International AIDS Conference 2015: PrEP Demo Project

Contributor: Dr. Oliver Bacon

Dr. Oliver Bacon
Dr. Oliver Bacon

The Vancouver IAS conference shows that PrEP research using emtricitabine/tenofovir (FTC/TDF) or TDF alone has moved from the proof-of-concept stage (does PrEP work? Is it safe?) to the implementation science stage (how to make PrEP work for patients). To summarize some of the main findings:

  • In adults and adolescents, PrEP does not seem to facilitate risky behavior or STI’s, but is a response to risky behavior
  • Adherence to daily PrEP can be very high in US MSM although was suboptimal in specific populations, like black MSM, and particularly among black adolescent MSM
  • Adherence to daily PrEP in adolescents was higher during monthly visits compared to when they were seen less frequently (quarterly)
  • Intermittent PrEP dosing is feasible in a variety of populations; however, overall, daily PrEP leads to better coverage of sex acts than intermittent dosing
  • The Thai participants in the ADAPT trial showed equivalent coverage (# of sex acts covered by PrEP), adherence, and drug levels between those who took daily PrEP and those who took intermittent PrEP

PrEP Before IAS 2015

To briefly review the state of PrEP science before IAS 2015: seven randomized trials of PrEP, using either daily FTC/TDF (iPrEx, Partners Prep, TDF-2, Fem-PrEP, VOICE, PROUD), or daily TDF alone (Partners PrEP, Bangkok Tenofovir, VOICE) showed PrEP efficacy in reducing HIV infection risk to range from 0%, if adherence was low (Fem-PrEP, VOICE), to 86%, if adherence was high (PROUD). PrEP was also well-tolerated and largely nontoxic. Pharmacologic substudies further showed that taking 4 or more doses of FTC/TDF per week conferred 96% risk reduction when used for anal intercourse, based on intracellular drug levels. In addition, the IPERGAY study of non-daily FTC/TDF showed an 86% reduction in HIV risk vs placebo when PrEP was used by MSM intermittently, around the time of sex, although the frequency of intercourse was high enough that participants used a median of 16 pills/month, or 4 pills/week, raising questions about whether intermittent PrEP would have been as effective when used less often.


New Results from IAS 2015

Demonstration Studies of Daily PrEP in Adults and Adolescents

Albert Liu from the San Francisco Department of Public Health presented follow-up adherence, risk behavior, and HIV incidence results from the US PrEP Demonstration Project, an open-label 48 week study of FTC/TDF in 557 MSM and MTF transwomen at STD clinics in Miami and San Francisco, and a comprehensive health center in Washington DC. Previous results published by Stephanie Cohen had shown high PrEP uptake (61%) when offered. After 48 weeks of follow-up, adherence was very high: over 80% of dried bloodspot (DBS) samples—a biological measure of adherence– at each quarterly visit correlated with taking 4+ pills/week, and 63% of sampled participants had DBS levels consistent with taking 4+ pills/week at all visits. Adherence was lower among participants who were Black or in Miami; it was higher among those with stable housing or who reported higher risk (≥2 condomless anal sex partners in the past 3 months). Prevalence of condomless receptive anal intercourse was high before starting PrEP (66%), and remained stable during follow-up. 26% of participants had an STI (gonorrhea, Chlamydia, or early syphilis) at study entry; incidence of any STI was stable throughout follow-up, at 90/100 person-years.

Using RNA-based testing at screening allowed the detection of three persons with acute HIV infection (antibody and 4th generation antigen-antibody negative) before starting PrEP. Two participants were infected with HIV during follow-up, for an incidence of 0.43/100 person years (95% CI 0.05-1.54); both had stopped taking PrEP at least 4 weeks before testing HIV-positive.

Sybil Hosek reported data from ATN110, an open-label demonstration project of FTC/TDF in 200 adolescents (age 18-22) enrolled in 12 cities across the US (results from a companion study in 15-17 year olds, ATN113, are expected in 2016). Follow-up visits were monthly through week 12, then quarterly though 48 weeks. Like their older counterparts in PrEP Demo, participants in ATN110 reported high risk at baseline: condomless sex (81%), condomless receptive anal sex with their last partner (58%), and high rates of STIs (22%) at screening. Over 48 weeks of follow-up, STI rates remained stable. Adherence, as measured by percentage of participants with DBS levels equivalent to taking 4+ pills/week, dropped substantially over the course of the study, from 56% at week 4, to 34% at week 24. This adherence drop correlated with the decrease in visit frequency from monthly to quarterly. Adherence was also associated with race and higher risk sex. White and Latino participants had DBS levels consistently above 4+ doses/week, while Black participants had DBS levels consistently below 4+ doses/week. Those who reported condomless sex had consistently higher DBS levels of tenofovir than those who didn’t (p=0.005), suggesting those at higher risk were more likely to take their pills. The opportunity to access PrEP was a helpful probe for finding adolescent HIV cases: 11 established and 2 acute cases of HIV were detected during screening. Four participants seroconverted during follow-up, for an incidence of 3.29/100 person-years. None of those individuals who seroconverted had detectable levels of tenofovir by DBS sampling.


Studies of Intermittent PrEP Dosing

 Although daily PrEP offers the most comprehensive coverage of sex acts, accounting for unplanned as well as planned events, many potential PrEP users have planned sex events. The IPERGAY trial testing “on demand” PrEP in a group of MSM from France and Canada showed that they are able to take PrEP in a targeted fashion before and after sexual intercourse. Bob Grant, Timothy Holz, and Sharon Mannheimer presented findings from ADAPT (HPTN 067), which compared daily PrEP vs. two intermittent regimens in three populations: heterosexual women in Cape Town, South Africa; MSM and TGW in Bangkok, Thailand; and MSM and TGW in Harlem, NY. This was not an efficacy study: key outcomes included coverage of sex events (measured by weekly interviews), side effects, adherence (measured by use of the Wisepill® device), drug levels, safety and acceptability, and HIV infections.

Although data from each city were analyzed separately, all three sites followed the same protocol:

  • 6 weeks of daily FTC/TDF, followed by randomized assignment to one of three arms for 24 weeks:
  • Continued daily dosing vs.
  • Time-driven dosing (1 tablet 2x/week plus one tablet after sex) vs.
  • Event-driven dosing (1 tablet pre-sex and one tablet post-sex; no more than 2 tablets daily or 7 tablets/week)
  • Coverage was defined as ≥1 tablet within 4 days before sex and ≥1 tablet within 24h after sex.
  • Drug exposure was defined as having TDF-DP levels in PBMCs >9.1fmol/M cells, which correlates with at least 2 tablets/week and a risk reduction of approximately 84% in previous pharmacologic studies.

Overall, results in Bangkok were more supportive of intermittent dosing compared to Cape Town and Harlem, and time-driven rather than event-driven dosing seemed to provide higher coverage and yield higher adherence.

In both Cape Town and Harlem, coverage (use of PrEP around the time of sex) was significantly higher in the daily arm (75% and 66% respectively) than the time or event-driven arms, and there was no difference in coverage between the time or event driven arms (56% and 52% in Cape Town; 47% and 52% in Harlem). In Bangkok, coverage was no different across all arms (85% daily; 84% time-driven; 74% event-driven), although the difference in coverage between daily and event-driven dosing was significant. In all three sites, most missed doses were those that were to be taken post-sex.

Adherence was highest with daily dosing in all three cities, with some minor variations: in Cape Town, adherence fell significantly as dosing varied from daily to time-driven to event driven; in Harlem, adherence was significantly higher in daily vs. either intermittent group, but there was no difference between intermittent groups; in Bangkok, adherence was no different in time-driven vs. daily dosing, but adherence in event-driven dosing was lower than either of the other two groups.

Drug level data were available from Cape Town and Bangkok. In Bangkok, approximately 90% of participants in all three arms had detectable TDF-DP consistent with 2-3 doses/week at both 10 and 30 weeks, with no significant difference among arms. In Cape Town, detectable drug levels were significantly higher in the daily dosing arm (81% at week 10, 66% at week 30) than in either intermittent arm at both timepoints.

During the 6-week daily pre-randomization phase, there were 2 infections in Cape Town (incidence 8.9/100PY), 1 in Harlem (4.4 100PY), and 2 in Bangkok (8.7/100PY). All had no to low-detectable drug levels at the seroconversion or prior visit. During the randomized phase, there were 5 infections in Cape Town (2 time-driven, 2 event driven, 1 daily) for an incidence of 5.4/100PY. There was 1 infection in Harlem (daily arm) for an incidence of 1.2/100PY. There were no infections in the randomized phase in Bangkok. In all post randomization infections, drug levels were low to negligible.

Al Liu’s Presentation on the PrEP Demo Project at IAS 2015

Credit: The International Aids Conference 2015
Credit: IAS 2015
back to top