News and Resources

Study Finds Incongruities Between Drug-Gene Associations Listed in FDA’s Table of Pharmacogenetic Associations and Guidelines Published by Clinical Pharmacogenetics Implementation Consortium (CPIC)

A comparative analysis conducted by the Personalized Medicine Coalition and published today in the American Journal of Health-System Pharmacy reveals incongruities between the drug-gene interactions that appear in the U.S. Food and Drug Administration’s Table of Pharmacogenetic Associations and those that are referenced in widely consulted clinical guidelines published by the Clinical Pharmacogenetics Implementation Consortium (CPIC). The findings underline differing perspectives regarding the classification of certain gene-drug interactions that may deter clinicians from ordering pharmacogenetic (PGx) tests whose results would be difficult to interpret.

A hallmark of personalized medicine, PGx tests are designed to help maximize medication regimens’ benefits and minimize their side effects by ensuring that drugs are prescribed and dosed based on the individual patient’s biological characteristics. After raising concerns about the strength of the evidence supporting certain drug-gene associations through a series of communications to PGx testing providers in 2018 and 2019, FDA published the first version of its Table of Pharmacogenetic Associations in February 2020. The table highlights pharmacogenetic associations the agency believes are “supported by sufficient scientific evidence.”

In comparing the drug-gene associations listed in FDA’s table with the guidelines published by CPIC, a team of authors including PMC Senior Vice President for Science Policy Daryl Pritchard, Ph.D., Atrium Health Cancer Pharmacology and Pharmacogenomics Chair Jai N. Patel, Pharm.D., PMC Program Manager Lindsay Stephens, LabCorp Associate Vice President and Scientific Director for Molecular Genetics Annette Taylor, Ph.D., and Geriatric Oncology Consortium Medical Director Howard L. McLeod, Pharm.D., found that the same drug-gene associations and dosing implications were reported for only five of the 126 drugs that were listed by one or both sources. An additional 34 drugs were listed by both sources but with differing gene associations and dosing implications. The remaining 87 drugs were listed by only one of the two sources.

“This study reveals conflicting information regarding potential drug-gene interactions that undoubtedly cause confusion among health care providers, thereby slowing the integration of pharmacogenomics, one of the pillars of personalized medicine, into clinical care,” said PMC President Edward Abrahams.