Women in sub-Saharan Africa are disproportionately affected by HIV for biological and sociocultural reasons. Many are unable to negotiate condoms, or unwilling to do so as they wish to conceive, or both. For two decades scientists have been searching for a method to prevent HIV that women can initiate and be responsible for. The female condom addresses some of these needs, but prevents conception. Delivering drugs locally to the genital tract in the form of a vaginal microbicide has long been envisaged as a way to fill the gap, and in 2010 there is finally proof of concept for this method, with the report of a significant reduction in HIV infections in women using tenofovir 1% vaginal gel compared with those using placebo Citation[1].
Lessons learned from efficacy trials
From the 11 trials that preceded the Center for the AIDS Program of Research in South Africa (CAPRISA) 004 study, we learnt that HIV microbicide trials are difficult to execute, but can be completed to international standards in resource-limited settings Citation[2–4]. An accurate estimate of HIV incidence in the target populations is essential to avoid an underpowered trial Citation[5], and intense efforts to trace participants that miss visits are required in order to achieve satisfactory levels of retention. In contrast to vaccine and circumcision trials, microbicides are user dependent and so adherence to gel use needs to be high in order to minimize dilution of the size of the effect observed. Self-reported adherence was high in all the trials. Although there will always be skepticism surrounding the accuracy of these reports, the positive feedback provided through qualitative research suggests that gel is highly acceptable to women and their partners Citation[6,7].
Seven products that reduced HIV replication in preclinical experiments, including ex vivo cervical tissue, did not provide benefit in clinical trials. These were surfactants, polymers or acidifying agents, and even at high concentrations fell far short of the potency demonstrated in the same assays by much lower concentrations of antiretroviral (ARV) drugs, such as tenofovir and dapivirine. Now there is evidence of clinical protection with tenofovir gel, this provides guidance on the level of activity required in preclinical experiments to justify clinical development.
What have we learnt from CAPRISA 004?
The pharmacokinetic and pharmacodynamic studies were not completed in May 2007 when CAPRISA 004 started, leading to criticism regarding the decision to proceed and the choice of dosing schedule Citation[8]. The coitally dependent regimen selected was complex. ‘BAT24’ required women to apply gel within 12 h before sex, to apply a second dose up to 12 h after sex, or as soon as possible after sex if they had not applied a dose before, but no more than two doses in a 24 h period. This two-dose coital strategy was based on the reduction in HIV transmission from mother to child following a single dose of nevirapine for the mother with onset of labor, and a second dose for the child following delivery. In addition, early macaque experiments dosed with oral tenofovir and challenged intravenously suggested that the postexposure dose was important.
At the fifth World AIDS conference in Vienna, 2010, the CAPRISA 004 investigators reported a 39% reduction in HIV incidence (incidence rate ratio = 0.61; 95% CI: 0.40–0.94; p = 0.017) and received a standing ovation Citation[1]. This provided proof of concept for microbicides, ARV prophylaxis, and somewhat unexpectedly, protection against herpes simplex virus-2 transmission. The data presented were consistent across the duration of follow-up, and supported by the higher levels of protection against HIV observed in women who reported adherence greater than 80% (54%), and in women who were high gel users, returning greater than the median number of used applicators (47%). Although unexpected, the 51% protection against herpes simplex virus-2 is biologically plausible as tenofovir shares molecular parentage with cidofovir, a drug used in the treatment of herpes simplex infection. A preliminary pharmacokinetic analysis revealed a compelling inverse relationship between the number of HIV sero-conversions and the concentration of tenofovir detected in a selected sample of cervico–vaginal lavage specimens collected during the trial.
The consistency across analyses and supporting pharmacology increase confidence in the conclusion that tenofovir 1% vaginal gel, administered in a coitally dependent two-dose strategy, is biologically efficacious. Whilst the criticisms at the launch of CAPRISA 004 may have been valid, this clinical evidence of protection advances the efforts to put products into women‘s hands by the 3 years it has taken to complete the supporting studies.
What next for microbicides?
A single positive result in a single trial population would need to reduce HIV by 50% or more, and have a p-value of less than 0.001 to be sufficiently outstanding to support licensure. A 39% reduction in HIV with a p-value of 0.017, and wide confidence intervals that permit a true level of protection as low as 6%, does not qualify as outstanding. Further placebo-controlled evidence is required.
The US NIH-funded Microbicides Trial Network‘s (MTN‘s) 003 Vaginal and Oral Interventions to Control the Epidemic (VOICE) trial is currently enrolling and will report on a daily dosing strategy of tenofovir 1% gel compared with placebo in 2013. In this trial, approximately 5000 women from four countries will be randomized to one of five groups:
Oral tenofovir
Oral truvada
Oral placebo
Vaginal tenofovir 1% gel
Vaginal placebo gel
Frequent HIV and pregnancy testing and routine laboratory parameters are being collected and there are safety substudies to assess bone mineral density and renal toxicity. As well as the extensive data on the safety of tenofovir gel at higher levels of exposure in broader populations than CAPRISA 004, VOICE will report on the effectiveness of daily gel, and the relative effectiveness of vaginal and oral dosing. Higher concentrations of tenofovir are achieved in the genital tract after vaginal application compared with oral dosing, and it will be interesting to see if this translates to a greater level of protection in women using gel.
If VOICE observes a higher level of protection than 39%, this could be explained either by a biological advantage or by greater adherence. Daily dosing should result in drug levels being maintained in the genital tissue, even if one or two doses are missed. However, it is also reasonable to expect adherence to gel to be greater in the VOICE study populations as participants now know the result of CAPRISA 004. Together, the data from CAPRISA 004 and VOICE studies may be sufficient for licensure. The US FDA has provided the opinion that VOICE could serve as the second trial providing pivotal data to support licensure, and the agency has agreed to conduct an accelerated review when data from both trials are available. A final decision and, importantly, determination of the instructions for use, must await this review, which is projected to take place in 2014. There is a risk that the two trials will prove insufficient, and to mitigate against this risk, a consortium of investigators is proposing to repeat the BAT24 comparison to placebo gel in broader populations within South Africa, including 16–17-year-olds, and aims to report in 2013, subject to funding.
An important issue to understand is why BAT24 did not provide greater protection – was it due to poor adherence or inadequate biological efficacy? The used applicators that were returned by participants fell over the duration of the CAPRISA 004 trial, which could be due to a fall in adherence, and would explain the declining level of protection observed from 50% at 12 months to 39% at 30 months.
Research completed in parallel to CAPRISA 004 has confirmed rapid absorption of tenofovir 1% vaginal gel, which reached much higher levels of drug in the genital tissues compared with oral dosing, sustained over at least 24 h in sexually abstinent women Citation[9]. Furthermore, repeated cycles of single-dose tenofovir 1% gel administered 30 min prior to low-dose intravaginal challenge protected all macaques Citation[10]. The correlation between declining drug levels at intervals after dosing in macaque tissue lymphocytes and breakthrough infections with intravaginal challenge provides insight into the ideal timing of gel administration Citation[11]. Taken together, these experiments in women and macaques suggest that it is not possible to improve on single dose applied prior to sex, and imply that the dose after sex is redundant.
It is critical to know whether one dose applied within a short time before sex does protect women, as it is very likely that women will use gel this way, for convenience and in order to preserve gel supplies.
A simple regimen of one dose prior to sex would be easier and cheaper to roll-out than the two-dose coital strategy. However, one dose may not be potent enough to provide benefit in women. Knowing this, women are far more likely to adhere to the two-dose strategy, however inconvenient. The most efficient and accurate way to assess the relative benefit of the one- or two-dose coital strategies is to compare each to placebo in a parallel group design in the same study population. The Microbicides Development Program (MDP) investigators are seeking funding to conduct this trial in four countries, including three that are not participating in the VOICE trial, and to include 16–17-year olds. The trial is designed to have sufficient power to detect small differences in adherence between the two dosing strategies. The hypothesis is that adherence to the one-dose strategy will be significantly greater. This may or may not predict effectiveness. If it does, we know adherence is the most likely explanation for the lower than expected protection seen in CAPRISA 004; if greater adherence is not predictive of protection then we will be more confident that the way forward is to find more potent and/or longer-acting agents.
The next steps for ARV prophylaxis
The WHO/Joint United Nations Program on HIV/AIDS convened a meeting in August 2010 to discuss the next steps for tenofovir gel, bringing together donors, regulators, researchers and advocates. There was general consensus that the window for placebo-controlled microbicide trials was open, but likely to close after VOICE reports. The agenda for the meeting focused on tenofovir, and the funding gap was estimated to be US$100 million, $40 million for each of the two proposed efficacy trials described above, and $20 million for further studies in the CAPRISA communities supplying open-label gel to HIV-noninfected participants, and following HIV-infected participants for response to ARV and emergence of resistance. The donor agencies that were present expressed their reluctance to fund more than one trial, given the current economic climate. Furthermore, the dapivirine ring could enter efficacy trials next year, subject to funding, and cannot be excluded from decisions regarding the immediate ‘next steps’ for ARV microbicides.
Further support for daily use of ARV microbicides is expected from the five trials of daily oral dosing assessing tenofovir or truvada in the prevention of HIV acquisition, in diverse populations. The first of these, the Pre-exposure Prophylaxis Initiative (iPrEx), reported on 23 November 2010 a 43.8% reduction in HIV incidence (95% CI: 15.4–62.6) in men who have sex with men allocated to truvada tablets compared with placebo. An analysis of drug levels in the truvada arm revealed detectable drug in only three out of 34 (9%) seroconvertors compared with 22 out of 43 (51%) HIV-noninfected controls matched for clinical center, suggesting that poor adherence could explain the lower than expected protection Citation[12].
These are exciting times, and a prevention method that women can initiate and control looks very likely. However, the critical path is not yet clear, and there is a grave danger that tenofovir gel will be the first licensed microbicide in 2014, but the evidence to inform and justify roll out in programs is insufficient. With four further trials to report, the strength of evidence will inevitably favor daily oral ARV prophylaxis, but these regimens are more costly as the drugs are still on patent and carry more risks in terms of adverse events and the emergence of resistant virus. Furthermore, distributing ARVs to the large pool of HIV negative individuals at risk to prevent acquisition, will not compare favorably to strategies to scale-up testing and offer immediate ART to all HIV-infected individuals to control viral load and onward transmission. There is an opportunity for the microbicide field to determine and agree the best use of scarce resources to ensure that HIV-noninfected women have access and a choice of products they can protect themselves with. This is an opportunity that will not last long.
Financial & competing interests disclosure
Sheena McCormack is a Co-Principal Investigator of the Microbicides Development Programme. The author has no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.
No writing assistance was utilized in the production of this manuscript.
Additional information
Funding
Bibliography
- Abdool Karim Q , Abdool KarimSS, FrohlichJAet al.: Effectiveness and safety of tenofovir gel, an antiretroviral microbicide, for the prevention of HIV infection in women. Science3, 329(5996), 1168–1174(2010).
- McCormack S , RamjeeG, KamaliAet al.: PRO2000 vaginal gel for prevention of HIV-1 infection (Microbicides Development Programme 301): a Phase 3, randomised, double-blind, parallel-group trial.Lancet16, 376(9749), 1329–1337(2010).
- Abdool Karim SS , ColettiA, RichardsonBet al.: Safety and effectiveness of vaginal microbicides buffer gel and 0.5% PRO 2000/5 gel for the prevention of HIV infection in women: results of the HPTN 035 trial. Presented at: 16th Conference on Retroviruses and Opportunistic Infections. Montreal, Canada, 8–11 February 2009.
- Skoler-Karpoff S , RamjeeG, AhmedKet al.: Efficacy of Carraguard for prevention of HIV infection in women in South Africa: a randomised, double-blind, placebo-controlled trial.Lancet6, 372(9654), 1977–1987(2008).
- Feldblum PJ , AdeigaA, BakareRet al.: SAVVY vaginal gel (C31G) for prevention of HIV infection: a randomized controlled trial in Nigeria.PloS ONE3(1), E1474 (2008).
- Pool R , MontgomeryCM, MorarNSet al.: A mixed methods and triangulation model for increasing the accuracy of adherence and sexual behaviour data: the Microbicides Development Programme.PLoS ONE23, 3(1), E1474 (2008).
- Pool R , MontgomeryCM, MorarNSet al.: Assessing the accuracy of adherence and sexual behaviour data in the MDP301 vaginal microbicides trial using a mixed methods and triangulation model.PLoS One5(7), E11632 (2010).
- Check E : HIV trial doomed by design, say critics.Nature12, 448(7150), 110–111(2007).
- Schwartz J , KashubaA, RezkNet al.: Preliminary results from a pharmacokinetic study of the candidate vaginal microbicide agent 1% tenofovir gel. Presented at: Microbicides 2008. New Delhi, India, 24–27 February 2008 (Abstract BO11–61).
- Parikh UM , DobardC, SharmaSet al.: Complete protection from repeated vaginal simian-human immunodeficiency virus exposures in macaques by a topical gel containing tenofovir alone or with emtricitabine.J. Virol.83(20), 10358–10365(2009).
- Dobard CW , SharmaS, HolderAet al.: Protection by TFV gel against vaginal SHIV infection in macaques three days after gel application and its relationship to tissue drug levels. Presented at: 5th International Workshop on HIV Transmission. Vienna, Austria, 15–16 July 2010.
- Grant R , LamaJ, AndersonPet al.: Preexposure chemoprophylaxis for HIV prevention in men who have sex with men.N. Engl. J. Med. doi: 10.1056/NEJMoa1011205 (2010) (Epub ahead of print).