The drug may work, but how well?

Image via Wikipedia

There’s a great perspective piece in the October 29 edition of the  New England Journal of Medicine on how significant information about a medication sometimes never makes it to the drug label.

In brief:  when a pharmaceutical company submits a drug to the FDA for approval, the company must demonstrate that their drug works better than placebo, and that the benefits of its use outweigh the harms.  In so doing, they submit to the FDA the results of preclinical studies and any phase three studies they have conducted.  The FDA then reviews the materials, and after up to a year of deliberation, renders their decision.

However, once the drug goes to market, much of the information from the studies, such as the drug’s efficacy, that had been presented to the FDA is not included on the label.  The authors of this article use as an example the drug Lunesta.  The drug label states that it is superior to placebo but offers no additional efficacy information.  Yet when the authors examined the material submitted to the FDA, buried in the 403 pages of information they found that patients taking Lunesta fell asleep 15 minutes faster and slept 37 minutes longer than those taking placebo, but “still met criteria for insomnia and reported no clinically meaningful improvement in next-day alertness or functioning”.  In other words, although it worked better than placebo, it really didn’t substantially fix the underlying problem.

The authors state that part of the problem is that the wording of the drug label is left up to the pharmaceutical company, which is then approved by the FDA.  Unfortunately, while reporting of efficacy data on the label is encouraged, it is not required.  Apparently the FDA is reconsidering this position, and in my opinion, rightly so.  We need to know how well a medication works if we are to make informed decisions about treating our patients, and shouldn’t have to go hunting through FDA documents to find that information.