The impact of co-infection with hepatitis C virus and HIV on the tolerability of antiretroviral therapy

Approximately 750 000 Americans are infected with the HIV virus and 4 000 000 are infected with the hepatitis C virus (HCV). Both RNA viruses have similar modes of transmission with an approximate 9–10% co-infection rate [1,2]. Until recently, morbidity and mortality associated with co-infection were primarily caused by the rapidity of progression of HIV relative to HCV. However, as a result of the recent advances in the treatment of HIV, co-infected patients are experiencing fewer opportunistic infections and are increasingly developing serious complications associated with HCV. Although therapy for HCV has proved elusive, the overall mortality rate from HIV in developed countries has dropped dramatically because of potent antiretroviral therapy [3]. Unfortunately, therapy with many antiretroviral agents is potentially hepatotoxic. The effect of HCV co-infection on the tolerability of antiretroviral therapy was assessed. Specifically, it was hypothesized that co-infected patients will discontinue antiretroviral therapy at an increased rate secondary to hepatotoxicity. Sixty-one patients co-infected with HIV and HCV from the Northwestern University HIV Treatment Center were matched with one HIV-positive, HCV-negative control on the basis of sex and baseline CD4 lymphocyte level (± 50 cells/mm3 within 6 months). Antiretroviral therapy was defined as any three-drug antiretroviral regimen or the initiation of any protease inhibitor or non-nucleoside reverse transcriptase inhibitor. Demographic information, baseline laboratory values and the duration of antiretroviral therapy and, if applicable, the reason for discontinuation were noted. Hepatic toxicity was a treatment-induced increase in hepatic transaminase levels to more than or equal to five times the upper limit of normal, which directly led to the discontinuation of antiretroviral therapy. Both the two-sided t-test and the Mann–Whitney rank test were used to compare mean and median values, respectively. Because the two groups were found to differ with respect to baseline alanine aminotransferase (ALT) and aspartate aminotransferase (AST) levels, proportional hazards regression analysis was initially used to model the time to hepatic toxicity. When these two variables were found to have no impact on hepatic toxicity rates, the product-limit method was used to estimate the survival distributions in the two groups, and the log-rank test was used to compare them. Exposure times in patients who permanently discontinued antiretroviral therapy for a reason other than hepatic toxicity or who did not discontinue antiretroviral therapy were treated as censored observations. Forty-five of the 61 HCV-positive patients and 50 of the 61 controls began antiretroviral therapy during the observation period and were used in the statistical analysis. The two groups were similar with respect to age, sex, race, hepatitis B status, initial CD4 lymphocyte level and the initial plasma HIV RNA level. The HCV-positive group had significantly higher baseline median ALT (P < 0.05) and AST (P < 0.01). No differences were found in any other laboratory parameters, including alkaline phosphatase and albumin. The patients in the co-infected group were more likely to be intravenous drug users and hemophiliacs (P < 0.001). HCV RNA levels were not consistently available for the co-infected patients, and no correlation was attempted. There was no significant difference in baseline CD4 lymphocyte level or plasma HIV RNA level, and no significant difference in the change from baseline of the CD4 lymphocyte level or HIV RNA level during the period of observation between the two groups. Six of the 45 HCV-positive patients and two of the 50 controls experienced hepatotoxicity that directly led to the discontinuation of antiretroviral therapy. No significant difference was found in the incidence of non-hepatic intolerance. Baseline ALT and AST did not have any effect on the rate of hepatic toxicity (risk ratio = 1.002 and P > 0.9 in both cases). At 580 days, these rates were 17.0% (95% confidence interval 4.3–29.6%) and 6.8% (95% confidence interval 0.0–16.0%), respectively (P < 0.05) (Fig. 1).Fig. 1.: Product limit estimates of highly active antiretroviral therapy discontinuation rates as a result of hepatic toxicity. HCV, Hepatitis C virus.In contrast to the literature regarding HCV progression in co-infected patients [4–10], the impact of antiretroviral therapy on co-infection has not been widely investigated. Although some research has suggested that there is no impact of antiretroviral therapy on HCV [11], one case report [12] and several studies [13–15] have noted an intolerance of antiretroviral therapy in co-infected patients. A recent prospective study by Rutschmann et al. [13] revealed that in 19 co-infected patients, antiretroviral therapy with protease inhibitors may temporarily worsen HCV viremia and ALT levels. Vento et al. [14], in a 9 month follow-up of 51 co-infected patients, also observed an enhancement of HCV replication and subsequent liver damage. This case–control study revealed that co-infection with HCV and HIV decreases the tolerability of antiretroviral therapy, particularly in terms of hepatotoxicity. Although this investigation is limited in terms of the numbers of observed subjects as well as the direct impact of antiretroviral therapy on HCV viremia and on hepatic histological changes, the data suggest that co-infected patients are more likely to discontinue potent antiretroviral therapy than HCV-negative, HIV-positive patients. David C. Melvin Jenny K. Lee Elizabeth Belsey Jason Arnold Robert L. Murphy