ABSTRACT HIV incidence remains stable among men who have sex with men (MSM) in the United States, and increasing in some groups, despite highly effective antiretroviral-based prevention tools such as preexposure prophylaxis (PrEP). Open-label PrEP demonstration projects have yielded strong evidence for the “real-world” prevention benefits of PrEP, with high adherence in these settings leading to virtually no incident infections. Yet critical questions remain about the future epidemiological impact of PrEP among MSM at the population level. Our preliminary work used mathematical modeling to estimate the impact of CDC’s clinical practice guidelines for PrEP, with policy-oriented outcomes that supported comprehensive HIV prevention planning efforts. Because of limitations in model parameter data, however, these models required assumptions about the generalizability of sexual network structures, profiles of PrEP and ART adherence, and stability in other prevention strategies when simulating specific interventions. This proposed project will collect new data among a sample of MSM from 15 high-burden US cities and use those data for multi-level (national and city-specific) models of HIV transmission. These modeling activities will address the limitations of the earlier research by broadening its epidemiological and geographic scope, and provide new insight into novel targeting and adherence strategies informed by the structure and dynamics of MSM sexual partnerships. Our specific aims are to: 1) Conduct an internet-based survey (ART-Net) of MSM living in 15 high HIV burden metropolitan areas to collect sexual network structure and use of and adherence to PrEP and ART as prevention; 2) Extend our network-based HIV microsimulation model to incorporate new behavioral and clinical parameters from the ART-Net data, and use Bayesian statistical methods for parameter validation and model calibration to adjust ART-Net data to account for the internet-based sampling methods by triangulating with key secondary data sources (external behavioral and clinical data; PrEP demonstration projects; HIV surveillance information); and 3) Model the HIV transmission dynamics and implementation of combination antiretroviral prevention among MSM nationally and by city using the validated parameters, with the goal of estimating the future epidemiological impact (infections averted) and efficiency (number needed to treat) of scaling-up these tools according to different strategies that are specific to the local epidemiological context of HIV transmission among MSM in each city. Completion of these aims will provide critically needed empirical data and software tools for HIV modeling researchers, and robust model-based estimates of ART intervention impact geared towards supporting the development of locally tailored HIV planning activities.

Project Narrative The long-term public health impact of antiretroviral-based HIV prevention tools such as preexposure prophylaxis among men who have sex with men (MSM) in the United States will depend on local epidemiological context, driven by heterogeneity in the structure of sexual partnership networks in which these tools are implemented. This project will collect new partnership network data along with PrEP and ART uptake and adherence data among MSM for use in network-based modeling of ART-based prevention strategies among MSM. This modeling work will provide highly needed outcomes on intervention impact and efficiency to support comprehensive HIV prevention planning nationally and in 15 high-prevalence cities in the United States.