.
.,
Norman D. Grace, MD
The prevalence of cirrhosis and decompensated cirrhosis in patients with chronic hepatitis C virus (HCV) infection has doubled in the last decade.1 Patients with HCV cirrhosis have been shown to have a poor response to standard antiviral therapy. In registration trials establishing the efficacy and safety of the antiviral combination of pegylated interferon and ribavirin for treatment of patients with chronic HCV infection, decompensated cirrhosis was an exclusion criterion, and patients with compensated cirrhosis were limited to 10%–20% of the study populations.2, 3, 4 An additional exclusion criterion was platelet count <90–100,000, further limiting the inclusion of patients with cirrhosis and clinically significant portal hypertension (CSPH). In determining a sustained viral response (SVR), patients with bridging fibrosis or cirrhosis based on liver biopsy criteria were grouped together. Manns et al2 reported an SVR of 30%, Fried et al3 reported 33%–43%, and McHutchinson et al4 reported 22% for the bridging fibrosis/cirrhosis populations. In addition, patients with cirrhosis were more likely to have side effects requiring a dose reduction or removal from the studies. In a study looking specifically at the HCV cirrhosis population, Heathcote et al,5 in an effort to avoid the debilitating anemia associated with ribavirin therapy, reported an SVR of 32% with peginterferon alfa-2a monotherapy. The black population is even more problematic, with a multicenter trial reporting an SVR of 28% for blacks compared with 52% for whites.6 However, only 5% of the black population in this study had cirrhosis at baseline. Faced with these grim statistics, patients with cirrhosis at the time of diagnosis might choose to forgo treatment because of the poor response rate, the increased side effects, and the cost of treatment, estimated at $30,000–$40,000 for a 48-week course.7
Are there potential benefits for pursuing treatment in the cirrhotic population? Patients with cirrhosis who sustain an SVR have diminished progression of their liver disease, a lower risk of developing hepatocellular carcinoma (HCC), and improved survival.8, 9, 10 On the basis of the results of the HALT-C study, patients with cirrhosis who are nonresponders to initial treatment do not benefit from prolonged maintenance treatment with pegylated interferon in terms of diminished progression of their liver disease.11 The study did demonstrate a reduced incidence of HCC in the small group of patients achieving an SVR when compared with a control population, but the benefit was marginal, and the authors do not recommend maintenance therapy.12 Practice guidelines for the treatment of HCV chronic liver disease have stated that patients with bridging fibrosis or compensated cirrhosis should receive treatment but caution that decompensated cirrhosis is a relative contraindication for treatment.7 However, Iacobellis et al13 reported an SVR of 16% in HCV genotype 1 patients with relatively stable decompensated cirrhosis and 55% for HCV genotypes 2 or 3. SVR was associated with a decreased risk of further decompensation, a reduced hospital readmission rate, and improved survival. Patients who do not develop a rapid virologic response at 4 weeks or who continue to be HCV RNA positive at 24 weeks have less than 5% chance of obtaining an SVR, and treatment can be discontinued.2, 3, 4, 7, 14 With the recent advent of the protease inhibitors boceprevir and telaprevir, the criteria for treatment of HCV chronic hepatitis might change. Boceprevir, when combined with peginterferon and ribavirin, significantly increases the SVR for patients with bridging fibrosis/cirrhosis in both naive patients and previous nonresponders.15, 16
Can we identify a subgroup of patients with cirrhosis who are more likely to respond to antiviral therapy? In addition to the usual prognostic indicators that predict a favorable response including HCV genotype 2 or 3, low baseline viral load (HCV RNA <600,000 IU/mL), female sex, age <40 years, non-black race, body weight <75 kg, absence of insulin resistance, alanine aminotransferase elevation >3 times normal, and the absence of advanced fibrosis or cirrhosis, there have been attempts to develop indicators specific to the cirrhotic population.7 In evaluating changes in hematologic parameters from baseline values, only a significant decrease in neutrophil count was associated with a virologic response.17 An elevated pretreatment level of the non-ELR CXC chemokine interferon-γ–inducible protein-10 (IP-10) predicts nonresponse to therapy with pegylated interferon and ribavirin.18, 19 Presence of the interleukin (IL)-28B gene polymorphism has been reported to predict a more favorable treatment response to antiviral therapy.20 This gene is more common in whites than in blacks and might partially explain differences in therapeutic response rates based on race. In evaluating these pretreatment markers in the VIRAHEP-C study, the authors report that the serum IP-10 level is a strong predictor of SVR and, when combined with IL-28B genotyping, significantly improves the SVR predictive value.21 However, only 7% of their patients had cirrhosis at entry, raising questions about the usefulness of these tests in the cirrhotic population. Kurosaki et al22 reported that the presence of the genetic IL-28B polymorphism correlated with both early virologic response and an SVR that was independent of other covariates including platelet counts, fibrosis score, and serum levels of HCV RNA. Twenty-four percent of their patient population had bridging fibrosis or cirrhosis. A decision model showed that patients who had the minor IL-28B allele and a low platelet count (<140,000 IU/mL) had the lowest SVR (7%).
Are there noninvasive tests that can predict the severity of portal hypertension in patients with cirrhosis? A recent meta-analysis that included 40 evaluable studies with 8739 patients reported that the aspartate aminotransferase-to-platelet ratio (APRI) identified HCV-related fibrosis with a moderate degree of accuracy.23 By using an APRI threshold of 0.7, the sensitivity for moderate fibrosis was 77% with a specificity of 72%. By using an APRI threshold of 1.0 for severe fibrosis or cirrhosis, the sensitivity and specificity were 61% and 64%, respectively, for severe fibrosis and 76% and 72%, respectively, for cirrhosis. The APRI was less accurate for HCV/human immunodeficiency virus coinfected patients. The authors concluded that although the APRI has less diagnostic accuracy than some other noninvasive tests in predicting hepatic fibrosis, it is inexpensive, widely available, and might be used as a screening test for identifying HCV-related fibrosis/cirrhosis. The FibroTest is a panel of biochemical markers of hepatic fibrosis. In a study correlating the FibroTest with the degree of portal hypertension as measured by the hepatic venous pressure gradient (HVPG), the authors found a good correlation with hepatic fibrosis but a poorer correlation with cirrhosis.24 Several studies have shown a correlation between HVPG and connective tissue markers, serum laminin levels, and serum hyaluronic acid concentrations, but their ability to detect cirrhosis and severe portal hypertension is limited.25
Transient elastography can determine hepatic fibrosis or cirrhosis by assessment of liver stiffness. Several studies have reported an excellent correlation between liver stiffness measurements and HVPG.25, 26 The technique has the advantage of being noninvasive and safe, but it is operator-dependent and is less accurate in obese patients or patients with advanced fibrosis.27 Magnetic resonance imaging has also been used to measure liver stiffness, but the data are limited, and there are no published studies correlating magnetic resonance imaging measurements with measurements of HVPG.
Measurement of HVPG remains the gold standard for the assessment of portal hypertension in patients with cirrhosis. Although an indirect measurement of portal pressure, there is an excellent correlation between HVPG and direct measurements of portal pressure in patients with HCV cirrhosis as well as other causes of cirrhosis.28 The technique is safe, takes approximately 20–30 minutes in experienced hands, and has very few complications. In our experience, there is a very high degree of patient acceptance, even with repeat measurements over time. In patients suspected of having cirrhosis, we usually obtain a transjugular liver biopsy and a measurement of HVPG in the same sitting. However, the technique has to be performed according to a strict protocol by using a balloon catheter inserted into the hepatic vein, obtaining at least triplicate readings, and using a recorder to obtain permanent tracings of the pressure measurements as opposed to digital readouts.29
HVPG measurements predict the development of esophageal varices, the risk of variceal hemorrhage, hepatic decompensation, the development of HCC, and survival.30, 31, 32 Repeat measurements of HVPG are extremely useful in determining the effectiveness of pharmacologic therapy in the treatment of portal hypertension related to cirrhosis. Noninvasive methods lack the sensitivity to detect small (10%–20%) changes in portal pressure, changes that have significant clinical applicability.
In the present study by Reiberger et al,33 90 patients with Child–Pugh class A, HCV genotype 1 and 4 cirrhosis had HVPG measurements and upper endoscopy at baseline before initiation of antiviral therapy with pegylated interferon and ribavirin. Forty percent had esophageal varices, but other complications of portal hypertension, ie, ascites, were exclusion criteria. In patients who had CSPH, defined as HVPG >10 mm Hg, 72% had esophageal varices at entry to the study. Varices were absent in all patients who did not have CSPH. An SVR was achieved in 51% of patients without CSPH compared with 14% for those with CSPH (P = .0007). Multivariate analysis revealed that an elevated HVPG, baseline HCV RNA levels, and HCV genotype were the only independent risk factors for treatment failure. Moreover, the predictive value of HVPG was greater than measurements of liver stiffness, baseline HCV RNA levels, or IL-28B polymorphism; the latter was not associated with an SVR.
No comments:
Post a Comment