CROI 2019 Abstract eBook

Abstract eBook

Poster Abstracts

Undetectable urine TFV via the immunoassay was strongly associated with HIV seroconversion (OR 2.8; 95%CI: 1.5-5.4, p=0.002). Conclusion: Urine TFV levels measured by a novel antibody-based assay were associated with protection from HIV acquisition among participants in a PrEP demonstration project. Urine TFV levels were correlated with other pharmacologic measures (hair/DBS), with high specificity in detecting sub- optimal dosing. Since immunoassays allow for POC testing, this novel assay could detect low PrEP adherence detected in real-time, allowing immedidate intervention to optimize PrEP outcomes.

couples followed for two years. Despite declines in condomless sex shortly after PrEP initiation, roughly a third of the HIV serodiscordant heterosexual couples in the study continued to engage in condomless sex, emphasizing the importance of continued PrEP use to sustain HIV protection.

Poster Abstracts

949 Y-CHROMOSOME DETECTION & CONDOMLESS SEX IN SEX WORKERS IN THE SENEGAL PrEP PROJECT Geoffrey S. Gottlieb 1 , Stephen E. Hawes 1 , Mame D. Bousso Bao 2 , Anna J. Ndiaye 2 , Daouda Gueye 2 , Dana Raugi 1 , Moustapha Mane 2 , Aminata Mboup 2 , Ousmane Diouf 2 , Carlos Suarez 3 , Fatima Jones 3 , Josie Presley 4 , Coumba Toure Kane 2 , Moussa Sarr 3 , Souleymane Mboup 2 1 University of Washington, Seattle, WA, USA, 2 Institut De Recherche En Santé De Surveillance Épidémiologique Et De Formation, Dakar, Senegal, 3 Westat, Inc, Rockville, MD, USA, 4 Bill and Melinda Gates Foundation, Seattle, WA, USA Background: Oral Truvada (FTC/TDF)–based PrEP has the potential to reduce HIV acquisition in female sex workers (FSW). Condom use is still recommended with PrEP due to concerns about adherence, HIV and STI acquisition. If FSW think they are protected from HIV from PrEP there may be incentives to reduce condom use: costs, client preference and ability to charge more for condomless sex. Self-reported condom use measures have significant limitations and biases. Detection of male Y-chromosomes (Y-c) in FSW genital swabs is a potential biomarker for condomless sex. Methods: During the Senegal PrEP Demonstration Project, vaginal swabs were collected fromwomen at baseline (pre-initiation of PrEP) and quarterly from PrEP initiation for 1 year. Vaginal swabs were frozen and tested in bulk at UW- Seattle. A random sample of 165 swabs were chosen for Y-c testing throughout the study period. The Quantifiler® Duo DNA Quantification Kit was used for Y-c detection. We analyzed self–reported condom use, STI (N. gonorrhoeae and C. trachomatis by NAAT (GC/CT)) and Y-c detection. Results: 165 vaginal swab samples from 132 FSWwere tested for Y-c. 164 samples gave valid results. 35/164 (21.3%) samples from 32 (24.2%) FSW contained detectable Y-c. (Baseline: 7/42 (16.7%); M1-3: 11/39 (28.2%); M6: 6/21 (28.6%); M9: 4/25 (16.0%); M12: 7/37 (18.9%). Overall, there was no significant difference between baseline and PrEP use for detection of Y-c (P>0.05, Fisher Exact Test). In 32 FSWwith serial Y-c sampling, 59.4%were always negative; 9.4%were always positive; 3.1%were initially negative and then positive; and 28.1%were initially positive and then negative. Eight FSWwho were screened for Y-c presence had a positive NAAT for GC and/or CT on at least one visit, however only one FSW had concurrent tests positive for both Y-c, GC and CT. All FSWwith available data (N=27), whom had detection of Y-c in vaginal swabs, also self–reported consistent condom use in the preceding 7 days with all clients. “Main partner” condomless sex did not account for the majority of Y-c detection. Conclusion: A significant number (24.2%) of FSW had presence of Y-c on genital swabs suggesting lack of or inconsistent condom use. Y-c detection was consistent throughout the year–long study period, suggesting lack of risk compensation due to Truvada use. STI were infrequent. Condomless sex, as detected by Y-c in vaginal swabs, appears common in FSWwho self-reported consistent condom use, casting doubt on this proxy for measuring their use.

948 NO EVIDENCE OF SEXUAL RISK COMPENSATION AMONG HIV SERODISCORDANT COUPLES ON PrEP

Katrina F. Ortblad 1 , Randy Stalter 1 , Elizabeth A. Bukusi 2 , Kenneth Ngure 3 , Andrew Mujugira 4 , Connie L. Celum 1 , Jared Baeten 1 , Renee Heffron 1 1 University of Washington, Seattle, WA, USA, 2 Kenya Medical Research Institute, Nairobi, Kenya, 3 Kenyatta University, Nairobi, Kenya, 4 Makerere University, Kampala, Uganda Background: Recent studies suggest evidence of increased HIV risk-related sexual behaviors, such as condomless sex, following initiation of pre-exposure prophylaxis (PrEP) among men who have sex with men and female sex workers. We explored the effect of PrEP initiation on condomless sex among HIV serodiscordant heterosexual couples in Kenya and Uganda. Methods: We used longitudinal data from HIV-uninfected participants enrolled in the Partners Demonstration Project, an open-label study of PrEP delivered to HIV-uninfected members of HIV serodiscordant heterosexual couples in Kenya and Uganda from 2012-2016. Participants were encouraged to use PrEP until their HIV-infected partner had used ART for ≥6 months (expected to be commensurate with viral suppression). At each quarterly visit, participants self-reported the frequency of sex and condom use with their study partner in the past month. We used linear regression models with individual-level fixed effects to measure the effect of PrEP initiation and time since PrEP initiation on reports of any condomless sex, controlling for the frequency of sex and self-reported pregnancy desires. We restricted our analysis to participants who reported any sex with their study partner in the past month during follow up time prior to the HIV-infected partner using ART for ≥6 months. Results: Of the 1013 HIV-uninfected individuals enrolled in the study, 974 (96%) initiated PrEP and reported sex with their study partner in the past month. In the month following PrEP initiation, reporting any condomless sex decreased from 65% to 32%, a decline of 33% (95% CI -37% to -30%, p<0.001). The prevalence of condomless sex between study partners on PrEP then remained relatively constant over the next 20 months (median: 33%, IQR 30%- 35%), Figure 1. The overall effect of time since PrEP initiation on condomless sex between study couples was a decline of 33% (95% CI -35% to -20%, p<0.001). Conclusion: We found no evidence of sexual risk compensation following PrEP initiation in a cohort of Kenyan and Uganda HIV heterosexual serodiscordant

CROI 2019 371