Efficacy of preexposure prophylaxis for HIV-1 prevention among high-risk heterosexuals: subgroup analyses from a randomized trial

Abstract

Background: Daily oral antiretroviral preexposure prophylaxis (PrEP) is a promising strategy for prevention of HIV-1 acquisition. Three clinical trials demonstrated PrEP efficacy; however, two PrEP trials among women did not find protection against HIV-1. One hypothesis proposed for these divergent results is that PrEP efficacy may be reduced in populations with higher HIV-1 incidence.

Methods: Using data from the Partners PrEP Study, a randomized, placebo-controlled trial of daily oral tenofovir (TDF) and emtricitabine/tenofovir (FTC/TDF) PrEP among heterosexual HIV-1 serodiscordant couples from Kenya and Uganda, we assessed PrEP efficacy among subgroups at higher risk for HIV-1 acquisition, including subgroups of women with high HIV-1 incidence.

Results: The overall placebo arm HIV-1 incidence was 2.0 per 100 person-years. Among higher risk subgroups, placebo arm HIV-1 incidence ranged from 3.9 to 6.6 per 100 person-years. In all subgroups, PrEP was protective against HIV-1 acquisition, with efficacy point estimates ranging from 64 to 84%. Among subgroups of women with placebo-arm HIV-1 incidence more than 5.0, efficacy estimates ranged from 64 to 84%. Monthly visit attendance for PrEP refills and tenofovir detection in plasma were high.

Conclusion: Among higher-risk subgroups in the Partners PrEP Study, including groups solely of higher-risk women, both TDF alone and combined FTC/TDF PrEP had consistently high efficacy for HIV-1 protection. PrEP, when used with high adherence, is a highly effective prevention strategy for higher risk heterosexuals. Prioritizing PrEP for persons at high risk of HIV-1 will maximize its prevention impact.

Figure 1. HIV-1 incidence and PrEP efficacy overall and among higher-risk subgroups

Incidence rates are per 100 person years. STI = sexually transmitted infection. NNT = number needed to treat to avert 1 infection, calculated as 1 divided by the difference in the placebo arm incidence rate and active arm incidence rate. Plasma HIV-1 RNA concentrations were measured by PCR in samples collected at enrollment from HIV-1 infected study partners. * For time-dependent variables, N in these subgroups represents participants who were categorized as high-risk during at least one month of follow up. For unprotected sex, data were from the three months prior to each visit. The STI risk group included participants who themselves or their partners had signs (genital discharge or ulcers), symptoms (genital burning, discharge, sores, or lower abdominal pain in women), or a diagnosed STI (Neisseria gonorrhoeae, Chlamydia trachomatis, Trichomonas vaginalis) within the three months prior to each visit. Symptoms of sexually transmitted infections (STIs) were evaluated quarterly and also recorded if reported in the interim. A clinical evaluation of STI signs and laboratory testing for STIs [13] were conducted annually and as indicated. The total person-time for time-varying exposures (i.e., unprotected sex and STI within the prior three months) was too small to estimate efficacy among women alone. ** Composite risk score includes age of the uninfected partner, number of children, circumcision status of male HIV-1 uninfected partner, married/cohabiting, unprotected sex, and HIV-1 infected partner viral load [7]

Figure 2. Visit attendance and medication adherence by visit month

A) visit attendance among higher-risk subgroups including both men and women; B) proportion with tenofovir detected, among a subgroup of randomly-selected participants from the trial’s active PrEP arms, as detailed previously [2]; C) visit attendance among higher-risk subgroups of women; D) proportion with tenofovir detected among higher-risk of women from within the randomly-selected participants in figure B. For figures B and D, tenofovir was measured in plasma at months 1, 3, 6, 12, 18, and 24 in 198 randomly selected active arm participants. Ns represent the number at enrollment for visit attendance figures and the number with any tenofovir measured for medication adherence figures. Subgroups with ≤5 plasma samples in a month are excluded for that month. For all figures, follow-up is presented through 24 months due to limited numbers thereafter.