Does treatment sequencing matter in stage IV NSCLC with brain metastases? A National Cancer Database study.

Abstract

e20657 Background: The optimal sequencing of brain-directed radiation and systemic therapy in stage IV NSCLC with brain metastases remains unclear. We evaluated survival outcomes associated with treatment sequencing using NCDB across contemporary treatment eras. Methods: Adult patients (2010–2022) with stage IV NSCLC and brain metastases at diagnosis who received both brain radiation and systemic therapy were identified from the NCDB. Treatment sequence was defined by initiation order (radiation-first vs systemic-first); patients with same-day initiation or upfront surgery were excluded. Overall survival (OS) was assessed using Kaplan–Meier and multivariable Cox models stratified by diagnosis era (pre-2015 vs ≥2015). Immortal time bias was addressed using a delayed-entry sensitivity analysis. Temporal trends were evaluated with logistic regression. Results: Among 45,577 patients, 35,706 (78.3%) received radiation-first and 9,871 (21.7%) systemic-first therapy. In unadjusted analyses, OS did not differ between sequences (log-rank p=0.055). By era, no difference was observed pre-2015 (p=0.56), while systemic-first therapy was associated with improved unadjusted OS in patients diagnosed in 2015 or later (p=0.018). After multivariable adjustment, OS was comparable between sequences (aHR 1.03 [0.98–1.09]). Delayed-entry sensitivity analyses showed a small increase in hazard with systemic-first approach (aHR 1.07 [1.01–1.13]). Older age, higher comorbidity burden, Hispanic and Asian race, Medicaid insurance, and longer time to treatment initiation were associated with increased mortality. Use of systemic-first therapy increased over time (aOR per year 1.03 [1.02–1.04]). Conclusions: In this national cohort of patients with brain-metastatic NSCLC receiving multimodal therapy, OS was similar between radiation-first and systemic-first sequencing after adjustment. These findings suggest that, in appropriately selected patients, initiating systemic therapy before brain radiation may be reasonable without OS compromise, supporting individualized sequencing decisions. Temporal shifts toward systemic-first strategies likely reflect evolving practice patterns rather than a survival advantage. Variable Category aHR (95% CI) Variable Category aHR (95% CI) Treatment sequence (ref: radiation-first) Systemic-first 1.03(0.98,1.09) CCI (ref:0) 1 1.10 (1.04,1.17) Age group (ref: ≤64 years) ≥65 years 1.16 (1.08,1.24) 2 1.11 (1.00,1.23) Sex (ref: Male) Female 0.98 (0.94,1.03) ≥3 1.24 (1.09,1.42) Race (ref: NH White) Hispanic 1.35 (1.21,1.51) Histology (ref: Adenocarcinoma) SCC 1.10 (1.00,1.20) Black 1.06 (0.99,1.13) NSCLC NOS 0.78 (0.71,0.86) Asian 1.14 (1.04,1.25) Time from diagnosis to treatment 1.003 (1.002,1.003) Insurance (ref: Private insurance) Medicaid 1.15 (1.07,1.25) Medicare 1.01 (0.94,1.09) Self-pay 1.19 (1.06,1.35)