The Fenway is a community health centre with a special focus on sexual and gender minorities. Its director Dr Ken Mayer told the conference that while randomised controlled trials proved comprehensively that PrEP worked, more surveillance studies were needed to show what actually happens to HIV infections when PrEP is offered to a clinic population.
The study cohort was everyone who had attended the clinic between the start of 2012 and the end of 2017, and who had had at least two HIV tests (to show they were initially HIV negative) during that time. People who received a prescription for PrEP at any time during that period were defined as “PrEP users”, even if they only took it for three months.
During this time 16,128 patients were tested for HIV more than once. Just over 60% were white with reasonably equal proportions of the others being Asian, Hispanic or African American. Fifteen per cent were known to be female though in 24% of clinic users, gender was not recorded.
Of the 16,128 patients, 3965 (just under 25%) initiated PrEP. PrEP users were much less likely to be women (1.4%) and a little less likely to be black (6% of PrEP users vs 8% non-users) and under 25 (11% vs 13%).
In the whole clinic population, 163 (1.34%) of those never prescribed PrEP became infected with HIV and 17 (0.43%) of PrEP users. This represents a 68% reduction of infection in people ever prescribed PrEP.
Of the 17 PrEP users who became infected with HIV (all gay/bi men), nine (53%) had a bacterial STI infection at the same time and one had acute hepatitis C. Seven admitted to chemsex, and four had HIV with some degree of drug resistance including one with tenofovir resistance (K65R).
However, 12 of the 17 PrEP recipients did not in fact have a PrEP prescription at the time they were infected, having discontinued it at least a month before. Of these 12, four said they discontinued because of difficulty with adherence/scheduling, four had insurance problems, three said they felt at low risk and one said he stopped because of PrEP stigma.
So the infection rate among the current PrEP users was 0.13% – indicating an effectiveness of 90%.
Of these five, three almost certainly had acute HIV infection at the time they started PrEP with one being found to have a positive HIV RNA test on the day it was started. A fourth admitted to having taken a month’s break from PrEP at the time they were infected.
The fifth said he “only missed a day now and then” of PrEP, and could possibly be a true PrEP failure. He caught primary syphilis at the same time as HIV and had HIV resistant to emtricitabine, which might possibly indicate a breakthrough infection. This would equate to a population-level effectiveness of 98%.
However, no drug levels measurements were taken from this patient, his pharmacy record suggested some lapses in adherence and he admitted to not always taking his PrEP with him on business trips, so this case cannot be counted among the handful of cases of documented PrEP failure.