Abstract
PURPOSE: With an increase in approved targeted therapies for non-small cell lung cancer (NSCLC), tumor profiling that maximizes identification of patients eligible for these treatments is essential. In this study, we compare testing approaches for newly diagnosed advanced or metastatic NSCLC. METHODS: An economic model was developed to estimate the annual costs and clinical outcomes associated with testing alternatives. Patients were assigned alterations on the basis of rates observed in published literature. Treatment approaches included no genomic testing; sequential single-gene testing; or a whole-exome, whole-transcriptome (WES/WTS) comprehensive genomic profiling to inform use of targeted therapies. The budget impact and survival implications of WES/WTS were compared head to head with the other testing approaches, and uncertainty was assessed in sensitivity analyses. In scenario analyses, tests that incorporate both DNA and RNA were compared with tests using DNA sequencing alone when assuming the prevalence of fusions that can only be identified by RNA ranged from 2.5% to 14%. RESULTS: Compared with no testing, WES/WTS reduced costs by $8,809 in US dollars (USD) per patient tested while increasing median overall survival by an average of 3.9 months. When compared with sequential single-gene testing, use of WES/WTS decreased costs by $14,602 (USD) per patient and had a minimal survival benefit. A test that included both RNA and DNA sequencing reduced costs by $400-$1,724 (USD) and increased identification of actionable alterations by 2.3%-13.0% across the range of fusion prevalence. CONCLUSION: Use of WES/WTS improved clinical outcomes while reducing costs. These findings should be considered when choosing between tumor profiling approaches.