Effect of 6-Month HIV Preexposure Prophylaxis Dispensing With Interim Self-testing on Preexposure Prophylaxis Continuation at 12 Months: A Randomized Noninferiority Trial

• Published in: JAMA network open 2023-06, Vol.6 (6), p.e2318590
• Main Authors: Ortblad, Katrina F, Bardon, Ashley R, Mogere, Peter, Kiptinness, Catherine, Gakuo, Stephen, Mbaire, Sarah, Thomas, Katherine K, Mugo, Nelly R, Baeten, Jared M, Ngure, Kenneth
• Format: Article
• Language: English
• Subjects: Anti-HIV Agents - therapeutic use; Disease prevention; Female; Global Health; HIV; HIV Infections - diagnosis; HIV Infections - drug therapy; HIV Infections - prevention & control; Humans; Kenya; Male; Medical tests; Online Only; Original Investigation; Self-Testing
• ISSN: 2574-3805; 2574-3805
• DOI: 10.1001/jamanetworkopen.2023.18590

Daily oral HIV preexposure prophylaxis (PrEP) delivery requires quarterly clinic visits for HIV testing and drug refilling that are costly to health systems and clients. To evaluate whether 6-month PrEP dispensing supported with interim HIV self-testing (HIVST) results in noninferior PrEP continuation outcomes at 12 months compared with standard quarterly clinic visits. This randomized noninferiority trial was conducted from May 2018 to May 2021 with 12 months of follow-up among PrEP clients aged 18 years or older who were returning for their first refill at a research clinic in Kiambu County, Kenya. Participants were randomized 2:1 to (1) 6-month PrEP dispensing with semiannual clinic visits and interim HIVST at 3 months or (2) standard-of-care (SOC) PrEP delivery with 3-month dispensing, quarterly clinic visits, and clinic-based HIV testing. Prespecified 12-month outcomes included recent HIV testing (any in past 6 months), PrEP refilling, and PrEP adherence (detectable tenofovir-diphosphate concentrations in dried blood spots). Binomial regression models were used to estimate risk differences (RDs), and a 1-sided 95% CI lower bound (LB) of -10% or greater was interpreted as noninferior. A total of 495 participants were enrolled, with 329 enrolled in the intervention group and 166 enrolled in the SOC group; 330 (66.7%) were women, 295 (59.6%) were in serodifferent relationships, and the median (IQR) age was 33 (27-40) years. At 12 months, 241 individuals in the intervention group (73.3%) and 120 in the SOC group (72.3%) returned to clinic. In the intervention group, recent HIV testing was noninferior (230 individuals [69.9%]) compared with the SOC group (116 [69.9%]; RD, -0.33%, 95% CI LB, -7.44%). PrEP refilling in the intervention group (196 [59.6%]) was inconclusive compared with the SOC group (104 [62.7%]; RD, -3.25%; 95% CI LB, -10.84%), and PrEP adherence was noninferior in the intervention group (151 [45.9%]) compared with the SOC group (70 [42.2%]; RD, 4.96%; 95% CI LB, -2.46%). No HIV seroconversions were observed over the follow-up period. In this analysis of secondary trial end points at 1 year, semiannual PrEP dispensing with interim HIVST resulted in noninferior recent HIV testing and PrEP adherence compared with SOC quarterly PrEP dispensing. This novel model has the potential to optimize PrEP delivery. ClinicalTrials.gov Identifier: NCT03593629.