aDepartment of Medicine, Division of Gastroenterology and Duke Clinical Research Institute, Duke University School of Medicine, Durham, North Carolina bDepartment of Medicine, Division of Gastroenterology/Hepatology, Henry Ford Hospital, Detroit, Michigan cDepartment of Medicine, Alamo Medical Research, San Antonio, Texas dMerck Research Laboratories, Department of Infectious Disease, Whitehouse Station, New Jersey eDepartment of Medicine, Johns Hopkins University School of Medicine, Baltimore, Maryland, USA fDepartment of Gastroenterology, St Vincent's Hospital, University of Melbourne, Melbourne, Victoria, Australia.
Pharmacogenomic testing is important in developing individualized therapeutic approaches. In the phase 3 IDEAL (Individualized Dosing to Assess Optimal Pegylated Interferon Therapy) clinical trial, a subset of patients receiving peginterferon and ribavirin for treatment of chronic hepatitis C agreed to provide blood samples for genetic testing. Genome-wide association studies subsequently identified associations between IL28B polymorphism and sustained virologic response, and ITPA polymorphism and ribavirin-associated anemia.
To characterize the groups of patients who accepted or declined pharmacogenomic testing in the IDEAL study.
Clinical and demographic factors and treatment outcomes were compared at all sites that had approved pharmacogenomic testing. Differences between patients who consented to and declined pharmacogenomic testing were analyzed using Student's t-test and χ²-test.
In total, 109 of 118 sites participated in the pharmacogenomic substudy, and 1674 of 2949 (57%) patients enrolled at these sites consented to pharmacogenomic testing. More patients treated in academic medical centers than in community centers (60 vs. 52%, P<0.001) provided consent. More men than women (58 vs. 54%, P=0.04) consented to pharmacogenomic testing. There was no significant difference in pharmacogenomic participation between patients from different racial groups, including whites and African Americans (58 vs. 54%, P=0.07). Treatment outcomes were also similar according to pharmacogenomic participation.
In the IDEAL study, patient consent to pharmacogenomic testing did not introduce selection bias. Treatment at an academic center and male sex were associated with higher rates of pharmacogenomic testing consent. Efficacy and safety outcomes were similar in patients who accepted and declined pharmacogenomic testing.