(USA TODAY) -- In a simpler world, the news might be greeted with cautious jubilation. After 20 years of setbacks, the first full-scale study of an AIDS vaccine offered a tantalizing hint that it may protect some black people from AIDS.

Instead, the findings announced in late February triggered a racial face-off between the two groups that need a vaccine most.

The vaccine, AIDSVAX, had failed to meet its key test by failing to outperform a dummy injection in the 5,400-person trial. But when researchers looked at the results for blacks, Asians and people of mixed race, they say, they got the surprise of their lives.

"This race thing fell out," says VaxGen's president, Donald Francis. "It stood out like a sore thumb. We thought, 'Oh God, how are we going to deal with this?' "

The announcement instantly provoked a vehement rebuttal from predominantly white, gay AIDS-advocacy groups and some vocal scientists, who noted that the vaccine failed to protect the majority of volunteers, who were whites and Hispanics. The skeptics dismissed the finding as statistical sleight-of-hand by a company struggling to salvage a 10-year, $200 million research effort.

Black AIDS advocates were angered by the quick condemnation of the first evidence suggesting that an AIDS vaccine might work in humans, especially in the population burdened by half of all new cases of the disease.

Phill Wilson of the UCLA African-American AIDS Policy and Training Center says the study didn't create racial tension among AIDS advocates; it revealed a rift that has existed for years. "Quite frankly," he says, "the study simply forced us to look at it."

Wedged between the warring parties sits VaxGen, whose vaccine has been ridiculed by some of the firm's most vocal critics -- one critic went so far as to label the vaccine "dishwater."

At a meeting Monday in Banff, British Columbia, the firm reported that the vaccine failed to clear a second hurdle in its study by failing to boost the immune systems of people who already are infected with HIV.

But the newer, more detailed analysis presented this week suggests that the vaccine works best when a person is exposed to an HIV strain that closely matches the strain used to make the vaccine.

In addition, women in the study appear to produce more potent antibodies than men. "The bad news is that the vaccine isn't effective overall," he adds. "The good news is that we may have found a chink in the wall."

VaxGen's decisions may reverberate for months or years as the company explores its data and critics dissect VaxGen's handling of it. The sharpest criticism yet emerged last week in U.S. District Court in Northern California, where investors filed a lawsuit claiming that VaxGen defrauded them by hyping the vaccine's prospects. When the findings were announced on Feb. 24, the firm's stock plummeted 85% to $3 a share from a Nov. 18 peak of $23.25. VaxGen's lawyers dismiss the claim as lacking any merit.

Experts say the study raises several key scientific questions:

* Public service vs. corporate survival. Private research companies are obligated first to stockholders and regard their scientific data as a company secret. Scientific traditions, including the examination of research data by outside experts before being released, fall by the wayside. How, then, will the world be able to test the validity of scientific claims?

* The insider-trading problem. Publicly held firms must release information to everyone at once to assure that no investor gets an unfair edge in the stock market. But that may mean releasing data before every aspect of it can be analyzed.

* Truth, lies and subgroup statistics. The company asserts that its trial produced a statistically significant association between the vaccine and HIV immunity in blacks and possibly Asians, even though the vaccine flopped in the majority of volunteers. But statisticians question the significance of the finding, saying that VaxGen didn't apply rigorous enough statistical tools.

* The Tuskegee legacy, part one. If blacks' hopes are dashed, the study may further heighten blacks' distrust of the medical establishment because of such episodes as the government-financed Tuskegee study, in which white doctors withheld antibiotics from black patients with syphilis so they could study the course of the disease.

"If people don't trust the results, and they think there's some kind of scam going on, it does hark back to Tuskegee," says Cornelius Baker, director of the Whitman-Walker Clinic in Washington, D.C.

"Black people don't believe (the backlash over the trial results is really) about VaxGen; they believe it's about the HIV/AIDS power structure not wanting to benefit black people."

* Tuskegee, part two. If blacks' hopes are dashed, researchers could have an even tougher time than they have now recruiting blacks into vaccine trials. The VaxGen study failed to produce convincing results in blacks because researchers couldn't recruit enough blacks into the study. The Asian group was even smaller.

Stacked deck

Opposition to AIDSVAX has been simmering for a long time. The vaccine is an updated version of one made by Genetech in the early '90s. Even then, some scientists were skeptical that it would prove effective, partly because the company's tests weren't as rigorous as those done in other labs.

Early tests were conducted with a lab-tamed strain of HIV, not the wild strains circulating in patients. And primate studies involved a chimpanzee infected with HIV that was an exact genetic match for the one in the vaccine.

"The deck was stacked for the vaccine's success," says John Moore of Weill Cornell Medical Center in New York, a prominent AIDSVAX critic who was part of a consortium of experts that analyzed data from an earlier study of the vaccine. Then, in 1998, the Food and Drug Administration approved VaxGen's bid for a full-scale human trial, and the stage was set for more controversy.

On Feb. 15, a team of VaxGen data analysts checked into a hotel not far from the firm's offices to analyze the study's results.

Researchers found absolutely no difference between those who received the vaccine and those who received a placebo. Then they began examining the data by looking at certain subgroups, including sex and race. The vaccine appeared to be 78.3% effective in blacks and 68% effective in Asians. "We found remarkable efficacy, especially in blacks," says VaxGen's Phillip Berman, the vaccine's inventor. "The result was highly statistically significant."

But the subgroups were small, which almost always saps a study's validity. Blacks numbered just 314 of the 5,400 volunteers, half the number the firm would have needed for a representative population. The number of Asians was so small that researchers can't rule out the possibility that the positive finding was pure luck.

Science and the SEC

Nevertheless, the finding sent researchers at the NIAID-funded HIV Vaccine Trials Network in Seattle scurrying to see whether they can find any race-related differences in protection among volunteers in all previous trials.

"The real story here is that the scientific community is starting to take a long, hard look at what differences we might expect in men and women by race," says Steve Wakefield, the network's associate director for community relations and education.

Francis says VaxGen didn't have time to analyze all the data, and the firm couldn't withhold the results. That information, no matter what it was, would affect the price of VaxGen stock. If the information leaked to some stockholders, but not others, those who knew the results would gain an unfair edge.

"We had to announce this because of (U.S. Securities and Exchange Commission) requirements that investors all get the same information at the same time," he says.

On Feb. 23, Francis revealed the findings to key scientists and activists in an invitation-only conference call. The results were to be released through the news media the next morning before the stock market opened.

Because of the racial implications, Francis had called Wilson to invite him to take part. As Wilson listened, his concerns grew. "I called Francis back after midnight," he says, and told him that the "announcement is going to be problematic. A white company is going to say it has a vaccine that only works for black folks, so we're not going to vaccinate white folks?"

Cornell's Moore says he was blindsided by the way the company embraced such preliminary findings. "I knew they'd do something to keep themselves alive. I didn't know they'd focus on race," he says. "Once they played the race card, everything changed. Science took a back seat. I was appalled."

A consortium of AIDS advocacy groups, including the American Foundation for AIDS Research, the Gay Men's Health Crisis in New York, the AIDS Vaccine Advocacy Coalition (AVAC), Project Inform in Los Angeles and the International AIDS Vaccine Initiative fired off their rebuttal.

"We were concerned about being over-optimistic, and we felt it was important to emphasize caution in interpreting the results," says Chris Collins, AVAC's executive director. "It doesn't serve the field as a whole or the communities involved to draw any conclusions before we know what we have."

Search for vaccine continues

"They came down with a sledgehammer, saying the vaccine is a failure," Wilson says. "Then people got pissed. They felt if the data had shown any efficacy among white people, they wouldn't have so cavalierly dismissed it."

Collins, of AVAC, says the lesson hasn't gone unnoticed. "A lot of groups have grown from this experience," he says. "We're more often working together."

Baker, of the Whitman-Walker Clinic, says, "The reality is, the search for a vaccine is just beginning. It's better to get these issues out on the table now than four years from now, when we have another trial in the works."

Copyright 2003 USA TODAY, a division of Gannett Co. Inc