Management of brain metastases from large cell neuroendocrine carcinoma of the lung: improved outcomes with radiosurgery

Abstract

Abstract Objectives: Large cell neuroendocrine carcinoma (LCNEC) of the lung is a rare pulmonary tumor, having similar natural history and management strategy as small cell lung cancer. Therefore, the management of brain metastases in these patients has mirrored that of SCLC through the use of whole brain radiation therapy (WBRT). We used the National Cancer Database (NCDB) to look at predictors of stereotactic radiosurgery (SRS) and any potential differences in outcomes for patients with brain metastases from LCNEC. Material and methods: We queried the NCDB from 2004 to 2015 for patients with LCNEC of the lung with brain metastases that received brain radiation. Univariable and multivariable analyses were performed to identify factors predictive of SRS use and overall survival (OS). Propensity-adjusted Cox proportional hazard ratios for survival were used to account for indication bias. Results: Out of 9970 patients with LCNEC of the lung we identified 348 with brain metastases. Sixty-eight patients were treated with upfront SRS and 280 were treated with WBRT. Patients that were treated at an academic facility or received chemotherapy as part of upfront treatment were more likely to receive SRS. Univariable analysis revealed improved outcomes with SRS compared to WBRT, with a median OS of 11 months compared to 6 months, respectively (p\u2009=\u2009.007). Multivariable Cox regression with propensity score confirmed SRS to have improved survival (HR: 0.68, 95%CI: 0.51–0.91, p\u2009=\u2009.0093). Multivariable Cox regression with propensity score also identified younger age, receipt of chemotherapy, absence of extracranial disease and non-rural locations as additional predictors of improved OS. Conclusions: Treatment of brain metastases from LCNEC of the lung with SRS was associated with improved survival. For the appropriate patients, upfront treatment of limited brain metastases with SRS may be appropriate.