Jesse Raffa, M.Sc., and Harout Tossonian, M.D., and their poster on hepatotoxicity in injection drug users and non-injection drug users.

Based on data generated in clinical trials and observational cohorts, hepatotoxicity is a known cause of treatment interruption among HIV-infected patients receiving nevirapine (NVP, Viramune)-based highly active antiretroviral therapy (HAART) in up to 10%-15% of cases. This may be especially true for injection drug users (IDUs), who are almost universally coinfected with hepatitis C virus (HCV). With this in mind, the current observational study¹ was undertaken to estimate the incidence of hepatotoxicity in HIV-infected IDUs (n = 80) and non-IDUs (n = 82) initiating nevirapine-based HAART as their first non-nucleoside reverse transcriptase inhibitor (NNRTI)-based regimen.

At baseline, median alanine aminotransferase (ALT) levels were comparable in both groups (39 versus 38 U/L in IDUs and non-IDUs, respectively), while aspartate aminotransferase (AST) levels were slightly higher in IDUs (44 versus 31 U/L), with 94% and 30% of IDUs and non-IDUs, respectively, being coinfected with HCV. Median baseline HIV plasma viral load measures and CD4+ cell counts showed slightly more advanced disease in IDUs (64,750 versus 26,650 copies/mL and 190 versus 290 cells/mm³).

Over 12 months of observation, there was no difference in ALT levels (47 versus 35 U/L in IDUs and non-IDUs, respectively), but there were significantly higher AST levels (53 versus 33 U/L, P < .001) in IDUs. Overall, five-fold increases in ALT or AST were observed in 15% of participants in both groups, with treatment discontinued due to hepatotoxicity in 9% of both groups. Interestingly, the peak incidence of elevation of transaminase levels was observed after one month of therapy in IDUs, but was delayed until the third month in non-IDUs. This suggests that there is a separate pathogenesis for our observations in the two groups. In IDUs, this may be driven by HCV coinfection, and may represent some form of immune reconstitution or HCV re-activation that would resolve on its own even if nevirapine therapy were continued. In non-IDUs, it may be more attributable to nevirapine itself, as this group included patients with higher CD4+ cell counts, a phenomenon that is more common in this population than among IDUs. Using a rigorous statistical model, the factors that were associated with five-fold increases in transaminase levels were HCV coinfection (eight-fold risk), baseline elevation of transaminase levels (three-fold risk) and lack of previous exposure to HAART (four-fold risk).

These results suggest that the majority of HCV coinfected IDUs will not develop hepatotoxicity following the initiation of nevirapine-based HAART. If they do, the mechanism of the elevation of transaminase levels may be different from that previously reported in non-IDUs and may not preclude the continuation of nevirapine therapy once it occurs as it may not relate directly to the drug itself. Therefore, nevirapine should still be considered a viable therapeutic option in this population.

Footnote

1. Tossonian H, Raffa J, Grebely J, et al. Hepatotoxicity in injection drug users (IDUs) and non-IDUs receiving nevirapine-based HAART. In: Program and abstracts of the XVI International AIDS Conference; August 13-18, 2006; Toronto, Canada. Abstract WEPE0176.