Seattle, Washington
March 4–7, 2019


Conference Dates and Location: 
March 4–7, 2019 | Seattle, Washington
Abstract Number: 



David Haas1, Yoninah S. Cramer2, Catherine Godfrey3, Susan L. Rosenkranz2, Francesca Aweeka4, Baiba Berzins5, Robert Coombs6, Kristine Coughlin2, Laura E. Moran7, David Gingrich4, Carmen D. Zorrilla8, Paxton Baker1, Susan E. Cohn5, Kimberly K. Scarsi9

1Vanderbilt University, Nashville, TN, USA,2Frontier Science & Technology Research Foundation, Inc, Amherst, NY, USA,3NIAID, Bethesda, MD, USA,4University of California San Francisco, San Francisco, CA, USA,5Northwestern University, Chicago, IL, USA,6University of Washington, Seattle, WA, USA,7Social & Scientific Systems, Silver Spring, MD, USA,8University of Puerto Rico, San Juan, Puerto Rico,9University of Nebraska Medical Center, Omaha, NE, USA

Abstract Body: 

In ACTG A5316 women receiving efavirenz (EFV)-containing ART had 79% and 59% lower etonogestrel (ENG) and ethinyl estradiol (EE) concentrations, respectively, after 21 days of ENG/EE given as a vaginal ring (VR). Women receiving atazanavir/ritonavir (ATV/RTV)-containing ART had 71% higher ENG and 38% lower EE. These results are likely related to ART modulation of pathways responsible for hormone metabolism. We studied genetic associations with ART and hormone pharmacokinetics (PK) in A5316.


A5316 enrolled women living with HIV in Africa, Asia, South America and the US into one of three groups: controls (not on ART), EFV group (600mg daily with nucleoside/nucleotide reverse transcriptase inhibitors (NRTIs), and ATV/RTV group (300/100mg daily with NRTIs). On day 0, a VR was inserted, releasing ENG/EE 120/15 mcg/day. On days 0 (pre-VR) and 21 (during VR), intensive PK sampling for EFV, ATV and RTV was done. On days 7, 14 and 21, single plasma samples for ENG and EE analysis were obtained. We genotyped 27 single nucleotide polymorphisms (SNPs), including 3 that define CYP2B6 normal, intermediate and slow metabolizers, CYP3A4/5, UGT1A1 and CYP1A1/2 SNPs, and estrogen trait-associated SNPs.


Of the 74 evaluable participants in A5316, 72 (97%) had both PK and SNP data (n=25 controls; n=24 EFV; n=23 ATV/RTV). Of these, 35 (49%) identified as Black, 26 (36%) as Hispanic, 8 (11%) as Asian/Pacific Islander and 3 (4%) as White, with 22 (31%) CYP2B6 normal, 32 (44%) intermediate and 18 (25%) slow metabolizers. On both days 0 and 21, CYP2B6 genotype predicted EFV PK (e.g., p=4.5E-5 for day 0 log10 EFV AUC0-8h). In the EFV group, CYP2B6 genotype predicted lower day 21 ENG (p=1.7E-3) and EE (p=6.7E-4) concentrations (Figure), which persisted after adjusting for weight and/or age. Compared to controls, EFV reduced median day 21 ENG concentrations by ~75% in CYP2B6 normal and intermediate metabolizers yet by at least 93% in slow metabolizers. EFV reduced median day 21 EE concentrations by 41% in CYP2B6 normal and intermediate metabolizers, but by 75% in slow metabolizers. No other SNPs were associated with hormone or ART PK after correcting for multiple testing.


CYP2B6 slow metabolizer genotype worsens the adverse PK interaction of EFV with ENG and EE administered by VR, likely due to enhanced cytochrome P450 induction by higher EFV concentrations. Lower EFV dosing based on CYP2B6 genotype may reduce, but likely not eliminate, the impact of EFV on ENG and EE PK.


Session Number: 
Session Title: 
Presenting Author: 
David Haas
Presenter Institution: 
Vanderbilt University School of Medicine