Background: Implication of pharmacogenetic variations and efavirenz pharmacokinetics in concomitant efavirenz based antiviral therapy and anti-tubercular drug induced liver injury (DILI) has not been yet studied. We performed a prospective case-control association study to identify the incidence, pharmacogenetic, pharmacokinetic and biochemical predictors for anti-tubercular and antiretroviral drugs induced liver injury (DILI) in HIV and tuberculosis (TB) co-infected patients.
Methods and findings: Newly diagnosed treatment naïve TB-HIV co-infected patients (n = 353) were enrolled to receive efavirenz based ART and rifampicin based anti-TB therapy, and assessed clinically and biochemically for DILI up to 56 weeks. Quantification of plasma efavirenz and 8-hydroxyefaviernz levels and genotyping for NAT2, CYP2B6, CYP3A5, ABCB1, UGT2B7 and SLCO1B1 genes were done. The incidence of DILI and identification of predictors was evaluated using survival analysis and the Cox Proportional Hazards Model. The incidence of DILI was 30.0%, or 14.5 per 1000 person-week, and that of severe was 18.4%, or 7.49 per 1000 person-week. A statistically significant association of DILI with being of the female sex (p = 0.001), higher plasma efavirenz level (p = 0.009), efavirenz/8-hydroxyefavirenz ratio (p = 0.036), baseline AST (p = 0.022), ALT (p = 0.014), lower hemoglobin (p = 0.008), and serum albumin (p = 0.007), NAT2 slow-acetylator genotype (p = 0.039) and ABCB1 3435TT genotype (p = 0.001).
Conclusion: We report high incidence of anti-tubercular and antiretroviral DILI in Ethiopian patients. Between patient variability in systemic efavirenz exposure and pharmacogenetic variations in NAT2, CYP2B6 and ABCB1 genes determines susceptibility to DILI in TB-HIV co-infected patients. Close monitoring of plasma efavirenz level and liver enzymes during early therapy and/or genotyping practice in HIV clinics is recommended for early identification of patients at risk of DILI.