Prevalence of Pharmacogenomic Actionability in >11,000 Patients Receiving Psychiatric Care

Authors Sarah Morris, Joe Stanton, Marcus Badgeley, Jai Patel

Introduction: Less than 40% of patients with major depressive disorder (MDD) achieve clinical remission after first-line antidepressant treatment. Pharmacogenomic (PGx) testing may improve drug selection and/or dosing in patients with genetic variation(s) that alter drug pharmacokinetics through CYP2D6 and CYP2C19. The Clinical Pharmacogenetics Implementation Consortium (CPIC) guidelines and Food and Drug Administration (FDA) Table of PGx Associations provide recommendations on drug-gene interactions and how to translate PGx findings into actionable prescribing recommendations. Herein, we evaluate the prevalence of PGx actionability in a large real-world cohort of patients receiving psychiatric care and referred for clinical PGx testing via Tempus nP.

Methods: We retrospectively analyzed PGx test results patients who received Tempus nP testing as part of their standard psychiatric care between April 2020 and July 2022. Tempus nP interrogates polymorphisms ~20,000 genes using whole exome sequencing, but only CYP2D6 and CYP2C19 actionability were estimated for this analysis. The primary objective was to determine the prevalence of potentially actionable CYP2D6 and CYP2C19 phenotypes across the entire population genotyped. “Potentially actionable” results are defined as phenotypes associated with a prescribing recommendation per CPIC guidelines or FDA (section 1 of the Table of PGx Associations), regardless of whether the medication was prescribed or considered. The secondary objective was to determine the prevalence of actionable drug-gene interactions based on CYP2D6 and CYP2C19 results in the subset of patients with medication data. Actionable results are defined as presence of a drug-gene interaction using medications prescribed at the time of or prior to testing (i.e., attempted medications), or medications considered at the time of testing, using CPIC and/or FDA guidelines. In the subset with CYP2D6 copy number variants (CNVs), we estimated the impact of allele-specific CNVs on phenotype assignment.

Results: In 11,834 patients undergoing clinical PGx testing, 64% were female and the median age was 29. The most common diagnoses were Depression, Anxiety, and Attention Deficit Hyperactivity Disorder. The most common reasons for ordering PGx testing were Depression or Anxiety and a history of neuropsychiatric medication failure. 65% of patients had actionable phenotypes for CYP2D6 and/or CYP2C19; 48% and 33% of patients had potentially actionable CYP2D6 and CYP2C19 phenotypes, respectively. Of 865 patients who attempted medications related to CYP2D6 and/or CYP2C19, 22% (N=187) had at least one actionable phenotype. Of 1,141 patients considering medications related to CYP2D6 and/or CYP2C19, 18% had at least one actionable phenotype. The most common potentially actionable medication classes were tricyclic antidepressants (59%) and selective serotonin reuptake inhibitors (17%). Of 883 patients with CYP2D6 CNVs (7.5% overall), 369 patients (42% of all patients with CNVs) had genotypes in which knowledge of allele specific CNV would change the phenotype (most often *1/*4, *1/*10, and *2/*4).

Conclusion: Prevalence of CYP2D6 and CYP2C19 actionability is high in a large real-world cohort of patients receiving psychiatric care and referred for clinical PGx testing. Comprehensive genotyping, including allele specific CNV for CYP2D6, is critical for accurate phenotype assignment and can identify clinically significant drug-gene interactions in psychiatric care.