Grünenthal’s Accidental Pregnancy Test
In 1952, an especially bizarre medical experiment featured in the pages of the American Journal of Obstetrics & Gynecology. The author of the paper was John Thiersch, a doctor and pathologist who thought he might have found a way to perform non-invasive abortions by simply giving pregnant women a few doses of aminopterin, an anti-cancer drug. Thiersch had used the drug to produce abortions in animals, and if it had the same effect in humans, thought Thiersch, then it might be an effective path to population control. Women could regularly take a small dose of aminopterin, preventing pregnancies advancing beyond the first few weeks. (By contrast, the contraceptive pill, which came a few years later, prevents conception.) In effect, Thiersch was trying to develop an early abortion pill, which he believed would be useful in many countries with poverty, booming populations and inadequate food supply.
To put his thesis to the test, Thiersch ventured into a fraught arena. Over an extended period, he administered aminopterin to twelve pregnant women in whom terminations had been suggested for medical reasons: tuberculosis, mental illness, cancer and muscular dystrophy.
Thiersch’s experiment was radical, groundbreaking and in some respects a profound disaster. In ten of the twelve women the drug worked and the women expelled the dead foetuses. But this took as long as thirty days. The lengthy delay between foetal death and expulsion meant some of the embryos ‘underwent partial resorption and dehydration’. On examination one of these babies was found to have a brain/skull malformation. These cases, according to Thiersch, were the successes.
The failures, Thiersch wrote, were cases eleven and twelve. Here aminopterin did not induce foetal death and surgical abortions were required. Both foetuses, when extracted, were found to have been damaged by the drug. One had hydrocephalus (fluid on the brain) and the other had a cleft palate and harelip.
Viewed today, Thiersch’s report makes uneasy reading. In Thiersch’s defence, he made great efforts by the standards of the times to obtain informed consent and engage independent medical oversight for his experiments. It was also a different era, one of bold and aggressive experimentation. Siddhartha Mukherjee has reported in The Emperor of All Maladies on the controversy surrounding the brutal treatment of cancer victims in the 1950s and ’60s with increasingly toxic combination doses of anti-cancer drugs. The leukaemia ward at the National Cancer Institute became known as the ‘butcher shop’; Mukherjee reports two researchers being labelled ‘insane, incompetent and cruel’ for proposing the 1961 trial of a cocktail of four cytotoxic drugs on children with leukaemia, a trial which took a terrible toll on patients but led in some cases to unprecedented remissions.
But the historical context can only explain so much. Even Thiersch admitted that the audience was horrified when he detailed his human experiments at a 1954 conference. Perhaps in part because of its shock value, Thiersch’s report quickly became famous and was much cited. And the report—indeed Thiersch’s whole career—was clearly a weapon to be used in Lyn Rowe’s legal battle with the drug companies. Thiersch’s work starkly demonstrated that it was well known in the early 1950s—years before thalidomide went on sale—that a drug given to a pregnant woman could be non-toxic for the mother and fatal (and/or malforming) for the foetus. This was going to help us with one of the most bitterly contested issues in Lyn Rowe’s claim: should Grünenthal and Distillers have checked the effect of thalidomide on the foetus?
The companies have always claimed such tests were simply not required at the time, that such checks were not commonly done, and that there was no reason to suspect thalidomide might cause damage. In 2013 the following statement could be found on Grünenthal’s website: ‘In the period before and during the thalidomide tragedy, it was the pharmaceutical manufacturer’s responsibility to decide how to test new drugs. Medicines did not have to be tested for their propensity to harm unborn life.’ Distillers has made similar statements.
In a literal sense, the Grünenthal statement is correct. There were almost no laws governing what a pharmaceutical company had to do in terms of testing. But pharmaceutical companies then and now had a clear legal obligation to ensure their products were safe. And that means taking sensible precautions, whether or not obliged to by government regulation.
It’s also fair to say that Lyn Rowe’s legal team did not ascribe much weight to protestations of innocence from Grünenthal and Distillers. But scepticism does not equal proof, and we had to produce evidence. Our research tasks were clear. When did doctors realise that embryos and foetuses could be damaged by substances ingested by pregnant women? Was it before the thalidomide era? Did sensible, responsible drug companies of that period try to investigate whether their drugs might harm foetuses? What if the drugs were being promoted for use in pregnancy, as was the case with thalidomide? And had Grünenthal and Distillers chosen to investigate thalidomide’s effect on the foetus, what would the tests have found?
To answer these questions we spent months searching through medical texts and journal articles from the 1930s, ’40s, ’50s and ’60s, talking to elderly doctors and reviewing piles of material in various archives. Soon it became clear that some pervasive myths were still current. Thalidomide was such an epochal disaster that in its wake many doctors and scientists divided knowledge about drugs and pregnancy into pre-thalidomide and post-thalidomide eras. According to that view, pre-thalidomide was a time of darkness and ignorance when, supposedly, nobody realised that drugs taken by a pregnant woman could harm a foetus. When it was unknown whether drugs even passed through the placenta and reached the foetus. When drug companies just had to hope for the best.
But then came the thalidomide disaster and, according to the story, everything changed. Prompted by the tragedy, drug companies began assiduously testing on pregnant animals, conducted rigorous clinical trials and kept a careful watch on their drugs once they were released for sale.
It’s a good story, and one that pays appropriate homage to the shock and scale of the thalidomide disaster. But, we discovered, it’s largely fiction.
Human fascination with physical difference is ancient. Giants, dwarfs, monsters, mermaids, cyclopses, albinos and two-headed creatures. The name given to the study of birth malformations is teratology, from the Greek for the study of monsters, an unfortunate though accurate historical reference. Even in the 1960s the reporting of the thalidomide disaster was occasionally blighted by resort to the offensive ‘monster’ terminology. ‘One out of every two women who took the drug bore a freak,’ a US headline screamed in 1962.
In 1959 teratologist Josef Warkany surveyed the history of his profession, and noted that over the centuries enlightened thought had tussled with superstition in seeking to explain birth defects. For much of recorded history it was believed that what a mother looked at during pregnancy could cause defects in her baby. Pregnant women were sometimes encouraged, for example, to gaze at beautiful statues and avoid looking at monkeys. This idea of mental pictures moulding the unborn child finds expression in Genesis, where Jacob and Laban reached a deal entitling Jacob to all speckled and spotted lambs and goats born to Laban’s flock. Jacob cunningly peeled some branches, creating a streaked and speckled effect, and then ‘set up the branches in front of the flocks when they were in heat and came to drink’. The scheme worked. ‘When the sheep mated in front of the branches, they gave birth to young that were streaked or speckled or spotted.’
These notions were not confined to the lay population. The sixteenth-century French surgeon Ambroise Paré wrote On Monsters and Marvels, an illustrated treatise dealing with birth defects. Paré listed expectant mothers’ mental imagery as one cause of birth malformations, and also ‘Seated Too Long, Having Had Her Legs Crossed, or Having Bound Her Belly Too Tight While She Was Pregnant’. But even then Paré was headed in the right direction, devoting a chapter to ‘monsters that are created by hereditary disease’. He also speculated that the foetus’s vulnerability to maternal vision was confined to the first six weeks of pregnancy because after this the embryo was largely formed.
Other pre-modern ideas about the cause of birth malformations included the agency of demons, witches and divine anger. Breeding between humans and animals was another culprit. Warkany reported the execution in 1642 of the unfortunate George Spencer of New Haven, Connecticut. Spencer had a malformed face and eye and, unluckily for him, a piglet stillborn in the same town with a twisted face and single eye was said to bear Spencer a startling resemblance. Spencer was accused of having sired the pig and given a very dubious trial. Despite the lack of any evidence against him, the vengeful officials of New Haven executed both Spencer (by hanging) and his putative paramour, the mother pig (via the sword) on the same day in April 1642.
Eventually though, the intellectual forerunners of today’s embryologists and teratologists began to advance their theories. During the seventeenth century the interruption of embryonic development began to be seen as a cause of birth malformations. This was in part the work of English doctor William Harvey, more famously credited with properly describing the circulation of blood in the body, who spent much time dissecting pregnant deer killed during royal hunting expeditions.
By the nineteenth century science was ascendant. In France, Étienne Geoffroy Saint-Hilaire, often seen as the father of experimental teratology, created abnormal chicks by subjecting eggs to all sorts of indignities: shaking, pricking, turning upside down, and altered atmospheric conditions. Others took up this work with further animals and techniques. By the early twentieth century defects in mammals were being produced by measures including radiation and altered maternal diets. In 1933, for example, Fred Hale, of the Texas Agricultural Experiment Station, reported in the memorably titled article ‘Pigs Born Without Eyeballs’ that a pregnant sow fed a diet deficient in vitamin A had given birth to a litter of eleven pigs, none of whom had eyeballs. By this time heredity via genes (and, increasingly, genetic disorders) were also established as prime causes of birth malformations.
A seminal contribution was made by an Australian paediatric ophthalmologist, Norman Gregg. In 1941, Gregg noticed an exceptionally high number of babies with cataracts among his patients. The story goes that one day he overheard a conversation between two mothers in his waiting room. Each had a baby with cataracts and each mentioned having had German measles (rubella) during pregnancy. Gregg searched his records and questioned his infant patients’ mothers. The results were astonishing. Close to ninety per cent of the mothers of children with congenital cataracts had suffered rubella during pregnancy. Gregg published his work in a little-known journal, and then a follow-up paper connecting congenital deafness with maternal rubella.
Clearly the foetus was vulnerable—not only to maternal illness—and this was well known. In 1939 Dr Harold Speert at Johns Hopkins Hospital in Baltimore worried about the effect that sulfanilamide—used to treat infection—might have on the foetus when given to pregnant women. He conducted a series of sophisticated experiments with pregnant rats and found increased mortality in their offspring, decreased litter size and reduced birth weight. Dr Speert published his findings, and strongly advised that sulfanilamide should only be used with ‘extreme caution’ in pregnancy until further observations of the drug’s effect in human pregnancy could be made. This was almost twenty years before thalidomide, untested in pregnancy, was spruiked by Grünenthal as utterly safe.
During the late 1940s and ’50s any number of medical texts recognised the capacity of drugs taken by the mother to pass the placenta and reach and damage the foetus. Some texts also pointed out that a dose which was non-toxic for the mother could damage the foetus. ‘Many drugs pass from the mother to the child, and this is to be borne in mind, because quantities that are insufficient to poison the former may have more serious effects on the latter,’ one 1954 text warned.
It was also understood that the risk to the foetus was greatest in the early months of pregnancy (when medically the foetus is often referred to as an embryo). For example, a 1948 article in the Journal of the American Medical Association recognised the potential risk of medicating pregnant women during the first trimester. ‘We would be hesitant to advise administration of streptomycin during the first trimester because of the generally accepted sensitivity of the fetal structures in…that period.’
The effect of drugs on the foetus was also the subject of study in Germany. In 1956, working a short drive from Grünenthal headquarters, Dr Ernst-Albrecht Josten wrote a paper that appeared in one of Germany’s most widely read medical journals, Münchener Medizinische Wochenschrift, under the heading ‘The Effect of Medication on the Unborn Child’. Dr Josten’s six-page article included an English summary that began: ‘In cases where pregnant women undergo treatment by medicaments, one must consider whether any harm can thereby be done to the fetus.’
The examples from the German literature are too numerous to mention, but mirrored publications elsewhere. In 1960, more than a year before thalidomide’s deadly effect was exposed, ‘The Effects of Drugs on the Foetus’ by Dr J. Baker of the Charing Cross Medical School in London was published. It included references to 354 previously published articles on the same or related subjects. Such was the interest in birth defects prior to the thalidomide disaster that two international conferences on the subject were held in London in 1960. Both featured multiple presentations and discussions on the effect of drugs on the foetus.
One additional document surely demolishes the theory that the era prior to thalidomide was an age of darkness with an entrenched ignorance about drugs damaging the foetus. In May 1961, six months before the thalidomide disaster was revealed, the US Committee on Fetus and Newborn drafted a statement, ‘Effect of Drugs upon the Fetus and the Infant’. The statement was so prescient and wise, that looking at it down the telescope of history it appears the committee saw the thalidomide disaster coming.
The fetus and the newborn infant often behave so differently as to warrant consideration as separate categories of the human species…data obtained from tests in mature animals and human adults or older children cannot be accepted as a satisfactory basis for recommendations concerning the fetus and infant.
The committee called for careful animal testing of the effect of drugs on the foetus. ‘Physicians who administer drugs to the fetus and the infant must be alert to unusual effects in this subdivision of the human species.’
The connection between drugs and birth defects was clearly not a dark mystery. Yet Grünenthal and Distillers promoted thalidomide to doctors, including obstetricians, as specifically suitable for use in pregnancy, without ever having tried to check the drug’s effect on the foetus. And mere months after the statement of the US committee, when thalidomide was outed as a birth defect drug, Grünenthal and Distillers started claiming that they could have done nothing to avoid the disaster.
It was spin: a completely spurious claim. Yet it was accepted without investigation or query by large sections of the media, industry, government and public alike. This nonsense remained entrenched fifty years later and we understood it would pose a serious obstacle to Lyn Rowe’s claim.
We knew that Grünenthal and Distillers would find ‘experts’ to come to court and talk about the pre-thalidomide dark age of medical knowledge. But we believed we had assembled the evidence to demolish such claims. The documents referred to in this chapter are a tiny fraction of the material we gathered. Copies of articles flooded in from libraries in Australia and overseas; Amazon made a tidy profit as we searched for and collected ageing medical texts. The material piled up in our offices and we hired casual staff to sort and file it electronically.
The mountains of material made clear that by the time thalidomide went on sale it was well known and well publicised around the world that drugs taken by a pregnant woman could pass the placenta, reach the foetus and cause damage. Scientists and researchers had for many years conducted experiments on pregnant animals to determine whether drugs might damage the foetus. Many doctors had for years urged caution in medicating pregnant women. Anti-cancer drugs were most obviously capable of damaging the foetus but other widely used drugs were also considered potentially dangerous.
In the era before thalidomide, best practice was enlightened. But by no means had all pharmaceutical companies adopted this best practice. Many were, like Grünenthal and Distillers, unenlightened. What the thalidomide disaster really did was to make public in the most dramatic fashion what was already known in informed circles about the vulnerability of the foetus to drugs. It also rapidly accelerated study of the subject. ‘Thalidomide changed the lives of the small band of experimental teratologists who [previously] worked quietly in their laboratories,’ one leading expert later wrote. ‘Now government, industry, and the media called on them for advice.’
All of this led to another important issue. Was the complacency of Grünenthal and Distillers universal in the industry? Did any drug companies of that era try to check the effect of drugs on the foetus? The work of John Thiersch, the man who used an anti-cancer drug to procure abortions, held some of the answers.
Thiersch grew up in Germany before making his way to Australia via a teaching stint in China. In Australia he developed his interest in cancer and leukaemia, worked as a pathologist in Adelaide, and in 1946 authored an article about his attempts to establish whether acute leukaemia might be transmissible. He must have been a very persuasive man. In order to test the possibility, Thiersch took bone marrow from patients suffering acute leukaemia and injected it into the bone marrow of volunteers. In other words, Thiersch set out to infect people with leukaemia. Thiersch’s volunteers were suffering from various cancers and perhaps he persuaded them that their situations were dire and would not be significantly worsened by adding leukaemia to their burdens. Thiersch later said his volunteers were terminal patients with less than two years to live, and that he had obtained ‘informed consent’ and consulted with the hospital authorities. In any event, none of the recipients developed leukaemia.
Thiersch then left Australia and found a job at the Sloan Kettering Institute in New York, a key cancer research centre. In 1947 in Boston, the doctor and pathologist Sidney Farber had pioneered the use of aminopterin to produce remissions in children with leukaemia. Aminopterin inhibits the body’s absorption of folic acid (thus ‘folic acid antagonist’); folic acid stimulates the growth of leukaemia cells (and is vital for the health of the foetus). Inspired by Farber’s successes in Boston, researchers at the Sloan Kettering Institute were working hard in the same field. It was in this environment that Thiersch found work.
In 1948 Thiersch fed a dose of aminopterin to a dog as part of an experiment. Unknown to Thiersch, the dog was pregnant and to Thiersch’s surprise it bled vaginally and aborted. ‘I took this single observation seriously and followed it up with a number of experiments in mice, rats and dogs studying the effects of folic acid antagonists in pregnant animals and fetuses.’
Thiersch soon realised that mother and foetus responded differently to drugs at different stages of pregnancy. Further, drugs taken by the mother could cause embryos to die and be reabsorbed into the lining of the uterus (resorptions), increase the number of stillborn, reduce litter size, reduce the birth size of babies, and cause malformations. Thiersch published the results of many of these experiments, reports that were widely available, including to the drug companies. Much of Thiersch’s research was funded by influential organisations: the Population Council and Planned Parenthood. Finally though, Thiersch had to give up on his abortion pill. ‘The entire approach was too dangerous,’ he concluded.
By the time the thalidomide scandal broke in the early sixties, Thiersch was one of a small group of scientists in Europe, Japan and the United States at the forefront of experimental teratology. So, understandably, the lawyers for thalidomide victims sought Thiersch’s expertise. He was happy to oblige, ultimately giving evidence in Swedish, Japanese and US thalidomide lawsuits. Thiersch had strong opinions, and they were strongly critical of Grünenthal. More than forty years after those lawsuits, and almost twenty years after his 1993 death, we found copies of Thiersch’s advice and evidence in various archives around the world.
In a lengthy written opinion for the Swedish lawyers, Thiersch said that thalidomide could and should have been tested in animals for its effect on the foetus. If this testing had been properly conducted it would at the very least have resulted in reduced litter sizes and increased embryonic death. This would have put the drug under suspicion and led to further tests and inquiry, and the terrible effect of the drug would have been discovered. By skipping reproductive testing in animals, Thiersch believed, Grünenthal had in effect experimented on vast numbers of humans.
Thiersch also provided important evidence about drug companies. During the 1950s major firms including Merck, Lederle, Burroughs Wellcome, Parke-Davis and Smith Kline & French all investigated the teratogenic potential of their drugs, Thiersch said. Thiersch himself had been asked to perform reproductive work, including tests on Diamox (a diuretic) for Lederle and on Daraprim (an antimalarial) for Burroughs Wellcome. ‘Many drug houses were aware of a potential effect on the fetus of their drugs, and actively supported investigation of this nature,’ Thiersch wrote.
Thiersch’s research and his controversial abortion experiments occasionally caused him grief while in the witness box for the thalidomide victims. The drug company lawyers seized on his work, trying to paint him as an ‘abortionist’.
‘You went to Seattle and started using aminopterin in human beings?’ Thiersch was asked by Merrell’s lawyer in one US thalidomide trial.
‘Yes. This requires an explanation.’
‘I am sure it does, doctor,’ came the response.
On another occasion, Thiersch was being questioned about his Australian work experimenting with leukaemia on cancer victims. The lawyer for the thalidomide victim stepped in, objecting that Merrell was trying to portray Thiersch as ‘some sort of beast’.
But Thiersch was an impressive witness: tough and assertive and never willing to apologise for his work. In one case he gave evidence that it would have cost just one hundred dollars and taken less than a month to run a test with thalidomide in pregnant rats or mice—a test, he pointed out, that would have shown up signs of a damaging effect on the foetus.
None of this is to suggest that reproductive testing in animals for new drugs was mandatory, or even an industry standard. Rather, the point is that some responsible drug companies were doing it.
In 1954 Rhône-Poulenc performed reproductive testing on rats with chlorpromazine, which had an anti-nauseant effect, and the results were published in 1954. ‘This research seems to us to be more than ever justified since this product is frequently administered to pregnant women,’ the article observed.
A Smith Kline & French memorandum from April 1956 (prior to the sale of thalidomide anywhere in the world) about the drug prochlorperazine sets out some basic common sense that should have occurred to thalidomide’s manufacturers. ‘Since [it] may be a useful drug in the treatment of nausea and vomiting during pregnancy, it appeared advisable to determine the effect of daily administration… on male and female rats during mating and on pregnant females and their young after impregnation had taken place.’
In 1956 and 1957 Richardson-Merrell performed reproductive testing on a non-steroidal oestrogen antagonist and found severe interference in rat pregnancies: embryonic death, stillbirths and neonatal deaths. In 1959 Richardson-Merrell did reproductive testing with Mer-29 (Triparanol, the controversial anti-cholesterol drug) on rats, both in-house and at an external consultancy. Merrell of course had a close relationship with Grünenthal as its US thalidomide licensee. Had Grünenthal been interested it could have asked Merrell about reproductive testing.
When the Sunday Times journalists conducted their investigation into thalidomide in the 1960s and ’70s they questioned many pharmaceutical firms about their procedures. Hoffman-La Roche said that since 1944 it had routinely performed reproductive testing on all of its new drugs and published some of the work. For example, a 1961 article on the new tranquilliser Valium concluded: ‘The results of [the rat] breeding experiment indicate that Valium does not exert a deleterious effect on the process of reproduction or normal development of the newborn.’
Edward Paget, the former head of pathology at British company ICI, told the newspaper that during the 1950s ICI had conducted tests on pregnant animals with every drug that might be used by a pregnant woman. In the aftermath of thalidomide’s withdrawal, Paget was approached by Distillers to give evidence for the drug company. Paget told Distillers of ICI’s longstanding policy. He added that he considered Distillers at fault for recommending the use of thalidomide in pregnancy without having tested the drug in pregnancy. Understandably, Distillers did not pursue Paget as a witness.
There was also much reproductive testing that never saw the light of day. In 1959 the leading teratologist James Wilson wrote that ‘pharmaceutical laboratories test many new drugs on incubating eggs and pregnant rodents but do not publish the negative observations’. To be sure, testing new drugs on pregnant animals was neither universal nor even standard in the late 1950s. It was also an inexact science. A lot of run-of-the-mill drug companies did not bother with it. But many responsible drug companies were making the effort, knowing that it was the best way to investigate a crucial issue: what might the drug do to a vulnerable foetus?
Of course Grünenthal and Distillers would argue that these diligent companies were the exception. But that’s a largely meaningless legal defence. Just because lots of companies were negligent does not excuse that negligence. Asbestos companies all over the world were negligently exposing their employees and workers to deadly carcinogenic fibres throughout the 1960s, often without warning or precaution. Many of those companies have now paid out vast sums to their victims. It never assisted their legal position to argue, ‘Our competitors were also exposing everyone to cancer so it can’t be our fault that we did the same.’
And anyway there was nothing far-fetched about testing the drug in pregnant animals. Grünenthal and Distillers knew how to do it and each had animal laboratories. The first thing they and other drug companies did when forced to confront the likelihood that thalidomide was maiming foetuses was to test the drug on pregnant animals.
By January 1962, only weeks after thalidomide’s withdrawal, tests by Distillers pharmacologist George Somers showed thalidomide had a dramatic impact on litter size in rats and increased the number of stillbirths. Somers pressed on with further tests. Grünenthal, meanwhile, was less successful in finding an effect on animal foetuses. It is hard to understand this failure. Perhaps its scientists were not very skilful. Perhaps its test animals were freakishly resistant to the effects of thalidomide. Or perhaps, a cynic might think, Grünenthal was not trying as hard as it might have.
While Grünenthal’s unsuccessful tests dragged on, Somers at Distillers quickly cracked the case. A colleague later recalled ‘a very excited’ Somers telephoning in March 1962 and demanding he drop everything and rush to Somers’ laboratory. ‘Like a conjurer producing a rabbit from a hat, he removed the cloth covering a dish to reveal a malformed rabbit!’ Somers did not plan to keep his work quiet, despite a request from Grünenthal that he delay publication. ‘Now we have succeeded in producing deformities in rabbits remarkably similar to those seen in humans,’ he wrote to The Lancet in a letter, with photographs, published on 28 April 1962. Somers promised further detail on these and other experiments with mice, rats and hens’ eggs.
Grünenthal was outraged. Somers was behaving like a scientist! First, Heinrich Mückter (now rich thanks to his personal share of thalidomide profits) wrote Somers an insulting letter. Then in May 1962 another Grünenthal executive, like Mückter also later charged with thalidomide crimes, wrote to Somers’ boss. ‘Unfortunately, Dr Somers’ publication on deformities in rabbits…is more the work of a free, independent scientist than that of a researcher working within the compass of a pharmaceutical firm.’ The Grünenthal letter warned that Somers’ coming publications would ‘lead to harm being done to both our firms, which is neither in your interests nor ours’.
Grünenthal has never accepted that it should have tested its drug on pregnant animals prior to sale or even during the period it was on sale. And Grünenthal insists that even had it done so, tests in pregnant rats, the likely test species, would have given no cause for concern. This is yet more propaganda. John Thiersch was one of the world’s foremost experts and he has stated that proper tests on pregnant rats, mice or rabbits, using the appropriate dose, would have detected a significant rise in resorptions, where the foetus dies and is absorbed into the placenta, leaving behind a scar. Competent scientists would have seen this as evidence of a possible teratogenic effect on the foetus, said Thiersch. Further tests would have followed, and the danger would have been exposed.
Yet despite the weakness of its argument against pregnancy testing, Grünenthal prefers to focus on this rather than on what happened once thalidomide was on the market. It is easy to see why. The reports of nerve damage that reached Grünenthal (and Distillers) after the drug went on sale established that thalidomide was not as safe as claimed. Obviously pre-sale testing had not picked up this major problem. The drug was neurotoxic, a serious problem, but exactly the sort of problem that could emerge once a drug was in widespread use. Drug firms were supposed to keep their drugs under close surveillance for just that reason. FDA commissioner George Larrick said in 1960 that pre-marketing animal experiments and clinical studies were merely pilot studies. ‘Untoward reactions may not become manifest during the treatment of two or three thousand patients during the period of clinical study, but may show up only after the drug has been subjected to use by millions of patients,’ Larrick said. ‘The undesirable properties sometimes show up tardily.’
Once the nerve-damage effect showed up, it would have been sensible to ask what a neurotoxic drug might do to a foetus. Years earlier exactly the same concerns had arisen about streptomycin, an antibiotic which was neurotoxic to the eighth cranial nerve, thereby damaging hearing. Grünenthal and Distillers had both been in the streptomycin business and should have been aware of this issue. A 1948 article warned against the use of streptomycin during the first trimester of pregnancy because of the sensitivity of the embryo. In 1951, three doctors in the United States published their investigation into streptomycin’s effect on the foetus. ‘Since streptomycin crosses the placenta, the possibility of fetal damage from prolonged administration of the drug required consideration,’ the doctors wrote. The doctors then checked on a series of children exposed in utero to streptomycin and found that, at least in mid- to late pregnancy, it seemed to be safe for the foetus.
Why did a similar concern never motivate Grünenthal or Distillers in relation to thalidomide? Why did they not worry, like Frances Kelsey, what might thalidomide be doing to a vulnerable foetus? If the drug had been pulled at the point at which its neurotoxicity became clear, about ninety per cent of thalidomide’s death and malformation toll would have been avoided.
As well as a vast number of nerve-damage reports, Grünenthal received reports of possible links to malformations. But still nothing was done: no investigation, no belated animal tests, no survey of pregnant women. Even after the drug went on sale, proper care and concern could have avoided much of the tragedy. Grünenthal’s response, however, was to sell, sell, sell. It’s an approach that still amazes fifty years later. Even several years of exposure to Grünenthal’s historic modus operandi did not reduce our team’s astonishment at the company’s conduct.
There is one further—remarkable—footnote to Grünenthal’s position on animal testing, which only emerged late during our research in the German archives. In September 1961, not long before it was forced to pull thalidomide from the market, Grünenthal accidentally—accidentally—conducted a test with thalidomide in pregnant rats. How did it happen?
By this time, Grünenthal had become sufficiently worried about the nerve-damage issue that it had started testing the drug on rats to see if it did indeed cause nerve damage. Three of the rats turned out to be pregnant, a development that surprised the Grünenthal scientists. In early October 1961 the rats gave birth: the first two gave birth to six and five babies respectively, the third to just one dead baby. These were small litters, and in the case of the third rat, dramatically so.
The accidental test was a model in miniature (albeit inadequate and unintended) of the sort of research John Thiersch said should have been done before the drug went on sale. Thiersch said thalidomide would cause reduced litter size and increased embryonic death, leading to suspicion about the drug. But Thiersch had vigilant scientists in mind, not Grünenthal’s team.
While the documentary record is sketchy, it appears that nobody at Grünenthal registered any concern about the three anomalous rat litters. This was despite the fact that reports of human birth malformations possibly linked to thalidomide had already been brought to Grünenthal’s attention. Further, some people inside Grünenthal were by then calling for animal testing to determine whether thalidomide could have any effect on the foetus. Weeks later, in November 1961, Lenz blew the whistle and Grünenthal was forced to halt the sale of thalidomide.
At that point the Grünenthal scientists who had conducted the accidental pregnancy research must have looked back at the episode with growing concern. But, fortunately, exoneration was at hand. In January 1962 an internal Grünenthal report dealt reassuringly with the unintentional pregnancy testing: it speculated that the small litter size of the first two rats was probably a result of the advanced age of the mothers, and made no reference to the third rat’s single dead baby. Thalidomide was not identified as a possible culprit.
This episode perhaps gives a prism through which to view some of Grünenthal’s subsequent self-serving statements about thalidomide. The fact is that Grünenthal accidentally performed a small-scale pregnancy test with thalidomide at a time when alarm bells were ringing in Stolberg about the danger of its drug. The results of the accidental test might have worried another company. At Grünenthal HQ it was business as usual.