American Foundation for AIDS Research, December 2003
Kristen Kresge

For many years, research on topical microbicides for the prevention of HIV has been overshadowed by the development of drugs to treat the virus and its related opportunistic infections. Topical microbicides are creams or gels with active chemical ingredients that are applied in the vagina to block the transmission of HIV during heterosexual sex. The approval process for chemical substances that prevent HIV has not yet been established.

In the US, new drugs must satisfy the Food and Drug Administration's (FDA) stringent requirements before they can reach people in need. Medicines do not hit the nation's medicine cabinets, or even the shelves at the local CVS, without rigorous inspection. For HIV drugs this process has been clearly defined and expedited based on the emergency of the epidemic. Yet various controversies surrounding this process still exist. Now that 19 approved drugs are available, advocates are arguing for both faster and slower approvals for new HIV medications.

But the raging debates over drug approval seem tame when compared with those regarding microbicides, where no approved agent exists. Now that four candidates are nearing or entering the final phase of clinical research where they are tested for efficacy in thousands of people—phase III development—the FDA is being forced to make decisions on what criteria are required for their approval.

And this decision may not come easily. In late August the FDA brought the issue of microbicide approval before its independent Antiviral Drugs Advisory Committee, the first time that the agency has brought the issue to a public forum. After hours of discussion, many questions lingered, revealing substantial areas of disagreement. The meeting explored issues around the design, size, number, and duration of studies required, and the populations to include in such trials, often with no consensus.

After the limited data from advanced microbicide trials were evaluated and opinions were expressed, the onus fell upon the FDA to articulate a model for bringing a microbicide to market without unnecessary delay, cost, or risk. "I didn't think there was a consensus reached. I did think some interesting ideas were presented and discussed. We're willing to work with sponsors. It's an evolving area and we have to be very flexible. No one has paved the way for the field," said Dr. Debra Birnkrant, director of the Division of Antiviral Products at the FDA.

"It's hard, but we recognize the absolute need for these products," she added. Still the level of disagreement within the FDA advisory panel suggests that the path to approval is largely uncharted.

Urgent Need, Yet Slow Development

Despite a universally recognized need for a prevention method that can be controlled by women without a partner's consent, the road to developing an effective microbicide has been long and arduous. "I really feel badly," said Birnkrant. "I've been working in this field for more than 10 years and it's not progressing very quickly."

Dogging microbicide development from the start has been a lack of funding from the big pharmaceutical companies that bring HIV drugs to market. Viewed as an unprofitable venture, most companies ignored microbicide research, delegating it to government, privately funded organizations, and smaller biotechnology companies.

In these arenas, however, a strong commitment to microbicides exists. Earlier this year, the Bill and Melinda Gates Foundation donated $60 million to accelerate the discovery of an effective microbicide, acknowledging that women at high risk for acquiring HIV—including commercial sex workers—are unable to negotiate with their partners to use condoms consistently in many areas of the world.

In addition to sex workers, married and monogamous women make up a growing population of HIV-infected women, according to Lori Heise, director of the Global Campaign for Microbicides, a microbicide policy and advocacy group. Studies find that condom use with casual partners is reportedly higher than with regular partners, says Heise, likely because married women are trying to become pregnant and therefore choose not to use condoms. For this reason, microbicides that block transmission of the virus but do not act as a contraceptive are especially attractive.

Also, many women in monogamous relationships put themselves at risk, not only for HIV but also for physical violence, if they ask that their male partners use condoms. This request could be misconstrued as a woman's admission of promiscuous behavior, which is highly stigmatized in most cultures. Without the ability to use a condom, these women often contract HIV from their male partners who do not know or do not disclose their HIV status. In the US, only 20,000 new HIV infections are attributed to heterosexual sex per year. In the developing world, there are 16,000 new infections through heterosexual transmission every day.

The majority of these new infections occur in young women. In a single clinic in Hlabisa, South Africa, there was a 51% rate of HIV infection in women between the ages of 20 and 24 in 2001. This has jumped from only 7% in this same age group in 1992.

In light of this ever-expanding epidemic among diverse groups of women, the FDA is now facing pressure to establish just what it will take for a microbicide to come to market. "We need to know that a microbicide, when used, is reducing the risk of HIV. It was very proactive of the FDA to have organized that committee meeting. They now have the opportunity to think about these issues before they're faced with data," said Rosalie Dominik, director of Biostatistics at Family Health International, a leading sponsor of microbicide research. The first obstacle they now face is as simple or complex as the control arms in the phase III trials.

Control(s) Freak

Today's microbicide research is largely influenced by earlier trials with nonoxynol 9 (N-9), a component of spermicides that was evaluated in phase III trials as a vaginal microbicide. In one phase III study with COL-1492 (a version of the N-9 gel), researchers found that N-9 offered no protection against HIV infection amongst female sex workers. In fact there was a significantly lower incidence of HIV in the placebo arm (women who received the vaginal moisturizer Replens instead of N-9) than in the N-9 arm.

This result indicated that N-9 might actually be harmful. Researchers on the study found that the likelihood of becoming HIV infected was twice as high when women developed vaginal lesions, which occurred more often with N-9 use. However in earlier safety studies, the presence of vaginal lesions was the same for N-9 and Replens, and researchers could not rule out the possibility that disruption in the epithelial wall of the vagina was a side effect of a sexually transmitted infection itself. In the end, the study concluded that N-9 not only failed to prevent HIV infection, but was actually harmful. This led to a major reversal in prevention strategies that previously promoted the use of condoms coated with the spermicide N-9.

The problem with using a placebo like Replens is that its effect on HIV transmission—protective, harmful, or neutral—is unknown. A truly inert, or inactive, placebo would ideally have no effect on transmission of HIV. Since the microbicide consists of active chemical ingredients and carrier agents (ingredients that allow the microbicide to be formulated into a cream or gel), an obvious choice for placebo would be just the carrier agents. Then the placebo would be equivalent to the experimental microbicide and safe, but without any activity against the virus.

However, for various reasons that include the consistency and odor of the gel or cream, developers have argued that they cannot formulate a placebo that appears indistinguishable from the microbicide under investigation, but contains only the carrier agents. If the placebo looks, smells, or tastes different from the active microbicide, participants in clinical trials would know that they are receiving something different.

Such complications led researchers and regulatory officials to consider adding a second control arm to future microbicide studies. This arm would be the no-treatment, or condom only arm. Women in this arm would receive neither the microbicide nor the placebo. All trial participants would receive safe sex and condom counseling, but for women in the no-treatment arm this would constitute the only intervention. Adding a no-treatment arm would allow researchers to compare the placebo, as well as the experimental microbicide, with the best preventive method currently available: the condom alone. In theory, the no-treatment arm would reduce the potential to overestimate or underestimate the effectiveness of a microbicide as compared with only placebo.

But addition of a third trial arm introduces a wealth of other complexities. The first obvious difficulty is that you are introducing a third group that is unblinded; participants know they are not getting a microbicide. Those randomly placed in the placebo and microbicide arms will not know which gel they are receiving, but those assigned to receive only condoms will be asked to continue participation in a trial that does not offer them an experimental agent. Keeping participants in this arm will be more challenging. The only previous microbicide study to include a condom-only arm was a trial in Cameroon conducted by Family Health International. This three-arm trial enrolled 1,200 women. Twenty of them were lost to follow-up during the six-month study, and 13 of these were in the no-treatment arm.

Additionally, women who are receiving a microbicide or placebo may share their gel with those in the no-treatment arm, further confounding the results of the study. The success of prevention trials in general depends heavily on the behavior of the volunteers. Aside from sharing gel with those in the no-treatment arm, the results are easily shifted based on other choices made by the volunteers. Relatively small differences across trial arms in the rates of condom use could also drastically swing the results.

In one scenario, the microbicide group might use condoms less frequently because of a potentially false sense of protection, while the no-treatment group could be more adherent to condoms. In this case the microbicide, even if protective against infection, would not look efficacious in comparison with the condom group and could force researchers to abandon the microbicide. Conversely if condoms are used more frequently in the microbicide group, the microbicide would look more efficacious then it actually is, leading researchers to promote use of an ineffective product.

As the hearing demonstrated, these dilemmas in trial design are not easily resolved, and the implementation of a condom-only arm therefore was the subject of heated and lengthy discussion. Proponents for and against argued vehemently over whether the FDA should require it. Dr. Thomas Fleming, a biostatistician from the University of Washington and principal investigator for the statistical and data management center of the HIV Prevention Trials Network (HPTN), was the most vocal supporter of the condom-only arm, saying that it represented a "real world" analysis of how a microbicide will work.

Others claim that this is not even the question we are trying to answer; rather the goal is just to see if the application of a topical gel could block transmission of HIV. "We want a microbicide that works against the virus," said Dr. Zena Stein, Professor of Public Health at Columbia University and co-director of the HIV Center for Clinical and Behavioral Studies. "Adding the extra arm is asking a different kind of question. It's asking if you add a microbicide in the real world would it reduce HIV infection? You can't do two things at once. It came out at the meeting that proving effectiveness has to come first," she added. "Time matters in this field. We have many steps to go and proving the concept must come first. We must do it."

In order to provide this proof of concept, at least one phase III trial is currently scheduled to go forward without a no-treatment group. This study will assess the efficacy of Carraguard, a seaweed-based product capable of killing HIV, being developed by the Population Council. Another phase III trial, which will test the efficacy of a microbicide called Savvy—a detergent-based gel containing the active ingredient C31G—is still in the planning stage. The only trial currently planning to include a no-treatment arm is HPTN 035, a phase IIb study of two active products, BufferGel and PRO2000/5, conducted by Reprotect. BufferGel works by altering the pH in a woman's vagina, making it difficult for HIV to survive, while PRO2000/5 is an HIV entry inhibitor

Initially the HPTN 035 trial was slated as phase III, but in an attempt to satisfy the FDA's requirements for licensure, the condom-only arm was added. Upon addition of this second control group, the trial was scaled down to phase IIb because of cost. Few organizations are able to afford adding a third arm to a large phase III trial, which is already expensive. The National Institutes of Health, which is sponsoring HPTN 035, would not commit the necessary resources to run a large efficacy trial with as many as 2,000 women per group. "Nobody can afford it. It is very difficult to run these trials. If you add a third arm, you go beyond the inherent difficulties," said Stein.

Running several efficacy trials would involve recruiting several thousand women at clinical trial sites around the world. Adding an additional arm to these trials puts an enormous strain on these sites. "Site capacity isn't even good enough now for two-arm trials," warns Dr. Zeda Rosenberg, chief executive officer of the International Partnership for Microbicides, a group promoting microbicide research and access.

The FDA is sympathetic to the high cost and logistics of running such studies, but the nagging question facing the agency is if the effectiveness of a microbicide can be proven to their satisfaction in a trial without an inert placebo. And with no product already on the market, the FDA is forced to make tough decisions when evaluating microbicide trials. The agency does foresee one possible way to avoid the added cost of the condom arm: the development of a "universal placebo." If an inert substance was tested extensively and found to be safe by the FDA, this placebo could be used in all future microbicide studies and the true effect of the microbicide would be elucidated. Then a condom-only arm would be unnecessary, according to Birnkrant. But a universal placebo would not match the physical characteristics (odor, taste, consistency) of every microbicide and would therefore be easily distinguished from the product being evaluated.

Problems like these were unresolved by the advisory board, which could offer the FDA little in terms of agreement on the third trial arm. "In summary, we're not sure," said Roy Gulick, chair of the Antiviral Drugs Advisory Committee at the conclusion of the discussion.

Recording a Win

Another major point addressed by the committee was just how effective a microbicide must be to be considered successful. Microbicide trials generally target an efficacy in the range of 33% to 50%. This means that a microbicide would be considered successful if it reduced the rate of HIV transmission by only 33%. By comparison, male condoms, when properly used, are 90% to 95% effective at blocking HIV transmission. The concern has always been that women using condoms will switch to a less effective microbicide, an idea referred to as condom migration.

And if a condom group becomes a mainstay in future microbicide trials, the FDA implied that for approval a microbicide must fare better than both the placebo and the condom. "It would be highly appreciated if it beat the condom as well," said Birnkrant. "Condoms really aren't used that often, however they are the standard of care."

This troubles many microbicide advocates because it sets a very high bar for success. Few expect that any vaginal microbicide will ever be developed that is as effective as a condom. Several committee members repeatedly reminded the FDA that microbicides are urgently needed where condom use is not practical and, in reality, not a standard of care. The irony is that the FDA is making decisions on a product that is desperately needed outside of the US. For now though, it appears that the FDA is more supportive of trials that include a comparison with condoms.

"They decided to make it a requirement, which I don't think it should be," said Rosenberg. "They could end up ruling out a potentially good microbicide."

Yet others fear just the opposite. After the experience with N-9, many are happy to have the US regulatory committee strictly monitor any microbicide trials to prevent a potentially harmful product from becoming widely available to women around the world. For this reason the FDA also heard arguments on whom to include in clinical trials, how many phase III studies to require, and how long participants should be followed after enrolling in a study. On these issues the committee was able to reach some agreement. Most members of the panel suggested including as many types of women in the trials as would use a microbicide if it were approved. This excludes trials that only enroll commercial sex workers or highly sexually active people, who were often favored for testing the safety of a product intended for frequent use. The panel agreed that even casual microbicide users must be part of any trial.

After other arguments over the length of trials, the panel agreed to follow enrollees for as long as practical, but for no less than one year. Most committee members agreed that requiring two years of follow-up was too difficult. Finally, they also concluded that one large phase III trial that showed a statistically significant efficacy would meet their expectations.

Before any decisions are made though, a microbicide must be successful in preventing infection. "You first have to find the biological effect, and in the most expeditious way," said Dominik. "I think that the meeting should help to speed up the approval of microbicides."

031210
AM031201

Copyright © 2003 by the American Foundation for AIDS Research (amfAR) and first displayed on amfAR's Treatment Directory web site (http://www.amfar.org/gl). They appear on AEGIS with amfAR's permission. Organizations wishing to reprint or redistribute these materials should request authorization from amfAR's Department of Treatment Information Services (212/806-1600).

AEGiS is made possible through unrestricted grants from Boehringer Ingelheim, iMetrikus, Inc., the National Library of Medicine, and donations from users like you. Always watch for outdated information. This article first appeared in 2003. This material is designed to support, not replace, the relationship that exists between you and your doctor.

AEGiS presents published material, reprinted with permission and neither endorses nor opposes any material. All information contained on this website, including information relating to health conditions, products, and treatments, is for informational purposes only. It is often presented in summary or aggregate form. It is not meant to be a substitute for the advice provided by your own physician or other medical professionals. Always discuss treatment options with a doctor who specializes in treating HIV.

Copyright ©1980, 2003. AEGiS. All materials appearing on AEGiS are protected by copyright as a collective work or compilation under U.S. copyright and other laws and are the property of AEGiS, or the party credited as the provider of the content.