We demonstrate that oral Truvada administered peri-coitally protects pigtail macaques from vaginal SHIV transmission. We used pigtail macaques because they have regular lunar menstrual cycles and plasma sex hormone changes similar to humans 
. We show that a simple Truvada regimen consisting of only 2 doses (one 24 h before and another 2 h after exposure) resulted in complete protection from infection. These findings expand our previous observations in rhesus macaques where the same Truvada regimen was also highly protective against rectal virus transmission 
. The high prophylactic efficacy demonstrated with a coitally-dependent Truvada regimen against both vaginal and rectal transmission in macaques suggest that this PrEP modality might be an alternative to daily dosing, and informs clinical evaluation of exposure-driven intermittent PrEP regimens for preventing sexual HIV transmission.
Our analysis of FTC and TDF PK profile was key to ensure that we are modeling doses of Truvada similar to those used in humans. The doses of FTC and TDF were adjusted for the weight and size of macaques since smaller mammals usually eliminate drugs faster than larger mammals 
. At these doses, Cmax
and AUC values for TFV and FTC in plasma were within the range seen in humans 
. Also as in humans, the half-life of TFV-DP and FTC-TP was long in PBMCs (about 60 and 24 h respectively). However, several limitations precluded a better comparison with available human data. First, our FTC-TP and TFV-DP PK analysis in PBMCs and rectal tissues was done at first dose and did not allow a direct comparison with human data which is only available at steady state 
. It will be important to compare our absolute drug levels and drug half-life once first dose kinetics become available in humans. Second, a direct comparison of drug levels in vaginal secretions between macaques and humans was not possible due to differences in collected methods (wicks compared to direct aspiration), sampling methodologies, and testing procedures. However, we found that, as in humans, FTC and TFV peaked later in vaginal secretions than in plasma and remained higher than plasma levels over a 48 h period 
. Also as in humans, the concentration of FTC in vaginal secretions was higher than the concentration of TFV, and both TFV and TFV-DP levels in vaginal tissues were substantially lower than in rectal tissues 
. Interestingly, the differences in TFV-DP levels between vaginal and rectal tissues appeared to wane overtime, suggesting that TFV-DP half-life in rectal tissues might be reduced compared to vaginal tissues. If confirmed in larger number of animals and in humans, this observation suggests that rectal efficacy of intermittent PrEP based on fixed drug dosing Truvada regimens (e.g., weekly), may not necessarily benefit from higher rectal TFV-DP levels.
The differential penetration of FTC and TFV in the rectal and vaginal compartments raises questions regarding the relative contribution of FTC and TFV to the efficacy of daily Truvada seen in different risk populations 
. The finding that daily TDF alone prevented HIV transmission in highly adherent women demonstrates that the antiviral activity resulting from daily dosing is sufficient to prevent vaginal transmission 
. Given the low vaginal TFV exposure in women, it is possible that both the systemic and vaginal drug activity contribute to protection. Additional studies in macaques using for instance vaginal gels that result in tissue TFV-DP levels similar to those achieved after oral dosing may help to understand the relative contribution of local vs systemic TFV levels to the observed protection. It will also be interesting to determine in our monkey model if intermittent dosing with TDF only is equally protective to Truvada. Although the efficacy of daily TDF in preventing rectal HIV transmission in humans is not known, macaque studies with TDF or another TFV prodrug predict lower efficacy against rectal transmission and suggest that, for rectal efficacy, Truvada may be required 
Our study is subjected to several limitations. First, all virus challenges were non-traumatic and done in the absence of semen or semen-derived factors that may enhance susceptibility to HIV infection or other cofactors that may increase HIV transmission risk, such as sexually transmitted infections. Second, we did not model fixed-dose intermittent PrEP regimens containing for instance 2 weekly doses of Truvada followed by a booster post-exposure dose as in the ADAPT study 
. Although fixed-dosing PrEP modalities were highly efficacious against rectal transmission in macaques, rectal efficacy results cannot be extrapolated to vaginal efficacy given the differences in drug penetration between vaginal and rectal tissues 
. Third, we did not measure drug levels in the interior iliac lymph nodes. This is important since these nodes drain both the genital and rectal tract and are believed to be an important site of early virus replication 
Vaginal transmissibility studies in macaques have conventionally used the injectable contraceptive Depo-Provera (DMPA) to maximize infection of animals 
. Our experimental design did not include DMPA because we wanted to fully recapitulate the variabilities in susceptibility to infection associated with the menstrual cycle as well as evaluate protection over repeated virus exposures 
. However, the use of DMPA may be useful to model the potential impact of injectable contraceptives on the efficacy of PrEP with Truvada. This analysis will be important since recent studies have suggested an increased risk of HIV acquisition associated with the use of DMPA, and PrEP clinical trials with different patterns of hormonal contraceptive use have shown contradictory results 
. The development of a biologically relevant macaque model to determine the effect of DMPA on PrEP efficacy will first require careful selection of DMPA doses that mimic human exposure and biological effects 
. It will also be important to explore if other factors such as sexually transmitted infections might reduce efficacy among highly adherent PrEP users.
Our analysis of virus levels in vaginal and rectal tissues showed high levels of virus shedding accompanying peak plasma viremia after correcting for blood contamination in secretions. We found virus levels as high as 1×106
copies per wick despite the low volume of rectal and vaginal secretion contained in a single wick (30–35 uL). These results are consistent with the increased HIV transmissibility seen during the acute phase of HIV infection 
The acceptability of intermittent PrEP in different high-risk groups as well as the capacity of individuals to anticipate sexual activity and consistently adhere to intermittent dosing is not known 
. Reported adherence to intermittent PrEP was recently found to be lower than daily dosing among MSM and female sex workers 
. However, only a small number of men and women participated in this study and measurements of adherence to coitally dependent regimens using medication event monitoring systems or short message service was particularly challenging 
. An ongoing, larger Phase II trial among MSM and women who have sex with men is currently evaluating adherence to daily and intermittent Truvada dosing, and is also exploring intracellular drug levels at given rates of adherence during direct observed therapy 
. This study will better define behavioral aspects associated with intermittent dosing, coverage of sex events with pre-and post-exposure dosing, and if intermittent PrEP may achieve intracellular drug levels that cross the pharmacological barrier associated with prevention of HIV acquisition. The later aspect is particularly important since the same level of adherence might have a different impact on rectal and vaginal efficacy given the differences in FTC and TFV penetration among rectal and vaginal tissues.
In summary, we used a pigtail macaque model to evaluate a before/after exposure prophylactic modality that included 2 doses of Truvada within 24 h, and showed a high effectiveness in preventing vaginal SHIV transmission. These findings suggest that Truvada might prevent vaginal transmission in humans if taken peri-coitally. Less frequent drug administration would reduce cost and might decrease drug toxicities by reducing unnecessary drug exposure and frequency of mild side effects. It will be important to determine the acceptability of coitally-dependent PrEP based on ongoing studies among different high-risk populations, and if individuals will be able to anticipate sexual activity and consistently adhere 
. Specific populations such as serodiscordant couples planning to conceive a child might particularly benefit from coitally-dependent PrEP.