Background: Antiretroviral-based HIV prevention interventions, including pre-exposure prophylaxis (PrEP) and antiretroviral therapy (ART), showed high efficacy for HIV protection in clinical trials among African HIV serodiscordant couples. Assessing the effectiveness of these interventions, and their ability to complement one another, in implementation settings is a priority.
Methods: We are conducting the Partners Demonstration Project, an open-label, prospective study of PrEP and ART delivery among antiretroviral-naïve high-risk heterosexual HIV serodiscordant couples in Kenya and Uganda. High-risk couples are defined by a validated risk scoring tool (Kahle et al., JAIDS 2013). PrEP is offered as a ‘bridge’ to ART in the partnership – i.e., until ART initiation by the HIV-infected partner and for the first 6 months after ART is started; ART is recommended following national ART guidelines – initially CD4 <350 cells/µL but more recently all HIV serodiscordant couples regardless of CD4 count. To assess the impact of the PrEP as a bridge to ART strategy on HIV transmission, we compared observed HIV incidence to a counterfactual simulation model, using bootstrapping methods and constructed with data from a prior prospective study of HIV serodiscordant couples (the Partners PrEP Study, placebo arm), sampled selecting for a subset with an identical distribution of risk scores and duration of follow-up.
Results: Enrollment in the Partners Demonstration Project began in November 2012 and completed in August 2014, with a final sample size of 1013 couples. Given the risk score distribution of the study population, the counterfactual simulations predicted 21.7 HIV infections would be expected to date in this population (incidence 5.3 per 100 person-years, 95% CI 3.2-7.6). However, through July 2014, only one incident HIV infection has been observed during 440 person-years of follow-up, for an observed HIV incidence of 0.2 per 100 person-years (95% CI 0.0-1.3, p<0.0001 versus predicted) (Figure). PrEP was used during 47% of the 440 person-years of follow-up, ART 17%, both 25%, and neither 11%.The one transmission occurred in the absence of ART and with evidence of low PrEP adherence.
Conclusions: Early results from this demonstration project integrating time-limited PrEP and ART for HIV prevention in African couples show near elimination of HIV transmission, with an observed HIV incidence <0.5% per year compared to an expected incidence >5% per year.