Care model demonstrates high SVR rates with different levels of fibrosis, genotypes, and HCV treatment history

Retrospective Study

World J Gastroenterol. Oct 14, 2016; 22(38): 8558-8567
Published online Oct 14, 2016. doi: 10.3748/WJG.v22.i38.8558

Levin JM, Dabirshahsahebi S, Bauer M, Huckins E. 

Retrospective analysis of hepatitis C infected patients treated through an integrated care model

Core tip: There are new effective options for treating hepatitis C virus. To maximize their effectiveness our health system developed an innovative integrated care model to manage these patients. Through our original therapy algorithm we were able to closely monitor patients from time of insurance approval to the time of obtaining a sustained virologic response (SVR). This real world retrospective study analyses our patient’s SVR rate, adherence, and interventions made by the patient care team. Additionally it will provide a model for other systems to improve their care coordination and response with direct acting antiviral treatment.

ABSTRACT

AIM
To determine if our health system’s integrated model reflects sustained virologic response (SVR) outcomes similar to those in clinical trial data, maximizes adherence, and averts drug interactions.

METHODS
Subjects with chronic hepatitis C had their medical records reviewed from November 1st, 2014 through March 1st, 2016. Patients eligible for treatment were entered into an integrated care model therapy algorithm. The primary outcome was SVR12 based on intention to treat (ITT) analysis. Inclusion criteria consisted of both treatment naïve and experienced patients over the age of 18 who were at least twelve weeks post-therapy completion with any genotype (GT) or METAVIR score. Secondary outcomes included adherence, adverse events, and number of drug interaction interventions.

RESULTS
At the time of analysis, 133 patients had reached twelve weeks post therapy with ITT. In the ITT analysis 70 patients were GT 1a, 26 GT 1b, 23 could not be differentiated between GT 1a or 1b, 8 GT 2, 4 GT 3, and 2 patients with multiple genotypes. The ITT treatment regimens consisted of 97 sofosbuvir (SOF)/ledipasvir (LDV), 8 SOF/LDV and ribavirin (RBV), 7 SOF and Simeprevir (SMV), 6 3D and RBV, 1 3D, 11 SOF and RBV, and 1 SOF, peg interferon alpha, and RBV. The overall SVR12 rate was 93% in the ITT analysis with a total of 6 patients relapsing. In patients with cirrhosis, 89% obtained SVR12. All 33 patients who were previous treatment failures achieved SVR12. Drug-drug interactions were identified in 56.4% of our patient population, 69 of which required interventions made by the pharmacist. The most common side effects were fatigue (41.4%), headache (28.6%), nausea (18.1%), and diarrhea (8.3%). No serious adverse effects were reported.

CONCLUSION
Dean Health System’s integrated care model successfully managed patients being treated for hepatitis C virus (HCV). The integrated care model demonstrates high SVR rates amongst patients with different levels of fibrosis, genotypes, and HCV treatment history.

Discussion Only

Download

Full Article (PDF) 

Read Online

Full Article (HTML)

HCV treatment guidelines emphasize the importance of addressing adherence, adverse effects, and drug interactions with HCV regimens as clinically indicated. However, no specific recommendations are made regarding follow-up methods. Thus, effective real-world care models need to be identified for the newer DAA therapies to ensure the best HCV treatment outcomes are achieved in real-world practice settings. Our study describes an integrated multidisciplinary care team model with SVR12 rates comparable to those seen in controlled clinical trial settings. Overall SVR12 among patients in the current study was 93% in the ITT cohort and 95% in the PP cohort. Among patients with cirrhosis our SVR12 rates remained high at 89% for both PP and ITT cohorts, despite this patient population generally being more difficult to treat. Another patient population that achieved notably high SVR12 rates in our study was the treatment-experienced cohort with a 100% SVR12 rate for PP and ITT analyses. This cohort achieved a higher SVR12 rate compared to our treatment-naïve patients of which 91% in the ITT cohort and 94% in the PP cohort achieved SVR12. This was an unexpected finding we cannot explain. This was additionally unexpected because more patients in the treatment-experienced cohort were cirrhotic compared to the treatment-naïve cohort (57.6% and 28.9% cirrhotic, respectively).

SVR12 achievement rates were similar to clinical trial results based on the specific treatment regimen as well. Patients who completed LDV/SOF regimens achieved 95% SVR12 PP (92% ITT) in our study. The ION-1 study included GT 1 treatment-naïve patients with or without cirrhosis treated with a fixed-dose combination of LDV/SOF with or without RBV[16]. SVR12 rates were 99% with LDV/SOF. ION-2 included GT 1 treatment-experienced patients with or without cirrhosis treated with a fixed-dose combination of LDV/SOF with or without RBV[18]. SVR12 rates were 96% with LDV/SOF. The addition of RBV did not significantly impact SVR12 rates in our study or in ION-1 or -2; SVR12 remained high.

The SMV/SOF regimen resulted in 100% SVR12 PP (100% ITT) in our study patients. OPTIMIST-1 and OPTIMIST-2 investigated SMV/SOF among GT 1 treatment-naïve and treatment-experienced patients[19,22]. Patients without cirrhosis were included in OPTIMIST-1 and the resulting SVR12 was 97%. OPTIMIST-2 included patients with cirrhosis and the SVR12 was 84%. Although only seven patients total received the SMV/SOF regimen among our study patients, five out of the seven were cirrhotic. Our SVR12 rates of 100% were unexpectedly higher than those seen in the OPTIMIST trials.

Patients who completed the 3D plus RBV regimen achieved 100% SVR12 PP (100% ITT) in our study. Two of the six patients were cirrhotic. The SAPPHIRE I and SAPPHIRE II clinical trials included patients that were treatment-naïve and treatment-experienced, respectively, without cirrhosis treated with 3D plus RBV for 12 wk. SVR12 was 96% for both studies[17,23]. In the TURQUOISE II trial, 92% of treatment-naïve or experienced patients with cirrhosis who received 3D plus RBV for 12 wk achieved SVR12[20].

Compared to other real-world analyses of newer DAA treatments, our response rates are either higher than or similar to other studies, demonstrating the effectiveness of our model. A real-world analysis of treatment-naïve or experienced patients with HCV GT 1 with or without cirrhosis was conducted on patients in the HCV-TARGET cohort treated with SMV/SOF with or without RBV[24]. The overall SVR12 rate for SMV/SOF without RBV was 85%, which was lower than the SVR12 of 100% (PP and ITT) seen in our study for patients who were treated with the SMV/SOF regimen.

A real-world study from Israel included treatment-naïve or experienced HCV GT 1 patients with stage 3 or 4 fibrosis treated with 3D with or without RBV. Amongst the patients who completed therapy and retested 12 wk after completion, SVR12 rates were 97.8%[25]. Seven patients in our study received treatment with 3D plus RBV and only two were cirrhotic. Our SVR12 rates with this regimen were 100% for both PP and ITT analyses.

Another real-world effectiveness study from a large integrated health care system in the United States enrolled patients with GT 1 infection and receiving LDV/SOF with or without RBV. Patients were treatment-naïve or experienced and both cirrhotic and noncirrhotic. SVR12 for LDV/SOF was 93% in the ITT analysis[26]. The overall SVR12 in our study for patients treated with LDV/SOF was similar at 92% in the ITT analysis. The addition of RBV did not significantly impact SVR12 rates in either study.

Six patients in our study relapsed. One patient with GT 1b and underlying cirrhosis may have relapsed due to a 5-d break in therapy, another with GT 1a and cirrhosis was due to reinfection from reusing diabetes supplies, and one patient with GT 1a and cirrhosis relapsed for unknown reasons. The other three cases warrant further discussion. The GT 1a infected patient with advanced cirrhosis and HCC treated with LDV/SOF for 12 wk in the current study would also have been treated with RBV if evidence from the SOLAR-1 and SOLAR-2 Phase 2 trials were available at the time of treatment course selection, which may have prevented the relapse. SOLAR-1 and SOLAR-2 enrolled patients with HCV GT 1 or 4 with cirrhosis and moderate to severe hepatic impairment (Child-Pugh class B and C) with and without a history of previous liver transplant[27,28]. Patients were treated with 12 or 24 wk of a fixed-dose combination of LDV/SOF once daily plus RBV. SVR12 was 87% in non-transplant patients treated for 12 wk in SOLAR-1. In SOLAR-2, SVR12 was approximately 86% after 12 wk of treatment in non-transplant patients with GT 1.

The African American patient with GT 1a, cirrhosis, and HIV coinfection relapsed after 12 wk of treatment with LDV/SOF for an undermined reason. A recent study, ION-4, enrolled patients with HCV GT 1 or 4 coinfected with HIV-1. All patients received a 12-wk, fixed-dose combination of LDV/SOF for their HCV treatment regimen[29]. Thirty-four percent of patients in this study were black. Black patients had a lower SVR12 rate than other races (90% vs 99%, P < 0.001). Of note, 10 of the 335 patients in ION-4 relapsed and all were black. Seven of the relapsed patients had the TT allele in the gene encoding IL28B and 8 were receiving efavirenz as part of their HIV treatment regimen. Black race and presence of the TT allele were both significantly associated with relapse in ION-4. Among black patients in ION-4, 13% relapsed if they were also taking efavirenz and only 4% relapsed if they were taking other antiretroviral regimens. However, the difference was not found to be significant. It is possible that the patient in our case possesses the TT allele; however, we did not test patients in our study for the presence of this allele. Concomitantly taking efavirenz could have provoked the relapse in our patient, even though the role efavirenz plays in reduced effectiveness of HCV treatment remains unclear.

The non-cirrhotic, treatment-naive patient with GT2 who relapsed after being treated with 12 wk of SOF and RBV was somewhat surprising to us. The VALENCE trial confirmed that this same regimen is 96.7% effective in naïve, non-cirrhotic patients with GT2[30]. We cannot provide an explanation for why this particular patient relapsed.

In our study, 130 patients completed the analysis PP and 133 were in the ITT analysis. The high percentage of PP patients represents a high engagement between patient and clinical staff monitoring in our model. Furthermore, in our model, a high percentage (79.1%) of patients were 100% adherent on their treatment regimen and only one patient missed more than three doses. Other real-world studies looking at adherence demonstrated about 14% of patients were non-adherent to their treatment regimen and 18% had gaps in therapy of greater than 14 d[31]. A second study reported that 89.3% of patients completed treatment and 9% were non-adherent to therapy in a real-world setting[25].

The specialty pharmacist in our model identified drug interactions in 56.4% of patients. Sixty-nine drug interaction interventions were made with the most prevalent intervention being PPI dosing changes. Overall, drugs to lower gastric pH accounted for about 44% of all drug interaction interventions made. A study from Europe of drug-drug interactions identified that between 12%-19% of patients being treated for HCV were taking a drug that was contraindicated with one or more drugs in their HCV treatment regimen[32]. This same study showed that 29%-39% of patients were on two or more drugs that were either contraindicated or required additional monitoring or dose reduction with their HCV regimen. Similarly to our study, a high percentage (27%-38%) of interacting drugs in the Marra et al[32] study were drugs that target the gastrointestinal tract. The frequency and severity of drug-drug interactions with HCV therapies supports the workflow in our model where a specialty pharmacist consistently screened all patients for drug interactions.

Adverse reactions reported by our patients were consistent with those reported in DAA clinical trials and real-world experience with fatigue, headache, and nausea being the most common[24]. No serious adverse events were recorded. Drop-out rates due to adverse effects tend to be low with the newer generation DAAs, but no patients discontinued treatment for this reason in our study. One possible reason for this may be due to close follow-up by the pharmacist on adverse effects and management strategies.

Our study had some notable limitations. A major limitation is the lack of a control group to allow a statistical comparison of the effectiveness of our integrated model compared to a non-integrated model. Only qualitative comparisons to clinical trial data and other real-world data could be made. A second limitation is that this is a single-center study and results may not be generalizable to patient populations with different demographics. The population at our site is primarily Caucasian and insured. A third limitation is that the methodology of fibrosis determination was not standardized in our protocol. A fourth limitation was that adherence was self-reported by patients via tablet counts. There are inherent limitations with using patient-reported information in a study. The high adherence rates reported in our study likely reflected reality as shown by the high rates of SVR12 in our patients.

The results of this study have demonstrated the need to continue to manage patients using the integrated care model in our current practice. However, the limitations of this study have showed that future research is needed to find causation for patients that relapsed on DAAs. In the scope of our practice, follow up studies will be pursued to assess the impact of adherence and how new technology may assist in increasing adherence to therapy. Additionally, future studies at our practice will analyze if there is correlation of NS5A resistance associated variants and treatment efficacy in our patient population. Furthermore, additional focus will be put on the financial savings that the integrated care model has on the system and the patient.

In conclusion, there is a scarcity of published trials that describe real-world integrated care models for successful treatment of patients with the newer DAA HCV therapies. Dean Health System’s integrated care model helped successfully manage the patients being treated for HCV. The results of our study demonstrated favorable outcomes despite not being able to statistically compare across other studies. The integrated care model demonstrates high SVR rates amongst patients with different levels of fibrosis, genotypes, and HCV treatment history. The integrated care model assisted in catching and evading potential drug interactions that may have impacted treatment efficacy and tolerability. Overall, the evidence from this retrospective analysis demonstrates the benefits and value of treating HCV patients in an integrated care delivery model.

Continue reading...