VaxGen's AidsVax: paper urges "watch the tycoon and tout" as results row deepens

The failure of AidsVax to prevent infection with HIV - in clinical trial results published in 2003 - triggered an intense debate about the controversial product and its manufacturer, VaxGen Inc of Brisbane, California. Mail to this website, maintained by Brian Deer, shows that existing material on a VaxGen-AidsVax index is read by significant numbers. This page seeks to further inform the discussion

In a news report published on Friday February 28 2003 by the San Jose Mercury-News - a near neighbour of the South San Francisco-based VaxGen - the company's start-of-week position seemed to be crumbling. Under straplines "Biotech trouble" and "VaxGen stands by its vaccine", the paper's business section reported:

STOCK PRICE FALLS ON NEW QUESTIONS OVER DISAPPOINTING VACCINE RESULTS

Under the by-line Paul Jacobs, the story opens:

VaxGen executives stood firm Thursday under mounting criticism from scientists who questioned the way the biotech company analyzed its results finding its AIDS vaccine may protect blacks and Asians more than others.

Jacobs reports that one "academic consultant" to the company, "while not ruling out" that there may be racial and ethnic differences in response to the vaccine, as VaxGen had claimed, said "it should have corrected its findings." He said that a corrected analysis suggested that any protective effect in minority volunteers could be much smaller than VaxGen had claimed and was possibly non-existent.

``It could be a real effect, or it could be a fluke,'' said Steven G. Self, a biostatistician of the Fred Hutchinson Cancer Research Center and University of Washington in Seattle. ``It's interesting enough to look at and find out which it is.''
But he said it was ``premature'' for the company to take the results to the Food and Drug Administration in an effort to license the vaccine for use by specific ethnic or racial groups.
The Brisbane company, responding to criticisms published in Science magazine and elsewhere, stood by its statistics Thursday: ``The results VaxGen reported on Monday remain accurate as stated, and the analysis continues.''
Earlier this week, VaxGen announced that the first large test of its AIDSVAX vaccine in more than 5,000 volunteers, mostly gay men, showed that the vaccine failed overall to offer protection against the human immunodeficiency virus or HIV, which causes AIDS. Altogether, 5.7 percent of those who received the vaccine became infected over the course of three years compared to 5.8 percent of those who got dummy shots, or placebos. The difference was not statistically significant.
But the company reported that the vaccine showed a significant reduction in infections among certain vaccinated groups -- a 67 percent lower rate among non-Hispanic minorities, and 78 percent lower among blacks.
Four AIDS advocacy groups issued a statement accusing the company of playing ``the numbers game'' to serve ``the commercial interests of the company.'' Central to their argument are the small numbers of non-Hispanic minorities in the study -- just 498 -- and the even smaller number of HIV infections, just 13 among 314 black participants and four among Asians.
Self and other statisticians say that the small numbers limit what they call the ``power,'' or the reliability, of the findings.Among the blacks who were infected were four women, who all received placebos and not the vaccine -- from a Chicago test site.
That raises the possibility that at least some of the apparent effectiveness of the vaccine in minorities is the result of gender and not race, says Neil J. Risch, a professor of genetics at Stanford University.``My opinion is they did not study enough African-Americans,'' he said. The results don't prove that there was an effect, he said, but they do suggest the need for further stud.

The Mercury-News concludes its news report, published four days after VaxGen released what it said were initial results of the AidsVax trial, by noting that the company's stock had continued to dive, closing at $4.25, down 12 percent for the day and 67 percent for the week.

Menwhile, the Mercury-News continued its coverage on the same day with a column by Scott Herhold, who encapsulated recent events surrounding VaxGen - including the hyping of its stock by "investment analyst" Porter Stansbury, cited at this website. Under the headline "Firm's players show self-interest can taint views", Herhold points out that behind every major stock market drama there are dozens of cameos:

In that spirit, let me introduce you to an intriguing sidebar to the meltdown early this week of VaxGen, the Brisbane company that lost more than half its value when its AIDS vaccine fell short. It's the story of the tycoon and the tout.

He tells his readers that they know the background, before filling them in on what they may not have followed as closely as Herhold himself clearly has:

What you may not know is how intimately the tycoon -- Microsoft co-founder Paul Allen -- was involved with the company, or what his departure from the stock in recent months signaled.And unless you're a VXGN investor, you probably don't know how vigorously a newsletter writer named Porter Stansberry pushed the stock. Even as the results came in, he stayed positive, saying, ``I have remarkable news.''
The opposing directions taken by these two men can teach investors a fundamental lesson: Self-interest bends our point of view.Take the tycoon first. Through his investment arm, Vulcan Ventures,
Paul Allen was an early investor in VaxGen, which spun off from the better-known Genentech in the early '90s. In December, 1999, after VXGN had already gone public, Allen stepped up his investment in a big way, putting $25 million into the company to fund its AIDS research. At $11.50 a share, he bought nearly 2.2 million shares. By mid-2000, he owned more than 20 percent of the company. A Vulcan Ventures biotech analyst, Ruth Kunath, took a board seat.
But over the past year, Allen has edged away almost completely from VaxGen. Kunath quit the board in mid-2002. And by the end of 2002, Allen had only 664,682 shares left, shares he filed to sell by the end of January. (While SEC records don't indicate those sales were completed, you can assume they were.)
Newsletter writer Stansberry, meanwhile, was headed in the opposite direction. On his site, Pirateinvestor.com, he's been bullish on VaxGen for at least two years. In January, about the time Allen was bailing, he sent out a note urging investors to buy his newletter (special price: $74.50) detailing his research on the AIDS vaccine.
Although he didn't mention VaxGen by name, it was clear Stansberry was describing the company when he said, ``I've uncovered a business currently worth $250 million that will soon be worth several billions. . . . In a matter of weeks, this stock will skyrocket when a story about it appears in the Wall Street Journal.''
In fact, much the opposite happened. But even as VaxGen was announcing its disappointing results Sunday night, Stansberry refused to surrender. Noting seemingly hopeful numbers among Asians and blacks, he concluded, ``AIDSVAX seems to work.''
So whom to follow, the tycoon or the tout? At a glance, you'd say Allen in a heartbeat. And you'd be right, though the record is mixed. Over time, Stansberry has picked a number of winners. And Allen has taken some notorious losses, particularly in cable and fiber optics.
What makes the choice clear, however, is the context. Remember that the purpose of Allen's investment was to further the company's AIDS vaccine research. Remember, too, that he has a biotech analyst who sold her own shares. When Allen leaves, it matters. (Neither Vulcan nor Stansberry responded to requests for comment.)
The newsletter writer, meanwhile, bought into the stock himself recently, though he says the purchases came after his newsletter appeared. So it's safe to assume he had a vested interest in keeping VaxGen afloat even after the flood started.
Given the company's slender prospects, it's an ever grimmer task. Since Sunday, further questions have come up about VaxGen's claim of good results among some minorities. The stock finished Thursday at $4.25, down 79 percent from its January high of $19.94.

Scott Herhold's Stocks.comment appears every Monday and Thursday. Write him at the San Jose Mercury News, 750 Ridder Park Drive, San Jose, Calif. 95190; e-mail sherhold@sjmercury.com; phone (408) 920-5877..