This developmental research grant award (R21) requests funds to explore the feasibility and impact of a public health system PrEP intervention in a recently emerging HIV epidemic in Athens Greece. The investigators propose a modeling approach using an Agent Based Model (ABM) that moves beyond basic pathogen and transmission patterns to dealing with complex social interactions, including overlapping social and sexual networks as well as implementation realities, like finite PrEP resources, delayed linkage to PrEP care and early PreP care retention based upon empirically collected data in Athens Greece.

This developmental research grant award (R21) requests funds to explore the feasibility and impact of a public health system PrEP intervention in a recently emerging HIV epidemic in Athens Greece. As HIV epidemics in most settings globally have plateaued or are in decline, emerging epidemics, re-emerging epidemics or outbreaks will become more common, particularly when social, political or other "shocks" that impact HIV prevention resources occur. One well characterized example is the recent epidemic among people who inject drugs (PWID) that started in Athens following austerity measures in 2010. While some success in limiting the epidemic within PWID has been observed, recent phylogenetic and surveillance analysis demonstrates that the HIV strain from this most recent PWID epidemic (CRF35_AD, CRF14_BG, subtypes A and B) has spilled over into MSM in 2013 (see Preliminary Studies).

In emerging epidemics, oral chemoprophylaxis is a commonly-used public health strategy to prevent infectious diseases in susceptible persons. For example, among US and European MSM, antibiotic prophylaxis for sex-partners in outbreaks of invasive meningococcal disease has limited emerging outbreaks. Similarly, pre-exposure prophylaxis (PrEP) has the potential to be used in emerging epidemics to prevent onwards HIV transmission. While this approach would seem intuitive, current conceptualizations of PrEP implementation narrowly use a clinical model focused on individualized intervention between health provider and client. This is problematic because the public health impact of PrEP may be limited due to lack of proper targeting. In fact, evidence from "real-world" PrEP use suggests that lower risk clients are accessing PrEP. As compared to resource intensive clinical trials or demonstration studies, careful modeling approaches can provide insight into who within a new HIV epidemic should be targeted for PrEP to prevent onward transmission as well as the strategies used to identify these individuals and link them to care. The investigators propose a modeling approach using an Agent Based Model (ABM) that moves beyond basic pathogen and transmission patterns to dealing with complex social interactions, including overlapping social and sexual networks as well as implementation realities, like finite PrEP resources, delayed linkage to PrEP care and early PreP care retention based upon empirically collected data in Athens Greece. Specifically the investigators aim to: 1) Characterize a bridging MSM network (n=308) by measuring individual-level risk factors, network-level connections, and HIV phylogenetic clusters; 2) Measure early PrEP cascade outcomes (HIV testing, PrEP linkage to care) of a sub-sample (n=50) of HIV uninfected MSM over the short term; and 3) Model the effects of this targeted public health PrEP intervention on HIV transmission in Athens. Agent-based models that account for empirical network structure are state-of-the-art in modeling HIV transmission and are flexible enough to address fundamental questions of who should receive PrEP and ultimately how a network-PrEP intervention can impact emerging/reemerging HIV epidemics.

Inclusion Criteria:

• identify as male,
• are between the ages of 18-39,
• report sex with a man (oral or anal sex) in the past 12 months,
• are willing/able to provide informed consent,
• are willing to provide biological samples,
• are Greek or English speakers and
• are able to lucidly respond to interview questions

Exclusion Criteria:

-