The Long Shot

What the world needs now, more than ever, is a good AIDS vaccine. Why don't we have one?

By Mark Caldwell|Sunday, August 01, 1993
We’re now well over a decade into the AIDS pandemic, and the dreadful clock hasn’t stopped ticking. According to the World Health Organization, as many as 14 million adults and 1 million children have been infected by the AIDS virus, and a stunning total of up to 120 million is predicted for the year 2000. True, medicine has gotten better at treating the opportunistic infections of late-stage AIDS and has come up with a few stopgap drugs like AZT. And reports of vaunted new therapies regularly erupt in the press. But many of these remedies have proved disappointing at best and completely illusory at worst. Antibiotics have spoiled us with their power to knock bacteria to the canvas with a satisfying roundhouse punch. Drugs against viruses, unfortunately, have a hard time even landing their blows.

Vaccines should be another story. Vaccination has a long track record against viral diseases, from smallpox to polio to measles--Europe had a crude smallpox vaccine as long ago as the 1700s.

And, in principle, vaccination seems so beguilingly simple. By introducing the body to a harmless version of a virus (usually one that’s been crippled or killed outright), you can whip the body’s defense system into a state of active vigilantism. If the real virus later tries to invade, it will be smartly repulsed.

At least that was the way it had always appeared to work. Thus in 1983, when researchers confirmed that the cause of AIDS was a type of virus--the human immunodeficiency virus, or HIV--you could almost hear the sigh of relief. There seemed little reason to doubt that you could deploy the immune system’s power against this new virus, no matter how peculiar it might be. (Our immune systems, after all, have been battle-trained by eons of evolution to fight off all kinds of nasty foreign agents.) A vaccine in ten years seemed a sensible forecast. Indeed, in early 1984 researchers told the New York Times that an AIDS vaccine might be ready for testing within two years. A decade later, of course, that kind of talk has begun to look like innocence, not to say naïveté. While about a dozen prototype vaccines have finally begun inching their way into human trials, they’re purely experimental. Nobody’s ready to claim they’ll actually fend off AIDS.

So what went wrong? Where’s our AIDS vaccine? Why are we still waiting?

It’s not, after all, as if researchers have been sitting on their hands--never before has so much been learned about a virus so fast. Unfortunately, what they’ve learned is not nearly enough. Molecular biology can tell us a great deal about HIV’s structure, its genes, and its rarefied life inside cells growing in lab dishes, but it tells us next to nothing about how the virus interacts with a warm human body, and in particular with the body’s immune system. That’s one of the major problems at the core of AIDS, of course--the virus attacks the very cells that are supposed to defend us from infectious invaders. Back in 1984, though, nobody could foresee just how insidious HIV was, how many feints it had to parry the thrusts that the immune system deployed against it. Nobody knew at first how genetically variable HIV was, constantly changing even after it infected a host. It was an incredibly elusive moving target, hard for any potential vaccine to hit. Nor was it understood how the virus could hide out, often for years at a time, invisible but nonetheless preparing the massive assault on the immune system that characterizes late-stage AIDS. It’s a terrible disease for patients, and so far it hasn’t proved a forgiving one for scientists.

John Moore, who studies the structure of HIV at the Aaron Diamond AIDS Research Center in New York, puts it this way: The guys who were hoping for a vaccine in a couple of years were really working on the assumption that they would be able to take the obvious approach. In other words, they trusted dumb luck, the way they had with a lot of other, earlier vaccines. Unfortunately, with hiv it’s not that simple. You can’t just inject this virus into people and hope their immune systems will make the right antibodies against it. What’s emerging pretty clearly, says Moore, is that the immune system is multifactorial and complicated. Antibodies are just one part of that complicated defense system, and in the case of HIV they may not be enough. When humans are infected with HIV, their immune systems make antibodies galore to the virus, but apparently those antibodies don’t ultimately protect them. The question is: What other parts of the immune system will we have to stimulate to get protection against AIDS?

We’re really out there in the unknown, admits Murray Gardner, the director of the Center for AIDS Research at the University of California at Davis. It’s ten years since this thing started and in many ways we’re still groping in the dark.

How did scientists involved in the vaccine race slip from the confidence of 1984 to the more chastened and tentative attitude prevalent today? It’s a long story, but we can start by understanding just why virologists regard HIV as a far nastier customer than those they’ve traditionally dealt with.

All viruses, including HIV, are primitive creatures: tiny bundles of genes wrapped in protein, quite unable to reproduce until they’ve infected their chosen host. Once inside the host’s body, however, they show a remarkable range of different behaviors. Some viruses, like polio and smallpox, stumble into the immune system like dim-witted thugs, setting off alarms everywhere. But the AIDS virus is different. It’s a retrovirus, one of the very few so far known to infect humans, and it steals into our cells like a master criminal. Once there, it hides in the cells’ DNA, undetected by the immune system, covertly copying itself every time the cells divide. Worse, it eventually kills off our key sentinels against infection: the white blood cells, called T cells, that are crucial to an effective immune response. Worse yet, nobody knows exactly how the virus goes about its killing spree. Research published in the British journal Nature in March suggested it may hide for years in the lymph nodes (small but important immune system checkpoints found in the neck, armpits, and groin), then emerge, either gradually or suddenly, to destroy T cells in the bloodstream.

Whatever the virus’s modus operandi--and chances are it may have many tricks up its sleeve--by the late stages of the disease it’s wiped out virtually all T cells. Lastly, and perhaps worst of all, HIV is genetically unstable. There are dozens of different known strains; in just a few years within a single individual, the virus can transmogrify itself so often that the human immune system can no longer keep up with it.

So unstable and dangerous is the virus that by the late 1980s most researchers felt you couldn’t justify putting whole viruses--whether killed or disabled--into a vaccine. As Dani Bolognesi, a virologist at Duke University, puts it: With a killed-virus vaccine, the worry is that you don’t kill everything--meaning that if even a single virus survives, it may sneak into a T cell, replicate, and begin the infection the vaccine is designed to prevent. With an attenuated virus, Bolognesi continues, the risk is different. Assume you really have made the virus nonpathogenic, unable to cause disease. Even so, we don’t know the long-term consequences of having any retrovirus inside you for 20 years. Retroviruses can affect your genes, turning them on or off, and nobody’s sure what tricks of this sort HIV might play, even if it didn’t bring on AIDS. Over time, Gardner explains, it might turn on oncogenes--genes that cause cancer--or God knows what.

That fear led to the first pothole on the road to a vaccine. To avoid such dangers, most researchers concluded, they’d have to turn to a brand-new vaccine technology. Rather than relying on whole virus, they pinned their hopes on using chunks of viral protein as harmless HIV stand- ins. These proteins--called antigens, because they generate antibody response in the host--could then be spliced into a harmless carrier organism like vaccinia, the non-disease-causing virus traditionally used in smallpox inoculations.

In fact, a successful vaccine against hepatitis B, using more or less this recombinant technology, had debuted in 1986. And by 1990 it looked as if the approach might really work for HIV too, thanks to a promising study by Marc Girard of the Pasteur Institute in Paris and Patricia Fultz of the University of Alabama at Birmingham. Until that point, the most encouraging results had been achieved in macaques, using injections of whole virus to ward off a monkey version of AIDS. Fultz and Girard, working with chimpanzees, showed for the first time that you could raise a protective response to HIV using just pieces of protein from the outer envelope of the virus.

Their finding seemed to lift a huge cloud from the entire AIDS field. But the respite didn’t last long. Within a year hope began to recede again, succumbing to the recurrent waves of pessimism that have characterized the AIDS battle. Experiments with recombinant vaccines just weren’t going--and still aren’t going--all that swimmingly. We’ve been trying, Fultz says, but the major problem has been that the chimps’ response wanes very rapidly following immunization. The protection we have achieved requires multiple vaccinations. That’s disappointing if you’re trying to envision a useful human vaccine. It would be impractical-- particularly in developing countries--if you needed four initial vaccinations and then a booster every year, Fultz concedes.

Trying to make a recombinant vaccine work against this heterogeneous virus is asking a lot of a little tiny clone, which is all a recombinant protein really is, adds Gardner. I’m not saying recombinant work is invalid, but the fact is that there’s something like six times as many studies in monkeys where recombinant envelope approaches haven’t worked as there are studies where they have. And even where they have, it’s usually when the experiments stack the deck in their favor by optimizing conditions. Recombinant vaccines have worked in the lab, but none of them has shown long-lasting, strong, or broad protection against mucosal challenge or cell-associated virus. In other words, they haven’t demonstrated protection where it may count the most--against virus entering through mucous membranes, the major route for sexual transmission of AIDS, or virus transmitted in infected cells.

Even if the recombinant approach does eventually pan out, there are plenty of hurdles on the way. Which viral proteins will make the most potent vaccine? What strain (or strains) of virus should the proteins come from? There’s a vast range of choices, and nobody really knows how many will have to be tried before somebody hits on the magically right combination. For a successful vaccine we’ll need to stimulate all the arms of the complicated immune system, says Moore. But in the meantime, he says, biotech companies doing vaccine research have taken the what-you- need-is-what-we-can-give-you approach.

And just what are they giving? At the moment, most of the 11 recombinant vaccines in human trials use all or part of gp (for glycoprotein) 160, a large, sugary protein found on the virus’s outer surface. Without this protein, HIV can’t lock onto--and therefore can’t infect--T cells. Gp160 consists of two major portions--a protruding knoblike structure, called gp120 (which the virus uses to attach itself to host cells), and a smaller protein, gp41, which anchors the knob in the virus’s outer wall. These envelope proteins act as particularly strong stimulants to the immune system, so they’re natural candidates for a recombinant vaccine. Gp160, for instance, is the mainstay of the highly controversial vaccine from MicroGeneSys, a private biotech company.

MicroGeneSys first drew flak last year when it lobbied Congress to become the sole candidate in a massive $20 million trial to be conducted by the U.S. Army and paid for by the U.S. taxpayer. The move--widely criticized by scientists--didn’t succeed; the Department of Health and Human Services stepped in and proposed instead that the National Institutes of Health run a study comparing the vaccine from MicroGeneSys with two gp120 vaccines, produced by Chiron and Genentech. Meanwhile, in February, in a stinging letter to Nature, Moore and two colleagues argued that the MicroGeneSys vaccine used a misshapen form of gp160. Our opinion, these researchers concluded, is that there could not be a worse choice from the current envelope glycoprotein vaccine candidates than MGS gp160 to stimulate at least one important arm of the human immune system--that is, the production of antibodies to a particularly crucial part of the virus.

Another researcher, who would speak only off the record, points out yet another potential problem, this one common to virtually all the vaccine prototypes based on envelope proteins. All these proteins, this researcher notes, are derived from only three sets of lab isolates--that is, harvested from only three individuals among the millions infected worldwide. Some of these viruses have been cultured and massaged in labs for as long as ten years. Could such hothouse products provoke an effective immune response against all the multiple types of HIV found in the real world? The question is particularly troubling when you consider that the viral gene that produces gp120 changes at the rate of about 1 percent a year in any given virus. Mightn’t the real-world virus already be one step ahead of the vintage proteins slated for vaccines? Such concerns--and the worry that recombinant vaccinces may not provoke a sufficient immune response to infected cells--have led at least one famed vaccinologist to reconsider what was unthinkable: Jonas Salk is conducting trials with a vaccine made from whole inactivated virus. Salk’s approach is so far being tried only in HIV-positive patients, to see if it can give their ailing immune systems a boost. Still, his work could offer glimpses into how the virus interacts with living humans--insights that might also help researchers design better vaccines to protect the uninfected.

There are other troubling questions in the air. Beginning in late 1991 a series of research reports appeared that, according to an alarming headline from the pages of Nature, turned AIDS research upside down. The most immediately eye-opening of these reports came from James Stott, a virologist at England’s National Institute for Biological Standards and Control. He and his colleagues had been working on simian immunodeficiency virus, or SIV--the close analogue of HIV that infects monkeys, producing a rapid AIDS-like disease. As had previous investigators, they found they could protect macaques by vaccinating them with inactivated SIV; no surprise there. They also found they could achieve protection if they inoculated the monkeys with SIV-infected human T cells. Not amazing either: such cells most likely flaunt telltale signs of the virus lurking inside them and could thus prick the monkeys’ immune systems into a defensive response.

The real stunner was Stott’s third finding. As a control, he had also injected four macaques with uninfected human T cells. Amazingly, two of these monkeys produced a protective immune response not just against the foreign-looking injected cells (which you’d expect) but against SIV. What in thunder, Stott and his colleagues wondered, was going on? How could a normal, uninfected human T cell raise immunity against a monkey virus that the macaques had never seen before? It was almost as if you’d given a prisoner in a locked cell a cupcake and then found he’d suddenly acquired a bazooka.

Many investigators suspected the answer, but it was immunologist Larry O. Arthur of Program Resources (a contractor with the National Cancer Institute) who finally showed what was happening. Neither HIV nor SIV grows on its own. Both have to be bred in human cells growing in lab dishes. The virus reproduces and eventually buds out of these cells in large numbers-- mimicking its behavior in the body of a host animal. Could these lab viruses, Arthur wondered, be grabbing proteins from their human incubator cells and then carrying them into the macaques? If so, it wouldn’t be a surprise to find that the monkeys’ antihuman immune response would also work against SIV: the virus was, thanks to its kleptomaniac habits, carrying stolen human proteins. Arthur actually found the proteins in question and proved that in fact the monkey virus does pick them up while incubating in human cells.

Stott’s and Arthur’s work jolted vaccinologists. Had all their experiments been bollixed up by proteins the virus filched from its human incubator cells? The new findings underlined again HIV’s inherent deviousness--its ability to interact with living things in extraordinarily diverse and confusing ways.

The truth is that nobody really understands what’s going on between these invading retroviruses and the host’s immune system. Which of the many proteins in the virus set off which of the many trip wires in the immune system? Where do those proteins come from--the virus or the human cell it was bred in? And which of the immune responses directed against the proteins fend off disease?

Even when researchers have raised some sort of defense against the AIDS virus, as they’ve done in chimpanzees, they don’t really understand how it’s worked. Your body has at least two major kinds of immune weapons, and they’re quite distinct from each other. One system fastens antibodies to a foreign invader circulating in the blood, thus marking it for destruction. Another system relies on a network of killer cells to seek out the virus hiding inside infected cells. Until researchers know which of these systems is more important for neutralizing HIV infection, the quest for a vaccine will rely to a disquieting degree on trial and error.

Finally, as if all these doubts weren’t enough, yet another lurking issue surfaced with a vengeance in 1991: autoimmunity. One of the fundamental mysteries of AIDS has been how HIV manages to destroy so many more T cells than it seems to infect--it’s rather scarce in the bloodstream until the late stages of the disease. Moreover, you can produce large quantities of antibodies against the virus yet still succumb to the disease. The revelation this past March that HIV hides in the lymph nodes for years and multiplies there begins to explain some of these discrepancies--even in the early stages of infection, there’s a surprisingly large covert reservoir of virus. But not everyone is convinced yet that these findings are sufficient to account for the massive T cell destruction and other cell damage caused by AIDS. Some researchers suspect that the virus may do part of its damage indirectly, by destabilizing the immune system, tricking it into an assault on itself, or causing T cells to commit suicide.

There are several models for an autoimmune component in AIDS, but Geoffrey Hoffmann--a maverick theoretical immunologist at the University of British Columbia in Vancouver--offers one of the more intriguing. Hoffmann, a disarmingly straightforward Australian, frankly admits his ideas are difficult: I have to do a lot of hand waving to explain it, he concedes. But the rudiments aren’t too hard to grasp. They rest on a long-standing mystery of the immune system. Our T cells are designed to coordinate an attack against foreign invaders. But how do they know what’s foreign? Why don’t they attack our own tissues as readily as they do marauding viruses?

Left to themselves, Hoffmann argues, they would. Each of us harbors some T cells that spring into action every time they meet a cell bearing a set of distinctive proteins that--just as a nation’s flag identifies it--brand you as yourself. Many immunologists think our bodies destroy most such antiself cells before they mature. But Hoffmann thinks they’re merely kept in check, diverted from committing mayhem by other cells, which carry proteins very similar to our own self-identifying badges. These counterbalancing cells preoccupy the potentially autoimmune cells and thereby stop them from launching an attack on us.

Fascinating--but what does it have to do with AIDS, autoimmunity, and vaccines? According to Hoffmann, the answer lies in an odd feature of a protein we’ve met before--gp120, the knob found on the envelope of the AIDS virus in many of the recombinant vaccines now under development. Previous research has suggested that a key part of this protein strongly resembles the ID badge carried on the cells that restrain our autoimmune cells. Hoffmann argues that this similarity can, in the human body, lead to a Keystone Kops-like series of mistakes that results in the collapse of the immune system. Think about it: to the body, a key part of the AIDS virus looks like--of all things--the self badge on a crucial subset of its own cells. The body, reasonably, launches an immune attack on HIV. But that means the immune system’s attack on HIV also destroys some of its own cells. Worse, these cells are the very ones that restrain your potentially autoimmune cells from targeting the rest of your T cells. The immune system, its network of checks and balances disrupted, self-destructs.

This scenario has led Hoffmann to raise a frightening possibility. Might not a recombinant vaccine backfire? If it raises a strong immune response against gp120, couldn’t it also trick the immune system into a devastating attack on itself, bringing on the very collapse the vaccine was designed to prevent? Hoffmann believes that it could. At least, he says, we need to be concerned about the possibility. This has led him to propose a counterintuitive strategy for AIDS vaccines. Alternative vaccines are conceivable that do the opposite of what conventional vaccines do. That is, they’d make you tolerate something as well as fight against it.

A vaccine that makes you accept unprotestingly part of the germ that causes the disease? That would be a first, and many immunologists are frankly skeptical. The experimental evidence is extremely poor for any of these autoimmune models, comments Moore. Hoffmann’s best support so far comes from some intriguing experiments in mice that he reported in Science two years ago. He found that mice that hadn’t been exposed to HIV could make antibodies against the virus if they were exposed to cells bearing proteins similar to their own self badges. That strongly suggested at least some kind of resemblance between HIV and mammalian self proteins-- the sort of resemblance that could lead to the tragedy of mistaken identities Hoffmann thinks may happen in AIDS.

Yet despite their skepticism, few researchers will dismiss Hoffmann’s ideas out of hand. AIDS, after all, is a difficult, many- faceted, still largely baffling disease. Maybe those interested in autoimmunity are opening up the door to another piece of this huge picture, says Gardner. And it’s not unreasonable to suggest that vaccines could cause autoimmune phenomena--vaccines have been known to enhance infections before. Those ideas are perfectly reasonable as hypotheses. The problem is proving them.

By now you may think vaccine research has become hopelessly mired in a morass of doubts, conflicting theories, and complications. Yet the need for a vaccine seems more urgent than ever. The news that the virus hides in the lymph nodes during the early years of infection--invisible, growing, apparently unimpeded by the immune system--gave many researchers pause. What does such early, large-scale infection imply for drug treatments and vaccines? It suggests you should probably do as much as you can as early as you can to prevent the seeding of the host, says Bolognesi, because once the virus is in, it’s hard to imagine any kind of treatment that would keep it from progressing. So it shows us once again that an AIDS vaccine will have to be extremely effective at blocking the virus from entry into the host.

At times it almost seems as if the only predictable theme in AIDS research is that AIDS is a subtler, more labyrinthine, and meaner disease than we’d thought. Does that imply we’re not much further than we were in 1983, when we began? If you ask researchers for a vaccine timeline, you can’t help noting a hint of déjà vu: For a vaccine in widespread use, I’d say ten years, Fultz hazards, echoing what seemed like a safe bet a decade ago. Kathelyn Steimer, who led the research to develop Chiron’s recombinant vaccine, one of the vaccines slated for the NIH trial, is a bit more optimistic. I think it’s not inconceivable to look at the year 2000 and see a first-generation vaccine--one that’s, say, up to 50 percent effective. It wouldn’t offer complete protection, but it would offer enough protection to have an impact on the epidemic, she says. A lot can happen in seven years.

In fact, the mood among vaccine researchers, while chastened, isn’t as pessimistic as the problems facing them might suggest. There have been some results that indicate it’s worth gamely plodding on. Late in May, for example, researchers at the New Mexico Regional Primate Laboratory announced they had protected monkeys against infection in conditions that simulate sexual transmission--now the most common way of contracting the disease--by using the traditional approach of a whole but weakened virus. A certain professional stoicism has set in. Maybe, says Moore, the public gets skeptical because too many scientists shoot off and say, ‘If you back me, I’ll come up with the goods.’ But if you’re a scientist, you have to be optimistic. You have to do your experiments--without shouting out to the press every time you get a good result. You have to keep going. You can always think of a million reasons why something won’t work. But if that stops you, you’ll never get a result.

Gardner agrees. Some people are pessimistic about ever getting a vaccine, he says, but who knows what might work? What worries him more in the meantime is the difficulty of reconciling the pressing needs of desperate patients, the competitiveness of companies, and the rigors of good research. If it weren’t for the urgency of the situation, right now we’d be demanding much more basic science, he says. In other words, more testing should really be done with animals before vaccines are tried in humans. Even given the urgency, I think we should be careful--not skimp research at the basic level, not rush to spend all our money on premature clinical trials. If we’re careful and go slowly, we won’t wreak havoc and we’ll prevent a great disaster. Nothing could be worse, from a vaccine researcher’s point of view, than a dicey vaccine that gives people a false sense of security.

Success, in the end, is likely to come the slow, hard, punishing way. And--barring dumb luck--it won’t come tomorrow or next month. Moore puts it very succinctly: If we could have done this by the seat of our pants, we wouldn’t be sitting here now.
Comment on this article

Discover's Newsletter

Sign up to get the latest science news delivered weekly right to your inbox!

Collapse bottom bar

Log in to your account

Email address:
Remember me
Forgot your password?
No problem. Click here to have it emailed to you.

Not registered yet?

Register now for FREE. It takes only a few seconds to complete. Register now »