LONDON - British Biotech plc suffered yet another setback with the announcement that its oral anticancer drug marimastat had failed in its fourth successive Phase III trial. The share price fell by 5 pence to 19.25 pence on the news.

The study compared the effect of marimastat, a matrix metalloproteinase inhibitor, with placebo in prolonging survival of patients with glioblastoma. Although there was an effect it was not statistically significant.

Tony Weir, finance director, told BioWorld International, "We said when we announced that marimastat had failed in pancreatic cancer that there was little chance of success in this study in glioblastoma, or one of the other studies due to report this year in ovarian cancer.

"If marimastat is to succeed it will be in the two other Phase III studies due to report this year in small-cell lung cancer.

Weir said the company's partner, Schering-Plough Corp., has the same view. "They expected this outcome when they entered the collaboration in September 1999. Their view is that small-cell lung cancer is the indication with the most chance of success." The drug is licensed to Tanabe Seiyaku Co. Ltd. in Japan and the Far East.

British Biotech is optimistic about the lung cancer studies because they have more patients, 360 in one and 540 in another, and because they include subsets of patients with less extensive disease. "We had a near miss [in terms of reaching statistical significance] with gastric cancer, and a significant result in a subset of patients. This is consistent with the way marimastat works, by stopping tumors spreading," Weir said.

"The ethics of cancer trials say you have to recruit patients with more advanced tumors, but marimastat is unlikely to be effective in patients with widespread disease."

There are three further studies in progress in non-small-cell lung cancer as an adjuvant treatment in pancreatic cancer and in breast cancer. But these are not due to report until 2002, and Weir said once the data from the small-cell lung cancer trials are available, British Biotech, based in Oxford, will decide whether to file or terminate the project.

The extensive program of Phase III trials was put in place three years ago under the previous management. "If you go back to the status of the company then, all the old team here thought marimastat would be a wonder drug," Weir said. "They thought it would be effective in all kinds of cancer, leading to the studies being inadequately powered.

"What we are living through is a range of studies that were not put together on the basis of reality."

Although the trials program was rationalized 18 months ago by the current management, it was not possible to change protocols or add more patients because the trials were fully recruited.

The string of Phase III failures particularly exposes the use of levels of cancer antigens as the key endpoint in the Phase II studies. "This measure is not approved for regulatory purposes, and you can't assume a cause and effect relationship," said Weir.

British Biotech's follow-up matrix metalloproteinase, BB-3644, has completed Phase I in healthy volunteers, and now is in Phase Ib in patients to assess the maximum tolerated dose. From animal studies British Biotech believes this compound will not cause tendonitis, the dose-limiting side effect of marimastat.