A real-world data platform for comparative analysis of treatment patterns, costs, and cost-effectiveness among ALK TKIs in advanced NSCLC.

Abstract

e20688 Background: Real-world evidence (RWE) directly comparing Anaplastic Lymphoma Kinase Tyrosine Kinase Inhibitors (ALK TKIs) is crucial for optimizing treatment strategies and healthcare resource allocation in advanced non-small cell lung cancer (aNSCLC). This study leverages a comprehensive medical insurance database to establish an RWE platform capable of comparing clinical outcomes, healthcare resource utilization (HCRU), and economic impact across multiple ALK TKIs. Methods: We conducted a retrospective cohort analysis using the Beijing Medical Insurance Database (01/2019-02/2025). Patients with aNSCLC initiating first-line ALK TKI therapy (Alectinib, Lorlatinib, Crizotinib, Ensartinib, Ceritinib, Brigatinib) were included. For each agent, we extracted real-world metrics on HCRU, direct medical costs, treatment duration, and sequencing patterns. These data were used to populate a partitioned survival model for cost-effectiveness analysis (CEA). Using Alectinib vs. Lorlatinib as a proof-of-concept pairwise comparison, the CEA calculated lifetime costs (2025 CNY), quality-adjusted life years (QALYs), and the incremental cost-effectiveness ratio (ICER), with a 5% discount rate. The model inputs were derived from the real-world cohort. The framework supports future head-to-head comparisons among other TKI pairs. Results: The cohort included 862 eligible patients. RWE revealed significant variations across TKIs in median duration of therapy (Alectinib: 569 days; Lorlatinib: 388 days; Crizotinib: 252 days), annual HCRU, and cost structures. Alectinib was associated with lower average annual non-drug direct medical costs, resulting in a saving of ¥8,657 compared to Lorlatinib. In the proof-of-concept CEA (Alectinib vs. Lorlatinib), Alectinib provided more QALYs (4.74 vs. 4.53) at a higher total cost (¥859,738 vs. ¥853,019), yielding an ICER of ¥31,471 per QALY gained, which is below the 1x GDP per capita threshold (¥95,749). The economic outcome was driven by Alectinib's longer treatment duration and associated utility benefit, alongside its lower real-world progressed disease phase costs (¥2,721 vs. ¥4,275 per cycle) compared to Lorlatinib, which offset differences in progression-free phase costs. Conclusions: This study establishes a robust RWE platform using real-world costs and resource use data that enables comparative effectiveness and economic evaluations among all ALK TKIs. The proof-of-concept analysis demonstrates that Alectinib is a cost-effective option compared to Lorlatinib from a Chinese payer perspective, supported by its clinical benefits and favorable cost structure, including lower non-drug medical costs. The developed framework supports future head-to-head comparisons for other TKI pairs, providing critical evidence to inform clinical decision-making and optimize aNSCLC care.