Pioneering virologist Don Francis is convinced it's the great Aids breakthrough - a safe vaccine that will save many thousands of lives. So why has the American government, with the active support of the gay communit y, suppressed gp 120? JESSE GREEN reports on the bitter policies of Aids resear ch community, suppressed gp120? JESSE GREEN repo...
Some 15,000 vials of a colourless, odourless, water-like liquid are lined up in trays in a room so cold, so remorselessly quiet, that you can't help thinking of it as a tomb. In a way, it is. Each vial contains a one-millilitre dose of a product known as gpl2O, which has cost its manufacturer, Genentech, $100 million and untold grief. Is gp12O, as some scientists believe, a crucial first approximation of the desperately needed vaccine to prevent Aids? Or is it, as others say, nothing more than expensive dishwater? No one knows. Neither Genentech's product nor a similar vaccine made by Chiron Biocine has yet been tested enough to tell.

So far, no vials have passed through the enormous steel doors to the loading dock beyond, and from there into syringes and into the arms of the 10,000 volunteers whose bodies alone can answer the question. Instead, the vials sit and wait, with nothing to interrupt their silence but a workman's boom box, which suddenly sings out from the dock: "Who'll Stop the Rain?"

"I was in Nigeria for cholera in 1970," Don Francis says fondly, as if recalling a world's fair. "The Sudan for Ebola virus, Phoenix for hepatitis B, just about everywhere for smallpox." The rsum of his long career with the Centers for Disease Control (CDC) reads like an itinerary of miracles. In Africa, India, Bangladesh, Francis helped eradicate terrible illnesses by the bold application of vaccine programs. But here in the United States, in his dishevelled office at Genentech's headquarters on San Francisco Bay, miracles keep getting derailed. To Francis, the company's vaccine evangelist, gpl2O represents the best hope yet of beating Aids, and he is determined to test it and see.

Francis has been at work on Aids since before it had a name. In 1981, as the head of the CDC's successful vaccine effort against hepatitis B - which relied on the co-operation of the gay community - he was on hand when inexplicable outbreaks of Kaposi's sarcoma were first reported in the same population. What happened during the next 11 years still brings the bile of indignation to his voice. Scraping for money, pounding on tables, writing memos that were never answered, Francis spent as much time doing politics as science.

Since he surmised that Aids would prove to be transmissible through blood products, he begged the blood banks to screen donations; they did not believe him and infected thousands. Since he saw that gay bathhouses were a major route of sexual transmission, he begged local boards of health to shut them; an alliance of owners and gay activists kept them open. Francis's enemy wasn't so much the virus itself as the perverse intransigence of human nature that gave it room, and time, to manoeuvre.

By the time Francis joined Genentech in 1992 to run its vaccine clinical trials, gpl2O had already suffered so many disasters - and unlikely resuscitations - that Phil Berman and Tim Gregory, the pair who led its development team, had chosen Indiana Jones as their mascot. The failure of their first attempt in 1986 had pushed the vaccine "out of development" and into the corporate equivalent of Siberia. To rescue the project, Berman and Gregory kept working, sub rosa and in their spare time.

Finally, on 3 January 1989, in a Texas lab, Berman watched as two chimps were injected with a revamped version of gpl2O; two with a different product, called gpl6O, and two with a placebo to serve as controls. Nine months and three injections later, the "challenge" took place: all six were injected with live HIV - the virus that causes Aids in humans. By December, blood tests showed that the two control animals and the two given gp160 had been infected. The two chimps given gpl2O, though, were completely clear. "We waited longer and longer and longer and they still didn't break," Berman says. "Finally, we knew it was true."

Chimpanzees are not an ideal model for Aids in human beings. Though they get infected, they never show signs of disease. And humans do not get "challenged" with HIV through single massive injections in laboratories: they have sex without condoms repeatedly or share dirty needles. But the team had demonstrated for the first time that such a vaccine at least was pos-sible - changing the mood among scientists from bleak to hopeful. It would prove to be the high point in the strange life of gpl2O.

"If you have the cowpox, you cannot take the smallpox" - that's what 18th-century milkmaids said. When Edward Jenner set out to test the hypothesis, he did not appeal to government-grant panels before exposing an eight- year-old boy named James Phipps to one disease and then the other. Two centuries later, Jenner's arrogance shocks our sensibilities, but arrogance can be effective: except for two frozen keepsakes, one in Atlanta and one in Moscow, smallpox no longer exists on earth. The milkmaids were right. Jenner's vaccine worked because the cowpox virus is similar enough to smallpox that Phipps's immune system learned from one how to fight off the other.

Modern vaccines employ the same trick but are all man-made. Some (like vaccines for measles, mumps and rubella) are made from live virus that's been artificially weakened, some (like Pasteur's for rabies and Salk's for polio) use virus that's been killed but left whole.

The advent of genetic engineering in the late Seventies opened up a new avenue for vaccine development; in a 1985 paper, Francis outlined how that technique could be used to make a gp120 Aids vaccine. Avoiding the core of the virus, where the disease-causing genetic instructions lurk, scientists could now synthesise only those portions of the virus's coat that would stimulate an immune response. In the case of HIV, the most promising portion was the "sub-unit" called gpl2O: a chain of protein that sticks out from the virus like a fist. (The gp stands for glycoprotein; the 120 for its molecular weight.) The key advantage to this approach, which had already proved successful in the new hepatitis B vaccine, was safety: like a toy gun with no bullets, the vaccine could scare you into defending yourself but it could not actually shoot.

Genentech's version of gp120 was first tried as a preventive in humans in March 1992. "Because of the way these vaccines are made, we didn't expect any problems in phase I," Francis says. "And we didn't get any, except for a few sore arms here and there."

For any product, promotion from one trial phase to the next is a major milestone; manufacturers proudly note the event in annual reports. But it's not entirely a scientific decision. The National Institute for Allergy and Infectious Diseases, or Niaid, which was financing the trials of gpl2O, could terminate further testing by withholding funds. But Niaid's decision to promote gp120 from phase I, which had concerned itself mostly with safety, to phase II, which would begin to show whether the vaccine might work, was enthusiastic. Of the 14 experimental vaccines tested since 1988 at Niaid's six Aids vaccine evaluation units, the two gpl2O candidates were by far the most promising.

Indeed, phase II for the most part bore out that promise. Volunteers produced strong immune responses, and, although it could not be determined until phase III whether those immune responses would actually protect anyone, the results were good enough to convince a government panel to make an initial recommendation in favour of promotion. It seemed that, nine years after Francis's 1985 paper, the United States was about to find out if it was on the path to a vaccine for Aids.

This optimism was short-lived, however. On 29 May last year, the Chicago Tribune ran a front-page story under the headline new doubts on aids vaccine; 5 study volunteers infected; us debates future of trials. The article accurately reported that volunteers had become HIV-positive during the first two phases of testing. But by the time the news wires and television networks picked up the story, it had become hopelessly garbled. Even people who should have known better ended up drawing the wrong conclusion: a fund-raiser for a major Aids organisation told friends that gp120 had "made people sick" because it was "made of live HIV" - gpl2O contains no live HIV and cannot cause infection.

In fact, according to Niaid, the volunteers had all contracted HIV as a result of "high-risk behaviours" - that is, through unsafe sex or injection- drug use. The infections had long since been reported at conferences and scientists were hardly surprised; indeed, if no infections had been recorded, even in the control group, little would have been proved. Designers of Aids vaccine trials are confronted with a unique paradox: if safe-sex and safe- needle practices are not recommended, volunteers who believe they are protected by the unproven vaccine (which may actually be a placebo) may take more risks and increase their chances of infection. If such practices are recommended - and modern ethical standards dictate that they must be - volunteers can cut their risk so effectively that they are never exposed to HIV, leaving the vaccine with nothing to fight.

To get around this problem, Phase II trials of an Aids vaccine must rely on populations in which the number of infections remains high regardless of education: namely young, high-risk gay men and injection-drug users - precisely the groups most scared by the news story. In the early years of the epidemic, these groups came to believe, with justification, that every dollar spent on the "innocent", uninfected masses would be a dollar less spent on treatments for the sick. The world has changed since then - for one thing, the activists have fought their way into the system. But the habit of outsiderness dies hard: though no longer hostile to vaccines, they remain ignorant of them.

For the people running the vaccine evaluation units, therefore, the Tribune's news story was a nightmare. New recruits dried up; long-term volunteers needed calming; reporters were buzzing everywhere. But one group with few doubts about the trials was the group of infected volunteers. Most agreed, despite their bad news, to give what a researcher called "buckets of blood" for further testing. One, a gay man known as BIS303, even flew to Washington for a lymph node biopsy that scientists thought might be revealing. Altruism like that of BIS303 was the kind of reaction Don Francis had imagined when he thought of how a trial would work. As a result, he had no idea what he was up against when he went before the Aids Research Advisory Committee on 17 June last year.

It used to be that Aids policy was made on the streets and in closed government meetings - or from the conflict between them. Now it is forged in chandeliered hotel conference centres like the so-called "NIH" Hyatt near National Institutes of Health headquarters in Bethesda, Maryland. The scene is always the same: hyper-floral carpets, gleaming urns marked "caf" and "decaf", presenters with carousels of slides eyeing the name tags of other participants.

What has changed is who those participants are. Here in the Hyatt's Crystal Ballroom, 40 or so panellists sit around the perimeter of four tables arranged in a large square with nothing in the middle. Surrounding the official participants are supporters and contestants from the various camps. If you regard Anthony Fauci, Niaid's director, as north, then government scientists are generally to the east, gay community activists are huddled to the south, Genentech and Chiron Biocine representatives, including a grimacing Don Francis, are stacked to the west. The media - 55 reporters and 15 camera crews, hoping for scandal - are kept to the back.

Try as they might to work together, the groups speak different languages. The scientists "present" their information while the com-munity representatives "share" theirs. The basic researchers require "proof of concept", while the manufacturers speak of "empirical results". But the language conflict is trivial compared to the conflict of culture. Depending on whom you talk to, the basic researchers, usually affiliated with academic institutions, are ivory-tower do-nothings, jealous of their grants and remote from the world of suffering; the manufacturers' scientists are sell-outs and cowboys, grubbing after profit; Niaid staffers are cowardly paper-pushers who'd rather do nothing than fail; the community representatives represent no one but themselves, and at tiresome length. And everyone hates the press.

Still, it is no longer seemly to make decisions unilaterally. Considering what's at stake, some displays of antagonism, or at least real debate, might prove useful, but the mood is sullenly respectful. During the half- hour set aside late in the day for "public comments", community activists rise one by one to lash out at gp120; their opinions, presented as scientific fact, are left unchallenged. And when the costs of various versions of a trial are discussed, no one points out that, compared with the enormous cost of Aids treatment, even the most expensive trial would be cheap: a small-scale trial would cost as little as $9million over two years, while America spends $13million a day on Aids care.

Perhaps that's why Don Francis keeps raising his hand (to no avail), like the kid in class who knows too much: at one point he even hands up a note that reads, i want to be called on. But he's out of order, which is just as well. Some of the panellists are muttering curses about his self-righteous badgering.

"Remember, this is not the Virgin Mary proposing a vaccine," I'm told later by Martin Delaney, a panellist from Project Inform, a San Francisco Aids organisation. "To some degree, this is about a couple of companies who want to be the first to get their product to market."

Delaney says that Project Inform has always made vaccines a priority and that he would be prepared to volunteer for a trial himself. But unless a vaccine is largely effective, it "doesn't make sense" when the money could be spent more directly on education and other preventions. "If your concern is purely public health, you have to ask: What are we going to get out of a vaccine trial of a product that's a complete shot in the dark, based on the most primitive understandings of the disease? And if you blow your one chance on the wrong product, where does that leave you? Then come the other issues. Like, does risky behaviour increase because people think they're protected? And the answer to that has to be yes."

Data from the testing sites suggest that, if anything, risky behaviour among volunteers decreases, but Delaney and others discount it. "You can survey and ask people all you want, but you're not going to get a straight answer," he says. "As a gay man, I know this personally."

"But", I ask, "are you comfortable making such a big decision when you're not a scientist?"

"I don't feel that science is magic," Delaney answers. "It's knowledge. It's quantifiable and transferable. You do worry about any decision to not go ahead, but the outcomes in Aids research have been so disappointing that" - he chuckles ruefully - "no, it doesn't give me any huge pause."

And that's how the meeting plays out. Apparently stunned by the passion of the gay community representatives, the members, one by one, vote no or abstain - even four who had previously endorsed expansion of the vaccine trials. Fauci annnounces that, since the recommendation is so strong, he will follow it. Though the Niaid news release will later say that the panel voted for "continuing but not expanding at this time", that's pure euphemism.

Don Francis looks as if he could strangle someone, but there's no one left to strangle. Within three minutes of the decision, scientists are upstairs with drinks in their hands, activists are on the subway and someone has turned off the chandeliers in the Crystal Ballroom.

As it turns out, Anthony Fauci had already decided what he was going to do. "I was convinced before the meeting even started that the science did not justify expansion," he says.

We are sitting, two months after the meeting, in Fauci's office. "I had spent a lot more time than them looking at the data," he says. "It was comforting to me that a large and diverse group of my peers came up with the same conclusion. But I still would have made the same decision. Let me be perfectly honest, which I always am: it would have been more difficult to say no to an expanded trial if the community said, `We're ready if you want to do it.' Because I know how quickly community views can change. But I would have done it, anyway."

Fauci's curriculum vitae says he's 54, but his nervous energy, his runner's trim, his Rotarian attire suggest a graduate student, circa 1966. To all appearances, he's pure science in a suit - not unlikeable, but not really there to like. He talks fast, and you sense that he's still got plenty of brain capacity left over for other things. In the middle of a sentence, he stands abruptly, goes to the door and barks at his secretary for a Diet Coke; he then sits down and resumes, quite calmly, the interrupted sentence.

Whether because of his manner or his power, Fauci stirs up a great deal of anger. Anne-Marie Dulige, associate director of clinical research at Chiron Biocine, which has been testing its gpl2O product in tandem with Genentech's, expresses openly what others say only in private: "The Niaid decision was a purely political, manipulated decision." The advisory committee meeting was stacked, she says, with people who knew little or "nothing at all" about vaccines. "Besides, isn't it curious how the news of the sero-conversions [the volunteers who became infected] got released to the Tribune reporter?"

Fauci denies having anything to do with the leak. In any case, you need not resort to Machiavellian scenarios to understand his decision. "With a preventive vaccine," he says, "you're talking about an otherwise normal person who does not have HIV infection and you're saying you want to inject that person with a product that, as a scientist, you can't in good conscience say there is even a more than reasonable chance of it being effective. And there is the possibility that there could be a deleterious effect."

No such effect has yet been demonstrated, but for Fauci there's no compelling reason to risk finding out. "Fauci was responding to fear that the trial might fail," says Dulige. "From a scientific point of view, no trial fails." But from a political point of view? Is it worth taking a fall for a vaccine that might not do much, that the activists hate, that virtually no one speaks up for? Unvaccinated against political fall- out, Fauci may have felt he had little choice but to practise "safer science" - which in this case may have meant no science at all.

Still, Fauci thinks of himself as an activist. "Aids is my life," he says. "I have many, many patients we've taken care of here; my social circle has evolved into an Aids social circle; my wife is an Aids nurse. It's all around me. We're in the middle of an era of incredibly sophisticated science and there's this goddamn virus that has rapidly evolved to stay always a step ahead of us. You feel that you're very smart and have a lot of smart people around you - and it's sticking its tongue out and saying: `I don't care how smart you are. You don't have me yet.'"

"Aids is eminently - and imminently - beatable," Don Francis says, still beating the drum for gpl2O. "If we sat around with smallpox and said, `I don't know, it's going to be awfully hard,' it would still be raging today. So you just do it. What's failure? The worst you're going to do is waste some needles and syringes and some people's time. That's a small cost. I hope we will be allowed to find out."

For Genentech and Chiron Biocine, that will have to be a business decision. It's not just about the vaccines' merits; it's about priorities. Continuing to fund gp12O may mean not funding a breast cancer product. Fauci's decision has had a chilling effect - some call it "devastating" - on the very few companies around the world that had bothered with an Aids vaccine at all.

Genentech is officially "re-evaluating" the situation, but seems close to dropping the project. Chiron Biocine and another company, Therion Biologics, have cut their efforts in half. "If you're the CEO of a company that has fiduciary responsibilities," Francis says, "yes, you'd love to make an Aids vaccine. But how much love can you afford?"

According to Martin Delaney, Genentech or Chiron Biocine can afford a lot. "Despite their whining," he says, "these are not poor companies. [Genentech's 1994 revenue was nearly $800 million.] I don't understand what's keeping them, if they feel that strongly, from going ahead with trials. Would we ask Merck or Abbott [major drug manufacturers] to wait for the government to pay for their trials? I don't think so."

Delaney is glossing over the economic difference between preventions and cures. Extremely expensive to develop and test, and riddled with legal liability risks, vaccines are not nearly as profitable as people might think. In any given year, the combined revenues of the 20 or so vaccines currently available are dwarfed by the revenues of a single good ulcer drug. Add to this the ethical uncertainties of a global epidemic - how much do you charge in Africa? - and it's no wonder that the manufacturers, and Fauci too, insist that the development of an Aids prevention vaccine will depend on close financial collaboration between industry and government. Which means collaborating with democracy.

Did democracy derail gpl2O? When Franklin D Roosevelt marshalled Americans to defeat polio (which he suffered from), he did it without permission from anyone. But the age of individual arrogance - and with it, perhaps, individual genius - has been superseded by the age of consensus. Instead of Roosevelt, we have panels: panels that are destined to disagree and cause more panels. Which isn't to say the process can't work. Even as Fauci announced his decision, the Food and Drug Administration was permitting Phase III trials of an experimental Lyme disease vaccine to go forward - based on the same kind of ambiguous evidence that sank gpl2O. And in January, the FDA permitted Phase III trials of Jonas Salk's Aids treatment vaccine, despite what it considered shaky data.

But Salk's vaccine is beyond the mayhem of democracy. A treatment, not a prevention, it could prove so profitable - in the event that it works - that its manufacturer, Immune Response Corporation, has been willing to foot the bills to test it. And Lyme disease, typically transmitted by tick bites into the bloodstreams of wealthy suburbanites, excites no controversy. Because Aids is so much more deadly than Lyme disease - and because preventions are so much more difficult to finance than cures - the trial of a vaccine like gp120 becomes almost impossibly political: everyone has expertise and everyone has a point. Activists are right to question a vaccine that may be 30 per cent effective, because sex isn't calculus and people aren't wise. But public health scientists are right, too: such a vaccine might save thousands of lives, and a nurse who sticks her finger with a needle will be glad to have any chance - even 30 per cent - of being protected.

Vaccines, however, have an eerie way of rising from trash cans. The recent success of the hemophilus influenza type B vaccine came after nine efficacy trials of five different products in 330,000 children over 17 years. Since many scientists assume that gpl2O will eventually be part of the cocktail of an effective Aids vaccine, the all-or-nothing mentality that halted testing may prove to have been shortsighted.

At least four countries have since sought permission from the World Health Organization to host gpl2O trials. Natth Bhamarapravati, a pathologist from Thailand - where Aids, he predicts, will soon kill 100,000 people a year - has pleaded eloquently for the development of a vaccine Marshall Plan, in which American money and expertise would be deployed to help such countries as his start testing something.

And so Genentech began making plans with the Thai government to conduct vaccine trials. For those who had championed gp120, this was at best a bittersweet victory. In order to prepare enough Thai volunteers and to ensure that the vaccine will work in them the same way it did in Americans, most of Phase I and Phase II must be repeated - delaying results by several years.

But at least they're starting. On 6 February, 45 cartons containing 270 vials of gp120 finally left Genentech's cold room, destined by plane for Bangkok. They arrived at the vaccine testing centre at Mahidol University four days later, and the first 14 volunteers, who are recovering drug users, were injected on 21 February. Are they safer from Aids than they would have been without gp120? In a few years - a few years longer than necessary - we'll know.

"I know there are limits to our ability as humans to accomplish important things," says Don Francis. A sign pinned to the wall by his desk reads, joyful participation in the sorrows of the world. But he doesn't look so much joyful as resigned. "It drives me crazy. I'm weary. I yearn for logic and a way to make good on all of this wasted time. But that's life. Well, that's the intellectual answer.

"On the other hand, San Francisco is losing four per cent of its population. My Rolodex, I don't take people's names out of it. I'm scared to. The frustration - I want to scream sometimes. But it doesn't help."

He points to another sign near his desk: a scientist who is also a human being cannot rest while knowledge which might be used to reduce suffering rests on the shelf. The quotation is from Albert Sabin, who developed the oral polio vaccine. "I guess I keep doing it because I'm an optimist," Francis says. "I think it'll break. Nothing can take your eyes off that goal; if I possibly thought that I might fail, someone might stop me. It's like a bulldozer: your top speed might not be very fast, but you keep on moving. And then you come to a slope and you'll move like crazy. That's the history of all of our progress."

But why do people resist that progress?

"Did I tell you about the frog?" Francis asks. "I hope it's true. If you take a pot of boiling water and dump a frog into it, he'll do all in his power to get out. But if you put it in a pot of cold water and put the pot on the stove and slowly bring up the heat" - Francis looks over the top of his half-glasses - "the frog will just sit there. And boil himself."