Tenofovir-based HIV Pre-exposure Prophylaxis

Lynn A Paxton


Future Virology. 2013;8(12):1207-1218. 

In This Article

Summary of Human Oral PrEP Efficacy Trials

To date, there have been six large, well-designed efficacy studies of oral PrEP with either TDF or TDF–FTC that have been published. Four of these showed strong evidence of efficacy, while two did not (Table 1).

iPrEx Study

This was a multinational study conducted among 2499 MSM or transgendered women in which participants were randomized in a 1:1 ratio to receive either daily oral TDF–FTC or placebo, then followed for a median of 1.2 years.[3] Participants reported high levels of risk behavior, and had a mean number of 18 sexual partners in the 3 months prior to enrollment. Over the course of the study, 110 individuals were found to be HIV infected; ten were determined to have been infected prior to study enrollment, and thus, were excluded from the modified intention-to-treat (mITT) efficacy analysis. The mITT analysis showed that 36 men were infected in the TDF–FTC arm compared with 64 in the placebo arm, representing a reduction in HIV acquisition of 44% (95% CI: 14–63%; p = 0.005). Further analysis of drug levels among all HIV seroconverters and a set of matched controls revealed that, although reported adherence was good, actual drug adherence was much lower; only three out of the 34 (9%) HIV-infected individuals in the TDF–FTC arm had detectable drug levels in plasma or peripheral blood mononuclear cells compared with 22 out of 43 (51%) of participants who remained seronegative. Participants with detectable drug levels had a 92% reduction in their risk of HIV acquisition (95% CI: 40–99%). Serious adverse events were similar between study arms. Nausea and weight loss were more commonly reported by individuals taking TDF–FTC but diminished after the first few weeks of drug therapy. There were no cases of HIV resistance among participants who seroconverted during the trial, but among the ten randomized participants who were infected at enrollment, three had the M184V/I mutation indicative of FTC resistance. Two of these participants had been randomized to the TDF–FTC study arm and one to the placebo arm. The resistant virus present in the placebo arm participant was presumed to represent transmitted, rather than acquired, resistance.

Partners PrEP Study

This study was conducted among heterosexual serodiscordant couples in Kenya and Uganda. A total of 4747 couples were followed; the seropositive partners were ineligible for ART per national guidelines and the seronegative partners were randomly assigned in a 1:1:1 ratio to receive daily TDF, TDF–FTC or placebo, and followed for a median of 1.9 years.[5] Over the course of the study, 96 participants were found to be HIV infected but 14 were later determined to have been infected at enrollment. Of the 82 infections that occurred during the trial, 17 were in the TDF arm, 13 in the TDF–FTC arm and 52 in the placebo arm. This corresponds to infection risk reductions of 67% (95% CI: 44–81%; p < 0.001) for TDF and 75% (95% CI: 55–87%; p < 0.001) for TDF–FTC; the difference between TDF and TDF–FTC efficacy was not statistically significant. Efficacy was similar for men and women. Adherence as assessed by pill counts was over 90%, and detectable levels of TFV were found in 82% of samples taken from a random selection of 198 participants in the active drug arms who did not acquire HIV. Only nine out of 29 seropositive individuals in those arms had detectable drug in plasma. This subset analysis estimated the relative HIV infection risk reduction for those with detectable drug levels to be 86% (95% CI: 57–95%) for TDF and 90% (95% CI: 56–98%; p = 0.002) for TDF–FTC. As in the iPrEx trial, none of the participants who seroconverted during the trial developed resistance mutations. Eight out of the 14 participants who were seropositive at enrollment were randomized to active drug arms. Two of these developed resistance – the K65R mutation in one person randomized to the TDF arm, and the M184V/I mutation in one person randomized to TDF–FTC. There were four cases of NNRTI mutations among seroconverters in the placebo arm, which were thought to reflect background resistance patterns of transmitted infections. Similar to other oral PrEP trials, both TDF and TDF–FTC were well tolerated by participants with no difference in serious adverse events between active and placebo arms.

TDF2 Study

This was a randomized trial of daily oral TDF–FTC versus placebo conducted among heterosexual men and women in Botswana. A total of 1219 participants were randomized to receive either daily oral TDF–FTC or placebo and followed for a median of 1.1 years.[4] Adherence by self-report and pill count was 94% for both arms [Kebaabetswe PM, Unpublished Data]. Among the study participants, 36 were found to be HIV infected over the course of the study, of whom three were later determined to have been seropositive at enrollment, and therefore excluded from the mITT analysis. Of the 33 infections that occurred during the study, nine were in active-arm participants; this corresponded to a relative risk reduction of 62% (95% CI: 22–83%; p = 0.03). In the as-treated analysis, in which follow-up data for participants were censored 30 days after their last reported dose of study medication (thereby restricting the analysis to four participants in the TDF–FTC group and 19 in the placebo group), the protective efficacy was 77.9% (95% CI: 41–94%; p = 0.01). Subgroup analysis of the four HIV-infected individuals in the active arm matched to 69 individuals who remained HIV negative showed that only two of the four (50%) seroconverters had detectable drug levels compared with 56 of the nonseroconverters (81%). The mean detectable plasma concentration was also much lower among the seroconverters (0.3 ng/ml [95% CI: 0.01–8.02] vs 30.6 ng/ml [95% CI: 16.3–57.5] for TFV [p = 0.007]; and 0.5 ng/ml [95% CI: 0.01–25.3] vs 103.3 ng/ml [95% CI: 45.4–234.9] for FTC [p = 0.009]). Adverse event rates were similar for both groups, with a slight excess of nausea, vomiting and dizziness in the active drug arm, which decreased after the first month on drug. TDF–FTC was associated with minimal but statistically significant declines in z- and t-scores for bone mineral density loss among a subgroup of participants who underwent serial DXA scans over the course of the study. There were no cases of HIV resistance among participants who acquired HIV during the trial. However, of the three participants who were already infected at enrollment, two were found to have resistance mutations: one in the TDF–FTC arm who developed K65R, M184V and A62V mutations, and one in the placebo arm who was found to have the M184V mutation.

FEM-PrEP Trial

The FEM-PrEP trial was a study of 2120 women in Kenya, Tanzania and South Africa who were randomized to receive daily TDF–FTC or placebo. Participants were intended to receive 52 weeks of study drug, but the study was stopped early at a planned interim analysis because there was no indication of efficacy.[70] At final analysis, 33 out of 1024 women in the active drug arm and 35 out of 1032 women in the placebo arm acquired HIV during the study, which corresponded to a relative risk of 0.94 (95% CI: 0.59–1.52; p = 0.81). While 95% of participants reported that they had usually or always taken the assigned drug, and pill count data were consistent with ingestion of the study drug on 88% of the days on which it was available to the participants, drug level analyses revealed that less than 40% of the HIV-uninfected women had evidence of recent pill use at visits that were matched to the HIV infection window for women with seroconversion. The study investigators concluded that low adherence levels resulted in the trial being substantially underpowered to detect an effect of TDF–FTC on the risk of HIV acquisition.


This study evaluated the efficacy of daily oral TDF, oral TDF–FTC and vaginal 1% TDF gel among 5029 sexually active women in Uganda, South Africa and Zimbabwe.[71] During the study, 334 women were found to be HIV infected; 22 of the 334 women were later determined to have been HIV-infected at enrollment. Overall HIV incidence during the trial was high at 5.7%, ranging from 0.8% in Zimbabwe to nearly 10% in South Africa. While no safety concerns were identified for any of the active drugs in the trial, two of the arms (1% TDF gel and oral TDF) were stopped early at interim analyses for lack of efficacy, and at final analysis TDF–FTC was also shown to not be effective in reducing HIV risk. While adherence by self-report and pill/applicator count was high at approximately 90%, actual product use was low for all study arms. Drug level analyses of a subset of 773 participants, including 185 who acquired HIV, detected drug in only 29% of blood samples from the TDF–FTC group, 28% from the oral TDF group and 23% from the TDF gel group. Younger, unmarried participants were both more likely to become HIV infected and less likely to use study products than older, married participants.

Bangkok TFV Study

This is to date the only Phase III trial to investigate PrEP efficacy among individuals who inject drugs. This trial recruited and enrolled 2413 male and female participants at 17 drug-treatment centers in Bangkok (Thailand); mean follow-up time was 4.0 years.[6] To be eligible for enrollment, participants must have injected drugs within the prior year. Volunteers were randomized 1:1 to receive daily oral TDF or placebo. Participants chose daily directly observed therapy (DOT) or monthly visits without DOT, and could switch between the two. Participants were on DOT an average of 86.9% of the time. Adherence estimates based on participants' study drug diaries showed that participants reported taking the study drug an average of 83.8% of days. Most participants (80%) were men. TDF and placebo recipients reported similar rates of injecting, sharing needles, similar numbers of sexual partners, and all risk behaviors decreased during follow-up: injection drug use fell from 63% at enrollment to 23% at month 12 and to 18% at month 72; sharing needles decreased from 18% at enrollment to 2% at months 12 and to 1% at month 72. Reports of nausea or vomiting were higher in the TDF group but between-group differences resolved by the second month of follow-up. Fifty two participants were found to be HIV infected during the course of the trial. Two participants assigned to the placebo group were later found to have been infected at the time of enrollment and were excluded from the efficacy analysis. Of the 50 incident HIV infections, 17 were in the TDF arm and 33 in the placebo arm, corresponding to a protective efficacy of 48.9% (95% CI: 9.6–72.2%; p = 0.01). In a case–control analysis of participants in the TDF arm, TFV was detected in the plasma of five out of 13 (39%) HIV-positive and 93 out of 138 (67%) HIV-negative participants. The reduction in HIV acquisition risk associated with having a detectable drug level was 70% (95% CI: 2.3–90.6%; p = 0.04). No participant was found to have TFV-associated resistance mutations.