Efavirenz (EFV) is one of the most commonly prescribed antiretrovirals for use in the treatment of human immunodeficiency virus (HIV) infection. EFV is extensively metabolized by cytochrome P450 to a number of oxygenated products; however, the pharmacologic activity and distribution of these metabolites in anatomic compartments have yet to be explored. The systemic distribution of EFV oxidative metabolites was examined in blood plasma, seminal plasma, and cerebrospinal fluid from subjects on an EFV-based regimen. The 8-hydroxy EFV metabolite was detected in blood plasma, seminal plasma, and cerebrospinal fluid, with median concentrations of 314.5 ng/ml, 358.5 ng/ml, and 3.37 ng/ml, respectively. In contrast, 7-hydroxy and 8,14-hydroxy EFV were only detected in blood plasma and seminal plasma with median concentrations of 8.84 ng/ml and 10.23 ng/ml, and 5.63 ng/ml and 5.43 ng/ml, respectively. Interestingly, protein-free concentrations of metabolites were only detectable in seminal plasma, where a novel dihdyroxylated metabolite of EFV was also detected. This accumulation of protein-free EFV metabolites was demonstrated to be the result of differential protein binding in seminal plasma compared with that of blood plasma. In addition, the oxidative metabolites of EFV did not present with any significant pharmacologic activity toward HIV-1 as measured using an HIV green fluorescent protein single-round infectivity assay. This study is the first to report the physiologic distribution of metabolites of an antiretroviral into biologic compartments that the virus is known to distribute and to examine their anti-HIV activity. These data suggest that the male genital tract may be a novel compartment that should be considered in the evaluation of drug metabolite exposure.

Efavirenz (EFV), one of the most widely prescribed drugs used to treat HIV, is extensively metabolized by the cytochromes P450 to a number of oxygenated products; however, the pharmacologic activity and distribution of these metabolites in anatomic compartments relevant to HIV infection beyond blood plasma have yet to be explored. The systemic distribution of EFV oxidative metabolites was examined in blood plasma (BP), seminal plasma (SP), and cerebrospinal fluid (CSF) from subjects on an EFV‐based regimen. The 8‐hydroxy EFV metabolite was detected in BP, SP and CSF, with median concentrations of 314.5ng/ml, 358.5ng/ml, and 3.37ng/ml, respectively. In contrast, 7‐hydroxy and 8,14‐hydroxy EFV were only detected in BP and SP with median concentrations of 8.84ng/ml and 10.23ng/ml, and 5.63ng/ml and 5.43ng/ml, respectively. Interestingly, protein‐free concentrations of metabolites were only detectable in SP, where a novel dihdyroxylated metabolite of EFV was also detected. The metabolite concentrations in 34 paired BP and CSF patient samples were not strongly correlated with viral RNA concentrations indicating that the metabolites of EFV are pharmacologically inactive. These results suggest there may be important compartmental differences in EFV metabolite distribution, and the male genital tract may be a novel compartment in understanding local drug distribution and efficacy of oxidative metabolites.

Efavirenz (EFV), one of the most commonly prescribed drugs for HIV, has been associated with liver toxicity; however the mechanism( s) for this is not well understood. In non‐stressed conditions, BiP (binding immunoglobulin protein) binds IRE1α (inositol‐requiring enzyme 1α), PERK (PKR‐like endoplasmic reticulum kinase), and ATF6 (activating transcription factor 6), rendering them inactive. BiP dissociates when ER stress is sensed, allowing these proteins to become activated. Immunoblotting revealed that primary human hepatocytes treated with 10μM of EFV expressed IRE1α at a 3‐fold increase at 6 hours, while no significant expression changes were observed at 3 and 24 hour time points. A commensurate increase in IRE1α mRNA levels was observed using reverse transcriptase PCR. The chaperone proteins calnexin and grp94, whose expression is upregulated by ER stress, demonstrated modest increases (1.5 fold) after 6 hours of EFV treatment, as did BiP (1.5 fold). No significant change in expression was observed in PERK or ATF6 with EFV treatment. Interestingly, results obtained from human samples were not recapitulated in primary mouse hepatocytes, despite being a conserved pathway, indicating a possible species difference. The results above suggest that EFV may exert its effects through the IRE1α pathway specifically, not by inducing the global unfolded protein response.