P3.109 Point-of-care testing for sexually transmitted infections in hiv prevention trials

Introduction Accurate HIV risk assessment is essential when screening volunteers for HIV prevention studies. STI testing plays a key role, but has traditionally been conducted in central laboratories resulting in reporting delays, which can impact on screening and enrolment decisions, and on participant care during follow-up. Here, we outline the implementation of a 2 hour point-of-care (POC) STI care model for clinical trial participants. Methods As part of a Phase I/IIa vaccine study, CAPRISA implemented POC STI testing for chlamydia, gonorrhoea (Xpert CT/NG, 90 min) and trichomonas (OSOM, 15 min) at a site laboratory of a research clinic in Durban, South Africa, in July 2015. Since then, the POC model has been adopted by 5 other prevention studies, becoming the main STI testing model at CAPRISA. Results A total of 1426 Xpert CT/NG assays were run on two 4-module Genexpert machines between July 2015 and November 2016. Chlamydia was detected in 206 (14.4%), gonorrhoea in 79 (5.5%) samples, and 52 (3.6%) samples showed both infections. No infection was detected in 1070 (75.0%) samples. 143 (8.6%) samples showed either an error message (5.0%), an invalid result (3.0%) or no result (0.6%) requiring repeat testing, either on the same or next convenient visit. Trichomonas testing was conducted simultaneously on all female participants (n=1093, prevalence 2.7%). Gram staining to diagnose bacterial vaginosis and candida infection was performed on-site in one of the studies. Implementation of this model allowed early detection of screen failures for Phase I/IIa vaccine studies. During follow-up, participants with STIs received enhanced risk reduction counselling and immediate treatment on the day of sample collection. This resulted in improved care, early partner notification, and cost-savings by avoiding unnecessary screening procedures and repeat visits. Conclusion A 2 hour POC STI testing model can streamline screening and follow-up of participants in HIV prevention studies, and should be considered for implementation by other research sites. Support: Cepheid loaned two 4-module Genexpert machines to the study team free-of-charge, but did not contribute to the preparation of this abstract