At the end of 1998, British Biotech (Oxford, UK) terminated a phase III trial of marimastat, its matrix metalloproteinase inhibitor cancer drug, after realizing that the design of the trial would not generate data suitable for regulatory approval. According to a company spokesperson, the end point of the trial—an unusual composite measure involving a combination of CT scans, levels of cancer antigens, and patient "well–being" (as opposed to survival)—would not produce data that the European Medicines Evaluation Agency (EMEA; London) and the US Food and Drug Administration (FDA; Rockville, MD) would accept for approval. However, the company points out this is only 1 of 11 trials; the remaining 10, which cite survival as an endpoint, will continue as they are. "Three–quarters of a million pounds will be saved" as a result of halting the trial, says the company positively, although £250,000 had already been spent on recruitment for the trial. The inadequacies of the marimastat trial were spotted during an external assessment of the company's technology following the firing of its head of clinical trials, Andrew Millar, and the resignation of CEO Keith McCullagh last year ( Nat. Biotechnol. 16:609, 1998).