Taking the Medicine: A Short History of Medicine’s Beautiful Idea, and our Difficulty Swallowing It

by Burch, Druin

  From the Nuremberg code, the World Medical Association developed their 1964 Declaration of Helsinki. It has undergone five major revisions since, the last in 2000, and is intended to represent guidelines for best practice in medical research. It represents today’s rough consensus. The declaration notes that research, like medical treatments themselves, must balance ‘risks and burdens’. It is an impressive document, and an imposing summary of the degree to which modern medicine takes experimental method – and the ethics attached to it – seriously. When it comes to informed consent, this is what it has to say:

  In any research on human beings, each potential subject must be adequately informed of the aims, methods, sources of funding, any possible conflicts of interest, institutional affiliations of the researcher, the anticipated benefits and potential risks of the study and the discomfort it may entail.

  That is an awful lot to tell to people. Its effect, like that of many of the restrictions over experimental therapies, is often destructive. Rather than always making sure that patients are kept safe, it frequently helps make sure that the necessary experiments never happen. Protecting people from the risk of trials is important. It fails to acknowledge that protecting them from the risks of treatments that have never been reliably trialled is even more so.

  The later history of thalidomide remains curious. For drug companies, surprises during a drug’s pre-clinical development are often full of promise. An unexpected effect can be the key to an entirely new use. Surprises during drug-testing, in contrast, are usually disastrous. The damage that thalidomide did to nerves and to embryos fell into the second category. Some doctors were astute enough, however, to realise that the drug’s harmful side effects might have their own promise. Selective toxicity, the idea uncovered by Ehrlich, was a proven method of aiming a drug at a particular target. If thalidomide stopped a developing embryo from growing properly, that suggested it was toxic to cells while they were dividing. Therapies that are specifically poisonous to rapidly dividing cells are the main line of attack against cancer.

  Cancerous cells are not those of a parasite. They are not like bacteria, full of molecules and receptors never found on human cells and therefore excellent targets for selective destruction. Cancer cells are our own. Mutations lead to their escaping from the normal checks and controls that the body exerts on cell division. So cells begin to grow and multiply in ways they are not supposed to. The reason so many treatments for cancer are highly toxic is that cancer cells are too similar to those around them for magic bullets to easily make a clean hit. Radiation and anti-cancer drugs are generally most harmful to cells in the process of division, since many of these will belong to the cancer. But not all. Collateral damage means that hair falls out, the lining of the gut sheds, wounds fail to heal; all those bits of the body that are constantly regenerating are harmed. The hope is that the therapy acts like the streptomycin that Orwell took for his tuberculosis. It half sinks the ship in order to get rid of the rats.

  In the years immediately after thalidomide was withdrawn, three trials involving a total of around 200 people were performed in America to see if it had any promise as a cancer treatment. No useful effect was found, and interest lapsed. Then there was a piece of good fortune.

  The doctor Jacob Sheskin, having fled Europe for Venezuela during the Second World War, had moved to Israel once the country was founded. He became the director of the Jerusalem Hospital for Hansen’s Disease. Set up by German Protestants in the nineteenth century, it was there to look after those that others shunned. Hansen’s Disease is another name for leprosy.

  Leprosy is in some ways a similar disease to tuberculosis; the two germs have more than a little in common. And leprosy, like tuberculosis, can take many forms. Sheskin was looking after a man sent to him from the University of Marseilles in France. The patient was suffering from a condition called erythema nodosum leprosum. It is a form of leprosy in which the body’s own antibodies, reacting with the invading organism, crystallise. These solid clumps of antibody and germ get stuck in blood vessels. They clot off the body’s normal circulation. Wherever the blood has been prevented from flowing, flesh decays. Testicles, joints, bone marrow, kidneys, eyes, nerves, skin – gradually and often with a terrific amount of pain, these bits of the body have their blood choked off and begin to die. Where the circulation to a patch of skin has been cut, the skin breaks down and weeps, sections of it dying away. Boils and ulcers appear on the outside of the body. They are mirrored by similar ones, hidden and internal.

  All drugs that work have side effects; anything that has an action on the human body sometimes causes unwanted reactions. The body’s own responses to disease also have their own side effects. Too mild and they are not enough to keep people healthy, too fierce and the damage they do to a person’s body can outweigh the harm they do to its disease. Erythema nodosum leprosum, ENL, can be an exquisitely awful way to die. It is not a manifestation of leprosy itself so much as a side effect of the body’s attempt to fight it off. Even the magic bullets provided by billions of years of natural selection are not perfect.

  When lots of small blood vessels clot off, they cause characteristic agony. The pain of a cramp is created by a temporary and mild failure of the body to supply enough blood. The pains of ENL are similar but worse, coming from a blood supply that is failing permanently. The man who arrived in Sheskin’s Jerusalem hospital had not been able to leave his bed for almost two years. Neither sedatives nor morphine gave him more than a few moments’ relief. The pain and the lack of sleep were crucifying him.

  Searching for something to give him, Sheskin came across a single bottle of thalidomide pills, left forgotten in the hospital’s pharmacy. Aware of their history, and remembering that they were once recommended as a sedative, Sheskin thought there was little to lose by giving them a try. Pregnancy was not a worry, and the man’s nerves were already being destroyed by his disease. Thalidomide might be dangerous, but his leprosy was killing him anyway.

  The effects were miraculous. Within an hour or two, the man slept properly for the first time he could remember. In the days that followed his wounds began to heal and his pain to melt away. Sheskin tried stopping the drug, and when he did so the man’s symptoms returned.

  Convinced that he was not witnessing a coincidence, Sheskin went on to try thalidomide with five other ENL sufferers. When they all seemed to benefit, he set up a placebo-controlled randomised trial. Published in 1965, it showed that the vast majority of ENL patients responded quickly and safely to thalidomide. (Safety, as always, is relative. The effects of the drug have to be balanced against those of the disease it treats.) A larger World Health Organisation trial went on to confirm Sheskin’s results in 1971.

  The main heritage that thalidomide has left us with is poisonous. Not because of what it is, but because of how it seems. Rather than a precise understanding of the dangers of flawed and insufficient evidence, we are left with a vague feeling of the general dangers of drugs. Rather than realising the essential nature of trials, we are suspicious of drug companies. Governments could have reacted to thalidomide by making sure that every drug was fully tested before being used; tested in such a way that reliable information about effects and side effects was always available. Instead they have made testing so difficult and expensive that we often do not understand the impact of the medication we take.

  Thalidomide has now been found useful for some cancers. Multiple myeloma, a disease in which antibody-producing cells become cancerous, is the best example. Other benefits were found with Behçet’s Disease, a rare disease of the blood vessels.

  The next curious step in thalidomide’s history was that people suffering from HIV found it seemed to help. It got rid of the mouth ulcers associated with their disease. In molecular terms, that was no mystery. Thalidomide seems to work in ENL (the leprosy variant) and in Behçet’s by damping down inflammation – subduing the body’s own response to disease, exactly as aspirin does when it eases a fever. Some of th
e most dangerous symptoms of HIV come when the immune system is disabled. Others, and they can be lingeringly unpleasant to live with, are due to the chronic inflammation that the disease provokes. Thalidomide can help with that. The curious part of the story was what it revealed about drug regulation.

  The Food and Drug Administration is not the world’s only regulatory authority, but it is America’s. Any product hoping to do big worldwide business needs to comply with FDA rules. An organisation that was set up with relatively small goals – reducing adulteration of foodstuffs and the false advertising of medical remedies – now regulates a market that involves a quarter of all money spent in America. That has given the FDA unexpected power, influence, and the criticism that those things draw.

  Although the FDA regulates what drugs are allowed to be sold, it has less control over the behaviour of doctors. Like its equivalents in other countries, the FDA can decide on the medical reasons that justify a drug’s use. Thalidomide, it can say, is reasonable to use in order to treat ENL. What it cannot do, again like its overseas peers, is control whether doctors take any notice of it. Doctors can prescribe a drug for whatever they please. They cannot guarantee that their prescription will be honoured – that their country will sell the drug and that a nearby pharmacy will stock it – but they can prescribe it all the same. And since that means a pharmacy somewhere can make some money by honouring the prescription, then so long as the drug is legal, that is what normally happens.

  In 1998 the FDA approved thalidomide for the use of leprosy patients suffering from ENL. That was a little curious, since America does not really have a problem with leprosy. But the approval meant that the pharmacies had an excuse to stock the drug. Doctors could therefore prescribe it, without paying any attention to the reasons that the FDA officially said were appropriate. The approval of thalidomide for ENL meant that American doctors could prescribe it for their AIDS patients, without any drug company having to go to the trouble of getting it approved for that use. It was a regulatory short cut, a way of avoiding the expensive trials of thalidomide in AIDS that no one was willing to fund and few people thought necessary. It appeared to be for this reason that the FDA approved the use of thalidomide for ENL.

  Currently a surprising amount of prescribing is ‘off-label’ in this way. A 2006 survey of American physicians, published in the Archives of Internal Medicine, put the figure at 21 per cent. Some specialities rely on this sort of prescription more than others. Paediatricians, for example, often find that drug companies are only willing to fund trials on adults: investigating the same drugs on children is more difficult, more expensive, and not required in order to get the medications stocked by pharmacies. Once they are there, it is much easier to persuade paediatricians that the adult trials probably provide accurate predictions of how the drugs will work in children. Other minority groups, like pregnant women, are equally unattractive targets for drug trials, offering only small potential market shares. People are naturally frightened of the idea of testing drugs on children and pregnant women. What happens instead is that doctors are forced to prescribe treatments that have never been properly tested in these groups.

  It seems reasonable to think that doctors should have these sorts of freedoms; more so than the drug companies trying to sell products to them. Doctors, traditionally, are trusted – the pharmaceutical industry is not. And in some ways doctors, compared to other professions, deserve the bulk of the faith that people put in them. Money can be a factor in their behaviour, but we generally trust that intelligent compassion will be more important. It is a reasonable view, and doctors work hard to deserve it. It ignores the extent to which the sound morals of doctors have not, historically, been matched by an understanding of what they were doing. Incompetent doctors, foolish doctors and mistaken doctors can give out drugs that do not work, and prescribe therapies that harm. If someone can afterwards prove that they have acted badly, they can end up in court. But unlike the disastrous effects of thalidomide, most drug errors are not easily detectable.

  There is a battle between doctors and those who want to regulate them, to control the way they behave. It is not entirely clear who should win. The doctors want to preserve their freedom over their own actions, pointing out that you need to tailor even well-justified treatments to suit particular people. What if aspirin, for example, makes them feel sick? Or they have a history of bleeding when they take it?

  The other side of the argument is to point out how superbly bad most doctors are at being efficient. We know from reliable trials what drugs someone should be on if he or she is at high risk of a heart attack – aspirin, a beta blocker and a statin, for example. Surveys always show that too few people are being prescribed the medications they need. Not through any great objection on the part of their doctors, but through inefficiency. Going through checklists and ticking boxes is not something that doctors seem to be good at. (Specialist nurses, more methodical about following protocols, are often better.)

  In the meantime, governments try to control doctors a little bit on the sly, through the financial rewards they give them and the way they regulate the borders of medical practice. Family doctors in Britain are now rewarded with extra money for treating certain people with certain drugs. Whenever certain prescribing practices have been tied to a reward in their wage packet, doctors appear capable of changing their behaviour greatly.

  Making doctors into employees – reducing their freedoms, limiting their choices – gives governments and businesses a method of getting doctors to do what they want, a way of making them cheaper and more efficient. It offers real benefits, both for reducing the amount that a society spends on health care, and also in improving the effectiveness with which some of it is delivered. There is something horrible about the prospect of it, all the same.

  The FDA has now rejected the latest revision of the Declaration of Helsinki, from 2000, objecting to two parts of it. One of the paragraphs that the agency dislikes discourages placebo trials. It suggests that new treatments should be compared with the best available alternative – being a placebo only if there isn’t really anything at all.

  Discouraging placebo-controlled trials can be perfectly reasonable. What doctors need to know about the latest antidepressant, antibiotic or sedative is not whether it works better than a sugar pill. They need to know how it compares to the best available treatment. Once one medication has been proven to be of benefit, all new therapies should be compared to it, not a placebo. Two active drugs, though, are likely to be more similar than one and a placebo. So trials comparing them need to be much bigger if they are going to show a difference. They need to last longer, recruit more patients, cost more. And at the moment, with the FDA’s support, that is money that drug companies do not need to spend. They can take the short cut of showing that their new pill is better than a placebo, then leave it up to their marketing department to convince you (without decent proof) that it is better than its competitors.

  A large amount of medical research is undertaken for marketing purposes. Since doctors still too rarely demand to see evidence that is reliable, these trials often produce enough appearance of knowledge to sway a drug’s sales figures. In that they are insincere and misleading, of course, they breach the conventions put in place after the Nuremberg trials, the clauses that say that doing bad research is as much an outrage as experimenting on people against their wishes. Arguably, since it affects many more patients, bad research is worse. As medical crimes go, it is certainly more common – and it goes almost universally unpunished.

  The other paragraph rejected by the FDA states that those involved in a trial should, when the trial finishes, be given access to the best available medicines. This is something that the FDA views as unrealistic and damaging. They say it purports to demand best-quality care for Third World trialists, but will actually just lead to fewer companies being willing to test anything in the Third World. It ties a company to providing top-quality care to people simply because they have already t
ried to improve the knowledge about what care is best. It rewards those patients willing to be in trials, but penalises the companies offering to do them. Forcing drug companies to deliver large amounts of care for free, in order to test their drugs, is likely to make them back away from new treatments altogether, particularly those aimed specifically at treating Third World conditions. It encourages companies to introduce their drugs with as little testing as they can get away with, which is often a dangerously small amount.

  The worldwide system of drug regulation, in fact, seems set up partly to discourage medications from being properly tested. In America, there are no national ethics committees. Testing a drug requires you to persuade each hospital involved that it is reasonable to do so. Each will have different opinions, forcing you to modify your trial in different ways at different places – as though morals and experimental methods were different in Kansas from Connecticut. And in America, as in other countries, the bureaucracy of running a trial is stifling. Side effects are recorded with an eagerness that veers into insanity. If some people taking a new therapy feel nauseated, or get a headache, then that has to become part of the drug’s declared side effects, something that everyone is warned about ever afterwards – even if there were more headaches and nausea among the people taking the placebo. If someone gets flu while on your tablet, that too can end up on the list of side effects – even if you happen to be doing the trial in the flu season and the rates of infection in your trial are lower in those on the active drug than on the placebo.