Comparative Effectiveness Approaches to Evaluate Pharmacogenomic Technology

Abstract

Background Recent advances in genomic science present opportunities to "personalize" pharmaceutical treatments and improve health outcomes. However, such improvements are impeded by sparse evidence assessing pharmacogenomic technology clinical utility. This uncertainty constrains stakeholder decision-making, and has resulted in both sub-optimal and premature clinical translation. Timely assessment of health outcomes and clinical utility is critical to overcoming these translational issues. Comparative effectiveness research (CER) can meet this need by applying epidemiologic, health economic, and bio-statistical methods to generate timely evidence in "real world" conditions. Objective To demonstrate the utility of CER methods to evaluate pharmacogenomic technology in case studies in warfarin therapy, lung cancer, and breast cancer. Methods First, I utilize a case-control design to evaluate the association between genetic variants and risk of major bleeding in warfarin therapy patients. Second, I conduct a value of information (VOI) analysis to assess the value of a future trial investigating the clinical utility of ERCC1-guided adjuvant chemotherapy in lung cancer. Third, I explore the potential application of benefit-risk modeling to inform regulatory review of pharmacogenomic technology, with a specific demonstration in gene-expression profiling in early-stage breast cancer. Results In the evaluation of warfarin pharmacogenomics, I demonstrate the first association between CYP4F2 variants and decreased risk of major bleeding, and provided exploratory evidence suggesting differential CYP2C9 variant bleeding risk based on geographic setting. In the ERCC1 VOI analysis, I demonstrate that a trial is expected to cost $30m, create value of $80m, and thus provide $50m in societal net-benefit. In my evaluation of regulatory benefit-risk modeling for pharmacogenomic technology, I examine a range of insights that can be gained through incorporation of systematic and quantitative evaluative methods in regulatory review processes. Conclusions In each case study, CER methods provide key evidence about health outcome impacts of pharmacogenomic technologies. These results have direct implications for clinical utility, and can inform stakeholder decision-making. These types of CER applications will likely play an important role in future assessment and translation of pharmacogenomic technologies due to accelerating research and development, and inability of traditional controlled trial designs to provide timely, adaptable, and readily applicable evidence.