New data from PrEP study shed light on adherence, bone mineral loss and resistance

Bob Grant, lead investigator of iPrEx. Photo by Gus Cairns/
Gus Cairns
Published: 02 March 2011

Near-perfect adherence to oral pre-exposure prophylaxis – taking HIV drugs to prevent HIV – may be achievable in the right settings, the Eighteenth Conference on Retroviruses and Opportunistic Infections heard yesterday.

Participants from the US sites of the international iPrEx study of tenofovir/FTC (Truvada) pre-exposure prophylaxis (PrEP) in men who have sex with men and transsexual women had near-perfect adherence, compared with 50% adherence from other sites, new data presented at the conference shows.

Analysis also found that adherence in men who had the highest risk of acquiring HIV, by having unprotected receptive anal sex, was, at 76%, far higher than those at lower risk, so participants were tempering their pill-taking to their perceived risk.

Another substudy has found that taking Truvada resulted in a small but significant loss in bone mineral density in participants. But it also found that participants’ bone mineral density at the start of the study was considerably lower than would have been expected in men their age.

Resistance tests uncovered no drug resistance in men who seroconverted (became infected with HIV) on the trial, but did find that one participant who took placebo had a small amount of virus with the K65R resistance mutation to tenofovir.

At a discussion forum on iPrEx and on the CAPRISA 004 microbicide trial, lead investigator Bob Grant announced that the US Food and Drug Administration had agreed that the iPrEx trial findings were sufficient for the FDA to move ahead and consider changing the indication for Truvada to include using it to prevent HIV.

PrEP, as a result, might be approved in the US by the end of this year.

Updated trial results

In the paper published on the trial findings in the New England Journal of Medicine last November, data were given up to May 2010. Bob Grant presented final figures for the trial, whose last participants left the trial in August 2010.

The final tally was that 130 HIV infections were seen in the 2499 men taking part in the trial, 48 of those taking Truvada and 82 on placebo, a rate of 2.6% a year. There were also 10 infections in men who had acute HIV symptoms at the time they enrolled, two of whom appear to have acquired resistance to FTC, and six infections in the three months immediately following the trial, four of them in men who had taken Truvada.

This means that the final efficacy figure in the ‘modified intent to treat’ (MITT) analysis, which excluded the men who had HIV at the start of the study and ignores factors like adherence and sexual risk, was 42%.

Efficacy was greater in men over 25 (56%), in men who reported greater than 90% adherence (68%), and, for reasons that are unclear, in the relatively small number of men who were circumcised (76%).


One of the surprises of the original trial was that while mean reported adherence was 95%, a study of drug levels found that only 50% of a random sample of men who did not acquire HIV had detectable drug in their white blood cells. Only 9% of those who seroconverted had any drug in their cells either. Drug can be detected in cells for several weeks and this shows, in the words of Bob Grant, that “people were either taking it daily or not at all”.

A new analysis of drug levels in 179 participants by Peter Anderson of the University of Colorado confirmed that only 50% had detectable tenofovir in their cells and 62% detectable FTC. However 97% of the 227 participants from the two sites in the USA, in San Francisco and Boston, had detectable drug, showing that adherence was near-perfect in this more treatment-literate group. It also confirmed that adherence was better in men over 25 (73%) than men under 25 (44%).

Adherence analysis also showed that participants tempered their adherence according to their level of risk; 76% of those who had had unprotected receptive anal sex in the previous twelve weeks had significant drug levels versus 36% of those who had not and 25% of those who had had no sex at all.

Rivet Amico of the University of Connecticut presented an analysis of how self-reporting and pill counts correlated with actual adherence in the study. There were four measures of adherence taken during the study: self-reports by interview and by computer-assisted interview, pill count of the number of pills dispensed minus those returned, and the medication possession ratio (MPR), which was a measure, based on refill rates, of how many pills the participants had relative to the number that would last exactly till the next visit (thus an MPR of 1.25 meant they had 25% more pills than the minimum needed).

MPR was the most accurate guide to actual adherence. Only 68% of those claiming 100% adherence by self-report were actually adherent; only 62% of those claiming 100% adherence by computer-assisted interview; and only 59% of those assessed as having 100% adherence by pill count. But 75% of those assessed as having 100% adherence by MPR actually were so. The overall ‘positive predictive value’ of a claim of 100% adherence was 69%, meaning that 31% of those assessed as having 100% adherence did not. The ‘negative predictive value’ of a claim of less than 100% adherence was 95%, however, meaning that self-reports of poor adherence could be relied on.

Bone mineral density

One concern in trials using tenofovir as post-exposure prophylaxis has been that all antiretrovirals seem to cause a transient loss in bone mineral density, but that this seems to be larger and in some studies persistent in people taking tenofovir.

A study of 2045 iPrEx participants using DEXA scans (Mulligan) found that bone mineral density (BMD) scores at trial recruitment were already lower than would have been expected in men of the same age. Twelve per cent of participants had a ‘Z score’ for BMD in their spine of more than minus two, meaning that they fell into the lowest 5% of BMD values for an average population. BMD scores in the hip were also lower than expected.

During the trial BMD declined approximately 0.6% further in the spine (though not in the hip) in people taking Truvada but did not decline in subjects taking placebo, during the first six months. In the minority of subjects so far measured out to 18 months it declined by 1%. Nine per cent of subjects on Truvada had a loss of more than 5% in their spinal BMD compared with 4% on placebo.

This 1% average decline compares with an average 2 to 4% decline in patients taking tenofovir for treatment. There appeared to be no clinical effects, but at present we only have data on most subjects for the first six months, so it is not known if this bone loss is progressive or will stabilise.  

Another study of BMD was presented from a different PrEP study: the safety trial of tenofovir PrEP in 400 US gay men which ended in July 2010 (Liu). This substudy only looked at the 200 participants from San Francisco. It found a loss of about 2% BMD in the neck of the hip bone over two years, though not in the spine.

This study also found lower-than-expected BMD in study entrants. It found than men using amphetamines were six times more likely to report low BMD at baseline and users of poppers 4.5 times as likely; conversely men who took vitamin D, calcium or multivitamins were 70% less likely to report low BMD.   


Grant R et al. Pre-exposure chemoprophylaxis for prevention of HIV among trans-women and MSM: iPrEx study. 18th Conference on Retroviruses and Opportunistic Infections, Boston, abstract 92, 2011.

Anderson P et al. Interpreting detection rates of intracellular FTC-TP and TFV-DP: the iPrEx trial. 18th Conference on Retroviruses and Opportunistic Infections, Boston, abstract 96LB, 2011. 

Amico R et al. Adherence indicators and PrEP drug levels in the iPrEx study. 18th Conference on Retroviruses and Opportunistic Infections, Boston, abstract 95LB, 2011.

Mulligan K et al. Effects of FTC/TDF on bone mineral density in seronegative men from 4 continents: DEXA results of the global iPrEx study. 18th Conference on Retroviruses and Opportunistic Infections, Boston, abstract 94LB, 2011.

Liu A et al. BMD loss in HIV men participating in a TDF PrEP clinical trial in San Francisco. 18th Conference on Retroviruses and Opportunistic Infections, Boston, abstract 93, 2011.

Abstracts and webcasts

You can view the abstracts from this research on the official conference website:

Abstract 92:

Abstract 96LB:

Abstract 95LB:

Abstract 94LB:

Abstract 93:

You can also watch a webcast of the presentations made at this conference session, including the speakers Robert Grant, Peter Anderson, Rivet Amico, Kathleen Mulligan and Albert Liu.

Webcast from Advances in PrEP.

We rely on donations to continue our work to help people with HIV, support us today at:

Community Consensus Statement on Access to HIV Treatment and its Use for Prevention

Together, we can make it happen

We can end HIV soon if people have equal access to HIV drugs as treatment and as PrEP, and have free choice over whether to take them.

Launched today, the Community Consensus Statement is a basic set of principles aimed at making sure that happens.

The Community Consensus Statement is a joint initiative of AVAC, EATG, MSMGF, GNP+, HIV i-Base, the International HIV/AIDS Alliance, ITPC and NAM/aidsmap

This content was checked for accuracy at the time it was written. It may have been superseded by more recent developments. NAM recommends checking whether this is the most current information when making decisions that may affect your health.

NAM’s information is intended to support, rather than replace, consultation with a healthcare professional. Talk to your doctor or another member of your healthcare team for advice tailored to your situation.