The first preliminary human testing of a highly anticipated new kind of AIDS vaccine offers tantalizing hints it may ultimately protect against the killer virus.

The study of Merck & Co.'s experimental vaccine is perhaps the most closely watched experiment in all of AIDS research. The approach seems highly effective in monkeys, and many believe it or something similar is the best bet for a shot that could slow the worldwide epidemic, which has already killed 20 million people and infected 40 million more.

Even though the vaccine is only part-way through first-stage safety testing, Merck's Emilio Emini was asked to update researchers in an address Tuesday at the 9th Annual Retrovirus Conference in Seattle.

His bottom line: At this stage, the vaccine appears to trigger the same immune system response in people that it does in newly immunized monkeys, though the volunteers have not been put to the crucial challenge of exposure to HIV.

"We are encouraged," said Emini, head of Merck's AIDS vaccine program. "Obviously, the big question is how effective this will be in preventing or mitigating infection. That will have to wait until we get into long-term studies."

If all goes perfectly, he said, it will be at least five years and probably longer before the vaccine reaches general use.

However, even if the vaccine works, many experts doubt it will stop infection cold the way most vaccines do. Instead, the hope is to fortify the body's immune defenses to hold the virus in check and prevent disease for many years, perhaps even a lifetime.

The difficulty of developing a vaccine has been one of the biggest frustrations in AIDS research, and experts were enthusiastic Tuesday about the Merck results.

"I would say full speed ahead for this particular research program. These are very encouraging results," said Dr. David Ho, scientific director of the Aaron Diamond AIDS Research Center in New York City.

Dr. Robert Schooley, infectious-disease chief at the University of Colorado, said the vaccine's safety and immune response are "what many of us hoped we'd be seeing when this trial started," though he cautioned that it is still "a long way from here to giving a vaccine that could protect people."

The Merck vaccine and several similar ones nearing human testing use a strategy called prime-boost. It involves first injecting one or more HIV genes, which are taken up by muscle cells and used as blueprints to make viral proteins. This is followed by an immunity booster, usually a harmless virus hollowed out to carry in more copies of the HIV genes.

In monkey experiments, this strategy fails to ward off infection but does keep the animals from dying from AIDS when they are exposed to the virus.

The new vaccines are designed to launch a quick counterattack by blood cells called killer cells against a new HIV infection. This way, HIV will eventually plateau at much lower levels in the bloodstream than typically occur when people catch the virus.

The Merck researchers plan to try the vaccine on about 600 volunteers in the first-stage tests. About 150 have been enrolled so far at 60 sites around the country. The goal is to make sure it is safe and to see if it triggers an immune response that mirrors that in monkeys.

So far, the researchers have tested the prime and the boost stages separately. About 100 people were given injections of the prime, an HIV gene called gag. As expected, the response was modest. About one-third of them showed an immune reaction, measured by their blood cells' production of the hormone interferon.

"You don't necessarily have to see the response for priming to occur," Emini said. "The fact we do see a response at all suggests a fair amount is going on under the radar."

The team is testing escalating doses of the boost stage, which consists of a disabled cold virus crafted to hold another copy of HIV's gag gene. Since virtually everyone has been exposed to ordinary varieties of this microbe, known as an adenovirus, the scientists worry the vaccine will be destroyed immediately by the immune system.

However, the new data suggest that when large doses are given, at least some of the vaccine survives to produce an immune reaction that could — if all works as planned — prepare the body to fight off AIDS.

Eventually the vaccine will be tested in people who are at high risk of catching HIV.

Many experts believe the approach pursued by Merck and several other teams has a better chance of success than do two others that are further along in development, although that assumption remains to be proven.

The AIDSVax vaccine, developed by VaxGen, has already been given to 7,900 volunteers in America, Europe and Thailand. It is a more traditional vaccine, made from the outer wrapper of the AIDS virus. Results are expected later this year.

On Monday, the National Institutes of Health abandoned plans for a large U.S. study of another two-step vaccine approach — Aventis Pasteur's genetically engineered canarypox followed by AIDSVax — but said it will still go ahead with a similar study of the combination in Thailand.