Exploring Strategies to Optimize the Value of Pharmacogenomic Testing

Thesis (Ph.D.)--University of Washington, 2018The emergence and continual advancement of genomic technologies offer numerous areas that warrant a critical appraisal for integrating new health care services to increase access and improve outcomes for patients. This dissertation examines two strategies in the context of pharmacogenomics (PGx) testing. In the Overall Introduction, we initially focus on identifying clinical utility and economic evidence gaps necessary to inform appropriate clinical adoption of PGx testing across diverse healthcare settings. In Chapter 1, we report our findings from an analysis of real-world evidence from a commercial PGx knowledge resource, which is comprised of data from patients who have undergone PGx testing. Additionally, we characterize gene-drug pair level of evidence as developed by expert groups, and present associated predictive factors that may inform clinical actionability. Following this, in Chapter 2, we focus on a hypothetical clinical cohort of acute coronary syndrome patients undergoing percutaneous coronary intervention. We use decision modeling methods to estimate the projected cost-effectiveness of a multi-gene panel to guide two treatment decisions for this clinical cohort from the payer perspective. We describe influential parameters in the model, discuss limitations of this work, and denote implications to health policy decision making. In the Overall Conclusions, we summarize and describe future considerations. To increase appropriate clinical PGx testing adoption, we provide evidence that healthcare entities may wish to consider the use of a commercially-developed PGx knowledge resource solution in lieu of delaying implementation awaiting publicly available PGx knowledge resource solutions. Additionally, for the aforementioned patient cohort, we provide evidence that the multi-gene panel estimates are projected to be cost-effective at a $100,000 willingness-to-pay threshold when compared to either a single-gene panel or to no gene testing. These findings build upon the available economic evidence for multi-gene panels that payer decision makers may consider during their health technology assessment evaluations to determine inclusion of covered services within their medical policies. Finally, the contents of this dissertation contribute to the broader discourse regarding value assessments in the interplay between precision medicine and clinical genomics