Efavirenz-metabolizing polymorphisms, viral suppression, and depression in HIV-infected individuals initiating antiretroviral therapy in southwestern Uganda

Abstract

Abstract Background: Single-nucleotide polymorphisms (SNPs) in CYP2B6 have previously been associated with a 10-fold range in trough plasma efavirenz concentrations, but associations between these SNPs and efavirenz (EFV)-mediated viral suppression and tolerability remain unclear. Methods: We evaluated three SNPs in CYP2B6 (rs3745274, rs28399499, and rs4803419, Illumina OmniExpress) among HIV-infected Ugandans observed in a cohort study every 3–4 months from 2005–2015. Genotypes from these SNPs were used to group participants into previously described pharmacokinetic strata: extensive (EXT), intermediate (INT), and slow metabolizers (Figure 1). The primary outcomes were viral suppression, defined by an undetectable viral load in the first measurement a minimum of three months after ART initiation, and incident depression in the first two years, defined by a mean score >1.75 on the Hopkins Symptom Checklist. We fitted standard and generalized estimating equations (GEE) logistic regression models for viral suppression and depression, respectively. Models were adjusted for clinical and demographic covariates that reached a significance of P < 0.25 in unadjusted models. Results: Among 103 participants with genotyping, there were no differences in pre-ART viral load or depression by metabolism strata (P > 0.5). Minor allele frequencies for rs3745274, rs28399499, and rs4803419 were 33%, 7%, and 4%, respectively. Approximately 79%, 78%, and 94% of participants were suppressed at their first viral load measurement in the extensive, intermediate, and slow metabolizer strata, respectively (Figure 2; P = 0.35). In adjusted models, metabolism strata were not associated with viral suppression (AORINT 0.81, 95% CI 0.26–2.56; AORSLOW 3.92, 95% CI 0.39–39.40) or with depression (AORINT 1.95, 95% CI 0.75–5.09; AORSLOW 0.72, 95% CI 0.17–3.02; Table). Conclusion: We did not identify an association between efavirenz-metabolizing polymorphisms and viral suppression or depression in a cohort of HIV-infected individuals initiating ART in southwestern Uganda. Future work should reassess these relationships with larger samples and longer-term outcomes and explore additional polymorphisms that may be associated with efavirenz metabolism in this population.