Influence of serosorting and intervention-mediated changes in serosorting on the population-level HIV transmission impact of pre-exposure prophylaxis among men who have sex with men: A mathematical modelling study

摘要: ABSTRACT Background HIV pre-exposure prophylaxis (PrEP) may change serosorting patterns. We examined the influence of on population-level transmission impact PrEP, and how could if PrEP users stopped serosorting. Methods developed a compartmental model parameterized with bio-behavioural surveillance data among men who have sex in Canada. separately fit without (random partner-selection proportional to availability by HIV-status (sero-proportionate)), reproduced stable epidemics (2013-2018) HIV-prevalence 10.3%-24.8%, undiagnosed fraction 4.9%-15.8%, treatment coverage 82.5%-88.4%. simulated PrEP-intervention reaching year-1 compared absolute difference relative HIV-incidence reduction 10-year post-intervention (PrEP-impact) between: models vs. sero-proportionate mixing; scenarios which continued sensitivity results PrEP-effectiveness (44%-99%) (10%-50%). Findings Models predicted larger PrEP-impact mixing under all assumptions (median (inter-quartile-range): 8.1%(5.5%-11.6%)). users” stopping reduced when serosorting: reductions were minimal (2.1%(1.4%-3.4%)) high (86%-99%); however, be considerable (10.9%(8.2%-14.1%)) low effectiveness (44%) (30%-50%). Interpretation assuming underestimate due PrEP. PrEP-mediated changes lead programmatically-important (e.g. poor adherence/retention). Our findings suggest need monitor sexual patterns inform implementation evaluation. Funding Canadian Institutes Health Research RESEARCH IN CONTEXT Evidence before this study searched PubMed for full-text journal articles published between Jan 1, 2010, Dec 31, 2017, using MeSH terms “pre-exposure (PrEP)” “homosexuality, male” key words (“pre-exposure prophylaxis” or “preexposure “PrEP”) (“men men” “MSM”) titles abstracts. Search (520 records) reviewed identify publications cost-effectiveness high-income settings. identified total 18 modelling studies (MSM) four based same minor variations (thus only most recent one was included). Among 15 unique impact, three included A nine assessed individual-level behaviour those its Specifically, increases number partners condom use. Most that realistic partner decreases use would not fully offset, but weaken, PrEP”s reducing transmission. did any estimated at population-level, what happen incidence Added value used mathematical estimate MSM. found higher serosorting, comparable mixing. Under (while other continue serosort themselves) we serosort. The magnitude high; programmatically-meaningful context (e.g., adherence retention) coverage. To our knowledge, is first directly examine underlying mechanism. Implications available evidence do consider baseline MSM potentially impact. In addition monitoring behavioural such as use, highlight their over time design evaluation implementation.