Abstract
BACKGROUND: Oral HIV pre-exposure prophylaxis (PrEP) can effectively reduce HIV incidence. A 2020-21 demonstration project assessed the feasibility and health outcomes of offering oral PrEP to men who have sex with men (MSM) in Cotonou, Benin. We evaluated the epidemiological impact and cost-effectiveness of this project and the potential scale-up of oral HIV PrEP for MSM in Cotonou. METHODS: We calibrated an HIV transmission-dynamic model structured by age and risk within a Bayesian framework to MSM-specific HIV prevalence and treatment data, parameterised with project behavioural and cost (including PrEP drug, implementation, and HIV care costs) data. We estimated the impact and cost-effectiveness of the 2020-21 Cotonou demonstration project (PrEP coverage, 5-10% of all MSM who are not living with HIV in Grand Cotonou; and adherence, 13-21% taking at least four of seven required doses [ie, at least four doses per week for daily users and at least four of seven expected doses given reported sexual activity for on-demand users]) and of its potential scale-up over 5 years (from 2022 to 2027), reaching 30% coverage of MSM in Grand Cotonou and with demonstration project adherence levels. We additionally modelled ideal PrEP adherence (100% taking at least four of seven required doses). We estimated the percentage of cumulative new HIV infections averted among participating MSM over 1 year and among all MSM in Grand Cotonou and their female partners over 20 years, and cost-effectiveness as cost per disability-adjusted life-year (DALY) averted over 20 years. Costs and DALYs were discounted 3% annually. FINDINGS: We found that the demonstration project averted an estimated 21·5% (95% uncertainty interval 16·6 to 26·2) of HIV infections among participants over 1 year. With ideal adherence, cases that would be averted increased to 95·2% (90·8 to 98·8). A 5-year PrEP scale-up could avert 3·2% (1·6 to 4·8) of HIV infections among all MSM and female partners over 20 years, at US$388 (36 to 2792) per DALY averted. With ideal adherence, this decreased to -$28 (-126 to 589) per DALY averted. INTERPRETATION: Low adherence to PrEP restricted the impact of the demonstration project. At 30% coverage among MSM by 2027, PrEP scale-up would be cost-effective at a $1225 threshold with 86·6% probability, and it could be more cost-effective if high adherence could be reached without substantially increasing costs. FUNDING: Canadian Institutes of Health Research and US National Institutes of Health. TRANSLATION: For the French translation of the abstract see Supplementary Materials section.