Washington

Statisticians are urging caution over intriguing results from clinical trials of an HIV vaccine, which suggest that some ethnic groups may benefit from the drug.

Biotechnology company VaxGen announced the results of its AIDSVAX trials on 24 February. The overall results were disappointing, with only a 3.8% reduction in infection among those who were vaccinated compared with those who received a placebo. But the vaccine shows more positive results for certain ethnic subgroups, the company claims. Among black volunteers, for example, only 2% of those who received AIDSVAX became infected with HIV, compared with 8% of those who received a placebo.

“The subset analysis showed surprising and provocative results in minorities,” says Anthony Fauci, director of the National Institute of Allergy and Infectious Diseases (NIAID).

But of the 5,009 participants in the study, only 314 were black, and some researchers warn against drawing premature conclusions. “Subgroup analysis has to be seen as more exploratory,” says biostatistician Steven Self of the Fred Hutchinson Cancer Research Center in Seattle, Washington. “Over the next few months, the detailed analysis will show whether there is an underlying biology that could explain the results.”

VaxGen, which is based in Brisbane, California, plans to investigate factors such as age, sex and behaviour among the subgroups to see if something other than the vaccine can explain the results. It says that a more detailed analysis of the trial's results should be available by the end of March.

Several other possible vaccines may be less than five years away from large-scale clinical trials. The NIAID is developing two candidates for testing, and drug company Merck, based in Whitehouse Station, New Jersey, is also working on a vaccine.