Much as I dislike firewalls, I don’t generally publish leghty direct quotations from behind them. But this column from the WSJ seems important enough to break the rule. [Who do you think you are making a decision like that - the editor of the New York Times?-ed. I think I'm someone old enough to worry about this.]:
Four years after Pfizer Inc. ended a clinical test of an experimental anti-aging pill and stopped its development for that use, the results of the study still haven’t been published in a scientific journal, where other researchers could take advantage of them.
The lag highlights an enduring issue in pharmaceutical research: the fate of data from trials of drugs that fail to live up to expectations. In recent years, drug makers have come under attack for failing to disclose negative research about medicines they have on the market. But there’s another twist to the data dilemma that concerns drugs that don’t get that far.
The research behind medicines that get nixed in the trial stage could be valuable to the scientific community. But that information may not immediately reach people, working in academia or at other companies, who might be able to solve the problems or otherwise build on the results.
A recent example is an anti-aging drug pursued by Pfizer that was the subject of a front-page story in The Wall Street Journal in 2002 after Pfizer hit a roadblock with the drug. The company’s study of a “youth pill” predates the latest public pushes for trial-result disclosures. Pfizer has allowed investigators on the project to make presentations of some data at scientific meetings, including a symposium on chemicals to boost growth hormone held in Italy in late 2004 and at a geriatrics conference in Florida last year.
But the critical step in medical progress is the publication of scientific findings that have been vetted by experts in the field, a process called peer review. So far that’s not happened for the intriguing anti-aging pill, called capromorelin.
And since drug trials use statisitically signifcant differences to claim one drug is better than another and/or placebo, failure to publish negative findings distorts this process. The negative findings never make it into meta-analytic work that is used to group the results (statistics) of various studies together to determine the efficacy of an itervention. There ought to be a public data base of all findings, including negative ones.