Validation of Pharmacogenomic Interaction Probability (PIP) Scores in Predicting Drug–Gene, Drug–Drug–Gene, and Drug–Gene–Gene Interaction Risks in a Large Patient Population

Abstract

Utilizing pharmacogenomic (PGx) testing and integrating evidence-based guidance in drug therapy enables an improved treatment response and decreases the occurrence of adverse drug events. We conducted a retrospective analysis to validate the YouScript® PGx interaction probability (PIP) algorithm, which predicts patients for whom PGx testing would identify one or more evidence-based, actionable drug–gene, drug–drug–gene, or drug–gene–gene interactions (EADGIs). PIP scores generated for 36,511 patients were assessed according to the results of PGx multigene panel testing. PIP scores versus the proportion of patients in whom at least one EADGI was found were 22.4% vs. 22.4% (p = 1.000), 23.5% vs. 23.4% (p = 0.6895), 30.9% vs. 29.4% (p = 0.0667), and 27.3% vs. 26.4% (p = 0.3583) for patients tested with a minimum of 3-, 5-, 14-, and 25-gene panels, respectively. These data suggest a striking concordance between the PIP scores and the EAGDIs found by gene panel testing. The ability to identify patients most likely to benefit from PGx testing has the potential to reduce health care costs, enable patient access to personalized medicine, and ultimately improve drug efficacy and safety.

Keywords

• pharmacogenomics
• population health
• adverse drug events
• phenoconversion
• pharmacogenetics
• precision medicine