Harvard AIDS researchers working with monkeys say the virus overcame an experimental vaccine by changing a single gene, killing one of the animals.
Scientists said the disappointment does not mean that AIDS vaccines are doomed to fail. But it illustrates how the virus remains a relentless opponent that will not be easily defeated or even contained.
HIV already is known to mutate and grow impervious to standard AIDS drugs in at least half of all Americans under care for the infection.
Now researchers have seen a similar outcome with an experimental vaccine that tries to stimulate immune cells to prevent the virus from multiplying. The mutation occurred in only one of eight vaccinated rhesus monkeys in the Harvard experiment.
The findings were published in Thursday's issue of the journal Nature.
"It is sobering to find that a single-point mutation within the virus can initiate a cascade of events resulting in a clinical vaccine failure and death," said Dan H. Barouch, a clinical fellow at the Harvard Medical School and lead author of the study.
More than one dozen experimental vaccines using different genetic strategies have been tested in various laboratories. Some have been successful for more than two years.
Unlike a flu shot, AIDS vaccines do not actually prevent infection by stimulating antibody molecules that neutralize the invading virus. That is because the HIV comes in many strains and it changes rapidly, overwhelming the antibody response.
Instead, the AIDS vaccines work to hold HIV infection in check. The vaccines are made with genes that carry the code for proteins in the virus. When the immune system sees these codes, it stimulates production of a broad class of virus-fighting cells known as cytotoxic T lymphocytes, or killer T cells.
Details of one such vaccine program developed by Merck & Co. were published in Nature along with the Harvard report. Company researchers discussed their progress at the first-ever AIDS vaccine meeting in September, and they have begun testing the vaccine in people.
In their animal study, the Merck researchers used the simian version of HIV. They said their best results were obtained by loading an SHIV gene for a protein known as gag into a genetically engineered cold virus. It stimulated the monkeys' immune systems to generate different types of killer T cells.
The Merck researchers subsequently injected lethal doses of SHIV into the animals. They said the vaccine has protected the animals and they have not seen signs of the Harvard problem, known as "gene-escape."
"We've gone 500 days and not seen any escape," said molecular biologist Emilio Emini, director of the Merck trials. "It's something to watch for. But if the vaccine elicits a sufficiently broad genetic response, this is going to be an issue that we can deal with."
The Harvard trial took a different approach. Rather than loading the genetic material from SHIV onto a bioengineered cold virus, the Harvard researchers injected gag DNA plus an immune cell growth factor directly into the animals. That technique known as a "naked DNA" vaccine.
Seven of Harvard's vaccinated animals have remained healthy for two years, even after injected with SHIV.
But the eighth vaccinated monkey, known as No. 798, responded differently. Six months into the trial, molecular studies showed the virus had started mutating the gag gene in the vaccinated monkey.
Within eight months, virus levels soared within No. 798 and mutated virus was rapidly replacing the original virus. Levels of killer T cell levels plummeted. No. 798 died of AIDS-related complications one year after vaccination.
Federal scientists who reviewed both the Harvard and Merck studies said SHIV strain used in the trials is very aggressive and hard to control. It may not reflect how vaccines will perform in humans, where HIV infection is more gradual.
Developing an AIDS vaccine will be "an uphill grade for the foreseeable future," Jeffery Lifson of the National Cancer Institute and Malcolm Martin of the National Institutes of Health wrote in an accompanying editorial.