Dashed hopes: A promising AIDS prevention tool has proven to increase the risk of HIV infection. Credit: Associated Press

AIDS researchers delivered a stunning blow to prevention efforts for the disease, announcing on 31 January that the microbicide Ushercell, a promising prevention tool, instead increases the rate of infection with HIV.

Ushercell's failure, the second time a microbicide has met this unwelcome fate, has come as a bigger shock to the scientists themselves—and illustrates the complex challenges scientists encounter in the final stages of microbicide research.

In earlier trials, Ushercell seemed both safe and effective. But in a large phase 3 trial of 1,333 sexually active women in South Africa, Benin, Uganda and India, the researchers discovered that more women using the gel had become infected with HIV than those using the placebo, and swiftly put an early end to the trial.

We have no idea what happened, we are totally baffled. Roshini Govinden, trial investigator

“We have no idea what happened, we are totally baffled,” says Roshini Govinden, lead researcher for the trial's South African arm.

Ushercell, a vaginal gel aimed at protecting women against sexually contracted HIV, acts as an entry inhibitor: its active ingredient, cellulose sulfate, binds to the viral envelope proteins and prevents them from fusing with host cells.

In 2000, trial results of the detergent nonoxynol-9 also showed that product increased the risk of sexually transmitted HIV infection. Several other microbicides have been found in phase 3 trials to be ineffective.

Trials of microbicides pose unique challenges for scientists. In many African cultures, men prefer dry sex, making the women in the trial reluctant to use the gels consistently. As part of the trials, researchers are also required to counsel the participants about condom use, leading some women to use condoms more often than the gel being tested. Counseling also generally lowers the rate of new infections, making it difficult to find a statistical difference between the microbicide and placebo arms.

All these factors, combined with treatment for sexually transmitted infections, make microbicide trials large and expensive. For instance, a phase 3 trial of the Carraguard, a seaweed extract that acts as an entry inhibitor, enrolled nearly 7,000 participants to measure a 33% decrease in HIV incidence.

Neither Carraguard nor the two others still in phase 3 trials are expected to be 100% effective. Ushercell's failure has bolstered skepticism about microbicides generally.

“I'm not at all surprised that Ushercell showed no activity,” says John Moore, an immunologist at Cornell University. The trial's outcome, Moore says, indicates the usefulness of monkey models “to weed out the less useful concepts before they get as far as Ushercell did.”

CONRAD, the US-based research organization that sponsored the trials, has thus far declined to explain the unexpected result. Detailed results are expected by the summer, says Annette Larkin, CONRAD's spokesperson.