I don't see a role for erythropoietin.
    The one other point that I wanted to make was that when -- well, let me retract that first sentence. Assuming this drug is approved and is widely available, the concern about anemia is going to be certainly greater, especially for patients with -- the oldest patient with cardiovascular disease. I am concerned about the incidence of angina and MIs in patients with anemia, particularly in the older population and those with co-morbidities, such as diabetes and hypertension.
    Thank you.

    DR. CARGILL: Dr. Van Dyke?
    DR. VAN DYKE: I likewise think that we can probably manage anemia, and I'm particularly reassuring that it resolves so promptly after the therapy is stopped. I think that's a good sign. I must say I have a little more concern, maybe just theoretically, about the neutropenia. It was relatively common. There were 22 percent with grade 3 and some percent with grade 4, and there were those three events that were life-threatening potentially; I mean, relatively uncommon. But I worry that as we treat more patients, those with co-morbidities, those with other medications, I think we might get into more problems with neutropenia, and I think those are potentially the life-threatening conditions that we need to be more aware of.
    DR. CARGILL: Thank you.Dr. Strader?
    DR. STRADER: I'd like to echo Dr. Ghany's comments. I, too, think that perhaps the use of EPO is not absolutely necessary. We don't have any data that shows that it really does improve SVR
    rates.
    I'm a little bit concerned about the difference between a magnitude of hemoglobin change and an absolute number. It seems like when we talk about grade 1 through 4, we're looking at numbers as opposed to a magnitude change. I'm not sure; maybe it doesn't matter whether your hemoglobin goes from 18 to 14 as opposed to goes from 12 to 8. But there may be some benefit to a recommendation about monitoring these patients on a regular basis. And I'm not sure whether the magnitude of hemoglobin change is as important as the absolute number.
    DR. CARGILL: Thank you.

    Dr. Ellenberg?
    DR. ELLENBERG: I'm assuming that as there's an ongoing study looking at the use of EPO, that the FDA is probably going to wait until they get those results before they say anything about it.
    But I will say that I'm glad that the sponsor did do a study that allowed us to see -- have some data with regard to the use of EPO
    because it would be inevitable, I think, with this kind of side effect that people will use EPO. So at the very least, we've got some initial data that I think is reasonably reassuring that we're not going to do a whole lot of damage if people use this to treat anemia.

    DR. CARGILL: Ms. Dee?
    MS. DEE: I'm a little bit more worried than I'm hearing people talk about. I'm not sure what's going to happen when less people get to use EPO. Maybe it won't affect SVR, but I'm wondering if we'll have more cases of anemia and whether they'll be more serious.
    I worry about the neutropenia, as well. I think there's a strong signal there that there are issues. And I also wonder about what's going to happen in Peoria when this gets out and people aren't really aware or don't understand -- the best thing Dr. Clay said, when primary care docs start to prescribe this and aren't really aware of the problems or issues that might arise. I really fear for patient safety.
    I'm also worried about the idea that -- I mean, I think these things are manageable, but I just wonder if we're going to have some issues. Two or three cases of deaths will set this whole field back a really long way, and that concerns me.
    So that being said, I think that it looks like this is manageable. I think that that PROVIDE study, though, I'm really disappointed that that wasn't done way before this, especially since EPO was allowed in this. It would have been very helpful to see the difference between EPO and ribavirin reduction way before this.
    I might be mistaken, but I don't think there is any data on what's going on with that yet, so I look very forward to seeing that data when it comes out. But I don't know, I don't think there is any data yet.

    DR. CARGILL: Ms. Young?
    MS. YOUNG: I just would like to agree with most of the comments that have gone on before in terms of using the reduction, ribavirin reduction versus the EPO, to start with, doing some more
    studies on that not only for the sake of the provider and the health care system, but also for the complexity of the treatment for the patient and the side effects. I think adding one more drug is not necessarily the best thing.
    DR. CARGILL: Thank you.

    Ms. Valbh?
    MS. VALBH: I also want to echo some of the thoughts that everybody has talked about today with the use of EPO. Even though it was managed very well in the clinical trial, I think, in practice, we have to understand that there are providers who don't quite follow the guidelines right now for management of ribavirin-induced anemia. And I would like to see tighter controls and guidance on how often CBCs are to be assessed in these patients and more education aimed at the primary care physicians who are prescribing and treating hepatitis C patients.
    In addition, even though 40,000 units once a week is used very well in this population, maybe we should talk about weight-based dosing of EPO to
    potentially avoid some of the side effects of EPO. We all know the adverse effects of EPO and recently with some of the tighter guidelines on the current indications for it. We don't know if that's going to happen with hepatitis C patients, but if we start using it quite often with hepatitis C patients, we may essentially see the same types of adverse effects. So maybe more of a weight-based dosing versus a standard 40,000 units once a week should be considered.
    DR. CARGILL: Thank you. Although I would again remind you that the use of EPO for the treatment of anemia in HCV is an off-label use of the drug and, therefore, I think that's beyond our purview right now.
    All right. So I'm going to try and summarize this. It appears that we have reached a consensus on this question, and that is that while there have been several concerns voiced about the hematologic effects which are felt to have been increased with boceprevir, they appear to be mainly manageable; that there are multiple options for
    dealing with these, especially if they're, in terms of anemia; first, ribavirin dose reduction and then, if necessary, proceeding from there.
    Several have voiced the opinion that this appears to be a great advance in hepatitis C management, but that it's going to be important to have appropriate patient selection. I think we've already heard one answer to one of the questions that's going to be coming up later on, which are several recommendations about dose reduction versus another agent for the management of anemia.

    Several of our panel members also pointed out the concerns about neutropenia and what this is going to mean as we go forward and if this drug is indeed approved, what this will look like when larger numbers of people are exposed to this agent, perhaps the experienced may have more profound adverse effects.

    But overall, I think what we heard is that the feeling was that these were manageable events, but that there would certainly need to be attention to patient selection was a recurrent theme, as well
    as to education of those who are going to use this, as this would be potentially used by individuals who do not have expertise or robust expertise with hepatitis C management in certain populations, as it may be used more by generalists.
    Question number 2: Considering the overall potential risks and benefits of boceprevir, do the available data support approval of boceprevir for treatment of patients with chronic hepatitis C genotype 1 in combination with pegylated interferon and ribavirin?
    I think I'm actually going to go ahead and take -- I'm sorry.
    We'll be using the new electronic voting system for this meeting. Each voting member has three voting buttons on your microphone: yes, no and abstain. You might want to take a minute to look at that now and make sure you find them.
    Once we begin the vote, please press the button that corresponds to your vote. You'll have approximately 20 seconds to vote. And after everyone has completed their vote, the vote will be
    locked in.

    After we have our discussion -- I just wanted to give you that information now -- the vote will then be displayed on the screen.
    So we're going to go ahead and open question 2 for discussion.

    Dr. Roland?
    DR. ROLAND: One of the questions I asked that did not get completely answered would potentially have an impact on my evaluation of efficacy. And so I asked the question about loss to follow-up after week 12 and was provided an answer on one of the studies, but not on the other.
    If the answer on the other study was similar to the answer on the first study, which was a very small number, sensitivity analysis showed no difference, then I would feel confident.
    So it would be helpful just to -- I mean, I'm assuming that FDA reviewers are relatively brilliant and would have picked up if that had been a problem, but I would feel sort of remiss in my responsibility by not confirming that.

    DR. ALBRECHT: Do you want me to answer the question?
    DR. CARGILL: If you can do it succinctly, yes. Thank you.
    DR. ALBRECHT: Yes. This is the RESPOND-2 study and I probably -- we can't put it on the screen maybe, so I'll read it.
    There were -- here it is on the screen. This is the same format as the one I showed you. And as you'll note on the bottom, there were no patients in the control, one patient in the arm, two response-guided therapy, and one patient in the boceprevir 48-week therapy that were lost to follow-up and carried forward. So the results in this study, there were even fewer patients than in the SPRINT-2 study.
    DR. CARGILL: Thank you.

    Dr. Clay?
    DR. CLAY: Maybe mine is more procedural than anything else, but in this sentence, as I read it, it doesn't specify the type of chronic hepatitis C infected person, and we're being asked
    to vote on it. But then we have subsequent questions in which we're being asked to provide an opinion for various subpopulations of this, but we're not being asked to vote on those.
    So I was curious if there was a way to amend this to reflect a particular population or if we were going to take a vote on the subpopulations after the discussion took place.

    DR. MURRAY: Well, we usually have one voting question which relates to does this drug make the cut for marketing approval, and I think then picking out certain subpopulations, duration for the subpopulations are more labeling issues, which, when we look at the advice, then we, FDA, looks at kind of the totality of the advice and what was said and take that all into consideration when we actually do the labeling.
    So this is just an overall risk-benefit for chronic hepatitis C patients, genotype 1, is it worthy of approval, and then the granular questions will help us with the labels that follow.

    DR. CARGILL: Dr. Friedman?
    DR. FRIEDMAN: Well, I just have to say, like many people on this panel, I've been taking care of hepatitis C patients for a long time and started out in the business when it was called non-A/non-B hepatitis, and there was no treatment.
    Then interferon was approved in 1991, and we were curing, and we didn't use the word "cure," but we had sustained responses in 10 percent. And for the past five or six or seven years, we've been having overall sustained response rates of 45 percent for genotype 1, 55 percent overall. And now, to go to 70 percent really seems like a dream come true for those of us who have been in this field.
    There are a lot of problems. This is difficult treatment. So is cancer chemotherapy. But this really, I think, is a major advance. And so I'm very enthusiastic about approving this drug, recognizing that there are a whole host of challenges, including the complexity of the regimens, the cost, and the need for additional data. But if you look at the big picture, I think
    this is a remarkable advance over the spectrum of the past 25 to 30 years.
    DR. CARGILL: Thank you. I'm actually going to take the chair's prerogative and actually talk about the question, and I would like to echo that and expand it a bit further.
    I haven't been taking care of patients quite that long, but I practice in a setting where, unfortunately, approximately 95 percent of our patients are hepatitis C co-infected; 99 percent of that population is African-American.

    So, in addition, the difficulties of trying to provide the standard of care for them, going through that standard of care and seeing the outcomes is very often disappointing.
    So while I would concur that I have concerns about the complexity of this regimen, I have concerns about people being able to adhere to pill burdens and to some of the other challenges, as well as the hematologic monitoring, which I think will be quite important, I think looking at the faces of a numbers of people who have had poor
    options and even poorer outcomes. And to see something like this is truly a chance to say that -- as caregivers, we often advocate and have to provide hope, and it makes you feel like you can come back to the office with some hope.

    Ms. Young?
    MS. YOUNG: I wanted to ask the FDA. If there is an approval, it's a blanket approval, correct? And then physicians can use it off-label. I guess I'm getting at the question of some of these concerns that we might have about drug-drug interactions and those kinds of questions that might change the standard of care that would be recommended, do we get to go back and look at those or is that something the FDA is going to monitor post-market surveillance?
    DR. MURRAY: Well, question 5 allows you to comment on what studies you would like to see. Yes, we will be requesting post-marketing drug-drug interaction studies and then also have a labeling that reflects some of the uncertainties for probably certain drug interactions or probably for
    certain 3A metabolized drugs, they'll be recommended not to use. So we are going to be kind of ferreting that out.
    DR. CARGILL: Okay. I think at this point, we really are ready to vote. So I would like to just read the charge again before you vote and remind you that do you have the three voting buttons on your microphone, yes, no and abstain. You'll have approximately 20 seconds to vote. After everyone has completed their vote, the vote will be locked in.
    The vote will then be displayed on the screen. I will read the vote from the screen into the record. And then next, we will go around the room and each individual who voted will state their name and vote into the record, as well as the reason why they voted as they did.
    So as this question is now complete and there is no further discussion on this question, we will now begin the voting process. Please press the button on your microphone that corresponds to your vote. You will have 20 seconds to vote, and
    please press the flashing button firmly. If you are unsure of your vote, please press the corresponding button again.
    [Voting.]
    DR. CARGILL: Okay. You can see the voting results here. There are 18-yes, zero-no, and zero-abstain.
    Dr. Knodell, if you could lead us off, please, with your vote and why you voted that way?

    DR. KNODELL: Yes. Robert Knodell. I would echo Dr. Friedman's remarks earlier. I think this is a tremendous advance and is going to give a lot of people hope that have already failed treatment once, but are likely to respond to treatment with these new drugs.
    DR. CARGILL: Thank you.
    DR. KORMAN: Louie Korman. I voted yes, for the same reasons. This is an important advance in treatment. I'm encouraged by the work but concerned about our ability to risk stratify and select patients, which is my misgiving. But that's always my misgiving taking care of patients.
    DR. CONNICK: Elizabeth Connick. My vote was yes and for the same reasons. There are risks, but the benefits for the people who do achieve SVR are fantastic. This is a marvelous advance.
    MS. VALBH: Pritybala Valbh. I voted yes because this is definitely a much needed advance in the treatment of hepatitis C. And for all those patients who have been on treatment before and who have not responded, this is the answer for them, I think, for the most part. So, therefore, I voted yes.
    DR. GHANY: Marc Ghany. I voted yes because I think the data speak for themselves. The efficacy, in my opinion, far outweighs the risks. And I think, actually, this was the easiest part. I think the challenge is now going to be how to use this medication in a broader population and in subpopulations of hepatitis C, who have probably the greatest unmet need and stand to benefit the most from treatment. So I think the easy part is over; the hard work is going to begin.
    MS. DEE: Lynda Dee. I would agree with
    that assessment. I think that boceprevir performed better in all the arms as opposed to the standard of care. That being said, I'm not sure what to do with everybody in these different categories, as everybody has described, and I think it's my job to be a pain in the neck about those things and to stress patient safety. But as one of my colleagues said at the open mic period, I think this is 21st century advancement, and it's very welcome.
    DR. MCGOVERN: My name is Barbara McGovern. I voted yes. I think that this is going to be a real game-changer for our hep C practices, and I can't wait to get back and to talk to my patients about it. I do urge the sponsor to move on with the trials in co-infection because some of my patients are HIV co-infected, and I would very much like to know how to approach that patient subset.
    It will also be very important to see how this drug will -- the triple therapy paradigm now will play out in the community.
    DR. ROLAND: I'm Michelle Roland, and I voted yes because of the clear incremental benefit
    of this drug and the manageable known risks. I have significant concerns about the unknown drug interactions, which I think will have potentially really big implications for the populations in which this drug could be useful and would urge the sponsor to complete those drug interaction studies as quickly as possible.
    DR. ELLENBERG: I'm Susan Ellenberg. I voted yes. I'm a biostatistician, so I don't treat patients, but I do look at data, and I think these data show pretty clearly that the risks, while clearly there, as for any effective treatment, they're clearly outweighed by the potential benefits.
    DR. CLAY: Patrick Clay, and I voted yes. This drug is going to exponentially improve the outcomes of people infected with hepatitis C. I do have concerns with regard to some of the additional testing that's yet to be completed, but given Merck's history of successful implementation in the community of a three-times-a-day drug and treating a viral infection, I have confidence in them that
    they're going to develop what is necessary to make this happen.
    DR. CARGILL: I'm Victoria Cargill. I voted yes for three reasons. One, I think it represents a clear advance over what we've had. Secondly, I do agree with Dr. McGovern, I think it changes the game completely and hopefully continues to advance the field. And, third, because I can look into the faces of the people that I have to see on Thursday afternoons and be able to offer them some hope.

    DR. STRADER: I'm Doris Strader. I voted yes because I think that the data does show that in patients with hepatitis C genotype 1, there is a very good improvement in sustained virologic response over pegylated interferon and ribavirin alone.
    I wanted to emphasize genotype 1 because we seem to be getting very excited as to all patients with hepatitis C, and we can treat with this. And these data were done particularly with patients with genotype 1, so we need to make sure we keep that in mind.
    I do have some concerns about some of the other populations that may not have been adequately represented in the studies, but I am sure that we can get to those patients and be able to provide them with care, as well.
    DR. VAN DYKE: I'm Russell Van Dyke. I voted yes. I think this drug is going to help eliminate infection from a substantial number of people, and I think that makes it worth dealing with the adverse events, which I think is going to be challenging, but worth it. I look forward to a pediatric formulation and pediatric studies.

    DR. GIORDANO: Tom Giordano. I voted yes. Basically, the risks are outweighed by the benefits. I think the risks are not trivial. We do know how to manage these risks, but there's clearly an increase with the experimental medication that is really not explained, and I think there's a lot to learn about how to use these drugs appropriately. But the benefit outweighs the risk at this point.

    MS. YOUNG: I'm Kathy Young, and I voted for
    Aapproval because I feel that the benefit outweighs the risk and there's a clear need and also to encourage the pharmaceutical companies to develop this kind of novel agent. I think it's important, but at the same time, working on the complexity of the treatment and having agents come down the pike that really would be simpler to use I think is very important with the health system as we are right now moving in toward national health insurance.

    DR. SCHECHTER: I'm Geraldine Schechter. And as a hematologist working in a VA hospital, which has about 20 percent incidence of HCV, and my hematologic cancer patients have that incidence and complicate their disease, the benefits really outweigh the risks. And the hepatologists, with some help from hematologists occasionally, will do very well, I think.

    DR. FRIEDMAN: I'm Lawrence Friedman. I voted yes for the reasons I outlined previously. I do think you have to be somewhat of a Talmudic scholar to prescribe this drug.
    [Laughter.]

    DR. FRIEDMAN: But that has a certain appeal to me.
    [Laughter.]

    DR. MURATA: I'm Yoshi Murata. I voted yes for the favorable risk-benefit profile that was discussed today in support of an approval.
    DR. CARGILL: All right. We'll move on to question number 3. Please comment on the strength of the evidence for use of boceprevir in combination with pegylated interferon/ribavirin in prior null responders, defined as less than 2 log decrease in HCV RNA at 12 weeks during previous course of peg/ribavirin therapy who were not included in the Phase 3 trial P5101 in subjects who had previously failed peg/ribavirin therapy.

    Dr. Korman?
    DR. KORMAN: This is where the Talmudist in Dr. Friedman is going to come to the fore as he leads us into a pill pull over whether to do this.
    My concerns about safety and efficacy and risk stratification really are brought to bear here, because this is the population where, if
    we're not correct and we're not careful, we may make these patients, who presumably may be at greater risk for progression -- I don't know that, but maybe -- may cut off an avenue of therapy in advance of newer agents that may be more effective.
    So this is of a concern, and I would prefer to see more data and a better understanding of risk stratification.
    DR. CARGILL: Thank you.

    Dr. McGovern?
    DR. MCGOVERN: I think that the FDA analysis with the 0.5 threshold was more convincing to me that what happens at week 4 bears out to week 12. So I thank them for that analysis.
    There is no question in my mind that there is a subset of these patients who are poorly interferon responsive that have a good response, about a 30 percent response, and that's real and something I can wrap my head around.
    I think that we need to target that, though, towards the patients who really need treatment today in terms of advanced liver disease, because
    although there's the 30 percent that are responding, there's the 70 percent who are at risk of viral resistance. In terms of the next wave of protease inhibitors, there are going to be issues of cross-resistance across the next four drugs that are close behind.
    So I think I would like to have some direction by the FDA or that we direct the FDA in terms of the labeling that these patients were not explicitly studied and that we are inferring and we are massaging the data. And I just wish that actually they were studied from the get-go, quite frankly, because the caution about not studying them was because we didn't want to give essential monotherapy. And yet we found out that, yes, those patients who are not interferon responsive within that four weeks, they are the ones at risk for resistance. So we did find what they were worried about finding.
    DR. CARGILL: Thank you.

    Dr. Strader?
    DR. STRADER: I would like to echo some of
    Dr. McGovern's comments. I, too, am suspicious of the -- albeit very elegant explanation of how the patients who were initially excluded from study may benefit from this drug, I think that we do know that there are drugs coming along that may be better and hopefully simpler to use. We do know that patients who are cirrhotic or who are null responders may be at increased risk for adverse events. There is some concern about resistance variants showing up in these patients. And I think that in our haste to want to do something, we should probably do no harm.
    There was a reason that these patients were excluded initially, and I think that we have to keep that in mind. We do want to do the best we can for the patients who are null responders. And so I think that we owe them the best that we can possibly do. And rather than inference as to how they might do based on some other group, we should really know how they're going to do before we subject them to a very complicated and potentially dangerous -- or increased adverse events.
    DR. CARGILL: Thank you.

    Dr. Giordano?
    DR. GIORDANO: If you actually read the question and take a literal interpretation of the question, the strength of the evidence for use of boceprevir in combination with peg and ribavirin in non-responders, prior non-responders, is zero. There is no evidence. They were excluded from the study. So we're left to extrapolate from data that rely on a number of assumptions and are based on small numbers.
    The science of health care has been -- or the science of medicine has been burned many times by making inferences from similar but not exactly the identical population that you're trying to treat. So I would say a strict interpretation is there is no evidence and a more generous interpretation is the evidence is weak to moderate, at best.
    DR. CARGILL: Thank you.

    Dr. Ghany?
    DR. GHANY: Yes. Actually, Dr. Giordano
    basically stole my thunder. That was how I was going to answer this question. But if you read the question literally, the evidence, in fact, doesn't exist. It still concerns me why the sponsor chose to not include these patients and then later come back and try to get an indication for this particular subgroup of patients.

    I think if we stretch the data and say that there is some data that the drug is efficacious in null responders, we really have to be careful. And I think I agree with Dr. Poordad that if you're going to consider treating a null responder, you really need to individualize therapy for each patient, because we can do harm, particularly if these patients develop antiviral resistance. There really isn't a lot of data on what the re-treatment of these patients will be with other agents, and we may be doing them more harm than good.

    So I think that is something that we all need to be cognizant of. There are going to be better treatments coming down the road. And particularly for individuals with mild disease, I

    would strongly suggest that they wait and not be treated and certainly wait until we have more information about treatment in larger groups of patients who are null responders.
    DR. CARGILL: Thank you.

    Dr. Friedman?
    DR. FRIEDMAN: So I wish a study had been done in which these patients were included prospectively, but that wasn't the case. I'm a clinician, not a scientist, so you were able to convince me that about 30 percent of these patients will respond.
    What's happened is we've built up a reservoir of non-responders to peginterferon and ribavirin, and they are waiting in the wings for the next generation of drugs to come along. And probably some of them can continue to wait, but some of them can't, and we're going to want to use them on some of these patients. It's going to be selective, and it's going to be the people who have advanced fibrosis. We're going to be concerned about resistance and an increased risk of side

    effects, but we're also going to be concerned that we're running out of time.
    So I personally would like to see this indication for the drug, but I'd like to see it used very selectively.
    DR. CARGILL: Thank you.

    Dr. Knodell?
    DR. KNODELL: I would like to again echo what Dr. Friedman has to say. I will accept these data. I think that they took people who had failed dual treatment and put them in a study, and 30 percent of them responded.
    I'm looking at these old milestones in therapy of chronic hepatitis C, and, I mean, we got pretty excited when we got a 6 percent response with interferon, and then 16 percent response treating, and then we added ribavirin and we got 34 percent.
    So we're talking about a response rate that certainly is 1 out of 3. And I would echo, again, you're not going to treat people with stage 1 fibrosis or a G1S1 Ludwig score. You're looking to
    treat these people with G3S4 or some of those folks where you see a very active hepatitis problem. You're not going to treat null responders with minimal disease.
    Again, you have your futility barriers in place, a 12-week and a 24-week, and I think the incidence of variants is not real high in these short treatment periods, and I think it's unlikely that you're going to remove these people from further therapy, therapeutic considerations.
    DR. CARGILL: Thank you.

    Dr. Clay?
    DR. CLAY: First off, I always get concerned when you hear that there are future things coming, because you never know -- I guess you don't know what you don't know. But one of the things that we can't control is should this drug get approved, who is going to use it and in what population. But what can be controlled, if that's the correct word, is if this drug is supplied, then that individual is enrolled into a data registry; that the elements of data that are captured would be that ideal to
    answer the questions of these experts around the table where you are simultaneously providing a drug to a patient that desperately needs it, even though the formal studies have not been done. But over time, we may gather enough information that we are able to make better decisions moving forward while the more rigorous structured trials are being conducted and/or other potential agents are brought to market.
    DR. CARGILL: Thank you.

    Ms. Dee?
    MS. DEE: The ideal data -- I mean, that seemed like torture the data until it confesses to me. I'm just not convinced by that at all.
    The 30 percent of people, they were really small numbers. And I didn't get to ask this question, but on page 12 of the sponsor's briefing document, it says that in vitro after 15 days, there was a 2 log drop of RNA with this drug.
    So I wonder -- I know that hepatitis resistance develops quickly, and I would think that resistance to this would develop quickly no matter
    what the futility rules are. And that might mean that patients are resistant to boceprevir and cross-resistant to other HCV BIs.
    I guess the question is how badly does a patient need it, and, again, I'm going to agree with Dr. Clay. In what population should we think about this?
    I mean, these people were not in the trial, and I think that the label should really clearly state that they were not included. I don't know. This is going to be a long label, is all I can say.
    [Laughter.]

    DR. CARGILL: Dr. Ellenberg?
    DR. ELLENBERG: I guess I'm more leaning toward the side of those who see the arguments that were made by both the FDA and the sponsor that, in fact, these patients were represented in the studies, and that I'm prepared to accept that it's more likely than not that there will be approximately a 30 percent response rate.
    Whether that's enough to treat a patient seems to me like a very individualized decision,
    and I would not like to tie hands of clinicians if I felt that this, in fact, will have a substantial response rate, to tie the hands of physicians who think it's appropriate to use in some cases.
    I understand the cautions that people have raised. I'm not exactly sure what would be put in the label to help that out. But it seems to me -- I'm ready to believe that there is a response rate in these that's sufficient that it ought to be available to this patient population.
    DR. CARGILL: Thank you.

    Ms. Young?
    MS. YOUNG: I guess I am leaning more towards the prior null responders not being included because the evidence wasn't there. Also, I think it's very important that this drug be used prudently to start with, so the risk stratification idea is something I'd support. And it would be nice to have a trial or a launch that was successful, and I think proving success in the groups that really would meet a risk stratification first and then maybe considering it for the other
    ones that are a little more questionable would be one way to go.
    DR. CARGILL: Well, this is going to be a little bit more challenging to summarize, because it seems our consensus really splits down into a fork in the road.
    We've certainly heard that there are concerns being raised about potential safety issues for null responders in addition to what potential other outcomes could happen to them; for example, not only just not responding, but perhaps closing future options, the development of viral resistance.
    We've heard from people who are hepatologists on the panel that they are comfortable with the notion that they could still treat null responders, for several reasons, not the least of which is that one would take into account individualized therapy; that, secondly, one would also take into account the clinical picture of the patient so that the therapy can be individualized, being offered mainly to those individuals who are
    clearly ill as opposed to those who could perhaps afford to wait.
    Finally, we heard some suggestion that perhaps some of the futility rules would be of assistance in this regard.
    But we also still did hear a minority opinion of those who felt that this was a patient population that should not be treated in this way or should be treated with extreme caution and perhaps even the offer of a registry, so that in exchange for being able to treat a patient that falls into this category, that there would be the opportunity to capture data so that we could learn something about this patient population.
    So I think we've seen that the committee remains as divided as it was at the outset of the discussion in this particular area.

    Yes, Debra?
    DR. BIRNKRANT: Given that we have all of this expertise here, I was wondering if I could ask if this were formally studied, is there an SVR rate below which you wouldn't feel comfortable using?
    Everyone is mentioning the 30 percent response rate. Is there a rate that if you had studied this formally, you would not be comfortable with using?

    DR. CARGILL: Dr. Knodell?
    DR. KNODELL: I'm going to sidestep this. The question is difficult to answer because it depends on how sick the patient is. You get somebody who is really -- the data is clear. If you get a sustained response rate in somebody who has cirrhosis, you absolutely cut their rate of hepatocellular carcinoma almost down to nothing.
    So when you've got that kind of clear-cut result that you can get clinically, you're willing to take a 1 in 5, 1 in 6 or 7 probably. If you get down in the low 10 percents, the low double figures, then you've got to ask yourself whether the benefit outweighs the risks. But, boy, once you get up in the 20 percent and certainly in 30 or 40 percent rate, when you start thinking of what you can accomplish by an SVR, it's really pretty great.
    ADR. CARGILL: Thank you.
    Ms. Dee?
    MS. DEE: I think that's right. I think that if you're going to exclude it from the label, then I think that's more important. But if you're not, as somebody said before, years ago, they were happy with 5 or 6. So I think it's up to the patient to decide whether they're willing to take a risk. I just would like to know what the data is and what the evidence is. So after that, I think I would like to talk to my doctor about whether it works for me.

    DR. CARGILL: Dr. Ghany?
    DR. GHANY: Actually, because of the known benefits of SVR, I think that the more relevant question is what is the acceptable risk that we're willing to take, and for that I don't think we have any data.
    So I don't think I can answer that question, because if you ask a hundred hepatologists, you're going to get a different number from everyone.
    DR. CARGILL: Did you want to continue to
    hear comments, Dr. Birnkrant, or did you get the answer you needed?
    DR. BIRNKRANT: I think I got the answer, for the most part.

    DR. CARGILL: Dr. Clay?
    DR. CLAY: I think the answer to your question would come from the registry. Over time you're going to see people's response rates. And then over time, you will be able to see what has been accepted and what has not been accepted. And the experts in the field, in conferences when that data is presented, they're going to begin to hone down which population it is over time. But it is an individualized basis until we have data to show us it's not going to be done.
    DR. BIRNKRANT: Do you think the futility rule should be different for this population then, because we're concerned about functional monotherapy perhaps?
    DR. CLAY: Time will tell. I'm sorry, we don't have the data here.
    DR. CARGILL: Dr. Van Dyke, do you want to
    have the last word on this?

    DR. VAN DYKE: Well, I'm going to skirt the question, as well, but it seems to me the downside is resistance, principally. So I think what we don't know is, A, does resistance persist, for how long. This isn't HIV where you've got pro-viral resistance there forever. So maybe if you wait long enough, the resistance goes away.
    I personally don't know how much cross-resistance there might be with some of the other agents then in development, but that would be an important issue, as well, because if there are different resistance patterns, then I think you might be more willing to take the risk of developing resistance if you -- even a 30 or 40 percent response I think would be great in these people that currently have no response at all.
    DR. CARGILL: All right. So we'll go on to our fourth question. As you can see form the screen, it's fairly complex, so we'll break it up into pieces; speaking about Talmudic scholars.
    Please comment on the strength of the
    evidence to support response-guided therapy with boceprevir in combination with pegylated interferon and ribavirin.
    Should certain groups of patients receive longer durations of boceprevir plus peg/ribavirin therapy than that evaluated in the response-guided therapy arms?
    So let's start with that, and then we'll go through the questions targeting, first, treatment-naive. But let's start with that.
    Ms. Dee?

    MS. DEE: All right. I'll go. Again, they were small numbers, and I remember Dr. Ellenberg's question about whether that was real or just an affect. But it also appeared to me that patients that stopped boceprevir after a certain period of time, that's when they had virologic failure. So it would seem to me that keeping on boceprevir longer as opposed to a longer course of peg/riba would make more sense, especially given the anemia considerations. And the duration of anemia might be an important factor. I think that's an
    important factor to consider.
    It seems like the agency and the sponsor have agreed on this, so I guess we still answer it, though, right?

    DR. MURRAY: Well, we put two options down for you to consider for duration, a full duration -- I guess when we get to talk about certain groups, A, 44 or 32, or you could suggest your own, I guess. But I don't think we came to any firm agreement because we wanted to hear what you had to say.
    MS. DEE: I think the 32 might be sufficient. The triple therapy might be better than longer, than 20 weeks of PR.

    DR. CARGILL: Dr. Roland?
    DR. ROLAND: I find this a very vexing issue, and I actually wish I had one table that showed all of the response rates across the three arms and then in the important subpopulations.
    The reason it's such a hard issue to address, it's not hard to address when the response-guided therapy rate was higher than the
    other rate, but when it's lower, the study was not designed to compare the two active arms. And so we have absolutely no idea what that means.
    So then I kind of wonder, well, then does it become another one of these individualized decisions where, for some patients, it's going to be more important to have the absolute best chance of an SVR, and for others, they are more than willing to take a little bit of risk, that maybe this isn't quite as effective and they're going to have a shorter duration of therapy. And without a head-to-head study of the two active approaches, I don't know how else to think about it.

    DR. CARGILL: Dr. Ghany?
    DR. GHANY: Actually, I agree completely with Dr. Roland that the studies were not designed to answer these particular questions. And what we're trying to do is do subgroup analysis on smaller cohorts of patients, and I think we're stepping on shaky ground here if we're going to use this as a labeling indication.
    I think one of the last questions put
    forward to the panel is what other additional studies are needed, and these are clearly studies that need to be designed to answer these important questions.
    Thank you.
    DR. CARGILL: Thank you.

    Dr. Knodell?
    DR. KNODELL: I don't know where that 32-week treatment came from, but it seems to me that you've got data that if you have a late responder, the only thing we have data on is a full 44 weeks of boceprevir, of triple therapy, and I don't know how you can possibly ignore that and shy away from it in terms of your labeling.
    You've either got people who respond very quickly, and they can be treated for the 28 weeks or you've got people that are slow responders, and the only thing you have any data on is a full 44 weeks of triple treatment.
    DR. CARGILL: Dr. Connick?
    DR. CONNICK: I agree with what everybody else has already said. I think without real data,
    a study that truly addressed this question, we're simply guessing.
    DR. CARGILL: Thank you.

    Dr. Friedman?
    DR. FRIEDMAN: So, again, I speak as a clinician, not as a scientist. The first thing I'd say is I'm glad we don't have to vote and we don't have to make this decision, because it's complicated. But my advice to the people who do have to make the decision is there clearly is a lack of complete data. And what clinicians are looking for is something that's simple and easy to remember and not complicated, and that's good for adherence for patients, too. And I think those factors have to be taken into account. Despite the appeal of Talmudic scholarship, I'd actually rather keep it in my head if I could.
    So I like the idea of response-guided therapy. Clearly, it's going to have a role. There are some people, the early responders, naive patients who are going to be able to come off therapy sooner. But when you get into all these
    subgroups, if it gets too complicated, we're going to be shooting ourselves in the foot. People are not going to be following it, following the protocols precisely, and patients are not going to be adhering precisely. So I'd consider that in the decision-making.
    DR. CARGILL: Thank you.

    Dr. Ellenberg?
    DR. ELLENBERG: I agree with most of what has been said. I'm not so worried about not giving very specific directions to people. I think there is a place for individual patient values in these things, and I the data that are here, I think it may help clinicians make decisions about somebody who's really, really miserable on the therapy but is in the category that looked like they did pretty well with a shorter course to help make the decision that maybe it's okay to go off earlier and to maybe push harder for people who are doing pretty well on the therapy and they were in the category that maybe they would do better.
    But I agree that we really don't have the
    kind of data that we would need to make a very strong recommendation one way or the other.

    DR. CARGILL: Dr. Strader?
    DR. STRADER: I will agree with some of those comments. Two things, in particular. I tend to be a purist. There is a reason that we require that studies be done and they be done in a certain fashion. There's a reason we're all sitting here around this table to discuss these.
    So I think that we should try to focus on the data that we have. It's nice to be able to make some inferences because we do want to take care of patients who have not been included in the studies, but it's important to realize there's a reason that we use the scientific method.
    Again, to echo Dr. Friedman's comments, putting on my gastroenterology hat for a second, there have been some studies that show that people are more likely to comply with colonoscopy screening if the recommendations are simple. You give people five or six different methods that they can choose to be screened and it's much more
    difficult for people to make a decision as to what to do as opposed to giving people very clear recommendations that say this is the preferred thing that we do. So I think that we should really try to keep that in mind when we're making these labels.
    DR. CARGILL: Well, with that, we're going to go on to the A part of the question, which is looking specifically at treatment-naive patients with detectable HCV RNA at week 8 and undetectable at week 24, or otherwise known as late responders; again, the strength of the evidence to support response-guided therapy and should certain groups receive longer durations of boceprevir.

    Dr. McGovern?
    DR. MCGOVERN: I think the bridging analysis that the FDA did was interesting and provocative. I would like to see a study to really sort that out. So I think this is an area that would require more patients.
    DR. CARGILL: Thank you.
    DR. MURRAY: May I ask -- I mean, if the
    answer is we need more studies and it's interesting, does that mean the default is basically arm 3 and not response-guided, not shorter therapy?

    DR. CARGILL: Dr. Roland?
    DR. ROLAND: So I'm trying to draw my table that I was looking for. I think what I'm seeing in the naive study is that in the two arms in the early responders, they are the same, 97/96. So how could you argue against response-guided therapy in that group? In the late responders, there is a potential difference in the direction we don't like, so 66 for the response-guided therapy and 75 for the long therapy.
    I would be much more uncomfortable -- actually recommending response-guided therapy in late responders would not feel comfortable to me, given that.

    DR. CARGILL: Dr. Ghany?
    DR. GHANY: I think to answer your question, Dr. Murray, the answer is yes. And I agree with what's just been said. In the absence of other
    data and from the small sub-analyses suggesting that even with a small dataset, that longer data gives you better SVR rates, that that's the only recommendation that you can come to at this point.

    DR. CARGILL: Dr. Clay?
    DR. CLAY: In the label, which is what I think we're sort of talking about here, it could be structured such that it is approved for use for this duration of time and you actually give the range for that, and that consideration can be given to discontinue this if you see this at this time point. I mean, that's more of a guideline statement, though, than what we actually would find in FDA labeling, because there's so many contingencies there.
    I think that's what you're looking for here, but I don't exactly know how to help you write the wording for the package insert.
    DR. CARGILL: Dr. Knodell?
    DR. KNODELL: I guess my previous remarks I thought were already dealing with section A here. But I would point out that over the years, we seem
    to have made a mistake on the short side duration of treatment. We initially recommended interferon for six months; then we went to 12 months. We initially recommended interferon and ribavirin for six months; then we went to 12 months. Now we're treating some people who are slow responders who don't have a negative RNA at 12 weeks but do show a 2 log increase; and, then, our negative at 24 weeks, we're carrying them out to 72 weeks.
    I'd hate to take one step forward and two back by stopping treating people early, assuming -- I mean, again, you have clinical considerations like how well they're tolerating the treatment. But I think that if we've got a situation where it looks like there is a difference between a little longer treatment duration and a higher sustained viral response, we should take that.

    DR. CARGILL: Ms. Valbh?
    MS. VALBH: I'm going to agree with that. These patients are typically harder to treat anyway, and I don't think that a shorter duration
    of treatment in this population is going to show that much more benefit. I think that they do deserve a longer duration of treatment. And then going back to what Dr. Friedman said, we have to keep in mind that within the labeling, we don't want too many variations of all these subpopulations. We need to be very clear and not have dosing recommendations or treatment recommendations that are not going to be easy to remember or practiced and out there.

    DR. CARGILL: Dr. Van Dyke?
    DR. VAN DYKE: I guess neither of these comments are very rigorous, but, first of all, to me, it makes sense that if someone is a slow responder, they need to be treated longer. I apologize for that, because that's an emotional statement, but it makes sense to me.
    Looking at the data and going beyond what we can really conclude from it, that table was already quoted in the treatment-naive that for the late responders, you went from 66 to 75 percent response by going to the longer duration of therapy, 48
    weeks.
    That difference is not seen in the same table looking at the previously treated group. Granted, it's 32 weeks, not 28 weeks in the response-guided therapy. So it's a little bit longer, and in that group, it's the same, 79 and 72 percent, almost the same; in fact, better for the response-guided therapy.
    So I guess what I'm saying is if you want to believe that longer therapy is better, I think you can get a suggestion that for the treatment-naive, there is a hint, a trend maybe, a trend that longer therapy was better and that with longer therapy in the previously treated ones, already, you didn't see that difference.

    DR. CARGILL: Dr. Ellenberg?
    DR. ELLENBERG: I can understand that somebody treating patients and looking at these data might feel like they might want to treat those late responders longer, but looking at the data themselves, the difference is so slight, it's not even close to a statistically significant
    difference. And, again, given that you've got -- in the other study, there really is no difference and there is no difference in the early responders.
    It's very hard for -- I don't see how we could make any kind of a strong statement that the shorter therapy is clearly to be avoided in these late responders. It's something that might be worried about. There are other medical areas where we gave more treatment, and it turned out that less treatment was just as good, and you avoided problems.
    So, again, I'm not comfortable recommending a strong recommendation that people have the longer course. I think people can look at the data. But I don't know from these data that the shorter isn't just as good, and it certainly has some advantages.

    DR. CARGILL: Dr. Ghany?
    DR. GHANY: In the spirit of keeping things simple, I tend to agree with the comments that we should just have sort of one regimen for these individuals. I wonder if the FDA even has to
    address this particular subgroup and whether this could be left for the practice guidelines to discuss some of these nuances of treatment.
    Would you like to comment on that?

    DR. MURRAY: Well, I guess we don't necessarily need to get that nuanced in the label, but if there are subgroups where you definitely would not go with shorter treatment, I mean, we'd certainly like to hear that. I mean, if you do feel strongly about a certain group, and I guess we'd be talking about cirrhosis in B, we'd certainly like that.
    I guess we certainly can wordsmith the label so that there is freedom for individualized therapy and treatment duration depending on the patient situation.

    DR. GHANY: The reality is that the physician taking care of the patient is going to make these decisions on whether to shorten the course of treatment or extend it for the recommended duration based on tolerance. So the labelings are really guidelines, and it's really a
    decision between the physician and the patient.
    DR. CARGILL: Thank you.

    Ms. Dee?
    MS. DEE: I don't want to be inconsistent about what I said about the null responders and no evidence being there, but I did see there was a difference -- excuse me. There seemed to be better SVR rates with the triple therapy.
    I'm a cancer survivor, and I was supposed to get nine treatments, and at eight, I was half dead and my doctor said, "Good enough, girl." So if you have patients -- maybe without cirrhosis, let's just put that qualification in there -- that you're seeing at a certain point in time you don't think are going to make it through, it may be a judgment call between you and that patient. But if it's not in the label that this might be efficacious, then the doc might not even try it.
    I mean, the idea that we should make this more simple is like a joke at this point. It's not simple in any way, shape or form. So if we could make it better for patients -- I mean, all of those
    extra weeks of interferon is a horror show, and I think that's getting as real as I can get. So if we have any evidence that that, in fact, might not be necessary, I think we ought to share that.

    DR. CARGILL: Dr. Korman?
    DR. KORMAN: Not being a scientist either, I was just looking at the numbers, and not being a statistician, and these numbers are two patients either way and they're equivalent. I also treat these patients. Sixteen weeks of longer therapy is really difficult, and we know that these regimens are a little bit more toxic than the other regimens.
    So I don't know how you wordsmith this, but if you give somebody a pass at 32 weeks because they're doing okay, then that seems to me to be a perfectly acceptable way to phrase the guideline so that clinicians are not obligated to treat for a full course of therapy.
    DR. CARGILL: Thank you.
    Well, as if that isn't complicated enough, we'll revisit our favorite group of null
    responders. And I feel like I'm a drowning man going down for the third time, but I'll stick in here.
    Does the panel want to give any additional advice to the FDA regarding null responders?
    I'm sorry. Let's look at blacks and those with advanced fibrosis or cirrhosis.

    Dr. Clay?
    DR. CLAY: I'll say it, but I guess you probably already talked to Merck about this. How are you going to better describe predicted response in that population? What is unique? What is different about that population that resulted in this response rate? And I don't have any crystal ball for you on that one.

    DR. CARGILL: Ms. Dee?
    MS. DEE: I think the cirrhosis area is much more clear than the case for African-Americans, blacks, whatever.
    I really just think that you need to say that the numbers were small or this is what happened with certain people or a certain amount of
    people and let docs decide what they want to do with their patients. I just don't feel comfortable that there were -- I mean, it seemed like there were a lot more cirrhosis cases in that one arm, if I recall, and that that obviously needs longer treatment. I think that's pretty clear. But I'm not as clear about what to do with, for instance, late responders who were black.
    DR. CARGILL: Thank you.

    Dr. Ghany?
    DR. GHANY: I would agree with what Ms. Dee said. The case is a little clearer for patients with advanced fibrosis or cirrhosis. The data does suggest that these patients would benefit from a longer duration of treatment -- well, a standard course of 48 weeks of therapy. But, again, the numbers are small. There are only 9 percent of patients with this in the naive trial.
    I'm sort of on the fence with what to do with African-Americans. If you believe that some of the difference can be explained by cirrhosis, I think that only explained 50 percent of the
    difference. It doesn't explain all the difference. So I'm not certain how to respond to this, whether they do as well with response-guided therapy as with the 48-week course of treatment.


    DR. CARGILL: Dr. Strader?
    DR. STRADER: I agree that the cirrhotic patients, it seems like an easier decision to make. I'm looking at the sponsor's slide 32, and it appears -- and it pains me to say so -- that when I look at the p-values -- when we looked at response-guided therapy for blacks, the p-value was .044, so we're just getting there, as opposed to when we talk about 48 weeks, it's .004.
    So it appears, even with these small numbers, that African-American patients probably do better with the full 48 weeks of treatment as opposed to the response-guided therapy. I would be a little bit more encouraged if the response-guided therapy p-value was a little bit lower. But, again, it's hard to say, but these are the data that we have.

    DR. CARGILL: Ms. Young?
    MS. YOUNG: Obviously, it's cost-benefit of covering your bases versus exposing that population to increased risk. So with the data we have, do you feel it's strong enough to go one way or another? I mean, are you leaning toward the --
    DR. STRADER: As a clinician, I would probably tell my black patients I think it's better to do the full 48 weeks, because I find that if you ask the patient a question, this came up on the other side of the table, "Do you want to take another 16 or 24 weeks of treatment and be reasonably assured you're going to get a good response or do you want to cut it short and it may work, it may not, but if it doesn't, we're going to have to do this all over again," people will say, "I'll do the full 48. I don't want to take the chance that it may not work and I have to do it again."
    So speaking personally, I would probably tell my patients, based on the data that I have here, my African-American patients, let's do the full 48.
    DR. CARGILL: Thank you.
    Dr. Roland?
    DR. ROLAND: This is almost a question for Dr. Ellenberg, not being a statistician. But I'm wondering, with this sort of unusual trial design, where we've got two comparisons, whether a multivariate analysis that includes black/non-black as a predictor, that includes fibrosis as a predictor, and includes treatment as a predictor could help us to understand what the variables are that are impacting the response.
    DR. ELLENBERG: I think the answer to that is maybe. The numbers here are quite small, so it might be more helpful. Again, I'm looking at that 42 percent versus 53 percent, remembering that that 42 percent was somewhat lower because of people who dropped out before they even got the drug.
    I think what Dr. Strader said is a perfectly reasonable approach to one's patients, but I think the data are just not strong enough for us to say this is the way everybody should treat their patients.

    I remember that the FDA asked us -- did express some concern about that 9 percentage point difference, but it's 22 out of 47 versus 29 out of 55. It's just a very unstable estimate.

    DR. CARGILL: Dr. Knodell?
    DR. KNODELL: Along these lines, I'm just curious. When Merck set this study up -- I address this to the sponsor.
    When you were treating treatment-naive people, you did a four-week lead-in and then 24 weeks of triple therapy. And then when you do your study design for the non-responders, you do a four-week lead-in and 28 weeks of treatment. Now, you obviously must have thought that this is a particularly tough group of people to deal with and that maybe a longer treatment period is reasonable.
    If you could comment on what made you make that decision, it may help us with our decision as to how long we should be treating these people.
    DR. CARGILL: Yes. Thank you.
    If the sponsor address that question.
    DR. GOTTESDIENER: Yes. I'd like to address
    it and also point out that the analysis that was requested for the multivariate analysis we have done. And if you give us one minute, we can actually share that with you about blacks.
    So it's important to realize how this late responder/early responder works. The way it happens is treatment week 8, people are either detectable or non-detectable, and if they make it through to the point at which the actual decision is made, which is week 28, they either stop therapy at that point or if they're a late responder, they go on additionally. And what we're really talking about is what is the duration of boceprevir treatment in those individual patients.
    So in practice, what we did in our trials is we said if you look like you're responding early, you should stop. If you look like you're responding late, you should get more. It really follows a very simple idea as one goes forward.
    What we did find is -- and as the FDA pointed out -- is that if you try to understand, in fact -- if you looked at some of those late
    responders and you asked how much boceprevir they need, could we stop that at 24 weeks or go on farther, in practice, there was a small difference there.

    What the sponsor really thinks is it's very clear that there's a difference between 24 and 48 weeks. The hard decision is where between those two time points to stop the boceprevir therapy. So one way to look at it would be to say we studied 24 and 48, 48 was better, and that settles the issue.

    I think what we and the FDA have done is we've said that isn't the only data we actually have. We also have data from the treatment failure patients, and when we look at those same type of patients in treatment failure patients, they look exactly the same between those two groups. So from our looking at the data, we would say, clearly, a bit more than 24 weeks is important, but the treatment failure patients tell us 32 weeks is enough.
    Now, having said that, the important
    question is can we actually tease out in the blacks what we should actually do there. And Dr. Koch, who is a professor of statistics, may be able to explain the multivariate analysis and help the committee understand why we think blacks should get, in fact, the ability to have RGT.
    DR. KOCH: I'll be very quick. Gary Koch, biostatistics department, University of North Carolina, a statistical consultant to Merck.
    As Dr. Albrecht pointed out, there were more patients in the 48-week regimen who actually showed a good response at week 4, and she then showed what the mITT analysis did when you accounted for that. But there also were more patients in the week 48 analysis -- I'm sorry -- in the 48-week group who had undetectable virus at week 8, 12, 16, 20, 24, and this is with both arms receiving the same treatment through the 28-week period.
    So there was a random tendency for the 48-week group to have more favorable status prior to 28 weeks among the blacks than in the response-guided treatment. When multivariate analyses were
    recently done to remove that random advantage, then these groups then became very close together, and actually there's an adjusted difference of about 2 percent in the all black ITT or mITT population.
    I could show a slide that revealed some of that, but basically if you do a multivariate analysis that adjusts out all the random imbalances that favor the 48-week group at all weeks prior to week 28, that difference shrinks.
    DR. KNODELL: You didn't address my question. The question was, your response-guided treatment for the people that hadn't responded before is 28 weeks, whereas your naive treatment period with triple drugs before you make a response-guided decision was 24.
    Why that four-week difference before you made a decision?
    DR. GOTTESDIENER: Actually, I'm not quite sure that that's factually correct the way you're describing it. In one study, there were four weeks of lead-in and 24 weeks of therapy, triple therapy, and that was week 28.
    DR. KNODELL: That's naive.
    DR. GOTTESDIENER: That's correct.
    DR. KNODELL: (Off microphone.)
    DR. GOTTESDIENER: I see.
    Jan, do you want to go ahead and answer that question? What you're really asking is why was the branch point for RGT 28 weeks in one study and 36 weeks in the other.
    DR. KNODELL: Yes.
    DR. ALBRECHT: The longer therapy for the response-guided therapy arm in the non-responder study was based on observational data that we made in an earlier study where we saw that patients that were very poorly responsive needed at least 24 to 28 weeks of therapy after they became negative.
    So using that data, we said in non-responder patients, that for the response-guided therapy arm, that we would make the duration of therapy longer. And so we used four weeks of lead-in plus 32 weeks of boceprevir and then added on the 12 weeks of two-drug therapy.
    So that data that supported that was from an
    early dose finding study that we did.
    DR. CARGILL: All right. I'm not sure that I can summarize all -–
    [Laughter.]
    DR. CARGILL: -- of that for my FDA colleagues. But I think that there are several points that have emerged from this. One is that we heard from the panel that it seemed that some of these questions could almost be jumping off points for additional studies that they would want to consider, and we will talk about that in our next question.
    Another is that it appears that from some of the data reviewed by the sponsor, that some of these differences at the -- for blacks, it appears, really are a function of the analysis, and if I understand you correctly, they were the same all the way along.
    The third point is I think we're still back to where we started, I hate to say, with our null responders in that we're still hearing a certain amount of discomfort around a firm recommendation
    with some expression of opinion that people should be free to be able to make the decisions that indeed we don't want to err on the side of being too short and others saying that we don't feel that we have the information.
    I think that's probably the best I can do from such a fairly complicated discussion.

    Our last question is, in addition to pediatric studies, are there other post-marketing studies you would recommend to further define risks or optimal use of boceprevir in clinical practice?
    I think given the many conversations that have gone on from the table, especially from my hepatology colleagues, I think this is a place where you can jump in with robustness and certainly list a number of things that you would like to see. So please have at it.
    Ms. Dee?
    MS. DEE: I'm going to leave the efficacy studies out, because I think that everybody else can comment on them. But I'm really interested in the drug-drug interaction studies. I'm amazed.

    I'm shocked that we don't have more information on antidepressants given the depression that occurs with interferon.
    Methadone, that's shocking to me that that hasn't been done already; the HIV protease inhibitors. It's just very disappointing that this stuff hasn't been done. And I heard today that they're going to be done, but I just need to say that I know that there's race in all of these drugs, but I think it's irresponsible that these studies haven't been done before now, and I hope they get done post haste.
    The other thing that worried me was the anti-diabetic studies. And I think there was one more thing. The contraceptives, the combined oral contraceptives. I mean, come on, we want women in these studies, and we want other populations. I mean, we really need to get this work done.
    I should have said this probably the last time around, but there are not a lot of blacks in these studies, and we have all these really complicated issues. Why aren't there more blacks
    in these studies? We need to do Phase 4 studies with just blacks like we did in HIV for GRACE and Latinos. I just don't see what -- I don't know what it's going to take for the sponsors to learn that they need to do this now or they're going to have to do it later, because I just really think that it's important, especially in HCV, to have these efficacy and safety questions answered in black people.
    I was very impressed with the slides that the doc from the VA Administration showed. I'm sorry, I don't remember her name. But they are the real people that have hepatitis C. So we really need to get all of this information so that the real world can benefit from these 21st century advances.
    DR. CARGILL: Thank you.
    For the record, that's Dr. Belperio. And I'm going to actually jump the line and list my own requests of studies.
    One of the things that I am quite concerned about is that our patients often are very eager to
    have these options, but I think it's like anything else; when you begin to take them, you realize the devil is in the details. And so having a feel for the adherence, the impact on adherence, which we always know changes after we get into broader populations, is really going to be quite important.
    In addition to the impact on adherence, what are the types of support that can help with adherence, because clearly this is going to be critical? We offer people an opportunity to have another shot at SVR, but, again, there are other mitigating factors. And, particularly, a regimen is going to require lots of pills, has other side effects. I think that's going to be something we'd want to know about.

    So I apologize to my colleagues for jumping the line, but I think after sweating through four questions, I could do it.
    Dr. Clay?

    DR. CLAY: I just want to echo some of the drug interactions that really I think are important. Obviously, conducting a study in women
    of childbearing potential on a variety of oral contraceptives is key because this disease is pervasive throughout the entire population.
    The lack of data on antidepressants is concerning given the fact that that is something that is used quite frequently in this population at the outset of therapy for hepatitis C, as well as the over-the-counter and complementary medicines that were mentioned earlier today by the members of the public, specifically milk thistle. And you do have some data that's been provided about the likelihood of interactions with methadone and buprenorphine, as well as other medicines.
    These are studies that are very easily done, very rapidly done in normal healthies, and you can provide an answer probably before this drug would need to be in the pharmacy shelves. So I think this is certainly a list of trials that could be conducted.
    One thing that is not drug interaction-related, and you may already have an answer to this -- and, if so, that's great, and I apologize
    for missing it -- would be if your data initially showed that you didn't have anemia reported in your normal healthy volunteer studies, and I'm not aware that normal healthy volunteer studies generally run for four weeks at a time. And if, indeed, you did four weeks' worth of normal healthy studies and you didn't see anemia, that's great, but if not, then I'd like to see that.
    Thank you.

    DR. CARGILL: Dr. Roland?
    DR. ROLAND: So for the record, I'd like to echo my support of all of what's been said so far. And following up on your request around adherence and supports for adherence, I think a very well thought out series of demonstration effectiveness studies are going to be really important for clinical guidance; obviously, very different from efficacy studies.
    I'm also really interested in the use of this drug in the post-transplant context. And so there's a series of steps that need to be considered for that; obviously, the interactions
    with the immunosuppressants and then all of the tolerability issues in the post-transplant setting when people are on all kinds of myelosuppressive agents.

    It seems like it would be pretty simple to do a virology study that looks at the ability of the triple therapy to re-suppress the virus in people who had resistance that then reverted to wild type. That wouldn't have to be an efficacy study, but that would give us some reassurance about whether there's antiviral activity.
    This is a little bit more of a question. I'm not sure if it's important to study this drug in the hepatitis B and C co-infected patients. Maybe some of the hepatologists have some thoughts about that.
    DR. CARGILL: Do any of our hepatologists want to address Dr. Roland's question?
    [No response.]
    DR. CARGILL: While we're thinking, Dr. Ghany?
    DR. GHANY: Well, again, I also agree with
    everything that's been said. And I'd also like to add a few other suggestions for post-marketing studies.
    Of course, we need to study this drug in all the special populations, so the post-transplant is an obvious one, larger populations of African-Americans, larger populations of cirrhotics, those with psychiatric issues and dependence issues.
    The final suggestion I had is in an effort to try to simplify this regimen, I'd like to see studies where there is no lead-in phase, because it's not clear to me that a lead-in phase is actually essential for this regimen. And that's a study that could be done very -- that's an easy study to do.
    DR. CARGILL: Thank you.

    Dr. Strader?
    DR. STRADER: I agree with that last one, Dr. Ghany. I think that that point is well taken, that there is not -- it may be that we don't need a lead-in phase.
    I think maybe resistance studies might be
    interesting just to know. The sponsor said that there doesn't appear to be any effect of baseline variants on SVR, but it may be interesting to know are they talking about the duration that variants remain positive and is there some cross-resistance that may affect your ability to respond to others.

    We also talked about EPO versus ribavirin dose reduction in patients. Is there a minimum dose of ribavirin below which we think we should not, with the triple drug therapy, be going and where EPO might be used.
    DR. CARGILL: Thank you.

    Dr. Friedman?
    DR. FRIEDMAN: I just felt shamed into commenting on the hepatitis B and C issue. It's a little complicated. Peginterferon, of course, can be used to treat hepatitis B. So you can do studies in patients who have combined infection, co-infection with B and C using this regimen to see what happens. But then I think if there is a response to C, but not to B, then you want to know about the interaction between boceprevir and the
    nucleoside and nucleotide analogue alternatives for the treatment of B. So you'd want to know something about those drug interactions.

    DR. CARGILL: Dr. Knodell?
    DR. KNODELL: Just a little expansion on Dr. Friedman's. Generally, when you have dual infection, one or the other of the viruses is the primary virus, and you can kind of look at either your HBV DNA or your hepatitis C RNA, and you'll generally see one will be high and one will be low. And I usually treat the one that's predominant. Generally, it seems like you don't have a dual infection with both viruses blasting the liver.
    My one comment on the additional studies. There hasn't been many -- there isn't much data on people over 65 who, one, will have some different levels of drug metabolism, and we're getting more aggressive, I think, in treating older people now with hepatitis C, particularly because of some data that was reported at the last liver meetings about the high rate of hepatoma that develops in people with hepatitis C over 65.
    So, again, it shouldn't be real hard to accumulate some data along those lines and see whether or not older people have a similar clinical response as your earlier treatment studies.
    DR. CARGILL: Thank you.

    Dr. McGovern?
    DR. MCGOVERN: If I remember the Kohl paper correctly, SPRINT-1 didn't show a major difference in efficacy whether you had a lead-in phase or not, but there definitely was a difference in terms of the incidence of resistance.
    So, again, it's getting back to is resistance important, and it keeps getting back to will wild type return for good. If you re-expose the patient to that drug, will those RAVs return? And we don't know that.
    I guess in terms of how that bodes to additional studies, I guess it would be nice if we do have data that these are not detected by population sequencing, maybe the sponsor could consider also looking at deep ultrasequencing, pyrosequencing.
    DR. CARGILL: Thank you.

    Dr. Clay?
    DR. CLAY: When you conducted your response-guided therapy, did you re-measure drug concentration at the point where they were failing?
    I didn't see a lot of heads nod that time.
    The last couple of questions that I was asking, it was like, "Oh, yeah, we did that."
    Why didn't you do that? That, to me, is a key thing to this. I mean, adherence is key. We've talked about adherence, and we've talked about the daily diaries. I run a Phase 1 through 4 research center. I'm aware of how to manipulate those patient diaries, because we generally incentivize our patients to complete the diary.
    But if you did not reassess the plasma concentrations, and, ideally, to make sure that the drug is getting where it needs to be, at the time of failure, I guess I would like to see that being done in some fashion.
    DR. CARGILL: Thank you.

    Ms. Valbh?
    MS. VALBH: I am very disappointed in the drug interaction studies and the lack of the studies, especially with oral contraceptives and with the antidepressants. And I would like to comment on what Dr. Clay said earlier, that maybe those studies are quicker to do and we could do those before the product hits the pharmacy shelves.
    In addition, and this is maybe later on down the road, even though there is good treatment right now for other genotypes -- I know we're looking at genotype 1, but there is a subset of patients who do not respond, like genotype 2 and 3 that do not respond to standard of care. I would like to see studies done on the non-responders for the other genotypes to see if this product will increase their chance of an SVR.
    DR. CARGILL: Thank you.

    Dr. Connick?
    DR. CONNICK: I wanted to echo some of the statements previously made. I think what Dr. Roland suggested, re-treating people with boceprevir who failed but whose resistance
    mutations have disappeared is absolutely key to address the significance of the resistance. And that study could probably be done pretty soon.
    In addition, I think adherence is extremely important and undervalued, and if we could figure out ways to perhaps improve adherence early on, maybe push more people into being early responders, if that is possible, that might be incredibly cost-effective rather than treating them longer.
    I also think we have another drug that may get approved and a nice head-to-head study might be warranted with that drug, if it gets approved.
    Lastly, I think we should start thinking about treatment trials for people who have failed boceprevir and are there ways that we can enhance those people's responses, giving them more interferon or ribavirin or what. Thanks.
    DR. CARGILL: Thank you.

    Dr. Giordano?
    DR. GIORDANO: I guess I have to go on the record and say I think there needs to be a study in null responders because I don't think that's
    actually been stated in response to this question yet.
    Similarly, I think it would be useful to have a study of different treatment durations in late responders. I think that question is not answered, and it would be useful to have that answered, in addition to supporting actually everything else that everyone else said.
    DR. CARGILL: Thank you.

    Dr. Roland?
    DR. ROLAND: I'm wondering if people with bleeding disorders were included in this study and, also, if they're being included in the HIV co-infection study.
    [No audible response.]
    DR. ROLAND: Neither? So I certainly heard from public comments that that's an important population to understand what the -- especially the risks.
    DR. CARGILL: Thank you.
    Dr. Van Dyke?
    DR. VAN DYKE: I think it would be
    worthwhile doing a study looking at the IL28B genotype up front, because it may well be that people with the CC genotype need shorter duration of therapy. So I think if you could define that, I think that would allow us to sort of customize therapy for individual genotypes.
    DR. CARGILL: And, Ms. Young?
    MS. YOUNG: I agree with everything that -- all the suggestions that have been made, and particularly the VA presentation had some very good suggestions about the drug-drug interactions and adherence and risk stratification in terms of the regimen and dosage and resistance.
    I'm not sure how you prioritize, because I know there's a cost to all of this, but maybe looking at the profile of the target population and figuring out what's most relevant to that, as well as the side effect issues that we're concerned about.
    DR. CARGILL: Well, compared to question 4, we have over 37 studies that have been recommended for our sponsors. So, obviously, you all are going
    to be gainfully employed for a long time. Just to revisit some of them, the drug-drug interaction studies alone, we'd all be as old as Methuselah getting through all them. But they would include everything from certainly our HIV medications, antidepressants, which I certainly echo. We've heard about other agents, such as milk thistle, certainly buprenorphine and methadone.

    Populations, specific, special populations, and in that group I would include both racial and ethnic minority populations, such as African-Americans and Latinos, as well as individuals who have diseases that place them in special populations, including those with bleeding diatheses or those who are null responders, cirrhotics, those who are co-infected with hepatitis B and hepatitis C.
    In addition to that, a population of individuals who are post-transplant and may become infected both in terms of the tolerability of the drugs, interactions with their immunosuppressives. Certainly taking a look at do we need a lead-in
    period, and, if so, what does that look like and why.
    Sequencing, a very important point made about people over 65, who especially, what I would concur, the more recent data on hepatoma in this population. Data on non-responders from other genotypes and what does that mean. Re-treatment with boceprevir after treatment failure and the viral resistance mutations have disappeared.
    So this is just the beginning of a very long list of some very important studies addressed in the questions that I think, again, point out that we can have robust efficacy trials, but as we've all learned in basic statistics 101, there is no perfect study. And so we have to always keep taking repeated bites at the apple.

    So I think at this point, I'm going to ask the division to give us their last words, being our resident, Talmudic scholars, I guess.
    DR. MURRAY: I don't know if we have any final words, unless Dr. Birnkrant does. But we do want to thank the committee for bearing through all
    of the complex presentations and doing everything they had to do get here. And we really appreciate your taking time out of your busy schedules to help us wrestle with some of these questions. We appreciate it a whole lot.

    Adjournment
    DR. CARGILL: Well, I will thank you from the chair, thank you to the sponsor, thank you to the FDA, certainly thank the participants, and also for the public who has come in to hear these. And we are now adjourned.
    (Whereupon, at 4:40 p.m., the meeting was adjourned.)