Association Between Treatment Sequencing and Overall Survival in Stage IV NSCLC With Brain Metastases: A National Cancer Database Study.
The optimal sequencing of brain-directed radiation and systemic therapy in stage IV non-small cell lung cancer (NSCLC) with brain metastases remains uncertain in the era of CNS-active systemic agents. We evaluated survival outcomes using national real-world data.
We conducted a retrospective cohort study of adults diagnosed between 2010 and 2022 with stage IV NSCLC and brain metastases in the National Cancer Database who received both brain-directed radiation and systemic therapy. Treatment sequence was classified as radiation-first or systemic-first based on initiation dates. Multivariable Cox proportional hazards models, stratified by treatment era (pre-2015 vs. 2015+), assessed associations with overall survival (OS), adjusting for demographic, clinical, tumor, and treatment factors. Propensity score matching and delayed-entry sensitivity analyses were performed to address confounding and immortal time bias.
Among 45 577 patients, 78.3% received radiation-first and 21.7% received systemic therapy first. Unadjusted Kaplan-Meier analysis showed no significant difference in OS (log-rank p = 0.624). In multivariable analysis, systemic-first sequencing was associated with a modest increase in mortality (adjusted hazard ratio [aHR] 1.06; 95% CI: 1.04-1.09), which was consistent in propensity-matched (HR 1.07; 95% CI: 1.04-1.11) and delayed-entry analyses (aHR 1.09; 95% CI: 1.07-1.12). The use of systemic-first therapy increased over time. Established prognostic factors demonstrated larger effect sizes.
Systemic-first sequencing was associated with a modest increase in adjusted mortality; however, the effect size was small relative to established prognostic factors and likely influenced by residual confounding and selection bias. These findings support individualized, multidisciplinary treatment decisions rather than a uniform sequencing strategy.
We conducted a retrospective cohort study of adults diagnosed between 2010 and 2022 with stage IV NSCLC and brain metastases in the National Cancer Database who received both brain-directed radiation and systemic therapy. Treatment sequence was classified as radiation-first or systemic-first based on initiation dates. Multivariable Cox proportional hazards models, stratified by treatment era (pre-2015 vs. 2015+), assessed associations with overall survival (OS), adjusting for demographic, clinical, tumor, and treatment factors. Propensity score matching and delayed-entry sensitivity analyses were performed to address confounding and immortal time bias.
Among 45 577 patients, 78.3% received radiation-first and 21.7% received systemic therapy first. Unadjusted Kaplan-Meier analysis showed no significant difference in OS (log-rank p = 0.624). In multivariable analysis, systemic-first sequencing was associated with a modest increase in mortality (adjusted hazard ratio [aHR] 1.06; 95% CI: 1.04-1.09), which was consistent in propensity-matched (HR 1.07; 95% CI: 1.04-1.11) and delayed-entry analyses (aHR 1.09; 95% CI: 1.07-1.12). The use of systemic-first therapy increased over time. Established prognostic factors demonstrated larger effect sizes.
Systemic-first sequencing was associated with a modest increase in adjusted mortality; however, the effect size was small relative to established prognostic factors and likely influenced by residual confounding and selection bias. These findings support individualized, multidisciplinary treatment decisions rather than a uniform sequencing strategy.