Whole-brain radiotherapy in NSCLC patients with brain metastasis: Who benefits and by how much?

Abstract

e18068 Background: NSCLC patients with brain metastasis have poor survival even with whole brain radiation therapy (WBRT) alone or in combination with surgery (S+WBRT). The aim of the audit was to identify any prognostic factors which can aid treatment decisions. Methods: A retrospective study of patients with brain metastasis from NSCLC, diagnosed between July 2008 and December 2010 was carried out. Patient characteristics, RPA score at the time of presentation, treatment modality, radiotherapy dose fractionation, time to progression, date of death and primary or secondary presentation were collected. The data was analysed using descriptive statistics; overall survival was calculated using the Kaplan-Meier estimates and compared with the Log-rank test. Multivariate analysis was carried using the Proportional Hazards Model. Results: Of the 156 patients included, 52% presented as primary manifestation and 48% as secondary. The mean age of the audit population was 64.4 year, with 60% of patients falling in the RPA II category. Histology was not available in 27 (18%) of the patients. Majority (79%) were treated with WBRT alone, 3 patients had stereotactic radiosurgery in combination with WBRT and 19% had S+WBRT. 9 (6%) patients did not complete WBRT. Median follow up is 3.8 months (0.5-34.5). Overall survival (OS) for the whole group is 4 months. Better survival was seen with S + WBRT (8 vs. 4 months, p <0.01) and primary presentation (5 vs. 4 months, p=0.009). RPA classification significantly influenced survival in the S + WBRT arm (12 months vs. 7 months for RPA I, p=0.009). On multivariate analysis of the various prognostic factors in the WBRT group, RPA classification and higher RT dose (30 Gy/10#) were associated with better survival (OR 2.3 95%CI 1.6-3.4). Patients treated with SRS have the best survival (11 months). Conclusions: RPA classification is the strongest predictor for outcome following treatment for brain metastasis. These findings can assist in better patient selection and optimising radiotherapy dose fractionation to improve survival outcome. It will be interesting to see if QUARTZ trial supports these findings and clarify the benefit of WBRT in patients with NSCLC.