Gristle for the mill

There’s an old saying in Washington DC — making laws is like making sausage; you’re better off not knowing the details. Well, that same spirit often applies to the coverage of science news. The focus is on the succulent final product rather than the process required to get there. But science journalists have found the need to take a break from that approach with one big biology story: the tale of angiostatin and endostatin. The fate of this story was cast last May, when the New York Times all but declared that victory against cancer was just a few years away, thanks to Harvard’s Judah Folkman and his amazing molecules (see Mediawatch, Curr Biol 1998, 8:R438). Angiostatin and endostatin together had eradicated tumors in mice by cutting off their blood supply; humans, seemingly, would be next. This, of course, had vast implications not only for human health but for the companies (EntreMed and Bristol-Myers Squibb) and shareholders with a stake in these potential drugs. So last November, a reporter from the Wall Street Journal checked up on the progress and made a rather startling discovery: Scientists at the National Cancer Institute had been unable to reproduce Folkman’s results. Simply reporting that result, though, could suggest either that Folkman was a charlatan, or that the NCI scientists were all thumbs in the lab. Of course, neither of those is the case. So the Journal article dug into the process of science, explaining the need for reproducibility and the trouble, sometimes, of attaining it. The stuff, it turns out, is hard to make, hard to transport, and hard to handle. “Says [Bristol-Myers Squibb] senior vice president of pharmaceutical development, Christopher Cimarusti: ‘There’s a difference between a lab curiosity and something you can take forward in man’. That distinction, while eminently clear to scientists such as Dr Folkman, is often lost in the public’s understanding of the immensely complicated quest for a cancer cure.” Genentech scientists also tried and failed to reproduce Folkman’s published results. The New York Times interviewed Arthur Levinson, the chief executive of Genentech, for a story that appeared one day after the Journal’s. “If we’re doing something wrong, I would like to know what it is, because we’re all hoping he’s right,” Dr Levinson said. “But things like this shouldn’t languish too long, because the whole basis of science is reproducibility. If they can’t be reproduced, at some point you have to ask yourself, are the initial claims correct?”