AT LEAST a billion new viruses every day. More than two billion white blood cells destroyed every 24 hours. These mind-boggling statistics, released to the world in January this year, have set the seal on one of the biggest rethinks in the history of AIDS research. The figures come from the first studies to measure how fast HIV reproduces in infected people, and they are hard to square with the idea of a virus quietly hiding in the immune system before springing into lethal action. Quite the opposite: from the first moment of infection HIV may be locked in a titanic struggle with the immune system, with the virus eventually triumphing through a dogged ability to sustain a phenomenal rate of reproduction, day after day, year after year.
Bolting horse
The full implications are still sinking in throughout labs and clinics around the world. But one thing at least is clear: if the new picture is right, the practice of delaying treatment with drugs until a patient has developed AIDS is a classic case of shutting the stable door after the horse has bolted. A more logical response to HIV鈥檚 war of attrition against the immune system is to begin limiting the proliferation of the virus as soon after infection as possible. And while clinical trials have yet to unearth a drug that can do this job well enough to delay the progression from HIV infection to AIDS, researchers have seldom been as optimistic as they are now. It is not just that there are several dozen new anti-HIV drugs being developed, but what this expanding range of drugs brings with it 鈥� the possibility of hitting on the right combinations of compounds to counter HIV鈥檚 tendency to evolve resistance to any one drug given in isolation.
Advertisement
AIDS has always been a difficult illness to explain. When HIV was identified as the cause of AIDS, many researchers reasoned that it must start off as a relatively benign virus, either reproducing slowly or switching itself off and sitting in a latent form inside infected cells. At some point, the argument went, HIV must be transformed into a vicious killer that wreaks havoc on the immune system and lays its victims open to sundry opportunistic infections. Over the past few years, however, AIDS researchers have slowly realised that there may be no hidden switch. And two papers in the 12 January issue of Nature brought this view squarely into the mainstream (This Week, 14 January).
Two teams, one led by George Shaw of the University of Alabama at Birmingham, the other headed by David Ho of the Aaron Diamond AIDS Research Center in New York, studied patients who had just started treatment with one or other of three experimental anti-HIV drugs. These drugs do not kill HIV: like existing AIDS drugs such as zidovudine (AZT), they merely slow down its reproduction. The difference is that they do it much more effectively, almost stopping the virus in its tracks until it finally produces mutations that allow it to develop drug resistance. And it was this potency that enabled the researchers to do what could not easily be achieved with AZT 鈥� measure how fast HIV reproduces in infected people.
The logic is simple: when a powerful inhibitor of HIV reproduction is first administered, the numbers of virus particles circulating in the bloodstream will decline because the virus is constantly being attacked and destroyed by the immune system. Eventually, the virus develops resistance to the drug and its numbers revive, but the rate of the initial decline can at least tell you how strongly the patient鈥檚 immune system is battling against HIV. It can also tell you how fast HIV must have been reproducing before the patient started taking the drug. At that stage, the virus would have been more than holding its own against the immune system so any new particles must have been produced at least at the rate they were being destroyed.
The results were astounding. Findings from both teams suggested that someone carrying HIV plays host to a monumental battle between HIV and the immune system, with more than 1 billion virus particles being produced and destroyed every day. Just how devastating this onslaught is for the immune system became clear when Shaw and Ho counted the CD4 cells in their patients鈥� bloodstreams. These are the white blood cells that become infected with HIV and dwindle in number to virtually zero in people with advanced AIDS. After the drug treatment began, the patients鈥� CD4 cell counts leapt up, suggesting that previously the cells were being continually destroyed and replaced in huge quantities. Indeed, the researchers calculate that someone carrying HIV must lose and replace at least 2 billion CD4 cells every day. Were these cells not being continually replaced it would take just 15 days for the virus to wipe them out.
The real significance of these figures, however, is that they explain how HIV could cause AIDS without undergoing a Jekyll-and-Hyde transformation from benign passenger to vicious pathogen. Instead, the virus may simply be reproducing as fast as possible from the start, while the immune system battles back as hard as it can. The result resembles the trench warfare of the First World War. Viewed from a distance, the battlefield looks static, with neither side making any significant advances. But all the time, both armies are suffering massive casualties and having to call up legions of fresh conscripts.
According to this view, there is nothing special about the transition from the outwardly healthy person who is HIV-positive to the terminally ill AIDS patient, just as there is no fundamental difference between someone with a tiny malignant tumour and a patient who is dying of cancer. 鈥淵ou don鈥檛 say that now someone suddenly has cancer just because they鈥檝e been admitted to hospital for the last few months of their lives,鈥� says Shaw. The fact that HIV-positive people can remain healthy for a decade or more, he says, just shows that the immune system, like most other organs of the body, has a huge excess capacity. 鈥淵ou can cut out three-quarters of the liver. You can remove feet and feet of gut and you do just fine,鈥� says Shaw. But remove too much of any organ, and a patient will become sick and die.
What is less clear is why the odds are ultimately stacked in favour of the virus. One possibility is that the strain of producing billions of new CD4 cells gradually exhausts the body鈥檚 ability to replenish the immune system. This could explain the gradually declining CD4 cell counts seen in HIV-positive people. But the real answer may be more subtle, says Martin Nowak, a theoretician at the University of Oxford who has developed mathematical models that describe the process of HIV infection. According to Nowak, the problem with HIV is not just its rapid and sustained reproduction, but also the huge genetic variability that is a direct consequence of this proliferation.
Unlike some AIDS researchers, Nowak and his colleagues at Oxford, Robert May and Roy Anderson, always believed that HIV was reproducing rapidly from the moment of infection. The clue, in their eyes, was HIV鈥檚 extreme variability: someone carrying HIV can harbour more than 1 million genetically distinct variants of the virus. Even though HIV seems to mutate unusually rapidly, says Nowak, it is hard to explain such extreme genetic diversity unless the virus is also reproducing rapidly.
This same diversity explains why HIV is a killer, argues Nowak, who likens the struggle between HIV and the immune system to the mythical combat between Hercules and the multi-headed Hydra. In the ancient myth, Hercules found that every time he cut off one of the Hydra鈥檚 heads, two more grew back in its place. In the same way, says Nowak, as the immune system destroys HIV, the remaining viruses become progressively more variable, and therefore harder for the immune system to recognise and hunt down. Gradually, the population of viruses inside the patient becomes so diverse that the immune system can no longer cope.
Nowak鈥檚 theoretical models can also explain why some people develop AIDS only a couple of years after they become infected. These patients, he suggests, have an unusually weak immune response to the virus. The result is that HIV is under less pressure to evolve a high degree of variability. Instead, in these patients HIV defeats the immune system by rapid reproduction alone. And as Nowak鈥檚 models predict, these 鈥渇ast progressors鈥� do indeed seem to mount weaker immune responses and harbour less variable populations of HIV than those found in other AIDS patients.
All of this theorising presumes that HIV is reproducing rapidly from the moment of infection. As yet, however, all that has been shown is rapid reproduction in the later stages of infection. The patients studied by Ho and Shaw had all been infected with HIV for some years, many were showing AIDS symptoms and all had low CD4 counts. In short, they were the people most likely to take part in clinical trials of experimental AIDS drugs. But checking that the virus is reproducing as rapidly in the earlier stages of infection is 鈥渃ritically important鈥�, says Shaw. Ho is now studying two infected people with higher CD4 cell counts who have been placed on experimental drugs, and other studies are planned.
Tip the balance
What makes these studies so important is that they will have profound implications for designing treatments. If HIV begins to reproduce rapidly from the moment of infection, it must make sense to begin fighting the virus as soon as possible, rather than waiting until a patient becomes sick with AIDS. 鈥淭he sooner you start the more cumulative benefit you鈥檙e likely to see,鈥� says John Coffin of Tufts University in Boston. 鈥淵ou may not have to do much to tip the balance in favour of the immune system, given the dynamics of the situation,鈥� agrees Jonathan Weber of St Mary鈥檚 Hospital in London. 鈥淭he question is, how do you do it?鈥�
Show stopper
Broadly speaking, there are two main approaches: you can try to stimulate the immune system into producing even greater numbers of CD4 cells, or you can give patients drugs that stop HIV reproducing. Stimulating the immune system could backfire, however: boosting the number of CD4 cells without stopping HIV could provide the virus with a larger pool of cells to infect. And in the end, argue some researchers, stopping HIV may be all that is necessary. In the patients studied by Shaw and Ho, drugs that dramatically slowed the virus down also dramatically increased CD4 cell counts. This was true even for patients who started with only a few tens of CD4 cells per cubic millimetre of blood (the normal range is between 800 and 1200 cells per cubic millimetre). 鈥淭o the very end, the immune system is still able to replenish CD4 cells,鈥� says Nowak. 鈥淚 think that reducing the virus load is everything that counts.鈥�
The problem is finding a drug 鈥� or combination of drugs 鈥� that can stop the virus reproducing over long periods, and so delay the onset of AIDS beyond the current average of about a decade after infection with HIV. The ultimate goal, of course, is to extend the symptomless phase of HIV infection so far that people carrying the virus are more likely to die of old age than AIDS. Today, the focus is on using combinations of drugs to avoid problems with drug resistance. This idea has been around for years among doctors treating patients with AIDS, but there is now real optimism that drug combinations could significantly delay the onset of the disease.
So far, only one drug has been rigorously tested in people with HIV who are symptom-free. In 1993, British and French researchers reported the sobering news that AZT, the first drug proven to benefit AIDS patients, does not seem to slow the progression from HIV infection to AIDS. This finding initially generated much pessimism, but today researchers are more upbeat about the possibility of delaying the onset of AIDS. The reason is that AZT has been joined by a new generation of drugs that are much more effective in slowing down HIV which opens the possibility of powerful drug combinations. One combination is already looking good.
The British drugs companies Glaxo and Wellcome, which are soon to merge, are now testing Wellcome鈥檚 AZT in combination with a drug developed by Glaxo called 3TC. Both drugs work by blocking an enzyme called reverse transcriptase, an essential part of HIV鈥檚 reproductive machinery. While the virus rapidly produces mutations to sidestep the effects of either drug given in isolation, it seems to struggle to evolve resistance to both drugs given together. Experiments have shown that when the most common AZT-resistant mutant of HIV is treated with 3TC, it soon evolves a mutation that makes it resistant to the Glaxo drug. For some reason, however, this second mutation makes the virus susceptible to AZT once more.
Fresh air
Things are also looking promising in AIDS patients being treated with both drugs. The combination has been given to more than 1000 people in Europe and North America; its effect on the virus has been studied in detail over a period of almost a year in some 30 French patients. Forty-eight weeks after they first took the drug combination, the blood of these patients still on average contains more than ten times less virus than it did before the trial began. 鈥淭his is the first therapy we鈥檝e seen where there is such a large and sustained fall in viral load,鈥� says Clive Loveday of University College London, chief virologist on the trial. 鈥淭his is a breath of fresh air.鈥�
The 3TC/AZT combination has not been tested in symptomless people, and Loveday stresses that it is unlikely to be the magic potion that can lift the death sentence that hangs over everyone who has been diagnosed HIV-positive. For one thing, AZT is highly toxic, so even if resistance does not become a problem, patients are unlikely to be able to stand years and years of treatment with a combination including AZT. And in any case, Loveday is sure that mutants of HIV will emerge sooner or later that are resistant to the 3TC/AZT combination. The encouraging thing about 3TC/AZT trial, however, is that it shows that combinations can achieve much more than individual drugs.
Today, there are tens of AIDS drugs at various stages of development. Many of these are much more potent than 3TC and AZT, particularly those that block another enzyme that is vital for viral reproduction called HIV protease. Two of the three drugs used in Shaw鈥檚 and Ho鈥檚 studies were protease inhibitors, but the problem is that HIV becomes resistant to these drugs even more rapidly than it does to AZT or 3TC: in some of Shaw鈥檚 patients, resistant mutants almost completely replaced the susceptible viruses within a month. By mixing up different combinations of the available AIDS drugs, however, and changing these combinations in individual patients every time drug resistance begins to become a problem, it might indeed be possible to extend the symptomless period of HIV infection for many years.
In an editorial that accompanied the papers published by Shaw and Ho in Nature, Simon Wain-Hobson of the Pasteur institute in Paris argued that the studies had removed much of the mystique surrounding HIV. 鈥淭hey show that HIV is behaving more and more like a virus, without frills or special effects,鈥� he wrote. 鈥淚t is unique and subtle, but a virus nevertheless.鈥� For the researchers who are battling a foe that at times has seemed to possess supernatural qualities, that is reassuring news indeed: the good thing about ordinary viruses is that they can be beaten.
More haste, less speed
REMEMBER the story of the tortoise and the hare? If Sebastian Bonhoeffer and Martin Nowak of the University of Oxford are to be believed, AIDS researchers could take a valuable cue from this most famous of Aesop鈥檚 fables.
Nowak is one of the main proponents of the idea that the problems with HIV stem, ultimately, from its rapid rate of reproduction. In the March issue of Immunology Today, he and Bonhoeffer argued that it may be possible to prevent AIDS by injecting people who carry HIV with a partially disabled, or attenuated, strain of the virus that can only reproduce slowly. The idea seems paradoxical, but Bonhoeffer and Nowak claim that an attenuated virus, which should not itself cause AIDS, could outcompete and ultimately replace fast-reproducing viruses that will definitely lead to AIDS.
Were two strains of HIV to infect the same person, Bonhoeffer and Nowak point out, they would compete for dominance in the body. And the outcome would depend only partly on how fast each strain reproduces. Just as important, argue the researchers, would be the ability of the immune system to recognise and destroy each of the two strains. If the genetic change that slows down an attenuated strain also makes the virus less likely to provoke an effective immune response, then the attenuated viruses will be destroyed more slowly than their disease-causing competitors. In theory, this could more than compensate for the sluggishness of the attenuated strain and allow it to dominate.
To demonstrate their ideas, the two Oxford researchers have run many computer simulations, varying the characteristics of hypothetical attenuated viruses. In some of these simulations, the crippled virus ultimately replaced the disease-causing virus. And even when the attenuated virus eventually lost out, its presence at least delayed the progress of the virulent strain. In real patients, say Bonhoeffer and Nowak, a dose of attenuated virus would act most effectively soon after infection, before the disease-causing strain has built up an overwhelming numerical strength. 鈥淭he sooner, the better,鈥� says Nowak.
So far, this radical idea has been given a cold shoulder by most AIDS researchers, many of whom are deeply suspicious of theoretical approaches. 鈥淭he chain of events should be that the data drive the models, not the other way round,鈥� says John Moore of the Aaron Diamond AIDS Research Center in New York. Even those who are not openly critical have trouble accepting the idea that a crippled virus could outcompete a fast-reproducing strain of HIV. 鈥淚 still don鈥檛 make the connection that it could work,鈥� says Anthony Fauci, director of the US National Institute of Allergy and Infectious Diseases in Bethesda, Maryland.
Nevertheless, there may soon be an opportunity to test the Oxford researchers鈥� ideas in animals. Over the past few years, a team led by Ronald Desrosiers of Harvard Medical School鈥檚 New England Regional Primate Research Center in Southborough, Massachusetts, has been developing attenuated versions of a closely related virus called SIV which infects rhesus macaques. His team has produced a range of SIV strains that lack parts of a gene called nef and which as a result reproduce only slowly and without causing disease in adult monkeys. Desrosiers and his colleagues have also found a human patient carrying a strain of HIV that lacks nef 鈥� someone who has been infected for more than a decade yet still has an intact immune system.
Desrosiers and his colleagues have developed their SIV strains as protective vaccines. In rhesus macaques, encouragingly, viruses lacking nef can prevent infection with disease-causing SIV. Despite this, most researchers are extremely wary of extending the research to human infection with HIV. 鈥淭he safety issues are absolutely terrifying,鈥� says James Stott of the National Institute of Biological Standards and Control in Potters Bar, Hertfordshire, whose group is also experimenting with nef-deficient SIV.
One concern is that an attenuated strain may spontaneously 鈥渞epair鈥� itself and recover the ability to cause disease (This Week, 17 September 1994). And in the 24 March issue of Science, a team led by Ruth Ruprecht of the Dana-Farber Cancer Institute and Harvard Medical School in Boston, announced that nef-deficient SIV can cause disease in infant macaques, raising the spectre of vaccinated women passing a potentially fatal virus to their fetuses.
Desrosiers is still pressing for the approach to be tried as a protective vaccine against HIV, arguing that viruses lacking nef are extremely unlikely to be passed from mother to baby in the womb. But most researchers believe the safety concerns are insurmountable. 鈥淓thically, about the only group you could test it in are people who are already infected with HIV,鈥� says Stott 鈥� which is where Bonhoeffer鈥檚 and Nowak鈥檚 ideas come in. One problem is that Desrosiers has so far been unable to infect SIV-infected macaques with his nef-deficient strains. But it may be possible to overcome any barrier to infection by smuggling in the virus inside cultured white blood cells.