Overview
Measuring the true survival benefit of a first-line (1L) cancer treatment can be challenging when subsequent second-line (2L) treatments vary across patient groups, practice settings, or countries. Using the US Flatiron Health Research Database, this study evaluated 20,309 patients diagnosed with stage IV non-small cell lung cancer (NSCLC) between 2017 and 2026 without a therapy-targetable genetic mutation and started 1L immunotherapy or platinum-based chemotherapy. Researchers used an advanced statistical approach (parametric mediational g-computation) to calculate the generalized direct effect (GDE) of 1L treatment on real-world overall survival (rwOS) under three hypothetical 2L scenarios (broad-chemotherapy-heavy, immunotherapy-heavy, and platinum-based-chemo-heavy).
The estimated 1L immunotherapy survival advantage was consistent over platinum-based chemotherapy across all tested 2L scenarios, including an 8.9% to 10.7% absolute survival improvement from 12 to 36 months under a chemotherapy-heavy 2L scenario.
Why this matters
This study suggests that electronic health record data can be used to separate the survival benefit of a 1L treatment from downstream choices in 2L care. By holding subsequent treatment to a common hypothetical scenario, this methodology may enable researchers, regulators, and health technology assessment bodies to evaluate initial treatment effectiveness more fairly across populations with diverse treatment practices.