Optimal Sequence of Local and EGFR-TKI Therapy for EGFR-Mutant Non-Small Cell Lung Cancer With Brain Metastases Stratified by Number of Brain Metastases

Purpose: It is unclear whether local therapy (LT) or epidermal growth factor receptor (EGFR) tyrosine kinase inhibitor (TKI) should take precedence for patients with EGFR-mutant non-small cell lung cancer (NSCLC) and brain metastases (BMs). The number of BMs is important in the choice of LT, including whole-brain radiation therapy, stereotactic radiosurgery, and surgery. Methods: We retrospectively evaluated cases of EGFR-mutant non-small cell lung cancer with BMs from a single site. Patients were divided into 2 groups based on up-front therapy-EGFR-TKI (TKI) or LTs-and subsequently stratified by the number of BMs. Results: Among 176 patients, 61% received upfront EGFR-TKI, and 39% received upfront LT. The number of patients with 1 to 4 BMs was similar (56% vs 52%; P = .61). All patients with 1 to 4 BMs in the LT group, except for surgical cases, received stereotactic radiosurgery (n = 31). Among those with >= 5 BMs, most (n = 27; 82%) received whole-brain radiation therapy. There was no significant difference in OS between LT and TKI groups (median overall survival, 28 vs 23 months; hazard ratio, 0.75; 95% confidence interval, 0.52-1.07). In patients with 1 to 4 BMs, the LT group showed significantly better OS compared with the TKI group (median overall survival, 35 vs 23 months; hazard ratio, 0.54; 95% confidence interval, 0.32-0.90). There was no difference in OS between the LT and TKI groups for patients with >= 5 BMs. Multivariable analysis showed that upfront LT yielded significantly better OS for patients with 1 to 4 BMs. Conclusion: Upfront LT followed by EGFR-TKI is more effective than upfront EGFR-TKI for the survival of untreated patients harboring EGFR mutations with 1 to 4 BMs. (C) 2019 Elsevier Inc. All rights reserved.