The long search for an AIDS vaccine has produced countless false starts and repeated failed trials, casting once bright hopes into shadows of disenchantment. The now familiar swings appeared in high relief this past fall, with news of the most recent, phase III trial in Thailand. Initial fanfare for a protective outcome gave way to disappointment after reanalysis showed that the protection could be attributed only to chance. But rather than dashing all hopes for an AIDS vaccine, the trial has heartened some researchers, who see new clues in the battle against the fatal illness.

Costing $105 million and enrolling more than 16,000 subjects, the Thai clinical trial was the largest AIDS vaccine test to date. It began in 2003, and early results released in September showed a slim but statistically sound benefit from the vaccine (a series of inoculations with drugs known as ALVAC-HIV and AIDSVAX B/E). But in October the full report, with various statistical analyses, was released in a Paris meeting to greater skepticism. Specifically, 74 people who had received the placebo became infected with HIV in the trial period, compared with the 51 people who became infected after receiving the vaccine, which makes for a protective effect of 31.2 percent. By including, however, the seven people who turned out to have had HIV at the start of the trial (two in the placebo group and five in the vaccine group), the effectiveness drops to 26.4 percent.

"There are still a huge number of uncertainties surrounding this trial," says Dennis Burton, an immunologist at the Scripps Research Institute in La Jolla, Calif. The subjects were in low- and moderate-risk groups, such as heterosexuals in monogamous relationships, rather than higher-risk groups such as intravenous drug users. "The numbers involved are small," he adds, noting that statistically the protective effects could be the result of mere chance.

Still, many researchers are convinced that the trial has provided plenty of data to run with. "This contributes more evidence that an AIDS vaccine may be possible," says Jerome Kim of the Walter Reed Army Institute of Research and co-author of the Thai trial study (which appeared in the New England Journal of Medicine in October). "We’ve taken a very small step," Kim says. "It's not a home run, but it opens the door to future work." Vaccine proponents also point to the lessons learned from the failed Merck STEP trial. That vaccine test, halted in 2007, got only as far as phase II, but even so it did not leave researchers back at square one. It suggested, he notes, how some HIV strains could be blocked from infecting cells and offered data that could help in the interpretation of the Thai results. And a new analysis of the stopped STEP trial, published online Monday in Proceedings in the National Academy of Sciences, provides a warning that the very vectors (adenoviruses, which are also employed in other vaccine development) used to distribute the inactive HIV strains can actually prime the immune system to be infected by recruiting susceptible T cells to mucous membranes, where they are more likely to be infected during sexual activity.

Finding a vaccine has become an increasingly urgent undertaking. Despite advances in therapies, HIV/AIDS is still incurable. Some 7,000 people worldwide contract HIV every day, and in the U.S. about 66,000 new cases are reported every year. Preventing people from getting the virus would save millions of lives as well as greatly reduce health care costs associated with treatment. "It’s really the only optimal method of control for this dreadful pandemic," Raphael Dolin, of the Beth Israel Deaconess Medical Center in Boston who also wrote an editorial accompanying the October paper, says of a vaccine.

Vaccines work by priming the immune system to recognize the target pathogen and attack it when detected. To fend off HIV, researchers introduced one vaccine (ALVAC-HIV) to induce a T cell response—thereby alerting the immune system—and another (AIDSVAX B/E) later to spur an antibody response. In a previous phase III trial in intravenous drug users, AIDSVAX did not work. ALVAC, from Sanofi Pasteur, had not been tested alone.

Using these two drugs together raised eyebrows in the vaccine community. Burton, along with 21 other researchers, co-authored a 2004 paper in Science criticizing the choice to proceed to phase III with two vaccines that had never demonstrated any effectiveness alone. The trial collaborators, however, based their decision on previous research that a combined approach can boost helper T cell response better than a single vaccine.

Despite his earlier doubts, Burton has been inspired by the trial results. "I feel more optimistic than I have in some time," he says. Researchers are now embarking on a host of new experiments to put the Thai findings to work. Volunteers from the trial will now be examined for immune responses—particularly neutralizing antibodies as well as cellular immunity in T cells—and some will get subsequent booster shots to see if protection can be sustained. In the lab, researchers will try to re-create the Thai results in monkeys to validate a new animal using multiple low doses. Other recent research has shown that the number of antibodies needed to provide protection is lower than previously believed, possibly making a vaccine easier to create.

Indeed, entirely new and promising candidates are now in animal trials, including those by the U.S. military to address subtypes A, C and E (rather than the Thai subtype B). Other organizations—including the International AIDS Vaccine Initiative (IAVI), the Karolinska Institute and the Swiss nonprofit EuroVacc—and manufacturers also have other vaccines in the works. "The science is really moving," says Seth Berkley, a professor in the Department of Epidemiology at Columbia University's Mailman School of Public Health and who is also president of IAVI. All those confronting the epidemic hope that the momentum leads to a payoff sooner rather than later.