Increased survival with the combination of stereotactic radiosurgery and gefitinib for non-small cell lung cancer brain metastasis patients: a nationwide study in Taiwan

PURPOSE Whole brain irradiation (WBRT) either with or without resection has historically been the treatment for brain metastases from non-small cell lung cancer (NSCLC). The effect of gamma knife (GK) radiosurgery, chemotherapy, or the combination remains incompletely defined. In this study, we assessed the outcome of brain metastases from non-small cell lung cancer treated by WBRT followed by GK, gefitinib, or the combination of GK and gefitinib. MATERIAL AND METHODS We retrieved the records of NSCLC patients with brain metastases from the National Health Insurance Research Database (NHIRD) of Taiwan from 2004 to 2010. WBRT either with or without resection was the first line treatment for nearly all patients. The decision to add GK and/or gefitinib treatment was at the discretion of the treating physician and based upon a patient's medical records and imaging data. These patients were classified into four groups including WBRT, WBRT + gefitinib, WBRT + GK, WBRT + gefitinib + GK. These data was evaluated for difference in survival and factors that portended an extended survival from the time of brain metastasis diagnosis. RESULTS Of the 60194 patients with newly diagnosed NSCLC, 23874 (39.6 %) developed brain metastases. The distribution of patients for the groups was WBRT for 20241, WBRT + gefitinib for 3379, WBRT + GK for 155, and WBRT+ gefitinib + GK for 99 patients. The median survival for the time of brain metastasis diagnosis for WBRT, WBRT+ gefitinib, WBRT+ GK, WBRT+ gefitinib + GK groups was 0.53, 1.01, 1.46, and 2.25 years, respectively (p < 0.0001). The hazard ratio (95 % CI) for survival was 1, 0.56, 0.43, and 0.40, respectively (p < 0.001). The adjusted hazard ratio (95 % CI) by age, sex and Charlson comorbidity index (CCI) was 1, 0.73, 0.49, and 0.42, respectively (p < 0.001). CONCLUSION Patients with brain metastases from NSCLC receiving GK or gefitinib demonstrated extended survival. The improved survival seen with GK and gefitinib suggests a survival benefit in selected patients receiving the combined treatment. Further Phase II study should be conducted to assessment these influence.