“On demand” PrEP dosing in IPERGAY: 86% reduced risk of HIV, no transmissions with active drug use
24 March 2015. Related: Conference reports, HIV prevention and transmission, CROI 22 (Retrovirus) 2015.
Simon Collins, HIV i-Base
Just as eagerly awaited as the PROUD study, the second oral presentation on PrEP at CROI 2015, with results that are equally important, were presented by Jean Michel Molina from the University of Paris Diderot. 
The French/Canadian approach to PrEP was to study a new “on demand:” dosing strategy which was individualised to the times people had sex. So while daily PrEP has theoretical benefits for optimal drug levels, maintaining continuous prophylaxis protection, this may involve greater drug exposure than many people need.
“On demand” dosing
A more flexible dosing option, if proved to have similar efficacy, would relax the need for daily adherence, and in the short-term (until generic PrEP is available) has the potential to reduce the cost. This would have a similar impact on cost effectiveness calculations, especially for people who have lower HIV risk.
The dosing regimen in IPERGAY involved taking a double dose of tenofovir/FTC (i.e. two pills) before having sex. The window period for this pre-exposure dose was from 24 hour to two hours before sex. Single doses of PrEP were then taken every day for each day that the participant continued having sex without a condom. A single dose was taken on the day after the last risk exposure. For someone who was only at risk from one exposure a month, this would be considerably different compared to daily PrEP. For a person who was sexually active once a week, their dosing would approach daily PrEP.
The strength of the IPERGAY study was the potential to discover a cheaper and more flexible way to use PrEP. The double-dose strategy was developed from earlier macaque research.  However, the choice of using a randomised study design that included a placebo, coming after the effectiveness of PrEP had already been established, might have been a factor that limited enrolment (less than half the predicted study numbers were enrolled when the DSMB-recommended changes to IPERGAY took place in October 2014, following early efficacy findings in the PROUD study). [3, 4]
Enrolment and baseline characteristics
IPERGAY screened 445 participants and randomised 414 to either active tenofovir/FTC (n=206) or placebo (n=208). The modified Intent-To-Treat (mITT) analysis included 199 and 201 participants respectively, based on seven people in each group not receiving study drug (due to withdrawn consent, lost to follow up or they were found to be HIV positive). The final analysis included 176 (88%) and 177 (88%) participants, respectively.
Baseline characteristics were well matched between groups and included: median age 35 years (IQR: 29-43), >90% white, ~90% completed secondary education, 85% employed, ~80% single, 20% circumcised, 46% psychoactive drug use (in previous year). Participants in each group had median 8 (IQR: 5-15) partners in the previous 6 months. The only slight differences were that PEP had been used by 28% of the active vs 37% of the placebo group and recent STI were in 22% vs 29% of participants, respectively.
Over a mean follow-up of 13 months, there were 16 cases of new HIV transmission: 2 in the active arm (incidence 0.94 per 100 PY) and 14 in the placebo group (incidence 6.6 per 100 PY). This resulted in an 86% reduction in the incidence of HIV (95% CI: 40 to 99), p=0.002. The number needed to treat (NNT) for one year to prevent an infection was 18.
Importantly, both transmissions in the active arm had not been using PrEP for many months and were both still sexually active.
Adherence and safety
Adherence reported by pill count was similar with median 16 pills/month in each group (IQR: 12-24 in active and 10-23 in placebo), p=0.84. PEP was used by 48 participants (12%) and was similarly balance (13% vs 11%), p=0.73.
Adherence patterns by month looked heterogeneous for individuals in both arms – with some months reporting low use and some months almost daily use, reflecting the potential for real life on demand use. When participant recall was reported, PrEP was used correctly for 43% (range 35-51) of times of last sexual experience (1212 times assessed in 319 participants). Suboptimal PrEP was reported 25% of the time and no PrEP was reported 25% of the time.
Tolerability was good with similar reports in active and placebo groups. For example, any side effect was reported by 92% vs 89%, with serious side effects in 9% vs 8%.
Side effects that were drug related were mainly nausea/vomiting or abdominal pain (13% vs 6%, p=0.013) but were generally mild. There was one safety report of deep vein thrombosis from a drug-drug interaction between tenofovir/FTC and dabigatran that resolved on discontinuation.
The only significant differences in laboratory abnormalities was Grade 1 creatinine increases in 14% vs 7% of active vs placebo group, p=0.042.
i-Base has also produced a nontechnical Q&A resource on the PrEP results from CROI. 
Both IPERGAY and PROUD were notable not just for reporting very similar efficacy results – 86% reductions in HIV transmission in arms receiving active PrEP compared to control groups (with slightly different estimated confidence intervals) – but for reporting a high probability that none of the few infections in people randomised to the active PrEP arms were likely to be on PrEP at the time of their infections.
Both studies provided convincing results that dampened concerns about risk compensation (the idea the perceived protection from PrEP would lead people to change their level of behaviour risk).
As with other PrEP studies, it appears likely that the increased focus on HIV and sexual heath that comes from engaging with a PrEP programme may in fact reduce their overall risk behaviour, and that this connection might have an additional prevention role in itself.
Both studies resulted in the lowest NNT of any PrEP studies so far: 13 in PROUD and 18 in IPERGAY.
Although efficacy was similar in both studies, most participants in IPERGAY were routinely using four or more doses a week. This correlates to the modelling data from iPrEX that suggests that four doses a week provided >96% protection for men.
Additional follow-up is needed to comment on whether the double-dose is sufficient when used much less frequently.
Additional follow up is also needed for the safety concern from routinely using a double-dose over a longer time in people who are using this strategy on a weekly basis.
- Molina JM et al. On demand PrEP with oral TDF-FTC in MSM: results of the ANRS Ipergay trial. CROI 2015. 2015 Conference on Retroviruses and Opportunistic Infections (CROI 2015), 23-26 February 2015, Seattle, Washington, USA. Oral late breaker abstract 23LB.
- Garcia-Lerma et al. Intermittent prophylaxis with oral Truvada protects macaques from rectal SHIV infection. Science Trans Med, 2010, Jan 13;2(14):14ra4. doi: 10.1126/scitranslmed.3000391.
- Collins S. IPERGAY PrEP study shows early efficacy in protecting gay men from HIV: all participants to switch to active drug. (29 October 2014).
- ANRS press release. A significant breakthrough in the fight against HIV/AIDS: A drug taken at the time of sexual intercourse effectively reduces the risk of infection. (29 October 2014).
- HIV i-Base. Q&A on results from the PROUD study (February 2015).