FDA should develop consistent evidentiary standards for post-market labeling changes to include pharmacogenetic information
This commentary in the Genomics Law Report’s ongoing series What ELSI is New? is contributed by Stephanie Devaney, Genetics and Public Policy Center at Johns Hopkins University, and Gail Javitt, Berman Institute of Bioethics at Johns Hopkins University and Sidley Austin LLP*
A prescription drug’s labeling is the primary vehicle used by drug manufacturers to inform providers of the conditions under which a drug is safe and effective for use, and must be approved by FDA at the time the drug is approved. Post-market labeling changes may also require FDA approval depending on the type of change. Pharmacogenetics is a relatively new and complex field that regulators and drug sponsors alike are trying to navigate. In recent post-market labeling changes to include pharmacogenetic information, FDA has been inconsistent in the type and amount of data it has required to support the new claim. A clearer regulatory path is critical to encourage drug sponsors to invest in pharmacogenetic research and to ensure that health care providers get the information they need to get patients the right drug at the right dose at the right time.
FDA encourages sponsors to include pharmacogenetic information as part of new drug applications. In 2005 the agency issued a guidance document titled “Pharmacogenomic Data Submissions”, which explains when and how pharmacogenomic data should be submitted during drug review. However, there are several hurdles to the incorporation of this information post market, especially when the efficacy of the drug is affected thus narrowing the patient population. Although FDA appears to be willing to permit pharmacogenetic claims relating to safety that are supported by a modest level of evidence, data from randomized controlled trials are usually required for claims of reduced or increased efficacy. However, conducting a randomized controlled trial for every potential predictive genetic variant would be costly and risks hampering innovation. On the other hand, absent appropriate data collection and analytical methods, retrospective analyses of data from completed clinical trials run the risk of creating biased results.
The recent debate between FDA and sponsors of the colon cancer drugs Erbitux (cetuximab) and Vectibix (panitumumab) over whether the drugs’ labeling should include KRAS mutant status as predictive of drug efficacy highlights FDA’s discomfort with using retrospective data to support efficacy claims. However, the rapid pace of genomics research means that genetic effects on drug response will frequently be identified post market. Therefore, clear guidelines on when retrospective analyses will be adequate and what criteria such data must meet are necessary to facilitate incorporation of pharmacogenetic information into drug labeling. This position echoes the recent comments of Larry Lesko, director of the Office of Clinical Pharmacology at FDA, during the National Conference on Personalized Healthcare in early October 2009. He commented that the KRAS experience could be used to set a framework to inform other drug sponsors on how to utilize retrospective analyses of genomic biomarkers and drug response.
————————-
*The views expressed in this article are exclusively those of the author and do not necessarily reflect those of Sidley Austin LLP and its partners.
