Tenofovir troubles

Just weeks after research on tenofovir gel, a preventative HIV treatment for women, won a prize from the African Academy of Sciences — as reported in the Manitoban — a follow-up study has been cancelled after failing to demonstrate the microbicide’s effectiveness. The new study found in a regular review that the rate of HIV infection was approximately the same in women taking the gel and those assigned to a placebo.

The original study, called CAPRISA 004, followed 889 women in the KwaZulu-Natal province of South Africa from 2007-10. It found that incidence of HIV in the gel arm of the study was 39 per cent lower than the placebo. Among high adherers, the women who used the gel more than 80 per cent before and after sex, incidence was even lower, by 54 per cent. The husband and wife team who led the study won the Olusegun Obasanjo Prize for their work earlier this month.

But the VOICE (Vaginal and Oral Interventions to Control the Epidemic) trial, which was intended in part to replicate CAPRISA’s results, found no decrease in HIV rates. As is standard practice in epidemiological studies, the VOICE data undergoes periodic review by a Data Safety Monitoring Board (DSMB), an independent ethics panel which alerts the team if there is clear evidence of ineffectiveness or of a danger to the participants.

The DSMB found that HIV incidence between the gel and placebo arms were almost identical — 6.1 per cent for the placebo group and six per cent for the gel group. The gel was considered safe, but not effective, and the board recommended cancelling that arm of the trial. “It didn’t work a little bit. It just flat out didn’t work,” said Sharon Hillier, head of the Microbicide Trials Network that conducted the study. Hillier had previously expressed cautious optimism about the CAPRISA 004 results.

Since the VOICE data will not be available until the entire study is complete — at least a year — the researchers can only speculate why the gel did not work in the meantime. “One has to assume it had something to do with adherence,” said John Moore, an immunologist at Weill Cornell Medical College in New York. “No matter how effective it is, if it stays in the tube, it’s not going to work.”

There are a few important differences between CAPRISA 004 and VOICE. CAPRISA tested a different gel-use regimen, instructing participants to apply the gel within twelve hours before and after sex. VOICE tested daily use. CAPRISA 004 studied a much smaller population than VOICE’s 5,029 participants, and was conducted in the province of South Africa with the highest HIV incidence. VOICE is being conducted at 15 different sites in South Africa, Uganda and Zimbabwe.

Connie Celum, an epidemiologist from the University of Washington, Seattle, points out other possible differences between the two studies, such as sexual behaviours, sexually-transmitted infections, vaginal inflammation, and contraceptive use.

The VOICE study began in 2009 and consisted of five arms of about 1,000 women each. One arm of the study, dealing with an oral tenofovir tablet, was cancelled in September at the DSMB’s recommendation, because it was determined that the study could not show a difference in effectiveness between the tablets and a placebo. This is a situation known in epidemiology as “futility.” The research will continue as scheduled with the two remaining arms — Truvada tablets and placebo.

An additional experiment called FACTS 001 has recently begun to test the effectiveness of tenofovir gel using the same regimen as CAPRISA 004. Another study on less user-dependent microbicide delivery methods, such as a vaginal ring containing dapivirine and dispensing it automatically, is planned.

illustration by s. arden hill