Optimal Timing And Technique Of Local Therapy For Brain Metastases From Non-Small Cell Lung Cancer With Driver Mutations

Abstract

Brain metastases (BM) are frequent in non-small cell lung cancer patients with driver mutations (dm-NSCLC). Since the availability of brain-penetrating tyrosine kinase inhibitors (TKI), the role of local therapy (LT) for BM from dm-NSCLC is frequently discussed. This analysis examines prognostic factors, particularly the effect of LT timing and technique, in patients with BM from dm-NSCLC. Between 2009 and 2019, 179 patients from one institution were diagnosed with BM from dm-NSCLC and subsequently received TKI-therapy. 78.8% (n = 141) were TKI-naive and selected for analysis. 108 patients (76.6%) had Epidermal Growth Factor Receptor (EGFR) mutations with exon19 deletion most prevalent (n = 61, 56.5%); 33 (23.4%) showed translocation of the Anaplastic Lymphoma Kinase (ALK) gene. 75 patients (41.9%) presented with ≥5 BM. 87 patients (61.7%) received early LT, whereas 54 (38.3%) received delayed LT upon progression (n = 34; 24.1%) or none (n = 20; 14.2%). LT consisted of stereotactic radiosurgery (SRS) (n = 40; 34.2%) or whole brain radiotherapy (WBRT) (n = 77; 65.8%); neurosurgical resection was performed with (n = 15; 10.6%) or without (n = 4; 2.8%) post-operative radiotherapy (RT) in 19 cases. TKI treatment comprised first (n = 93; 66.0%), second (n = 31; 22.0%) or third generation substances (n = 17; 12.1%). A univariate and multivariate Cox proportional hazards model was fitted for overall survival (OS) and intracranial progression-free survival (icPFS). Median OS was 23.0 [IQR 14.6-54.5] months for all patients, 49.1 [IQR 22.7-NA] months for the ALK subgroup and 19.5 [IQR 12.8-46.3] months for the EGFR subgroup (HR 2.6, 95%CI: [1.5-4.7], p = 0.001). Median icPFS was 15.7 [8.9-26.2] months for all patients, 15.7 [10.3-26.2] months and 14.0 [7.9-22.0] months for the subgroups, respectively (HR 1.1, 95%CI: [0.6-1.8], p = 0.814). On multivariate analysis, early LT improved icPFS (HR 0.5, 95%CI: [0.3-0.9], p = 0.009) but not OS (HR 1.4, 95%CI: [0.9-2.3], p = 0.140). This effect was reproduced in independent analysis of the ALK subgroup, but not in the EGFR subgroup. Here, exon 19 deletion (HR 0.4, 95%CI: [0.2-0.8], p = 0.009) and singular BM (HR 0.4, 95%CI: [0.2-0.7], p = 0.002) were independently prognostic of longer icPFS. In the combined cohort, RT technique (SRS vs. WBRT) and number of BM did not influence icPFS or OS. In this analysis, early LT improved icPFS but not OS in TKI-naive patients with BM from dm-NSCLC, compared to upfront TKI treatment. No benefit was shown for WBRT over SRS regarding either icPFS or OS. In light of the toxicities of WBRT, the choice of RT technique should be considered carefully in the context of overall prognosis and quality of life. Especially patients presenting initially with multiple BM may benefit from delaying RT or from individualized approaches like the SRS of multiple or only selected BM instead of WBRT.