A cold virus used to make an experimental HIV vaccine that was discontinued in September somehow may have caused volunteers to be more susceptible to the AIDS virus, the vaccine's developers said on Wednesday.
Researchers were doubly dismayed when it appeared that those who had been vaccinated were more likely to become infected, and cautioned more than 3,000 volunteers who had been testing the vaccine that they may be at higher risk of infection.
Merck and the National Institutes of Health agree that the vaccine itself could not possibly cause HIV infection. But initial data appears to indicate that people who got vaccinated had a higher rate of HIV infection that those who got placebo shots.
Experts are hashing out the confusing findings at a meeting in Seattle. Merck and the NIH say that men with the highest pre-existing levels of immunity to a cold virus called adenovirus 5 were the most likely to have also become infected with HIV after being vaccinated.
That adenovirus was modified to use as a so-called vector or delivery mechanism in making the HIV vaccine.
Dr. Keith Gottesdiener, vice president for clinical research at Merck Research Laboratories, said it is possible an earlier infection with adenovirus 5 primed the immune system to produce more of the cells that HIV attacks.
"That is the leading hypothesis — that the vector causes production of more CD4 cells that the virus could infect," Gottesdiener said in a telephone interview.
AIDS, which has killed 25 million people globally and which currently infects 40 million, attacks immune system cells called CD4 helper T-cells. The infection itself does not kill, but it runs down the immune system so that patients die of other infections or sometimes of cancer.
Attempts to make a vaccine against HIV have flopped because of its unique effects on the immune system. The Merck vaccine was aiming to stimulate production of T-cells called CD8 killer cells, which, it was hoped, would recognize and attack HIV.
Usually, vaccines use a whole or weakened virus to stimulate an immune response. This was too dangerous to do with HIV, so the Merck scientists used pieces of genetic material from the virus instead.
To carry them into the body, they used a virus that infects people easily — in this case, adenovirus 5, one of dozens of common cold viruses.
To test the vaccine, the researchers sought volunteers who had a high risk of HIV infection anyway, such as injecting drug users and men who have sex with men. The volunteers are counseled on ways to protect themselves, such as by using condoms, but some get infected anyway.
When they combed through the numbers, the researchers found that mostly men who had sex with other men became infected after vaccination.
They also noticed that people with more pre-existing adenovirus 5 immunity — meaning they had been infected in the past and mounted a strong immune response — were also more likely to become HIV infected after receiving the vaccine.
"Among the 778 male volunteers who had high levels of pre-existing immunity to adenovirus 5, 21 cases of HIV infection were observed in those who had received vaccine and nine cases of HIV infection were observed in the volunteers who had received placebo," Merck said in a statement.
But it was possible that people who happened to have high levels of adenovirus 5 immunity just behaved differently, having more risky sex, for instance, Gottesdiener said.
"It could be due to some biological phenomenon that we don't even understand yet or, honestly, it still could be due to chance," he said.
The trial, which began in 2004, had enrolled volunteers in the United States, Peru, Brazil, Dominican Republic, Haiti, Jamaica and Australia. A second trial had begun in South Africa earlier this year with 800 volunteers.