Revealed: Fatal flaws of drug that gave hope

By Brian Deer
The Sunday Times (London) 16 April 1989

If anyone can reasonably be described as the world's top doctor it is Samuel Broder, the director of the American government's National Cancer Institute, near Washington. He is the physician, scientist and politician who commands the richest medical research centre on earth.

His position is mostly due to his work on AIDS. Four years ago he unveiled laboratory tests on AZT the drug later licensed as the only treatment for the disease. He helped steer it through regulatory procedures from test-tube to patients in a record-breaking 19 months. He also led the campaign that won the drug formal approval in more than 60 countries.

His work, at the height of the AIDS panic, earned him Ronald Reagan's appointment to the top job at the cancer institute. With two others Dr Robert Gallo, co-discoverer of the Human Immunodeficiency Virus, and Dr Anthony Fauci, the official head of America's AIDS programme he now runs the medical world's global response to AIDS.

They work from the vast campus of the National Institutes of Health and, although there are rivalries, operate like a high command. "I view AZT as the battle of El Alamein," said Broder. "It is symbolic that we can do something against the virus that causes AIDS; that we can make progress; that those who preached that it was inherently untreatable were wrong."

For a nation traumatised by the epidemic, such optimism became as psychologically infectious as fear of the virus itself. By the time the drug was licensed, in March 1987, 36,000 Americans had been diagnosed with AIDS, a quarter of whom were already taking AZT in "compassionate" experimental trials.

If one drug could work against the virus, the message went out, there was hope that the corner could be turned. AZT was not a cure but, its supporters said, at least it fought the virus, improved health and extended life expectancy.

There have been impressive claims about AZT's beneficial effects. Dr Ian Weller, chairman of the British Medical Research Council's study of the drug, believes the average patient can gain up to two extra years of life. Others claim more. Specialists at London's St Stephen's hospital said in a recent paper that AIDS patients on AZT survive an average 3.5 years longer.

The drug is now prescribed to 50,000 patients in America and Europe. It is extremely expensive: in Britain, a year's full-dose treatment costs the National Health Service £ 4,500. In America the price is $8,000 a year, in Australia A$10,000.

Broder found the drug on the reject shelves in North Carolina of the Burroughs Wellcome company, a subsidiary of the British firm, Wellcome plc. It had been looked at in the 1950s as a potential cancer treatment but abandoned in 1964 as being too poisonous.

The drug, technically a "DNA chain terminator", is absorbed into cells and interferes with the replication of the basic building blocks of life. The benefit for people with AIDS is that the virus is very vulnerable to this effect.

When doctors began using the drug they were impressed with its immediate effects. Some severely ill patients put on weight and their mental state improved. Short-term gains were found in their immune systems. There was a general feeling that something beneficial was going on.

In recent months, however, a more worrying picture has started to emerge. A paper from a team of researchers at the Claude Bernard hospital in Paris, published in The Lancet, described the benefits as "disappointing" and said that because of the drug's damaging effect on blood cells, the average AIDS patient could expect only six months' benefit from the drug.

Half a year's treatment is better than nothing, but new research also shows that some people who have used AZT would have been better not taking it. About 50% of patients suffer severe side-effects, largely caused by the drug attacking bone marrow. Serious side-effects include anaemia and a collapse in the white blood cells that fight infections, making patients even more prone to disease.

Some can be rescued by abandoning the drug and having urgent blood transfusions; but others are not so lucky.

A series of disturbing reports from Britain, America and Canada have shown that the damage is not always reversible, that the virus becomes even more virulent after treatment and that discontinuation of AZT can provoke other illnesses.

Such evidence is causing unease among scientists and has prompted critics to attack efforts to increase its use. "This stuff is horseshit," said an internationally-recognised American researcher, who asked to remain anonymous.

AZT may be bad enough for those whose lives are shortened by it. But its critics claim that it is also obstructing more promising areas of research, and that those delays mean more lost life. With American AIDS research exceeding that in any other country by 100:1, the most extensive data on the virus are held by the American Foundation for AIDS Research, which monitors drug investigations in the United States.

"Given the very limited resources we have in physicians, volunteers and dollars the big emphasis on AZT research has delayed other drugs and there would be other drugs now were it not for AZT," said Dr Mathilde Krim, the foundation's director. "It has occupied too much of our attention, to the detriment of other treatments."

Broder's research programmes at the National Cancer Institute, which has a budget of £ 1 billion a year, are cited as the most glaring evidence that the sort of research needed to combat the AIDS virus is being "crowded out" by AZT.

Broder believes the development of treatments for AIDS means finding other compounds that are chemically very similar to AZT, but which lack its toxicity. Such drugs which include compounds called DDI, DDC and DDA, are being tested in laboratories in America, Europe and Australia.

But these compounds are nothing like the breakthrough that would be needed to overcome the devastation of AIDS. Like AZT, they are essentially off-the-shelf chemicals developed in the 1950s, and there is a growing sense that if anything is to tackle the virus, it will need to be "rationally designed", rather than discovered at random in the test-tube.

"To hear that these old compounds are being tested is very strange," said Dr Norbert Rapoza, a drug analyst at the American Medical Association in Chicago. "It's not how much money you put into research that counts, it's how you use it. There is a desperation and we understand that, but if you don't take account of modern technology, you are doomed to repeat the errors of the past."

At the more practical level of human tests of other treatments, research is in crisis. Despite the risks of AZT and its lack of efficacy, the American government is supporting twice as many drug trials and spending four times as much money on AZT as on all other potential AIDS treatments put together.

Behind these tests is a two-fold strategy which AZT's supporters believe will keep it in use for years to come. First are trials which combine AZT with other medicines, such as Wellcome's anti-herpes drug, Acyclovir. Second are projects such as the Anglo-French "Concorde" study, which aims to extend AZT treatment to people who are infected with the virus but who have not developed AIDS.

"The implications for the development of new agents are chilling," said Dr Stephen Marcus, medical director of the Triton Biosciences company in San Francisco. "The patients are not available. They are scooped up into clinical trials for AZT. Big clinics and the top medical centres are extremely busy with AZT, so they don't have the space and the staff to manage other projects." One drug being held back is Imuthiol, a compound of heavy metals which many researchers believe can help rebuild patients' immune systems. In America doctors have found it difficult to recruit patients, and in France, where a team based at five hospitals has already published a report in The Lancet on its beneficial effects, there are also difficulties.

Dr Jean-Marie Lang, a researcher in Strasburg who is studying Imuthiol, said: "Since most people have been dealing with AZT, we are having a big problem convincing people. There is considerable scepticism now about whether any treatment can be effective."

Despite the huge bias in research towards AZT, there has been little published in leading journals on what the work has achieved. To date, only two significant papers from independent researchers have been published on the general efficacy of AZT. One was the pessimistic Claude Bernard hospital study, the other was the original American trial that got the drug its licence.

That study was published in July 1987 in the New England Journal of Medicine, and is still the one solid item of scientific evidence cited by doctors. Its unique power lies in its claim to have been a "double-blind, placebo-controlled trial", in which one group of patients with AIDS or advanced AIDS-related illnesses received AZT and another group received an inactive dummy pill. "Double-blind" means that neither doctor nor patient knew who got what.

At face value, the results were impressive. Over an average 16-week period, of 137 patients receiving the placebo, 19 died. But of the 145 receiving AZT only one died.

The difference was so significant that the study was abandoned ahead of schedule in September 1986, and Broder took part in a celebratory announcement declaring a breakthrough in AIDS research.

But analysis of the study reveals that it was severely flawed and, in strict scientific terms, invalid. Because of the extreme toxicity of AZT, the double-blinding collapsed at an early stage, and it was possible for both doctors and patients to know who was taking the active drug and who was getting the placebo. As a result, the two groups may have been treated differently and serious bias introduced into the results.

One of the most forceful critics of the study is Joseph Sonnabend, a British doctor and scientist who worked at the Medical Research Council in London and now runs a medical practice at the epicentre of the AIDS epidemic in New York. He believes the "unblinding" led to the AZT group getting a better standard of care than patients taking placebo, and that the number of deaths on the trial was greater than in the normal progression of AIDS.

"The mortality was unacceptable," says Sonnabend. "I have treated many, many people with AIDS and they do not die in those sort of numbers over a 16-week period. People in that trial died because they were neglected not deliberately, but because things happen like that."

A further question-mark is that the trial was abandoned on ethical grounds at almost precisely the time that laboratory tests and more recent research suggests the benefits of AZT were wearing out. It was also the time when some of the most severe side-effects were becoming critical, raising the question of what would have been revealed if it had run its full course.

The speed with which AZT got its licence, and the fact that some of the usual scientific tests were skipped, has also led to attacks from outside the AIDS world. One powerful critic is Dr Seymour Cohen, a pre-eminent American biochemist, who believes the medical response to AIDS was "panic-determined".

"The whole idea we had in the Western world since the war of knowing what we were doing before we handed out the drugs has been abandoned," he said. "We have damaged our medical tradition and that is very serious."

Some of the most serious doubts came from the American Food and Drug Administration, the most influential agency responsible for protecting the public against drug risks. It confirmed that the double-blinding was broken, but nevertheless gave AZT its first approval for use.

Reports obtained from the FDA under the American Freedom of Information Act reveal that records compiled during the study were altered, without explanation, giving the drug a more favourable record; "multiple deviations" from the terms of the trial occurred; and FDA investigators argued for data from one of the centres to be dropped altogether from the results.

The reports also show that one of the FDA's senior staff, Dr Harvey Chernov, who analysed the pharmacological data on AZT, argued that it should not be granted a licence when it was. "The full preclinical, toxicological profile is far from complete," he told his superiors. "The available data are insufficient to support New Drug Application approval."

Why the FDA then licensed AZT is not in much debate among insiders. The now suspect research's 19:1 mortality ratio over 16 weeks was deemed to be better than nothing. And as an arm of the American government's public health service, the FDA was faced with Broder's endorsement, combined with powerful political and consumer demands not to hold back the first tangible evidence that something was being done against AIDS.

Dr Ellen Cooper, the FDA medical investigator who analysed the trial, is today adamant that the approval was justified. But she agrees that the science was only one factor in reaching a decision. "We're all under tremendous pressure and there's no question that politics is a much greater part of AIDS drug development than approval and availability of drugs in less publicly-visible diseases."

Criticism of the FDA approval which was followed by an effective rubber-stamping by Britain's Committee on the Safety of Medicines and other licensing bodies worldwide is rejected by Broder. He remains convinced that AZT is a vital "stepping stone" which has shown a sceptical world that progress is possible. *