INTRODUCTION The HIV epidemic is continuing to grow worldwide [1]. Consistent and correct use of condoms is recommended for prevention of sexually transmitted HIV, but often women are unable to negotiate condom use with their male partners. Safe, effective, and easy to use methods of HIV prevention are urgently needed, especially for women. Tenofovir disoproxil fumarate (TDF) [2], the orally bioavailable prodrug of tenofovir, is metabolized to a competitive inhibitor of viral reverse transcriptase. TDF was selected for clinical development as a treatment for HIV infection because of its (1) potency against wild-type HIV and some nucleoside-resistant strains of HIV [3–6], (2) low potential of selecting for TDF-resistant mutants [7], (3) low likelihood of metabolic/mitochondrial toxicity [8], and (4) pharmacologic profile supporting daily dosing [2]. TDF was licensed for the treatment of established HIV-l infection by the United States Federal Drug Administration in 2001, and the European Commission issued a marketing authorization in 2002 [9]. TDF has since been used worldwide for treatment of HIV infection, accounting for nearly 500,000 patient-years of observation [10]. We investigated the safety and effectiveness of a daily dose of 300 mg of TDF in preventing HIV infection among women at high risk for infection based on the following rationale: (1) initial prevention studies in simian models have provided support for both pre- and post-exposure efficacy of TDF in preventing retroviral infections [11–14]; (2) TDF has been shown to be safe in large numbers of HIV-infected persons [15,16]; (3) TDF is dosed conveniently once a day; (4) TDF has no known interactions with hormonal contraception [17]; (5) a high barrier to resistance was seen in clinical trials of HIV treatment, with the primary mutation identified (K65R) resulting in a reduction of viral replication to almost half that of wild-type [18]; and (6) the drug's sponsor, Gilead Sciences, is supportive of investigating the potential use of TDF as a preventive agent. Moreover, should it prove to be effective for HIV prevention, Gilead Sciences has committed to making TDF available in resource-poor settings for public health use, as they currently do for treatment of HIV, via no-profit pricing and licensing agreements. METHODS Participants Study participants were recruited from areas within each city that were considered high HIV transmission areas, including markets, bars, and hotels. Although we did not specifically ask as part of the clinical trial procedures if the participants were sex workers, most exchanged sex for money. Special ethical considerations were taken into account because of the potential vulnerability of this population. We developed strategies to protect the confidentiality and autonomy of the participants, increase/ensure comprehension of the informed consent and research methods, and promote access to resources and services during and after the trial. We enrolled HIV-antibody-negative women 18 to 35 y old who were at risk of HIV infection by virtue of having an average of three or more coital acts per week and four or more sexual partners per month. Participants had to be willing to use the study drug as directed and participate for up to 12 mo of follow-up. Because TDF has been associated with rare episodes of renal disorders, increased liver enzymes, and hypophosphatemia, participants were also required to have adequate renal function (serum creatinine 2.0 mg/dl) for renal function, grade 3 or 4 AST or ALT elevations (>170 U/l) for hepatic function, and grade 3 or 4 phosphorus abnormalities ( 170 U/l) ALT or AST elevations before product withdrawal, whereas two and three of the 368 participants in the placebo group had grade 3+ ALT and AST elevations, respectively. The percentage of grade 1 or higher (>42 U/l) ALT and AST abnormalities was greater in the TDF group, but the difference did not achieve statistical significance. One participant in the TDF group had a grade 3+ decrease in phosphorus ( 2.0 mg/dl). We did not find significant differences in laboratory abnormalities between treatment groups when the data were stratified by site, although significantly more grade 1 AST and phosphorus abnormalities occurred in Ghana than in Cameroon, and significantly more grade 1 creatinine abnormalities occurred in Cameroon than in Ghana. Among the 56 participants who tested positive for HBsAg, 23 were in the TDF group and 33 in the placebo group. The mean and median ALT and AST levels were not significantly different between groups immediately before or after discontinuation of study drug. Four ALT/AST abnormalities (none over grade 1 [>42 U/l]) occurred within 3 mo after discontinuation of study drug in HBsAg-positive participants; one was in the TDF group and three were in the placebo group. In Cameroon, study drug was stopped before the implementation of the protocol amendment that included HBsAg testing. We therefore monitored liver function for several months after product withdrawal in all participants, regardless of HBV status. The mean and median ALT and AST levels were not significantly different between groups immediately before or after discontinuation of study drug. Twenty ALT/AST abnormalities occurred within 3 mo after discontinuation of study drug; 14 were in the TDF group and six were in the placebo group. With the exception of one, all were grade 1 or 2 events (i.e., less than 85 U/l). One participant (who received TDF) had a grade 3 AST abnormality, which resolved within 1 mo. Adverse events. A total of 320 (75%) women in the TDF group and 310 (72%) women in the placebo group had at least one AE. The most frequently reported AEs (occurring in ≥5% of participants in either treatment group) are summarized in Table 3. There were no significant differences between treatment groups for any AEs within system organ classes. Similar results were obtained among HBsAg-positive participants. Twenty–two SAEs were reported during the study (nine in the TDF group and 13 in the placebo group) in 17 participants. The majority of the SAEs were hospitalizations due to malaria (nine events). No SAEs were considered related to study drug. Two deaths occurred in Cameroon during the course of the study; both occurred approximately 5 mo after study drug dispensation was suspended. One was due to an unspecified condition with anemia (the participant had been randomized to placebo), and one was suspected to be related to an induced abortion (the participant had been randomized to TDF). Study drug was not discontinued in any participant by the site investigator or study clinician because of an AE or abnormal laboratory result. Effectiveness: HIV Incidence For the primary effectiveness analysis, women in Cameroon, Ghana, and Nigeria contributed 232.6 person-years of follow-up in the TDF group and 241.3 person-years in the placebo group. Eight seroconversions occurred among participants while receiving the study drug. Two HIV infections were diagnosed in participants randomized to TDF (rate = 0.86 per 100 person-years) and six in participants receiving placebo (rate = 2.48 per 100 person-years), yielding a rate ratio of 0.35 (95% confidence interval = 0.03–1.93; p = 0.24). With the exception of two participants (one randomized to TDF and one to placebo), all seroconversions were detected on or after the second monthly follow-up visit. Blood specimens were available from one of the two participants who seroconverted while on TDF; standard genotypic analysis revealed no evidence of drug resistance mutations. An additional six participants seroconverted after study drug was discontinued in Cameroon (four had been randomized to TDF and two to placebo). DISCUSSION Interpretation and Overall Evidence Our data provide an encouraging rationale for additional research to evaluate oral antiretroviral drugs as prophylaxis against HIV infection. No significant differences in safety patterns occurred among participants receiving daily oral TDF compared with those receiving placebo, consistent with results seen in previous treatment studies [2]. No randomized clinical studies have been completed to evaluate the effectiveness of antiretroviral drugs as pre- or post-exposure prophylaxis against HIV infection. In a recent Cochrane review [20], only one case-control study of health-care workers after needlestick injury was identified. HIV-infected workers had significantly lower odds of having taken zidovudine prophylaxis after exposure than those who did not seroconvert (odds ratio = 0.19, 95% confidence interval = 0.06–0.52) [21]. The effectiveness of antiviral prophylaxis after sexual exposure is not known [22]. Non-randomized observations of post-exposure prophylaxis (PEP) are difficult to interpret because there could be underlying and uncontrolled differences between those who seek and use PEP and those who do not, including differences in behavior and access to information and medications. Furthermore, PEP or event-driven dosing is limited by the fact that failures can occur when treatment is not initiated because the risk of the exposure is not recognized [23]. We address this limitation of event-driven dosing in this study by recommending daily dosing for frequently exposed persons—a strategy referred to as pre-exposure prophylaxis. Because public health practice should be guided by evidence, especially in settings having competing demands for scarce public health resources, an urgent need exists for antiviral prophylaxis for HIV to be evaluated in randomized, placebo-controlled trials, as we report here. Further effectiveness studies in populations of women at high risk for acquiring HIV should proceed rapidly. The premature stopping of the study in Cameroon and Nigeria limited the amount of follow-up safety and effectiveness data obtained in this study. Furthermore, AEs and laboratory abnormalities in the TDF group may have been diluted by lower than expected product use due to missed visits, drug stoppage due to pregnancy, and other reasons for non-use of study drug. The overall rate of HIV infection while women were on TDF or placebo in Ghana, Cameroon, and Nigeria was too low to demonstrate a reduction in risk for those assigned to the TDF group. TDF is active against HBV, and is recommended for treatment of HBV infection in Europe and by many experts [24–27]. Transaminase increases have been observed in up to 25% of patients stopping anti-HBV drugs after receiving therapy for clinically important HBV infection, characterized by pre-treatment elevated ALT or AST or signs of liver fibrosis [28]. No flares of ALT or AST were observed among those with HBV infection in this study, although our analysis was limited to 23 TDF-treated women with reactive tests for HBsAg. The rate of HBV flares after withdrawing TDF may be low among people with normal baseline liver tests and no signs or symptoms of advanced liver disease, such as the women enrolled here. Additional data on the use of TDF in persons with circulating HBsAg are needed to confirm the results observed in this study. We expected the HIV incidence in the placebo group to be no less than five per 100 person-years, over twice that we observed in this study. Thus, our power was less than anticipated. The lower than expected HIV incidence may be due in part to at least four factors: (1) the incidence rate was estimated from both our experience in earlier trials in a similar population and from current prevalence data; it was not specifically measured in each population before starting the study; (2) intense and consistent HIV/sexually transmitted infection prevention services and messages were provided to the study participants during the trial; (3) the effect of taking a pill every day for HIV prevention may have served as a timely reminder of imminent HIV risk such that participants modified their behavior to incorporate precautions against infection such as increased use of condoms; and (4) participants who elect to join clinical trials may be more inclined to safer behavior than those in their community who do not participate. Indeed, reductions of risk behavior have been observed in open label studies of PEP [23,29]. Generalizability As a new HIV prevention approach, prophylactic use of TDF could be used with other prevention strategies such as condoms to reduce the number of people who become infected with HIV. Larger phase 3 studies to conclusively determine the safety, effectiveness, and feasibility of using TDF (either alone or in combination with other antiretrovirals) as chemoprophylaxis against HIV infection in both women and men are needed. SUPPORTING INFORMATION CONSORT Checklist Click here for additional data file. Original Trial Protocol Click here for additional data file. Amended Trial Protocol Click here for additional data file.
