When health authorities approve a new drug for use in Europe, they are unable to assess the potential safety risks of drugs that will be taken for extended periods because of a lack of long-term data, according to an analysis published in PLoS Medicine today.
Because US regulators use the same guidelines for the sizes of clinical trials of investigational drugs and often evaluate data from the same studies used by European regulators, the findings may also have implications for patients and physicians in the United States.
When drug regulators consider whether to approve the marketing of a new drug in their country, they look to preapproval clinical trials for evidence of the potential risks associated with taking the medication. When such trials are designed, the number of patients enrolled is typically based on how large the trial must be to assess the efficacy of a particular product, the authors explain. But clinical trials often last a comparatively short time, which means that at the time of approval, they provide limited long-term data on the risks of taking a medication intended for chronic use.
To assess the amount of information about a drug’s safety that exists at the time the drug is approved, a team of European researchers analyzed the clinical data available for new medications approved by the European Medicines Agency between 2000 and 2010. The researchers found that clinical trials of each new medication approved under the standard pathway had an average of 1708 participants; clinical trials of “orphan drugs” approved under a special pathway for rare conditions had an average of 438 participants.
Trials of medications that would be taken chronically typically involved larger numbers of patients (on average, 2338) than drugs intended for moderate-term use (878 patients) or short-term use (1315 patients). However, data on the longer-term safety and efficacy of new products were limited. For about half of the medications, fewer than 1000 patients were studied for at least 6 months while taking the medication, and for nearly 60% of the medications, there were fewer than 1000 patients studied for a year. Of the 84 medications for chronic conditions, about 82% met standards for clinical data set by the International Conference on Harmonization.
“For medicines intended for chronic use, the number of patients studied before marketing is insufficient to evaluate safety and long-term efficacy,” the authors conclude.
The authors note that conducting studies large enough to detect potential safety problems before approval may be difficult. For example, they note a 2007 meta-analysis found an elevated risk of heart attack among patients taking the diabetes drug rosiglitazone. However, to detect the magnitude of risk identified by the analysis, a controlled clinical trial would have to enroll 60 000 patients.
They argue that trial sizes and duration for preapproval studies should be reconsidered and approaches that would facilitate postapproval evaluation of drugs developed. For example, they note that a proposed strategy called “adaptive licensing” would allow regulators to conditionally approve a drug, with continued approval dependent on results of required postmarketing studies.