OBJECTIVES: Treatments for advanced non-small cell lung cancer (NSCLC) target genetic mutations driving cancer growth. Panel-based genomic testing streamlines targeted treatment selection by detecting mutations in several genes simultaneously. In British Columbia (BC), Canada, BC Cancer reimbursed panel-based testing beginning in September 2016. This study uses real-world data and machine learning-based quasi-experimental methods to determine the cost-effectiveness of publicly reimbursed panel-based genomic testing compared to single-gene testing for advanced NSCLC.

METHODS: This retrospective study used patient-level linked population-based health administrative databases provided by Population Data BC and BC Cancer. We considered adult BC residents with a panel-eligible lung cancer diagnosis between 2016-09-01 and 2018-12-31. We 1:1 matched patients receiving panel-based testing to controls receiving single-marker testing using genetic algorithm matching, a machine learning approach that maximizes balance on observed covariates. Following matching, we estimated mean three-year survival time and costs (public healthcare payer perspective; 2021 CAD) and calculated the incremental net monetary benefit (INMB) for life-years gained (LYG) at conventional willingness-to-pay thresholds. We applied inverse probability of censoring weighted linear regression to account for incomplete follow-up data when estimating mean outcomes, and nonparametric bootstrapping to generate confidence intervals.

RESULTS: We identified 942 individuals receiving panel-based testing and 942 matched controls (from n=1,238 patients). Mean incremental costs were lower for the panel-based group (–$10,009 [95%CI: –$41,802, $21,167]) and we observed a small survival difference (6.2 days [95%CI: –8.9, 28.6]). The INMB at $50,000/LYG was $10,862 (95%CI: –$19,394, $41,976) and at $100,000/LYG was $11,714 (95%CI: –$18,315, $42,784). We found that 57.2% and 59.3% of the 1,000 bootstrapped samples were cost-effective at either $50,000/LYG or $100,000/LYG, respectively.

CONCLUSIONS: In Canada and internationally, there is uneven reimbursement of panel-based sequencing. Using population-based real-world data, we found the cost-effectiveness of panel-based testing to inform targeted treatment for NSCLC to be finely balanced at conventional thresholds.