The Rise and Fall of Modern Medicine (41 page)

Stock markets rely on credibility, which is precisely what Swanson and Boyer had set out to acquire by choosing as their
first target one of the best-known of human proteins, insulin. Their strategy had paid off. In the unprecedented reaction to the flotation of Genentec it is possible to discern what was to be a central feature of biotechnology in the coming years – the credulousness of investors in believing in the commercial possibilities of something they did not really understand. Few, if any, of the punters buying Genentec shares at $89 apiece understood molecular biology well enough to grasp the limits of its therapeutic potential. They could only infer that something big was about to happen, ‘big' enough for
Time
to put Herbert Boyer on its front cover:

In a similar vein, James Erlichman, the
Guardian
's expert on the pharmaceutical industry, observed that ‘the rewards for genetic engineers will be immense', a rosy view of the future that reflected the prevailing opinion:

But Genentec's immediate priority was to sell sufficient human insulin to make it profitable. This was not a straightforward matter as there was certainly no sign that the supplies of the considerably cheaper insulin from pigs and cows would be incapable of meeting demand. Two tactics were adopted. First, the intrinsic superiority of human insulin was vigorously promoted on the lines that patients with diabetes, who might have to be injecting themselves with insulin for several decades, deserved ‘the best', even if it was structurally very similar and much more expensive. Second, just in case doctors did not ‘get' the message, Eli Lilly decided to ‘phase out' its production of animal-based insulin so it was less readily available.

The general verdict on ‘human' insulin could be that, besides being a triumph for biotechnology, its advantages are more apparent than real. It held out the promise that this new way of producing drugs, so different from and so much more ‘scientific' than what had gone before, it would seem, could not fail to deliver. But that is not how it has turned out. Rather, human insulin remains biotechnology's commercially most successful product (see opposite). Indeed, the few biotechnology drugs that can genuinely be described as representing a significant therapeutic advance are a vaccine against the viral infection of the liver hepatitis B; erythropoietin (EPO), a hormone secreted by the kidney that stimulates the production of red blood cells, and Gleevec for the treatment of chronic myeloid leukaemia.
¹⁹

Biotech products in use in 1995

(Adapted from a list provided by Dr Richard J. Wurtman, Department of Brain and Cognitive Sciences, Massachusetts Institute of Technology.)

It is testament to the power of the idea of genetic engineering that the limits to its therapeutic potential were not appreciated earlier, but the reason is quite obvious. Biotechnology may be a technically dazzling way of making drugs but it is severely constrained by the fact that the only
things that genes can make are proteins, so the only therapeutic use for biotechnology products is in conditions where either a protein is deficient and needs replacing (such as the use of insulin in diabetes) or where it is hoped that giving a protein in large enough doses might in some way or other influence a disease, such as cancer.

Genetic engineering by definition cannot come up with the sorts of surprises that drove the therapeutic revolution, completely novel chemicals that just happened, like chlorpromazine, to improve the symptoms of schizophrenia, or just happened, like azathioprine, to prevent the rejection of transplanted organs. Further, the technical complexities of biotechnology markedly constrained its innovatory capacity, in contrast to the ease with which medicinal chemists in the 1950s and 1960s synthesised thousands of variations of a single chemical. In 1996, a decade and a half after human insulin launched the genetic engineering revolution, the editor of
The Lancet
mordantly observed that despite the ‘millions of dollars poured into biotechnology research worldwide', there is ‘very little to show for such investment'. Perhaps, he speculated, a new anti-cancer drug, Marimastat, developed by the firm British Biotechnology at a cost of £150 million, would be the ‘breakthrough that the biotech industry has been waiting for'? A month later Marimastat was shown to be no more effective than no treatment at all.
²⁰

Throughout the 1980s The New Genetics was blossoming out in all directions, the cumulative effect conveying the impression that the possibilities of medicine were being transformed. Thus in the same year, 1982, that the molecular biologist Herbert Boyer and the venture capitalist Robert Swanson launched human insulin, Judy Chang and Yuet Wei Kan of the University of Southern California described a technique for diagnosing the blood disorder sickle cell anaemia in the foetus while still in the womb, thus opening the way to a whole new medical venture: the elimination of genetic disease by prenatal screening and the selective abortion of foetuses found to carry abnormal genes.
²¹

The genetic mutation involved in sickle cell anaemia takes the form of the substitution of one triplet of nucleotides, GAG, with another, GTG. The messenger RNA then carries this ‘faulty' message to the protein factory, the ribosome, which makes the haemoglobin protein, where the amino acid valine (coded for by GAG) is substituted with another, glutamic acid (coded for by GTG). This single substitution of the ‘wrong' amino acid alters the physicochemical properties of the haemoglobin protein, as a result of which the red blood cell collapses inwards (it assumes a sickle shape). The tissues are therefore deprived of oxygen, resulting in ‘sickling crises', which the patient experiences as pains in the chest and bones.

Chang and Kan's technique made use of the text-cutters that cut up DNA into fragments at the site of a particular sequence of nucleotides. Thus, a restriction enzyme that usually cuts the haemoglobin gene at the GAG sequence will not do so if it is
replaced by the GTG mutation, so the resulting fragments of the haemoglobin gene will be of a different size in sickle cell anaemia. In theory this method can be applied to virtually any genetic disease where the gene is known, with the obvious corollary that a ‘positive' prenatal diagnosis permits the prevention of genetic diseases by aborting those found to be affected. In reality, things, as might be expected, turned out to be a bit more complicated. So, to appreciate the principles of this type of genetic screening, it is necessary to take a step back and look at genetic diseases in general.

There are more than 5,000 genetic diseases. This might sound a lot but virtually all are staggeringly rare, being the result of a ‘spontaneous mutation' in the DNA a child inherits from its parents. Spontaneous mutations ‘just happen'; there are so many and they are so unpredictable that they cannot be ‘prevented' by genetic screening. This leaves a handful of commoner genetic diseases due to the inheritance of a faulty gene from one or both parents for which prenatal genetic screening might be appropriate. Most will be familiar and include the blood disorders such as sickle cell anaemia and thalassaemia; the bleeding disorder haemophilia, famously transmitted by Queen Victoria to the royal households of Europe and resulting from an abnormality of the gene that codes for the ‘clotting' protein factor VIII; cystic fibrosis, a disease of the lungs that predisposes to chronic infections which destroy the lung tissue, leading to respiratory failure; muscular dystrophy, which causes a progressive weakness of the muscles; and Huntington's chorea, which causes a dementing illness from the forties onwards and whose most famous victim was the American folk singer Woody Guthrie.

These inherited genetic diseases are currently incurable, with the obvious exception of haemophilia which, as pointed out in
the previous chapter, can be corrected by transfusions of the missing factor VIII. Their symptoms can sometimes be ameliorated, but they can only be prevented by prenatal genetic diagnosis and selective abortion. Put another way, most of these conditions can be diagnosed quite straightforwardly after birth, but by then the option of ‘prevention' is lost as the postnatal equivalent of abortion, infanticide, has not – since the German eugenics programme of the 1930s and 1940s – been permitted in Western countries.

The genetic screening of foetuses certainly can work, most obviously in the rather unusual circumstances where a particular genetic disorder is common in well-defined communities. Thus the abnormal haemoglobin gene involved in the blood disorder thalassaemia (responsible for a very severe form of anaemia) is common in Cyprus, with as many as a quarter of the population being carriers, resulting in fifty-one children being born with the disease in 1974. Ten years later, following the introduction of screening, this figure had fallen to two.
²²

This situation is, however, not typical and certainly cannot be compared to the problem of trying to find, for example, those foetuses carrying the gene for cystic fibrosis out of the tens of thousands of pregnancies in Britain every year. First, it is necessary to identify those pregnancies where the foetus might be affected, so, as a preliminary, both parents must be screened early on in the pregnancy to identify those couples where both mother and father are ‘carriers'. Prenatal testing can then be performed in these pregnancies and those foetuses found to be carrying the abnormal gene can be terminated. The complexities of this type of prenatal genetic screening are illustrated by a project that ran in Edinburgh over a ten-year period. During this time 25,000 couples were tested. In just twenty-two both mother and father were found to be carriers and thus the foetus
was ‘at risk' of having cystic fibrosis. The diagnosis was confirmed in eight of these twenty-two pregnancies and the foetuses aborted. But despite this massive screening programme, several babies with CF were ‘missed', because so many different mutations can give rise to the disease.
²³

Clearly this type of mass screening during pregnancy is an enormous undertaking, expensive in both laboratory services and the professional skills of those who administer the tests. Further, like all antenatal testing, the process of screening invariably generates much anxiety among parents. It is thus not entirely obvious that it is worthwhile screening 25,000 couples to terminate 0.03 per cent of pregnancies, or as
The Lancet
cautiously observed in commenting on the results: ‘We still have to think whether nationwide screening programmes are what we really want.' And it is a fair bet it will not happen. If this is the verdict for cystic fibrosis, then clearly prenatal genetic screening cannot be considered a valid option for preventing the many other much rarer inheritable disorders.

The practicalities of screening for cystic fibrosis have been discussed at some length because they illustrate so well a recurring feature of The New Genetics – the hiatus between anticipated benefits and reality. As the enthusiasm for prenatal genetic screening declined, so the focus shifted to ‘genetic testing', to identify those individuals at high risk of a serious disease in later life, such as cancer and heart disease. When heart disease and cancer ‘run in families', they almost invariably occur at a relatively young age and are often very aggressive. The ‘cause' in such cases is almost entirely genetic: the mutation of one gene or other involved in, for example, cholesterol metabolism (leading to heart disease) or breast development (leading to breast cancer when young).