The Downside: Potential Safety Problems May Remain Hidden Until Later
The "News roundup" section of the November 12, 2005 edition of the British Medical Journal ("BMJ") included an article by Michael Day about how an increasing number of new drug trials are ended early "on the grounds that apparent treatment benefits have emerged rapidly". While this may get the drug companies' new drugs to market quicker, an unfortunate result is that drug-safety issues associated with those new drugs might not be found until later -- at the expense of patients taking the new drugs.
Dr. Victor Montori recently led a group of researchers from McMaster University, Ontario, Canada, who reviewed 143 randomized controlled clinical trials sponsored by drug companies in connection with their respective new drugs. According to Mr. Day's article, Dr. Montori's group determined that (a) "the proportion of randomised trials halted early has doubled in the past 10 years" and (b) "the scientific validity of the data they provide is not proved".
In more detail, according to the November 12 BMJ article, Dr. Montori said that the drug companies "often fail to explain adequately why trials were stopped early and often claim 'implausibly large treatment effects, particularly when the number of events is small.'" In the final analysis, Dr. Montori expressed his belief that “[t]hese findings suggest clinicians should view the results of such trials with scepticism."
For his BMJ article, Mr. Day sought comment about the review done by Dr. Montori and his group of Canadian researchers from Stuart Pocock, a professor of medical statistics at the London School of Hygiene and Tropical Medicine. Professor Pocock asserted:
- “There is clear evidence from this review that some of these clinical trials have been stopped too soon. In some cases the researchers have shown a lack of judgment.” and,
- “It’s likely that in many cases doctors halted the trials because they wanted to give what they see as treatment benefits to all patients. But it could be that many more people are adversely affected in the long term if misleading results are given credence."
In concluding his BMJ article, Mr. Day made reference to an article this past spring by Richard Smith, a former editor of the BMJ, which appeared in PLoS Medicine. As characterized by Mr. Day, the message of that earlier piece by Mr. Smith was "that leading medical journals had become an 'extension of the marketing arm of pharmaceutical companies' by inadvertently publishing selective research and results that favoured the products of trial sponsors."
Along these lines, one may recall how the New England Journal of Medicine ("NEJM") handled the publication of a study report concerning Provigil earlier this year.
(Posted by: Tom Lamb)