Andrew E Grulich, Rebecca Guy, Janaki Amin, Fengyi Jin, Christine Selvey, Jo Holden, Heather-Marie A Schmidt, Iryna Zablotska, Karen Price, Bill Whittaker, Kerry Chant, Craig Cooper, Scott McGill, Barbara Telfer, Barbara Yeung, Gesalit Levitt, Erin E Ogilvie, Nila J Dharan, Mohamed A Hammoud, Stefanie Vaccher, Lucy Watchirs-Smith, Anna McNulty, David J Smith, Debra M Allen, David Baker, Mark Bloch, Rohan I Bopage, Katherine Brown, Andrew Carr, Christopher J Carmody, Kym L Collins, Robert Finlayson, Rosalind Foster, Eva Y Jackson, David A Lewis, Josephine Lusk, Catherine C O'Connor, Nathan Ryder, Emanuel Vlahakis, Phillip Read, David A Cooper, Don Smith, Catriona Ooi, Sarah Martin, Tuck Meng Soo, David Templeton, David Townson, Brad Forssman, Nick Doong, Gai Han Thai, Ben Anderson, Hugh MacLeod, and Nicolas Parkhill
Summary Background HIV pre-exposure prophylaxis (PrEP) is highly effective in men who have sex with men (MSM) at the individual level, but data on population-level impact are lacking. We examined whether rapid, targeted, and high-coverage roll-out of PrEP in an MSM epidemic would reduce HIV incidence in the cohort prescribed PrEP and state-wide in Australia's most populous state, New South Wales. Methods The Expanded PrEP Implementation in Communities–New South Wales (EPIC-NSW) study is an implementation cohort study of daily co-formulated tenofovir disoproxil fumarate and emtricitabine as HIV PrEP. We recruited high-risk gay men in a New South Wales-wide network of 21 clinics. We report protocol-specified co-primary outcomes at 12 months after recruitment of the first 3700 participants: within-cohort HIV incidence; and change in population HIV diagnoses in New South Wales between the 12-month periods before and after PrEP roll-out. The study is registered with ClinicalTrials.gov , number NCT02870790 . Findings We recruited 3700 participants in the 8 months between March 1, 2016, and Oct 31, 2016. 3676 (99%) were men, 3534 (96%) identified as gay, and 149 (4%) as bisexual. Median age was 36 years (IQR 30–45 years). Overall, 3069 (83%) participants attended a visit at 12 months or later. Over 4100 person-years, two men became infected with HIV (incidence 0·048 per 100 person-years, 95% CI 0·012–0·195). Both had been non-adherent to PrEP. HIV diagnoses in MSM in New South Wales declined from 295 in the 12 months before PrEP roll-out to 221 in the 12 months after (relative risk reduction [RRR] 25·1%, 95% CI 10·5–37·4). There was a decline both in recent HIV infections (from 149 to 102, RRR 31·5%, 95% CI 11·3 to 47·3) and in other HIV diagnoses (from 146 to 119, RRR 18·5%, 95% CI −4·5 to 36·6). Interpretation PrEP implementation was associated with a rapid decline in HIV diagnoses in the state of New South Wales, which was greatest for recent infections. As part of a combination prevention approach, rapid, targeted, high-coverage PrEP implementation is effective to reduce new HIV infections at the population level. Funding New South Wales Ministry of Health, Gilead Sciences.