CROI 2018 Abstract eBook

Abstract eBook

Poster Abstracts

Analyses were restricted to the 261 participants (87%) who were retained at 12-months. We examined baseline demographic, behavioral, and psychosocial predictors of STI diagnosis over the 12-month follow-up, as well as change scores (e.g., changes in condom use and number of partners). Results: 11% of participants had an STI diagnosis in the 6 months before starting PrEP. Over the course of 12-month follow-up, 44% of participants were diagnosed with an STI. Diagnoses per visit ranged from 16% (6M) to 10% (12M), and 23% of participants were diagnosed at interim (i.e., non-study) visits to the health center. STI diagnosis was associated with being under 25 (p < .02), but not with race/ethnicity, income, or education. In stepwise regression models including significant bivariate variables, four factors were retained as predictors of STI diagnosis: a) being under 25 (OR = 4.8); b) reporting more than 5 partners at baseline (OR= 3.5); c) STI diagnosis in 6-months prior to PrEP uptake (OR = 3.5); and d) increasing the number of partners from baseline to 12-months (OR=2.1). Average condom use decreased from baseline (60%) to 12-months (45%), but neither overall condom use nor change in condom use were associated with STI diagnosis. Conclusion: The strength of baseline factors in predicting STI incidence suggests that risk compensation may be less significant than underlying behavior patterns in post-PrEP STI diagnosis. The known association between STI diagnosis and HIV seroconversion suggests that PrEP is effectively preventing new HIV infections. Particular attention and support is needed for younger PrEP users. Although many PrEP prevention messages stress condom use, number of partners appears to be a more important predictor of STI diagnosis among PrEP users. 1026 INCIDENT HIV, HEPATITIS C AND OTHER STI IN DAILY AND EVENT-DRIVEN PrEP USERS Elske Hoornenborg , Liza N. Coyer, Roel C. Achterbergh, Amy Matser, Udi Davidovich, Maarten F. Schim van der Loeff, Henry J. de Vries, Maria Prins Public Health Service Amsterdam, Amsterdam, Netherlands Background: There are concerns that HIV pre-exposure prophylaxis (PrEP) use may lead to a rise in sexually transmitted infections (STI). We present interim results of HIV, hepatitis C virus (HCV) and bacterial STI incidence among men who have sex with men (MSM) and transgender persons (TGP) in the Amsterdam PrEP demonstration project (AMPrEP). Methods: AMPrEP enrolment started in August 2015 at the STI clinic of the Public Health Service of Amsterdam, the Netherlands. Participants could choose between daily (dPrEP) and event-driven PrEP (edPrEP, i.e., 2 pills tenofovir/ emtricitabine before sex and 2 times 1 pill after 24/48 hours). Switching between PrEP regimens was allowed at 3-monthly study visits. Participants were tested for HIV, HCV and STI at each study visit and additionally in case of symptoms or partner notification. Data until June 2017 were analysed. Incidence rates per 100 person years (py) were calculated overall and in each quarter since PrEP start. The association of time on PrEP with STI incidence was assessed in piecewise exponential survival models. STI incidence was compared between PrEP regimens by Poisson regression with random effects. Results: Of 372 AMPrEP participants with follow-up data, 271 (73%) initially chose dPrEP and 101 (27%) edPrEP. Median follow-up time was 15 months (IQR 14-18). We recorded 108 switches (n=53 dPrEP to edPrEP; n=55 edPrEP to dPrEP) in 81 participants. HIV incidence was 0.42/100py (n=2, 95% CI 0.11-1.69) (dPrEP: 0.57; edPrEP: 0). HCV incidence was 1.29/100py (n=6, 95%CI 0.58-2.87) (dPrEP: 1.16; edPrEP: 1.68, p=0.666). Incidence of bacterial STI (chlamydia, gonorrhea, syphilis) was 97.8/100py (95%CI 89.3-107.1) and was significantly higher in dPrEP users than in edPrEP users (111.4 vs 57.9/100py, IRR 1.87, 95%CI 1.40-2.51). We found no significant change in STI incidence over time (p=0.697); incidence varied from 92.3/100py in month 0-3 to 90.4/100py in month 15-18 (figure). The incidence of anal STI was higher in dPrEP users than in edPrEP users (75.4 vs 30.6/100py, IRR 2.32, 95% CI 1.56-3.44). Conclusion: Both HCV and STI incidence following PrEP start were high in this PrEP demonstration project for MSM and TGP. STI incidence was not increasing with time on PrEP. STI but not HCV incidence was higher in dPrEP than edPrEP users, possibly reflecting differences in sexual behaviour. HIV incidence was low in line with previous dPrEP and edPrEP studies. These data suggest that frequent screening for bacterial STI and HCV should be offered to PrEP users.

Poster Abstracts

1027 SEXUALLY TRANSMITTED INFECTIONS AND ADHERENCE TO PrEP Claire Bristow 1 , David J. Moore 2 , Michael Dube 3 , Katya Calvo 4 , Eric Ellorin 2 , Jill Blumenthal 2 , Sheldon Morris 2 1 University of California San Diego, La Jolla, CA, USA, 2 University of California San Diego, San Diego, CA, USA, 3 University of Southern California, Los Angeles, CA, USA, 4 Los Angeles Biomedical Research Institute at Harbor–UCLA Medical Center, Torrance, CA, USA Background: High levels of STI incidence have been reported in preexposure prophylaxis (PrEP) studies. We conducted an analysis from a PrEP adherence study to assess the incidence rates of STIs and determine if higher adherence was associated with higher incidence of STIs to suggest differences in risk taking by adherence. Methods: Men who have sex with men (MSM) and transgender women, age 18+ years, at risk for HIV were recruited for participation in a PrEP adherence study at four urban Southern California medical centers. Chlamydia (CT) and gonorrhea (NG) nucleic acid amplification testing was performed at baseline and follow-up visits at week 4 and every 12 weeks throughout the 48-week study period. Adherence was measured by intracellular tenofovir diphosphate (TFV-DP) using dry blood spots collected at week 12 and 48. adherent was considered >719 fmol/punch, consistent with ≥4 doses per week and >1246 fmol/punch, consistent with 7 doses per week. Plasma FTC levels were also measured and adherent was considered >350 ng/mL, consistent with dosing within the past 24 hours. Results: Three-hundred and ninety-four participants had 121 new CT/NG cases. The incidence rate of CT/NG was 43.5 per 100 person years of follow-up for first CT/NG infection in the study. Of the 75 incident NG cases and 98 incident CT cases, 16 (21.3%) and 33 (33.7%) had a positive NG and CT results at subsequent visits, respectively. Among those adherent to TFV-DP at week 12, the incidence rate of CT/NG was 44.9 per 100 person years (95% CI: 36.4, 53.4) compared to 35.2 per 100 person years (95% CI: 16.8, 53.6) among those not adherent at week 12 (IRR: 1.3 (95% CI: 0.7, 2.2)). Among those adherent to FTC at week 12, the incidence rate of CT/NG was 51.4 per 100 person years (95% CI: 37.4, 65.3) compared to 39.0 per 100 person years (95% CI: 29.8, 48.2) among those not adherent to FTC at week 12 (IRR: 1.3 (95% CI: 0.9, 1.9)). No STI rates were statistically different between groups. Similar results were found for adherence by cutoff of >1246 fmol/punch and for week 48 TFV-DP levels. Conclusion: The incidence of CT and NG infections in this PrEP demonstration project was comparable to other PrEP studies. Although point estimates were higher for STI rates in those who were more adherent, as defined by 7 doses a week, 4 or more doses a week or dosing in last 24 hours, this finding was not statistically significant and therefore does not support the association between higher sexual risk taking with improved adherence.

CROI 2018 393