The two-decade-long search for an AIDS vaccine suffered a major blow last week when researchers prematurely halted the much-anticipated trial of a new HIV vaccine after it failed to block or slow down infections.

The so-called STEP trial, sponsored by pharmaceutical giant Merck & Co. and the federally funded HIV Vaccine Trials Network (HVTN), was the first to test the idea of stimulating the immune system's killer T cells to hunt for the virus more aggressively, in this case using a weakened form of the cold virus to carry three genes from HIV.

Researchers did not expect the vaccine to prevent infection, but had hoped that it might hinder the growth of the virus enough to delay the onset of full-blown AIDS and make it harder for an infected individual to transmit HIV to others, creating a stopgap while they searched for a more effective therapy.

They now must await the outcome of the next candidate in the pipeline of more than 30 products in clinical testing and mine the STEP data for clues on ways to develop one that works.

"It was a very promising candidate," says Susan Buchbinder, director of HIV research for the San Francisco Department of Public Health AIDS Office, and STEP protocol co-chair for the HVTN. "Getting this information, while it's disappointing, is really what this trial is designed to do, because [it] will move the field forward when we understand more about why [the vaccine] was not protective."

Conventional vaccines work by triggering the immune system into manufacturing antibodies against an infectious organism, but such a vaccine has proved elusive for the rapidly mutating HIV.

Researchers for the past decade have focused on the T cell approach, based on studies showing that monkeys receiving such vaccines against simian immunodeficiency virus, related to HIV, lived longer or had lower viral levels than usual.

In V520, each of three HIV genes—gag, pol and nef—was inserted into a separate weakened adenovirus, one of the viruses that cause the common cold. Human cells infected by the viruses produced the gene products, giving T cells an advance exposure to them.

The STEP trial began in late 2004 and enrolled 3,000 uninfected people in North and South America and Australia, who will now be monitored but will no receive no additional vaccinations. The HVTN has stopped a second trial of V520 in South Africa.

Final results were set to be publicly released in 2008 or later, but an independent monitoring panel conducted a scheduled review of the vaccine's effectiveness and found that 24 of 741 participants injected with the vaccine had contracted HIV, compared with 21 of 762 given a dummy vaccine, Merck said in a statement last week. The two infected groups had nearly the same levels of virus in the blood.

In the coming months, scientists will have to evaluate the trial's full data to determine why the vaccine failed: Were the infecting viruses too different from the vaccine strain to offer resistance? Did the vaccine not stimulate enough T cells or for long enough?

"This particular vaccine approach didn't work, but there are lots of different vaccine approaches in testing that produce different immune responses," Buchbinder says.

Wayne Koff, senior vice president for research and development at the International AIDS Vaccine Initiative, notes that the adenovirus used in V520 was altered so it couldn't reproduce, which may have limited its effectiveness.

Koff says the next milestone will be later stage clinical testing of a vaccine developed by the federally funded National Institutes of Health Vaccine Research Center, which should begin next year. Also based on an adenovirus, it includes several versions of a fourth HIV gene—env.

Ultimately, researchers believe that a T cell–boosting vaccine by itself will not be enough to stave off AIDS. Although disappointed in the outcome of the STEP trial, Koff says the results will likely give a shot in the arm to a growing effort to identify antibodies capable of neutralizing the virus before it causes a full-blown infection.