Response rates to pegylated interferon and ribavirin in HCV/HIV coinfection: a research synthesis.

Several recent trials have determined varying rates of response to pegylated interferon plus ribavirin (peg-IFN + RIB) in hepatitis C virus (HCV)/human immunodeficiency virus-coinfected patients. We sought to identify a pooled response rate and sources of interstudy variability. A literature search was conducted to identify randomized or prospective studies evaluating response rates to peg-IFN + RIB in coinfected patients. A Bayesian hierarchical model was used to estimate overall response rate. Between-study variance was calculated and sensitivity analyses were conducted. Meta-regression was employed to identify sources of between-study variability. The literature search yielded seven studies of 146, which matched keywords and inclusion criteria. The combined patient total was 784. Individual intention-to-treat response rates ranged from 27.3% to 44.2%. The pooled Bayesian estimate of percent response was 33.3%. Significant interstudy heterogeneity was detected. Meta-regression yielded no significant effects of covariates on response rate. Subanalyses by CD4+, HCV viral load and genotype yielded sustained virological response (SVR) odds ratios of 0.73 for low CD4+, 0.41 for high viral load and 0.30 for genotype 1/4. The pooled response rate is not attributable to any one study. Response is poor compared with HCV-monoinfected patients. Interstudy variability is not satisfactorily explained by factors influencing individual response, but may be due to differences between studies unavailable for inclusion in this analysis. However, both genotype 1/4 and high HCV viral load at baseline were significantly associated with a reduction in odds of SVR in pooled subanalysis. Improved treatments are needed in coinfected patients, especially with genotype 1/4 and high viral load.