Microbicide reduces HIV infections by 30% in first success for field

This article is more than 15 years old. Click here for more recent articles on this topic

Campaigners were celebrating the results of a trial of a microbicide to prevent HIV that has produced a positive result, the first one to do so. The results of the HPTN 035 were announced at the Sixteenth Conference on Retroviruses and Opportunistic Infections in Montreal today.

Microbicides are chemicals that can be incorporated into gels, foams or devices that can be used in the vagina (and potentially the rectum) to prevent HIV transmission during sex. A number of previous studies in women in sub-Saharan Africa have failed to show a protective effect of microbicide candidates, including Carraguard and cellulose sulphate (UsherCell).

The study reported today, conducted by the US-funded Microbicide Trials Network, used the candidate microbicide PRO 2000 (polynaphthalene sulphonate) and involved 3099 women in Malawi, South Africa, Zambia, Zimbabwe and the USA.



A product (such as a gel or cream) that is being tested in HIV prevention research. It could be applied topically to genital surfaces to prevent or reduce the transmission of HIV during sexual intercourse. Microbicides might also take other forms, including films, suppositories, and slow-releasing sponges or vaginal rings.


A pill or liquid which looks and tastes exactly like a real drug, but contains no active substance.


How well something works (in a research study). See also ‘effectiveness’.

statistical significance

Statistical tests are used to judge whether the results of a study could be due to chance and would not be confirmed if the study was repeated. If result is probably not due to chance, the results are ‘statistically significant’. 


The last part of the large intestine just above the anus.

PRO 2000 prevented about a third of potential infections in women who used PRO2000, compared with women who used a placebo gel or no gel at all. These results were not statistically significant. This means that there was a one-in-ten probability that the 30% reduction in HIV infections seen in women who used PRO 2000, versus those who use a placebo, was due to chance (p = 0.1).

In the case of women who used no gel, this probability was one in 17 (p = 0.06). Scientifically, results are not regarded as ‘significant’ unless the chance that they are wrong is less than one in 20 (p = 0.05).

Investigators could therefore only call the microbicide ‘promising’ rather than ‘effective’. Principal investigator Professor Salim Abdool Karim of the University of Kwazulu Natal, South Africa, told a press conference: “We cannot reach the definitive conclusion that PRO 2000 is a microbicide, but it is a promising candidate”.

A larger study of PRO 2000, the UK-funded MDP 301 study, is due to announce results later this year and will be able to demonstrate efficacy if they find similar reductions in the HIV infection rate.

The fact that, despite these modest results, microbicide campaigners were buoyed up by the result is a measure of how disappointing results of trials in the last few years have been, with some microbicide candidates (such as cellulose sulphate) actually causing harm and others (such as Carraguard) proving to be ineffective.

"The results on PRO 2000 are a ray of hope for women," observed Lori Heise, Director of the Global Campaign for Microbicides (GCM). "It's not a home run, but this ‘proof of concept’ should invigorate the field."

There were four arms to the study. The first arm involved women who received 0.5% PRO 2000 gel. In the second, women were treated with BufferGel, a second microbicide candidate with a different mode of action. In the third and fourth arms women received an inert placebo gel or no gel at all. The gel arms were blinded so women did not know which product they were using. Monthly follow-up visits assessed pregnancy and safety and every three months the women were tested for HIV.

Average follow-up was 20 months. No significant side-effects were reported. Women reported using the gels 81% of the time they had sex. Condom use was 72% amongst the women who received either the treatment or placebo gels and 81% amongst women in the no gel arm, a statistically significant difference (p

Amongst the women who received PRO 2000 there were 36 HIV infections and an incidence rate of 2.7 infections per 100 women a year. This compared with around 4% a year in the other three arms (so BufferGel was not effective).

Women spent an average of 20 months in the trial but spent 6% of this time not using the trial products due to reasons like pregnancy. When the investigators performed a second set of analyses that excluded this time off treatment, they found that PRO 2000 provided a significant 36% reduction in the risk of HIV infection compared to no gel (p = 0.04).

The investigators performed an additional set of analyses that compared efficacy in women with higher-than-average levels of gel use versus those with lower-than-average levels. Higher than average use was defined as above the median self-reported level of use (>81%). This analysis found that the more often women used the gel, the higher the level of protection against HIV: there was an annual incidence of 2.4% in high gel users randomised to PRO 2000 compared with 4.25 in those using placebo and 44% of HIV infections were prevented in high gel users versus 9% in low gel users.

Because women who used high levels of gel might also be frequent condom users, the investigators then compared infection levels in women who used high levels of both condoms and gel versus women who used few condoms but often used the gel. In this latter group, whose protection against HIV solely or mainly consisted of gel, there were three infections in 299 women who used PRO 2000 (incidence rate 1%) and 15 infections in 324 who used placebo (incidence rate 4.6%). This meant that 78% of infections in non-condom users appeared to be stopped by PRO 2000.

If the MDP study also finds an efficacy of 30% for PRO 2000, this poses awkward questions as to whether the product should be developed and promoted for use as the microbicide's effectiveness could be cancelled out by reductions in the use of condoms, whose efficacy is 80 to 90%.

Professor Abdool Karim commented: “This may be a niche product for women with no other choices. For a woman with no other option – who is say faithful and trying to get pregnant with a high-risk migrant husband – what other options are there?”


Abdool Karim S et al. Safety and effectiveness of vaginal microbicides BufferGel and 0.5% PRO 2000/5 Gel for the prevention of HIV infection in women: results of the HPTN 035 trial. Sixteenth Conference on Retroviruses and Opportunistic Infections, Montreal, abstract 48LB, 2009.