Impact of HIV self-testing for oral pre-exposure prophylaxis scale-up on drug resistance and HIV outcomes in western Kenya: a modelling study

Sarah N Cox,Linxuan Wu,Rachel Wittenauer,Samantha Clark,D Allen Roberts,Ifechukwu Benedict Nwogu,Olga Vitruk,Alexandra P Kuo,Cheryl Johnson,Muhammad S Jamil,Anita Sands,Robin Schaefer,Christine Kisia,Rachel Baggaley,Joanne D Stekler,Adam Akullian,Monisha Sharma
DOI: https://doi.org/10.1016/s2352-3018(23)00268-0
IF: 16.07
2024-01-31
The Lancet HIV
Abstract:Summary Background Community-based oral pre-exposure prophylaxis (PrEP) provision has the potential to expand PrEP coverage. HIV self-testing can facilitate PrEP community-based delivery but might have lower sensitivity than facility-based HIV testing, potentially leading to inappropriate PrEP use among people with HIV and subsequent development of drug resistance. We aimed to evaluate the impact of HIV self-testing use for PrEP scale-up. Methods We parameterised an agent-based network model, EMOD-HIV, to simulate generic tenofovir disoproxil fumarate and emtricitabine PrEP scale-up in western Kenya using four testing scenarios: provider-administered nucleic acid testing, provider-administered rapid diagnostic tests detecting antibodies, blood-based HIV self-testing, or oral fluid HIV self-testing. Scenarios were compared with a no PrEP counterfactual. Individuals aged 18–49 years with one or more heterosexual partners who screened HIV-negative were eligible for PrEP. We assessed the cost and health impact of rapid PrEP scale-up with high coverage over 20 years, and the budget impact over 5 years, using various HIV testing modalities. Findings PrEP coverage of 29% was projected to avert approximately 54% of HIV infections and 17% of HIV-related deaths among adults aged 18–49 years over 20 years; health impacts were similar across HIV testing modalities used to deliver PrEP. The percentage of HIV infections with PrEP-associated nucleoside reverse transcriptase inhibitor (NRTI) drug resistance was 0·6% (95% uncertainty intervals 0·4–0·9) in the blood HIV self-testing scenario and 0·8% (0·6–1·0) in the oral HIV self-testing scenario, compared with 0·3% (0·2–0·3) in the antibody rapid diagnostic testing scenario and 0·2% (0·1–0·2) in the nucleic acid testing scenario. Accounting for background NRTI resistance, we found similarly low proportions of drug resistance across scenarios. The budget impact of implementing PrEP using HIV self-testing and provider-administered rapid diagnostic tests were similar, while nucleic acid testing was approximately 50% more costly. Interpretation Scaling up PrEP using HIV self-testing has similar health impacts, costs, and low risk of drug resistance as provider-administered rapid diagnostic tests. Policy makers should consider leveraging HIV self-testing to expand PrEP access among those at HIV risk. Funding The Bill and Melinda Gates Foundation.
immunology,infectious diseases
What problem does this paper attempt to address?