The trial, which began this week, will result in people at risk of Aids in 30 American cities being injected with a prototype vaccine made from a synthetic protein designed to mimic the outer coating of HIV.
VaxGen, a private biotechnology company based in San Francisco, has received approval from the US Food and Drug Administration to go ahead with the phase three trial, which is usually the last stage before going on sale. It is the first of more than 40 Aids vaccine trials to get this far in the testing procedure. It will be given to half of the volunteers, who are all considered at risk of HIV infection. The other half will be given a harmless placebo.
Neither the scientists nor the volunteers will know who has been given the vaccine and who has the placebo until the end of the 30-month experiment, when the infection rates of the two groups will be compared to see if the vaccine has prevented HIV transmission.
VaxGen said its vaccine is modelled on an HIV protein called gp120. This protein has been of interest to Aids researchers for nearly 15 years because of the possibility of it being used to make vaccines that will stimulate the body's immune defences to attack HIV.
Scientists this month completed the first detailed, three-dimensional images of the structure of gp120. However, the new findings suggest that the VaxGen vaccine will be unable to stimulate an effective immune response in the volunteers because it is too dissimilar to natural gp120.
Andrew McMichael, professor of immunology at Oxford University and one of Britain's leading experts in Aids vaccine research, said: "I think the leading immunologists and virologists in the field are pretty unanimous in their view. The likelihood that [the vaccine] is not going to do anything is quite high."
Professor McMichael said the reasons why the trial is going ahead are more to do with ``business and politics'' than with science.
The vaccine is supposed to work by stimulating a patient's antibodies to attack the gp120 protein on HIV. However, the natural protein and the synthetic vaccine are too different for this to work with any great effect, he said.
``On theoretical grounds the chances are low. But preliminary trials earlier this year suggested that some vaccinated people still became infected. So really there was quite good negative evidence that this was not an effective vaccine,'' he said.
The reservations about vaccines based on gp120 are shared by dozens of American Aids specialists who recently signed a letter to the journal Science supporting a decision not to begin trials until problems with the vaccine's potency can be resolved.
Attempts to develop an Aids vaccine have met with problems because of the unique nature of HIV. It mutates rapidly, even within one person, so evading the best efforts of a vaccine to stimulate effective immunity.
Albert Sabin, who pioneered the development of a polio vaccine, warned just before be died in 1993 that it would be virtually impossible for any Aids vaccine to work. The virus he said was also able to escape the immune defences by infecting the cells that are sent to destroy it.