Paul Allen is a co-founder of Microsoft ( MSFT), a high-profile multibillionaire and someone with a reputation as an active technology investor. So investors might want to take notice of his latest move: Selling a huge stake of his ownership in VaxGen ( VXGN) right before the small biotech firm takes the wraps off test results for its controversial HIV/AIDS vaccine, AIDSVAX.

Allen is not the only VaxGen shareholder poised to sell before anyone knows whether AIDSVAX works or not. Last month, the company filed a registration statement with the Securities and Exchange Commission allowing other preferred shareholders to also sell their considerable stakes in the company.

With the sting of Enron and ImClone Systems ( IMCL) still fresh, investors need not be reminded that the stock-selling activities of company executives and other insiders has become a closely watched measure of corporate health. Allen and the other stakeholders who could be unloading their VaxGen stakes might just be locking in profits -- the value of VaxGen stock has more than tripled since this summer -- but their apparent eagerness to get out now could also be a red-flag warning about the upcoming AIDSVAX study.

Results from the randomized, controlled, phase III study of AIDSVAX -- which enrolled 5,000 gay men and 400 women with HIV-positive partners in the U.S. -- should be released in the first quarter of 2003.

VaxGen shares have been on quite the roll, quadrupling in price since sinking to a recent low of $5.10 on July 12. The stock hit $20.80 on Dec. 2 and was trading at $17.10 recently.

Brisbane, Calif.-based VaxGen is sitting on a potential commercial gold mine if AIDSVAX proves effective in helping prevent infection from HIV (more on what that means later). Drug companies have been successful in developing medicines to manage the disease but not cure it. A vaccine to actually prevent transmission has similarly proven to be an incredibly difficult scientific challenge. VaxGen is not alone in searching for such a holy grail, but its approach has not been entirely embraced by other AIDS researchers. In the mid-1990s, government researchers, for instance, took a look at AIDSVAX but declined to fund studies.

Allen, through his Seattle-based investment fund Vulcan Ventures, has been the largest shareholder in VaxGen, at one time owning about 3 million shares. But not anymore, according to SEC filings.

Vulcan sold 780,000 shares of VaxGen common stock between Nov. 18 and Nov. 20 at prices ranging from $16.73 and $21.57. The sales reduced Allen's ownership stake to 1.27 million shares, according to three Form 4s filed with the SEC.

Since then, Allen has filed three Form 144s, signaling the possibility that he may sell another 580,000 VaxGen common shares. It is not known whether those shares have been sold yet.

VaxGen spokesman Lance Ignon says Allen's disposal of VaxGen shares are "part of a planned selling plan and not indicative of anything to do with AIDSVAX." Allen could not be reached for comment.

On Nov. 26, VaxGen filed a registration statement with the SEC to allow four other institutional investors to sell just under 2.4 million shares of common stock. In April 2001, these investors purchased $20 million worth of VaxGen preferred stock in a private placement. The largest of these investors, Halifax Fund, converted its preferred shares into almost 1.1 million common shares on Nov. 19 at a conversion price of $14.13.

Although these are significant institutional investors, unlike Vulcan they each hold less than 10% of the company, so they are not deemed to be "insiders" under federal securities laws.

The Palladin Group manages the Halifax Fund. By registering the stock on Nov. 26, VaxGen frees Palladin and the other institutional investors to sell their stock, although it's not yet known if any sales have been completed.

Insider selling as the vaccine results are approaching "is a concern," says Jonathan Moreland, of InsiderInsights, a research firm that tracks insider selling. (Moreland is also a contributor to RealMoney.) Under normal circumstances, it wouldn't be unreasonable to see some selling, given the stock's run-up since this summer, he adds.

Suggesting that AIDSVAX, if it works, would immunize someone against getting HIV would be incorrect, because even VaxGen scientists acknowledge that it will not be anywhere close to 100% effective. In the past, VaxGen has said its study will be successful -- and that a filing with the Food and Drug Administration would follow -- if AIDSVAX reduces the level of HIV infection by at least 30% compared with the rate among patients taking a placebo.

But as the study nears an end, VaxGen appears to be backtracking from this efficacy floor. Spokesman Ignon says the company is only looking for a statistically significant difference in the two arms of the trial and is not offering any guidance on what a minimum efficacy standard might be, or about what the FDA might deem approvable.

An HIV/AIDS vaccine that offers 30% protection would be useful, says AIDS prevention expert Thomas Coates, but the danger, of course, is in people thinking they're 100% protected and reverting to high-risk behaviors.

"It would be a step in the right direction. In the U.S., you could conceivably see it being used with people at high risk for infection, such as IV drug users, gay males and minority women, but it's not the magic bullet," says Coates, director of the AIDS Research Institute and the Center for AIDS Prevention Studies at the University of California, San Francisco.

He adds: "Any campaign for its use would have to be carefully structured to ensure that people understand they still need to take all the other preventative measures."

VaxGen was formed in 1995, spun out of AIDS vaccine research then being done by Genentech ( DNA), which still owns a stake in the company and retains the right to market the vaccine if it works. VaxGen went public in 1999 but has never garnered much attention from Wall Street's big investment banks, likely owing to the long odds against success. Its market cap hovers around $250 million.

Sharon Seiler, an analyst with the boutique investment bank Punk, Ziegel, says the odds of an AIDSVAX failure run about 65%, while the probability of minimal efficacy is about 30%. (She only gives 5% odds for a home run -- universal efficacy -- being hit in the study.)

But Seiler still has a buy rating on the stock and a $38 price target, because she believes the company is unfairly tagged as a one-note wonder. In recent months, VaxGen has done much to diversify, winning a government contract to develop a new anthrax vaccine and forming a drug manufacturing joint venture with Korean investors.

"If AIDSVAX doesn't pan out, VaxGen still has other business lines to fall back on," she says. Her firm has a banking relationship with the company.

Try telling that to short-sellers, who are salivating now that the AIDSVAX results are near. Short interest in the stock has risen to 2.4 million shares in November, from 1.5 million shares in July.

Says one particularly feisty -- and short -- hedge fund manager, "I don't care what else VaxGen has cooking, AIDSVAX has no chance of working, and that makes this thing a zero."

As originally published, this story contained an error. Please see Corrections and Clarifications.