One of the most unexpected and exciting results of the CAPRISA trial of a successful microbicide (see aidsmap article: Microbicides can work, CAPRISA trial shows, but more trials needed before approval) was that the gel also reduced the risk of infection with the genital herpes virus, HSV-2, by 51%. In other words, of trial entrants who did not already have HSV-2, women using the gel were half as likely to acquire herpes as women using the placebo.
The link with protection against herpes was only discovered at the end of the trial because, in order to avoid ‘unblinding’ it, herpes serology was not assessed during the trial but only from stored samples when it finished.
There were 29 new HSV-2 infections in women using tenofovir and 58 in women on placebo, yielding annual incidence rates of 9.9% and 20.2% respectively, meaning that tenofovir prevented 51% of infections.
CAPRISA principal investigator Salim Abdool Karim explained that the trial included assessments of herpes infection because tenofovir had a similar molecular structure to drugs used on herpes viruses, such as cidofovir.
Herpes is a global sexual health problem – not only because in some people it causes unpleasant symptoms but because a number of ecological studies have shown that having herpes at least doubles the chance of acquiring HIV. However randomised controlled studies using anti-herpes drugs to prevent people acquiring or transmitting HIV have produced disappointing results.
Herpes infection is extremely common, with 20% of sexually active adults having it globally and 50 to 60% in South Africa, and is the rule rather than the exception in people with HIV, with about 80% of HIV-positive people also having HSV-2.
Ninety per cent of herpes infections are asymptomatic and do not produce lesions. Instead the area near the initial infection site becomes subclinically inflamed: the skin sheds large amounts of herpes viruses and immune cells in the area become highly activated – and therefore both highly susceptible to HIV infection and much more productive of HIV if they are already infected.
In the CAPRISA trial, the HSV-2 infection rate at trial entry was 48.8%. There were 426 women in the trial who did not already have herpes at trial entry and who completed the trial, 202 of whom received the tenofovir gel and 224 the placebo.
There were 29 new HSV-2 infections in women using tenofovir and 58 in women on placebo, yielding annual incidence rates of 9.9% and 20.2% respectively, meaning that tenofovir prevented 51% of infections.
However, there was no evidence of a causative link between herpes and HIV prevention. Nearly half the women in the trial already had herpes at the beginning of the trial and these were not more likely to acquire HIV than women who started herpes-negative. Tenofovir prevented herpes and it prevented HIV infection, but it seemed to do so independently and there was no evidence that preventing infection by one virus caused fewer infections by the other.
However, the trial was not powered to detect such synergistic effects and Salim Abdool Karim could not rule them out.
Safety of tenofovir
There appeared to be no safety or behavioural concerns flagged up by CAPRISA 004, Salim Abdool Karim said.
One concern has been that ARV-containing microbicides might cause resistance, but there was no HIV resistant to tenofovir found in any woman who became infected, and no kidney toxicity (a potential tenofovir-related side effect). There was a somewhat increased rate of mild diarrhoea in women using tenofovir which was statistically significant (17% versus 11%, p= 0.015), of unknown origin.
There were 53 pregnancies in 52 women in the study. Women found to be pregnant were excluded from the study, so the CAPRISA 004 study was unable to establish whether tenofovir gel was safe during pregnancy.
Importantly, there was no deterioration in condom use rates reported by the women during the study. The blinding worked: participants were asked whether they thought they were getting tenofovir gel or the placebo, and exactly 50% of women who thought they were getting tenofovir gel were actually on placebo. Women who thought they were on tenofovir had the same baseline and longitudinal condom use rates as women on placebo.
Watch a webcast on the findings on the Kaiser Family Foundation website.
Watch interviews from global health representatives and NGOs on the Kaiser Family Foundation.
Abstracts and presentations about the trial are available on the official conference website.