Intermittent Prophylaxis with Oral
Truvada Protects Macaques from Rectal SHIV Infection
Sci Transl Med 13 January 2010:|
Vol. 2, Issue 14, p. 14ra4
J. Gerardo García-Lerma1,*, Mian-er Cong1, James Mitchell1, Ae S. Youngpairoj1, Qi Zheng1, Silvina Masciotra1, Amy Martin1, Zsuzsanna Kuklenyik2, Angela Holder1, Jonathan Lipscomb1, Chou-Pong Pau1, John R. Barr2, Debra L. Hanson1, Ron Otten1, Lynn Paxton1, Thomas M. Folks1,3 and Walid Heneine
1Division of HIV/AIDS Prevention, National Center for HIV, Hepatitis, STD, and Prevention, Centers for Disease Control and Prevention, 1600 Clifton Road, Atlanta, GA 30329, USA.
2Division of Laboratory Sciences, National Center for Environmental Health, Centers for Disease Control and Prevention, Atlanta, GA 30329, USA.
3Southwest National Primate Research Center, Southwest Foundation for Biomedical Research, San Antonio, TX 78245, USA.
*To whom correspondence should be addressed. E-mail: GGarcia-Lerma@cdc.gov
HIV continues to spread globally, mainly through sexual contact. Despite advances in treatment and care, preventing transmission with vaccines or microbicides has proven difficult. A promising strategy to avoid transmission is prophylactic treatment with antiretroviral drugs before exposure to HIV. Clinical trials evaluating the efficacy of daily treatment with the reverse transcriptase inhibitors tenofovir disoproxil fumarate (TDF) or Truvada (TDF plus emtricitabine) are under way. We hypothesized that intermittent prophylactic treatment with long-acting antiviral drugs would be as effective as daily dosing in blocking the earliest stages of viral replication and preventing mucosal transmission. We tested this hypothesis by intermittently giving prophylactic Truvada to macaque monkeys and then exposing them rectally to simian-human immunodeficiency virus (SHIV) once a week for 14 weeks. A simple regimen with an oral dose of Truvada given 1, 3, or 7 days before exposure followed by a second dose 2 hours after exposure was as protective as daily drug administration, possibly because of the long intracellular persistence of the drugs. In addition, a two-dose regimen initiated 2 hours before or after virus exposure was effective, and full protection was obtained by doubling the Truvada concentration in both doses. We saw no protection if the first dose was delayed until 24 hours after exposure, underscoring the importance of blocking initial replication in the mucosa. Our results show that intermittent prophylactic treatment with an antiviral drug can be highly effective in preventing SHIV infection, with a wide window of protection. They strengthen the possibility of developing feasible, cost-effective strategies to prevent HIV transmission in humans.
More than a quarter century after the description of the first cases of AIDS, HIV has spread to virtually every country in the world, infecting 65 million people and killing 25 million (1). An effective HIV vaccine remains elusive, although the recent finding of a modestly successful two-vaccine prime-boost vaccination strategy gives new hope to vaccine development (1, 2). A multicomponent approach that includes behavioral, structural, and biomedical interventions will likely be essential to curb the HIV epidemic (3, 4).
Daily preexposure prophylaxis (PrEP) with antiretroviral drugs is a promising intervention to protect high-risk HIV-negative people from becoming infected. Multiple lines of evidence favor this approach. Antiretroviral drugs effectively prevent HIV transmission at birth, during breastfeeding, and after occupational exposure (5–7). Daily PrEP with the HIV reverse transcriptase inhibitors tenofovir (TFV) disoproxil fumarate (TDF), emtricitabine (FTC), or Truvada (an FTC-TDF combination) can prevent or delay simian immunodeficiency virus (SIV) or simian-human immunodeficiency virus (SHIV) transmission in macaques in a dose-dependent manner (8–10). These observations have provided the basis for evaluating the efficacy of daily PrEP in preventing HIV transmission in humans, and several clinical trials to evaluate HIV transmission with TDF or Truvada are now ongoing in high-risk populations (3, 4). Although these trials will provide the first proof-of-concept data on the efficacy of PrEP in humans, intermittent drug administration is a far more feasible strategy than daily PrEP. Intermittent PrEP (iPrEP) is likely to be more practical to implement and more cost-effective; it may reduce the risk of drug toxicities and potentially increase adherence and minimize the emergence of drug resistance.
We hypothesized that intermittent dosing designed to deliver long-acting drugs at the viral entry sites will be as effective as daily dosing in blocking the earliest stages of virus replication and preventing mucosal transmission. We based our rationale on the proved efficacy of short-course prophylaxis with the long-acting drug nevirapine in preventing HIV transmission at birth (7), the relatively small size of founder virus populations that initiate an HIV infection in the mucosa (11–13), and the high FTC and TFV concentrations achieved in the male and female genital tract after oral administration (14, 15). Our recent findings in macaques showing complete protection from rectal SHIV exposures by a two-dose subcutaneous regimen containing a high dose of TFV with FTC provided the first proof of concept for iPrEP as a chemoprevention strategy against sexual transmission (8). However, the high drug doses and subcutaneous drug delivery system used may produce an overestimation of iPrEP efficacy. Therefore, more studies with orally administered drugs that provide drug exposures and distribution kinetics similar to those in humans are necessary to better evaluate iPrEP efficacy and inform clinical trial designs.
Here, we repeatedly exposed macaque monkeys rectally to SHIV to assess the efficacy of different iPrEP dosing regimens with oral Truvada and to define windows of protection for preexposure or postexposure dosing. The macaque model resembles human HIV transmission in many ways (16). First, the SHIV challenge dose is lower and more physiologic than what is conventionally used in the single high-dose challenge models. This SHIV contains an R5-tropic HIV-1SF162 envelope similar to naturally transmitted viruses. Second, exposures to virus are repeated up to 14 times to mimic high-risk human exposures. The single-dose approach only measures protection against one transmission event per animal, whereas the repeated-challenge model evaluates protection against multiple transmission events and may be a more robust model for testing biomedical interventions (8, 9, 17). We show that significant protection can be achieved with a single dose of Truvada given 1 to 7 days before exposure followed by a second dose 2 hours after exposure. We also show that drug administration around the time of exposure may be effective but that protection is lost if the first dose is delayed 24 hours after exposure.
Effect of preexposure and postexposure doses on iPrEP effectiveness
We recently showed that an intermittent regimen consisting of two subcutaneous doses of FTC (20 mg/kg) and TFV (22 mg/kg) given 2 hours before and 22 hours after virus exposure was fully protective in the repeat-exposure macaque model of rectal transmission (8). Although the dose of TFV used in these animals was higher by factors of 3 to 4 than that used in humans, these findings demonstrated that an iPrEP strategy might be effective. To explore whether the preexposure dose was sufficient for full protection and to assess the contribution of the postexposure dose, we gave a group of six macaques the same subcutaneous FTC-TFV dose 2 hours before each rectal exposure without the +22-hour postexposure dose. We also evaluated, in a second group of six macaques, the effectiveness of this regimen when given at 24 and 48 hours after each exposure [postexposure prophylaxis (PEP)] without the preexposure dosing. Protection from 14 weekly rectal virus challenges with SHIV162p3 was evaluated against 32 untreated controls; 5 were real-time controls and 27 were recent or historic controls from experiments done with the same virus stock, inoculum size, and inoculation protocol (8).
A single subcutaneous FTC-TFV dose given 2 hours before exposure (-2 hours) was not fully protective, and two of the six macaques were infected during the 14 challenges (challenges 8 and 13) (Fig. 1). This result indicates that protection can be achieved by the preexposure dose but that the treatment after exposure is necessary for maximal iPrEP effectiveness. In contrast, three of the six macaques receiving PEP were infected during the first two challenges; the challenge series was stopped at this point after an interim analysis showed absence of efficacy relative to untreated controls (P > 0.5, log-rank test for differences in survival curves). These findings show the inability of this short-course PEP regimen to control virus spread after mucosal infection occurs (Fig. 1). Overall, these results indicate that a preexposure drug dose that can block the initial establishment of infection is critical, highlight the importance of drug availability before virus exposure, and suggest that both a preexposure and a postexposure drug dose are necessary for maximum effectiveness.
Protection by subcutaneous FTC-TFV administered before and/or after virus exposure. Each survival curve represents the cumulative percentage of uninfected macaques as a function of weekly (-2-hour/+22-hour, -2-hour, and control groups) or biweekly virus exposures (+24-hour/+48-hour group). Uninfected animals remained negative after 14 exposures and a mean washout observation period of 13 weeks. The challenge series in the +24-hour/+48-hour group was stopped after an interim analysis done at week 2 showed absence of efficacy (P > 0.5). The full protection conferred by the -2-hour/+22-hour FTC-TFV regimen has been reported (8). The group of control animals includes 5 real-time controls and 27 historic and recent controls exposed to the same virus stock and dose with the same inoculation method. The five real-time controls were infected at challenges 1, 2 (two animals), and 3 (three animals).
Protection by preexposure dose followed by a second postexposure dose
To define the window of protection of the preexposure dosing, we designed several interventions in which we administered FTC and TDF at different times relative to the virus challenge. To better model the situation in humans, we gave FTC and TDF orally at doses that resulted in exposure to Truvada similar to that in humans (FTC, 20 mg/kg; TDF, 22 mg/kg) (Fig. 2A) (8). Macaques received the first dose of Truvada 22 hours (group 1), 3 days (group 2), or 7 days (group 3) before each virus exposure followed by a second dose 2 hours after exposure. An additional group of nine control macaques was maintained without drug administration at the same time. All animals were exposed weekly to SHIV162p3 rectally for up to 14 weeks.
Protection by exposure-independent and event-driven prophylaxis with Truvada. (A) Study design and interventions with oral Truvada. All macaques received two weekly doses of Truvada by oral gavage and were exposed once a week at the indicated time points, with the exception of macaques from group 5. Group 5 macaques were exposed and treated every 2 weeks to minimize residual drug exposure due to long intracellular FTC and TFV persistence. Truvada was administered at the indicated times (short black bars) relative to viral exposure. (B) Reduction in the risk of infection (HRs) by intermittent prophylaxis with oral Truvada. Each survival curve represents the cumulative percentage of uninfected macaques as a function of virus exposure. For comparison, the previously reported survival curve showing protection of daily PrEP with oral Truvada (continuous red line) is shown (8). The group of control animals includes the 9 real-time controls and 23 historic and recent controls exposed to the same virus stock and dose using the same inoculation method. The nine real-time controls were infected at challenges 1 (six animals), 3 (two animals), and 5 (one animal).
Consistent with previous data, all nine controls became infected within one to five challenges. In contrast, all three iPrEP regimens showed protection (Fig. 2B). Five of six animals from group 1 (-22 hours/+2 hours), five of six animals from group 2 (-3 days/+2 hours), and four of six animals from group 3 (-7 days/+2 hours) remained uninfected after the 14 weekly challenges [P < 0.0001 for each group relative to 32 untreated control macaques (9 real-time and 23 recent and historic controls)]. Relative to all untreated controls, the risk of infection by these three interventions was reduced by factors of 16.7 (group 1, P = 0.006), 15.4 (group 2, P = 0.008), and 9.3 (group 3, P = 0.003).
To better understand the relation between protection and drug pharmacokinetics, we performed a separate experiment with different animals to calculate the half-lives of active TFV-diphosphate (TFV-DP) and FTC-triphosphate (FTC-TP) in peripheral blood mononuclear cells (PBMCs). FTC-TP and TFV-DP were measured in four macaques for 12 days after oral administration of a single dose of Truvada. The estimated mean half-life for TFV-DP was 115 hours (range, 78 to 170 hours), similar to that seen in humans (150 hours; range, 60 to >175 hours) (18, 19) (Fig. 3). FTC-TP concentrations declined at a higher rate (half-life, 24 hours; range, 15 to 49 hours) and followed kinetics that were also similar to those seen in humans (20). These findings suggest that the high level of protection by the different preexposure dosing strategies may be related to the long persistence of TFV-DP and FTC-TP in PBMCs.
Intracellular FTC-TP and TFV-DP concentrations in macaque PBMCs. The mean (ĪSEM) FTC-TP and TFV-DP concentrations observed in four macaques after administration of a single dose of FTC and TDF by oral gavage are shown. Mean half-lives (t1/2) of FTC-TP and TFV-DP are also indicated.
Protection by prophylaxis related to viral exposure
To investigate the effectiveness of drug administration around the time of exposure, we administered orally to two groups of macaques a two-dose Truvada regimen starting 2 hours before (group 4) or 2 hours after (group 5) exposure (Fig. 2A). Animals from both groups received a second dose of Truvada 24 hours after the first dose. The -2-hour/+22-hour regimen conferred significant protection [hazard ratio (HR) reduction by a factor of 4.1, P = 0.02], with three of the six macaques protected during the 14 challenges (P = 0.0004 relative to controls) (Fig. 2B). Infection of these three macaques was not associated with lower plasma drug concentrations at exposure or with delayed or reduced drug absorption, as indicated by the analysis of area under the curve (AUC24 hours) and plasma drug half-life values (figs. S1 and S2). Significant protection was also seen in group 5 macaques that received the +2-hour/+26-hour PEP regimen (HR reduction by a factor of 4.0, P = 0.03), with three of the six animals protected during the 14 challenges (P = 0.0004 relative to controls) (Fig. 2B). We also explored whether a higher dose of Truvada could increase the effectiveness of exposure-driven iPrEP by orally administering to a group of six macaques (group 6) twice the human equivalent dose of both FTC (40 mg/kg) and TDF (44 mg/kg) 2 hours before and 24 hours after exposure. Five of the six animals were protected and only one was infected at challenge 4 (P < 0.0001 relative to controls) (fig. S3). An examination of plasma drug concentrations in the infected animal showed consistently low FTC and TFV concentrations during the first 7 to 9 weeks, which suggests that in this particular macaque, infection might have been facilitated by suboptimal drug absorption (fig. S3).
Pharmacokinetics of FTC and TFV in rectal secretions and tissues
Optimal drug concentration at the mucosa is essential for effective iPrEP. We evaluated the pharmacokinetic profiles of FTC and TFV in both plasma and rectal secretions by administering to four macaques a single dose of Truvada orally followed by collection of blood and secretions at 2, 5, and 24 hours (Fig. 4A). Concentrations of TFV at 2 and 5 hours in secretions were low or undetectable but increased substantially in all animals at 24 hours (median = 1186 ng/ml, range = 279 to 59,920). In contrast, FTC was detected in secretions in all macaques at 2 hours (median = 100 ng/ml, range = 41 to 618) and in three of the four animals at 5 hours (median = 114 ng/ml, range = 95 to 240), albeit at lower concentrations than in plasma (Fig. 4A). Similar to that observed with TFV, FTC concentrations in secretions also increased considerably at 24 hours (median = 743 ng/ml, range = 243 to 28,504) (Fig. 4A). Thus, the pharmacokinetic profile of FTC and TFV in rectal secretions was different from that seen in blood in which drug concentrations peaked around 2 hours and substantially decreased at 24 hours. At 24 hours, median FTC and TFV concentrations in secretions were higher by factors of 17 and 50 than those seen in plasma, respectively (Fig. 4A).
Extracellular and intracellular FTC and TFV concentrations in macaques after oral administration of a single dose of Truvada. (A) Extracellular FTC and TFV concentrations in plasma and rectal secretions from four macaques. Secretions were collected at three time points with rectal wicks as detailed in the Supplementary Material. Data represent the median (range) FTC and TFV concentrations seen in four animals. (B) TFV-DP concentrations in mononuclear cells from blood, rectal tissue, and axillary, mesenteric, and inguinal lymph nodes collected from seven macaques that received Truvada 2 hours and 1, 2, 3, or 7 days before scheduled necropsy. Individual TFV-DP concentrations in each specimen are shown. The number of animals evaluated at each time point is also indicated.
The kinetics of TFV-DP and FTC-TP were further evaluated in rectal tissue by orally administering to seven macaques a single dose of Truvada followed by necropsies to collect tissues after 2 hours and 1, 2, 3, or 7 days. Intracellular drug concentrations were compared to those seen in blood and axillary, mesenteric, and inguinal lymph node specimens collected in parallel. TFV-DP concentrations at 2 hours were similar in blood, rectal tissue, and lymphoid tissue (Fig. 4B). However, TFV-DP concentrations in rectal tissues collected from four macaques at 1 day (361 and 156 fmol per 106 cells), 2 days (1131 fmol per 106 cells), or 3 days (882 fmol per 106 cells) were substantially higher than those seen in blood or lymph nodes, a finding that was consistent with the high extracellular TFV concentrations in rectal secretions at 24 hours and long TFV-DP persistence in mononuclear cells (Figs. 3 and 4A). Degradation of FTC-TP during tissue processing was evident by the detection of FTC diphosphate, FTC monophosphate, and FTC and precluded an accurate analysis of FTC-TP concentrations. Nevertheless, we were able to measure FTC-TP in rectal tissues from two animals at 2 hours (358 fmol per 106 cells) and 24 hours (390 fmol per 106 cells).
Plasma virus loads and drug resistance emergence in iPrEP failures
The mean peak viremia was significantly lower in iPrEP failures (5.6 log10 RNA copies per milliliter) relative to untreated controls (7.1 log10 RNA copies per milliliter) (P = 0.009) (Fig. 5). Virus loads also declined significantly faster in iPrEP failures (0.32 log10 per week) during continued treatment (median = 13.5 weeks; range, 9 to 16) than in control animals (0.13 log10 per week, P < 0.0001) (Fig. 5). None of the animals infected during iPrEP developed the K65R or M184V mutations associated with TFV or FTC resistance during a median follow-up period of 13 weeks on PrEP. A detailed description of the infection kinetics for each iPrEP failure is shown in fig. S4.
Blunted acute viremia in macaques infected during iPrEP. Individual virus load kinetics in breakthrough infections under continued drug exposure (n = 6, red lines) and in untreated control macaques with sufficient follow-up (n = 22, black lines) are shown. Time 0 indicates the peak plasma virus load. Mean peak viremia was significantly lower in iPrEP failures (P = 0.009). Virus loads also declined significantly faster in iPrEP failures than in control animals (P < 0.0001).
We used macaque monkeys repeatedly exposed rectally to SHIV to investigate the effectiveness of intermittent treatment schedules with the antiretroviral drug Truvada orally administered at human equivalent doses. We tested dosing regimens initiated 1, 3, or 7 days before exposure and showed that, as with daily PrEP, all were protective against infection (8). We selected these drug-dosing intervals to evaluate whether one to two weekly doses of Truvada followed by a booster dose after exposure to the virus were sufficient to prevent transmission. We attribute the extended window of protection to the long intracellular persistence of FTC and TFV in PBMCs; drug half-lives were within the range of those observed in humans (18–20). We further show efficacy for prevention when drug treatment is initiated within 2 hours of virus exposure, thus providing evidence that event-driven drug dosing might also be effective. We saw increased protection in this scenario when the dose of Truvada was doubled. Together, these findings indicate a range of possible iPrEP designs and highlight the promise of this strategy as an alternative to daily PrEP in preventing HIV transmission. If the ongoing daily PrEP clinical trials show efficacy in humans, intermittent dosing such as described here may provide a next-generation chemoprevention strategy with potential benefits, including cost-effectiveness, greater patient adherence, and reduced risks of drug toxicities.
Our observations with a potent subcutaneous FTC and high-dose TFV combination suggested that treatment is needed both before and after exposure for maximum protection. We administered FTC and TFV subcutaneously to these animals to facilitate a rapid tissue distribution. The finding that a single subcutaneous dose of this regimen given 2 hours before exposure was protective highlights the importance of efficiently blocking early replication in the mucosa, an observation that was confirmed by the lack of protection seen when the first dose was delayed until 24 hours after exposure. The inability to protect these macaques further demonstrates that founder virus populations are rapidly established mucosally, as previously noted in macaques exposed vaginally to SIVmac251, and, importantly, the difficulty in controlling virus spread from productively infected cells (13, 21, 22). Thus, waiting 24 hours after exposure to provide FTC and TFV may have allowed the earliest mucosal infections to proceed because reverse transcription can be completed in 4 to 6 hours in activated T cells (23).
With the exception of one animal that showed unusually low drug concentrations, we could not establish an association between infection outcome and plasma drug concentrations. However, our analysis was limited because the number of drug measurements associated with infection was small and only a few animals were infected in our studies. Our observation that the efficacy of exposure-related iPrEP can be increased by doubling the Truvada dose suggests that drug concentrations may correlate with protection and also strengthens the possibility of using iPrEP tied to viral exposure if the risks of drug toxicity do not increase when the dose of Truvada is doubled. It will be important to define the threshold for plasma and intracellular drug concentrations that are associated with protection.
It is noteworthy that we found that FTC and TFV concentrations in rectal secretions were low during the first 5 hours and were highest at 24 hours. Low tissue drug concentrations shortly after drug administration may increase risks of infection and partially explain the apparent lower effectiveness seen when treatment was contingent on exposure (6 of 12 animals from groups 4 and 5 infected), although a lower efficacy of these regimens could not be statistically ascertained. The high TFV and FTC concentrations in rectal secretions 24 hours after oral drug administration may partially originate from degradation of TDF to TFV by esterases in the intestines and from elimination of FTC in feces (24). A combination of local and systemic drug absorption may protect more susceptible target cells at the mucosa and contribute to iPrEP effectiveness. High drug concentrations in rectal secretions at 24 hours may increase intracellular drug concentrations in rectal tissues and explain the higher (by factors of 3 to 5) intracellular TFV-DP concentrations at 24 hours in rectal lymphocytes when compared to PBMCs and the presence of TFV-DP in tissues 2 to 7 days after administration of TDF. We also show that FTC was more rapidly detected than TFV in rectal secretions, indicating more rapid absorption and tissue distribution. Together, these findings suggest that the rapid FTC distribution in tissues and the long intracellular TFV persistence complement each other to maximize iPrEP effectiveness.
We recently found that macaques failing daily PrEP with FTC or Truvada had lower acute viremias than untreated macaques (8). We also show here attenuated viremias in animals infected during iPrEP with only two doses of Truvada per week, likely reflecting the extended antiviral activity of FTC and TDF resulting from their long intracellular half-life. Such decreased acute viremias may have clinical and public health implications. Acutely infected persons with very high viral loads may play a key role in the epidemic spread of HIV-1 because they are more infectious than chronically infected persons who have lower virus loads (25–27). Therefore, a reduction in acute viremia during iPrEP treatment may contribute to decreases in HIV-1 transmission in the population and could add to the overall effectiveness of PrEP. Substantial reductions in virus loads during acute viremia could also reduce CD4+ T cell depletion, help preserve immune function, and attenuate the course of HIV infection (28).
Notably, none of the animals that failed iPrEP had detectable resistance even with sensitive testing for minority M184V or K65R mutants associated with FTC or TFV resistance, a finding that differed from our earlier observations in macaques that failed daily PrEP with FTC or Truvada in which two of six failures acquired M184V/I (8). The lack of resistance seen in all seven macaques failing iPrEP is encouraging and suggests that two weekly doses of Truvada may minimize selection of drug-resistant viruses. However, our data need to be interpreted cautiously because it is not known whether the risks of resistance emergence would increase with more frequent exposure to drug resulting from more frequent administration of Truvada (29). In addition, our study cannot test the emergence of resistance after prolonged (>12 weeks) intermittent drug exposure, because viremias in animals infected with SHIV162p3 tend to decrease to low or undetectable values over time.
Our study is subject to several limitations. First, all viral challenges were nontraumatic and were done in the absence of semen or semen-derived factors that may enhance HIV infection in vitro or other cofactors that may increase transmission risks, such as sexually transmitted infections. Our use of the less pathogenic SHIV162p3 isolate might also potentially overestimate efficacy. However, SHIV162p3 is easily transmissible to macaques at low [10 TCID50 (tissue culture infectious dose)] infectious doses both rectally and vaginally (8, 9, 16) and, thus, is well-suited for transmission studies. Second, we used a wild-type virus that is fully susceptible to both TFV and FTC. Because circulating drug-resistant viruses may be common, more work should be done to define PrEP efficacy on drug-resistant viruses. Third, our study was not powered to evaluate statistical differences between the different iPrEP modalities or between daily PrEP and iPrEP because of the limited number of animals per group. Finally, it is important to confirm the efficacy of these PrEP regimens against vaginal transmission in appropriate macaque models. Although many biologic similarities exist between rectal and vaginal HIV transmission, some differences in the early events of mucosal infection or changes in susceptibilities associated with the menstrual cycle and thinning of the epithelium are possible (13, 30, 31). The pharmacokinetic profiles of FTC and TFV in the female genital tract may also be different from those in rectal tissues and could potentially affect iPrEP effectiveness.
In macaques repeatedly exposed rectally to SHIV, intermittent prophylaxis with Truvada may be highly effective for preventing infection with a wide window of protection. All current PrEP trials are based on daily drug-dosing intervals that were selected because they are effective for treatment. The protection seen in our macaques suggests that Truvada might prevent transmission in humans if taken on the basis of exposure events or after a fixed weekly schedule of one to two doses with a booster dose after any exposure to virus. Less frequent drug administration would reduce cost and might decrease drug toxicities and foster adherence by reducing unnecessary drug exposure and frequency of mild side effects. Ongoing clinical trials with daily PrEP will shed light on the efficacy of PrEP as a prevention strategy. If the human daily PrEP trials prove effective, additional trials would be needed to evaluate whether iPrEP modalities may be ultimately sufficient to prevent HIV transmission.
Materials and Methods
Drug preparation and administration
For oral administration, TDF was first suspended in phosphate-buffered saline (PBS) and dissolved with NaOH followed by the addition of FTC. Drugs were given orally by gavage to anesthetized macaques via a gastric feeding tube (8). For subcutaneous injections, stock solutions of TFV and FTC were prepared in deionized water or PBS, respectively (8). TDF, TFV, and FTC were provided by Gilead Sciences.
Repeat-exposure macaque model
The efficacy of different iPrEP modalities was evaluated with a repeat-exposure macaque model of rectal transmission previously described (8, 9, 16). Rhesus macaques were exposed rectally once weekly or every 2 weeks (group 5 only) to a SHIVSF162P3 chimeric virus that contains the tat, rev, and env coding regions of HIV-1SF162 in a background of SIVmac239 [National Institutes of Health AIDS Research and Reference Reagent Program (32)]. The SHIV162p3 challenge dose was 10 TCID50 or 7.6 x 105 RNA copies, which is within the range of HIV-1 RNA concentrations in semen (103 to 106 copies per milliliter) during acute infection in humans and higher than the concentrations (102 to 104 copies per milliliter) seen after primary viremia (33). Virus exposures (up to 14) were done by nontraumatic inoculation of 1 ml of SHIVSF162P3 into the rectal vault via a sterile gastric feeding tube of adjusted length (16). Macaques were anesthetized with standard doses of ketamine hydrochloride. Anesthetized macaques remained recumbent for at least 15 min after each intrarectal inoculation. Virus exposures were stopped when a macaque became SHIV RNA–positive. Macaques infected during PrEP continued treatment for a median of 13.5 weeks (range, 9 to 16). All experiments were done under highly controlled conditions by the same personnel using the same virus stock, inoculum dose, and inoculation method. The animal handlers who administered drug or performed the virus challenges were not blinded. The Institutional Animal Care and Use Committee of the Centers for Disease Control and Prevention approved this study.
Infection monitoring by molecular and serologic testing
Plasma SHIV RNA was quantified with a real-time reverse transcription polymerase chain reaction (PCR) assay as previously described (9). This assay format has a sensitivity of 50 RNA copies per milliliter. Detection of low-frequency K65R and M184V mutants in plasma was performed with sensitive allele-specific real-time PCR methods as previously described (8, 34). Virus-specific serologic responses (immunoglobulins G and M) were measured with a synthetic peptide enzyme immunoassay (Genetic Systems HIV-1/HIV-2, Bio-Rad). Animals in the iPrEP arms were considered protected from systemic SHIV infection if they remained seronegative and negative for SHIV plasma RNA and SHIV DNA in PBMCs during PrEP and during the following 70 days of washout in the absence of any drug treatment (35).
Measurement of drug concentrations in plasma, PBMCs, and tissues
FTC and TFV concentrations in plasma and rectal secretions were measured by high-performance liquid chromatography–tandem mass spectrometry (MS/MS) (36). Intracellular FTC-TP and TFV-DP concentrations were measured with an automated online weak anion exchange solid-phase extraction method coupled with ion-pair chromatography–MS/MS. Procedures are detailed in the Supplementary Material.
The Cox proportional hazards model was used to estimate instantaneous risk for infection, as a HR, in controls relative to treated animals, assuming constant risk at all inoculations. Graphical methods of model assessment supported the use of Cox proportional hazards regression. The two-sided Fisher's exact test was used for a categorical analysis of number of infections per total exposures in each group relative to controls. Plasma FTC and TFV concentrations were compared with a mixed-effect model, with a random intercept and unstructured covariance to account for within-subject correlated measurements. Mean area under the curve (AUC24 hours) and plasma terminal elimination half-life (HL λz) values among infected and protected animals were compared with the Wilcoxon-Mann-Whitney test (two-sided). The WinNonlin software (version 5.2; Pharsight) was used to calculate AUC24 hours and HL λz values. The Wilcoxon rank-sum statistic was again implemented to test for group differences in magnitude of peak virus load; mixed-effects regression was used to assess differences between treatment and control groups in the rate of decline after peak virus load. All statistical analyses were performed with SAS software (version 9.1; SAS Institute).