Cellulose sulfate (CS) is a cotton-based compound that was formulated as a clear gel for clinical studies. A CONRAD-initiated Phase II trial of CS in 200 couples showed that CS functioned well as a contraceptive, with a six-month typical use pregnancy probability of 13.4% and a “perfect use” rate of 3.9%.
In vitro, CS also helps prevent HIV from entering the target cells. CS inhibits cell entry and multiplication of other sexually transmitted infection pathogens such as herpes virus, papilloma virus, and chlamydia.
Scientists scrutinized the safety data from 11 clinical studies conducted on CS in Africa, Europe, India and the United States. None of the data suggested that the product could increase the susceptibility to HIV and, in fact, results from the contraceptive trial suggested that it could possibly prevent infection.
Two Phase III trials of CS as a microbicide began in 2006, one in Nigeria conducted by Family Health International (FHI) and the other in Benin, South Africa, Uganda and India. A planned interim analysis of the CONRAD study data, presented to an independent data monitoring committee, showed that CS might be contributing to an increased risk of HIV infection. Although the number of seroconversions was small, the group in the CS arm of the study contracted HIV at higher rates than did those in the placebo arm, and therefore the trial was stopped. FHI subsequently decided to close its CS study in Nigeria due to the safety concerns raised in the CONRAD trial, even though the interim results from the FHI study did not show an increased risk of HIV infection among women using CS. Final results, which were presented at the International AIDS Society conference in Sydney, Australia, showed that the difference in HIV seroconversions between the CS group and the placebo group was not statistically significant.
Following the halt of the study, scientists from CONRAD conducted a study of CS in two groups of six monkeys each to try to understand why the microbicide was not effective in women. A placebo was used in one group, while 6% CS gel was used in the second group. Although 5/6 macaques were infected in the placebo group, none of the animals in the CS gel group seroconverted. These data suggest that consistent vaginal application of CS gel effectively reduced the infectivity of the virus, significantly decreasing the rate of vaginal mucosal SHIV transmission following repeated virus exposures.
Why does CS have a protective effect in monkeys and not in humans? There is not a clear explanation at this point. However, certain hypotheses are plausible. Women may not have consistently used the gel with every act of intercourse. They used the gels more frequently than in the monkey study, possibly irritating the vaginal mucosa or changing its microflora. Although scientifically sound, the monkey model has not been clinically validated yet and may differ from the actual human conditions of transmission in a way that affects its predictive power.
In spite of extensive preclinical characterization proving CS safety and efficacy, the CS-containing gel tested in Phase III clinical trials failed to effectively protect women from acquiring HIV infection. Comprehensive additional research is currently ongoing to try to verify hypotheses about the causes for such failure. The need for better, more predictive preclinical models and clinical biomarkers of microbicide safety and efficacy is evident.