Outcomes of Initial Therapy for Synchronous Brain Metastases from Extensive Stage Small Cell Lung Cancer

Abstract

Small cell lung cancer (SCLC) has a propensity for brain metastases, which is associated with poor prognosis. 20% of SCLC patients have synchronous brain metastases at initial diagnosis. We sought to determine predictors of overall survival (OS) and brain progression-free survival (bPFS) in SCLC patients with synchronous brain metastases treated with chemotherapy (chemo) and/or whole brain radiation (WBRT). Among the 836 SCLC patients treated at a single institution between 1998 and 2018, 109 patients with synchronous brain metastases at initial diagnosis were included in this retrospective analysis. Survival was estimated with the Kaplan-Meier method. Factors predictive of OS and bPFS were analyzed using Cox proportional hazards regression modeling and log-rank testing. Of the 109 patients, 18 received chemo without WBRT, 15 received WBRT without chemo, and 76 received chemo and WBRT. Of the 76 patients who received chemo and WBRT, 8 started both concurrently, 35 received chemo first, and 33 received WBRT first. Median WBRT dose was 30 Gy in 10 fractions. The median number of brain lesions at diagnosis was 3, and the median size of the largest metastasis was 2.0 cm. Median OS for the entire cohort was 9 months and median bPFS was 8.4 months. OS was 30.3% at 1 year and 14.4% at 2 years. Increased number of brain lesions was significantly associated with decreased OS. 59 patients presented with neurological symptoms, which was not associated with OS or bPFS. 46 patients had isolated brain metastases (no extracranial metastases), which was not associated with improved OS (p = 0.061) or bPFS (p = 0.17) compared to patients with other extracranial metastatic disease. The median age at diagnosis was 63. Patients younger than 63 years old had improved OS compared to patients age 63 or older, but not bPFS. Patients who received chemo and WBRT had improved OS compared to those who had either chemo or WBRT alone (p<0.001), but there was no difference in bPFS (p = 0.34). There was no significant difference in OS or bPFS depending on the sequence of therapy initiation. For patients who had chemo followed by WBRT, the median time between start of chemo and start of radiation was 103 days (range 6 – 182 days), which was not significantly associated with OS or bPFS. 13 patients received upfront brain metastasis resection, which was associated with improved OS but not bPFS compared to those who did not have surgery. 14 patients received SRS or partial brain radiation as part of their initial therapy, which was not significantly associated with OS or bPFS. 15 patients received salvage brain radiation after WBRT failure, which was associated with improved OS compared to those who did not have salvage radiation (25.5 months vs 8.0 months, p<0.001). The combination of chemo and WBRT was associated with improved OS compared to either modality alone. The sequence of treatment was not associated with OS or bPFS. Shorter time to WBRT after chemo was not significantly associated with OS or bPFS.