Title: “Cost-Effectiveness of Next-Generation Sequencing in Early-Stage NSCLC: A Future-Oriented Simulation as Emerging Biomarkers and Treatments Evolve”
Background: In early-stage non-small cell lung cancer (esNSCLC, stage IB-IIIA), emerging genomic biomarker targets underscore the need for an optimal biomarker testing strategy. While next-generation sequencing (NGS) is well established in advanced disease, its economic value in early-stage settings remains unclear. This study takes a forward-looking approach, evaluating how the economic outcome of NGS evolves as new biomarkers become clinically actionable.
Methods: We developed a decision-tree model from a US Medicare payer perspective comparing NGS with single-gene testing (SGT). Five scenarios were evaluated for SGT strategies to simulate the sequential addition of biomarkers. NGS, represented as small gene panels (5-50 genes), remained constant across comparisons. Actionable biomarkers included ALK (prevalence: 2.0%), EGFR (17.9%), KRAS G12C (12.4%), RET (1.1%), BRAF (0.9%), and ROS1 (0.4%), prioritized by clinical prevalence, product/trial pipelines and guideline relevance. Biomarker prevalence and test performance were informed by published sources, supplemented with conservative assumptions for NGS. Testing costs were obtained from the 2025 CMS Clinical Diagnostic Laboratory Fee Schedule; treatment costs were not considered. Primary outcomes included cost per correctly identified patient (CCIP)—derived by monetizing the number needed to predict (NNP), a function of positive and negative predictive values—to measure cost in achieving an accurate diagnosis of a patient’s genomic alterations, and cost per member per month (PMPM).
Results: Across all scenarios, NGS showed better diagnostic performance versus SGT, requiring fewer patients for accurate diagnosis (i.e., lower NNP). A key finding was that NGS achieved a lower CCIP beginning at the scenario with 3 actionable biomarkers (EGFR & ALK, + KRAS G12C), suggesting a diagnostic “tipping point” beyond which NGS becomes cost-saving (CCIP difference = -$459). The impact of NGS further improved as additional biomarkers were added (+RET, +BRAF, +ROS1). Incremental PMPM results showed a consistent downward trend: $0.01 with two biomarkers, -$0.05 with three, -$0.08 with four, -$0.11 with five, and -$0.14 with six.
Conclusions: Our analysis suggests that, in addition to its potential to improve diagnostic accuracy, broad-panel NGS will become cost-saving relative to SGT in esNSCLC as the number of actionable biomarkers continues to increase. These findings support a forward-looking strategy in which broader biomarker availability strengthens the clinical and economic rationale for NGS testing in early-stage patients.