CROI 2016 Abstract eBook

Abstract Listing

Poster Abstracts

864

Fertility Desire, Unprotected Sex, and Viral Load in HIV-Positive FSWs in Kenya Kate S. Wilson 1 ; Linnet Masese 2 ; Lucy Adala 2 ; Juma Shafi 3 ; Barbra A. Richardson 1 ; Jane Simoni 1 ; Kishor Mandaliya 4 ;Walter Jaoko 3 ; Scott McClelland 1 1 Univ of Washington, Seattle, WA, USA; 2 Kenya Rsr Prog at the Univ of Washington, Nairobi, Kenya; 3 Univ of Nairobi, Nairobi, Kenya; 4 Coast Province General Hosp, Mombasa, Kenya Background: In HIV-positive female sex workers (FSWs), fertility desire may increase secondary HIV transmission risk if it is associated with unprotected sex, and with poor ART adherence, leading to detectable viral load (VL). We conducted a prospective cohort study of HIV-positive FSWs in Mombasa, Kenya to examine fertility desire as a risk factor for semen detection by prostate specific antigen test (PSA) and detectable plasma VL. We also explored fertility desire as a risk factor for HIV transmission potential (TP), the co- occurrence of unprotected sex and detectable VL. Methods: Participants were HIV-positive FSWs taking ART, ≥18 years old, pre-menopausal, and not pregnant. Fertility desire ( wants children ) was assessed quarterly. Primary outcomes were semen detection in vaginal secretions by PSA (quarterly) and detectable VL (≥180 copies/ml, semi-annual). HIV TP was defined as visits with both positive PSA and detectable VL. We evaluated effect modification by contraceptive use (modern method vs. none/condoms only). We used log-binomial generalized estimating equations with independent working correlation structure and robust standard errors to obtain relative risks (RR) and 95% confidence intervals (CI). Final models were adjusted for age and other covariates identified through model building. Results: Overall, 211 women contributed 1,581 visits to the analysis. Fertility desire was common (72 women, 22.9% visits). Semen detection by PSA was similar at visits with and without fertility desire (67/360, 18.6% vs. 201/1,207, 16.7%, Table 1). The association between fertility desire and PSA detection differed in women not using contraceptives compared to those using a modern method besides condoms (p-value for interaction=0.06). Fertility desire was associated with significantly higher risk of PSA detection at visits with no contraceptive use (adjusted RR [aRR] 1.52, 95%CI 1.14-2.03), though not at visits with contraception use (aRR 0.63, 95%CI 0.28-1.40). Fertility desire was not associated with significantly higher risk of detectable VL (aRR 0.83, 95%CI 0.52-1.34) or HIV TP events (aRR1.22, 95%CI 0.47-3.11). Conclusions: Fertility desire was common in this sample of HIV-positive FSWs, and was associated with higher risk of unprotected sex in women not using contraception. However, given the high level of viral suppression, there was no substantial increase in events when HIV transmission would likely occur. Combination HIV prevention should better address fertility desire in this key population.

Poster Abstracts

865 Inaccurate Reporting of CondomUse AmongWomen Using Injectable Contraception Renee Heffron 1 ; Urvi M. Parikh 2 ; Kerri J. Penrose 2 ; Nelly R. Mugo 3 ; Deborah Donnell 4 ; Connie M. Celum 1 ; JohnW. Mellors 2 ; Jared M. Baeten 1

1 Univ of Washington, Seattle, WA, USA; 2 Univ of Pittsburgh, Pittsburgh, PA, USA; 3 Kenya Med Rsr Inst, Thika, Kenya; 4 SCHARP, Fred Hutchinson Cancer Rsr Cntr, Seattle, WA, USA Background: Observational analyses of HIV acquisition by women using injectable contraception versus no hormonal contraception are potentially biased by differential accuracy in self-reported condom use, a key confounding factor. Laboratory markers of semen exposure can be used to objectively determine whether over-reporting is occurring with different frequency among women using injectable contraception and no contraception. Methods: Using stored vaginal swab specimens from a random sample of HIV-uninfected women participating in a randomized clinical trial of pre-exposure prophylaxis (the Partners PrEP Study), we measured the presence of Y chromosomal (Yc) DNA among those who reported 100% condom use with all partners during the past month and sex within the past 7 days. Half of the samples tested were fromwomen reporting injectable contraceptive use at the visit when the swab was collected and half were fromwomen using no contraception aside from condoms. Multivariate logistic regression was used to estimate the association between injectable contraceptive use and Yc DNA detection. Results: The median age of women was 34 (interquartile range [IQR] 29-39), most (97.0%) had at least one child, and the median number of sex acts during the month prior to vaginal swab collection was 3 (IQR 2-5). Among 428 specimens tested (213 from injectable contraceptive users and 215 fromwomen using no contraception), 32.0% had Yc DNA detected with a mean of 193 copies/10,000 human cells (range 0.1-8201). The frequency of detection did not differ by contraceptive use: 34.2% of DMPA users versus 29.8% of women using no contraception, with an odds ratio [OR] of 1.3 (95% confidence interval [CI] 0.9-2.0) after adjustment for participant age, number of children, and sexual frequency. Conclusions: One third of women in this study potentially over-reported condom use, but the frequency of over-reporting did not differ among women using injectable contraception compared to women using no contraception. These results refute the hypothesis that residual behavioral confounding may entirely explain the increased risk of HIV

363

CROI 2016