Pre-exposure prophylaxis (PrEP) effectively reduces human immunodeficiency virus (HIV) transmission. We aimed to estimate the impact of different PrEP prioritization strategies among Black and Latino men who have sex with men (MSM) in the United States, populations most disproportionately affected by HIV. We developed an agent-based simulation to model the HIV epidemic among MSM. Individuals were assigned an HIV incidence risk index (HIRI-MSM) based on their sexual behavior. Prioritization strategies included PrEP use for individuals with HIRI-MSM ≥10 among all MSM, all Black MSM, young (≤25 years) Black MSM, Latino MSM, and young Latino MSM. We estimated the number needed to treat (NNT) to prevent one HIV infection, reductions in prevalence and incidence, and subsequent infections in non-PrEP users avoided under these strategies over 5 years (2016-2020). Young Black MSM eligible for PrEP had the lowest NNT (NNT = 10) followed by all Black MSM (NNT = 33) and young Latino MSM (NNT = 35). All Latino MSM and all MSM had NNT values of 63 and 70, respectively. Secondary infection reduction with PrEP was the highest among young Latino MSM (53.2%) followed by young Black MSM (37.8%). Targeting all MSM had the greatest reduction in prevalence (14.7% versus 2.9%-3.9% in other strategies) and incidence (49.4% versus 9.4%-13.9% in other groups). Using data representative of the United States MSM population, we found that a strategy of universal PrEP use by MSM was most effective in reducing HIV prevalence and incidence of MSM. Targeted use of PrEP by Black and Latino MSM, however, especially those ≤25 years, had the greatest impact on HIV prevention.
Keywords: agent-based model; human immunodeficiency virus (HIV); men who have sex with men (MSM); number needed to treat; pre-exposure prophylaxis (PrEP); simulation.