Radiosurgery for the Treatment of Brain Metastases: Comparison of Prognostic Indexes in a Brazilian Population

Abstract

Purpose/Objective(s)To define the most accurate prognostic index for patients with brain metastases in a Brazilian Public Hospital.Materials/MethodsWe retrospectively analyzed all patients with brain metastases treated with radiosurgery at Brazil’s largest public institution. Data was gathered from the first case treated (September 2008) until December 2010. In this period, 78 patients received radiosurgery for brain metastases. 4 patients were excluded due to having no follow-up data (missing completely at random). 74 patients entered our final analysis. Patients were treated with a 6 MV LINAC Varian 6EX with brainLAb micro-MLC or cones. Data on the prognostic indexes SIR (Score Index for Radiosurgery), GPA (Graded Prognostic Assessment), RPA (Recursive Partitioning Analysis) and BSBM (Basic Score for Brain Metastases) were collected on baseline and compared.ResultsThe median age was 56 years (range 20 - 82 years). Median dose was 1800cGy (1000 - 2200cGy). 44 patients died during follow-up, with a median follow-up of 6.23 months (1 - 25.6 months). The majority of primary tumors were lung (40.54 %) and breast (20.27%). The median survival was 9.53 months (1 - 25.6months). The median survival by RPA classes I, II, and III was 10.3 months, 9.53 months, 3.53 months respectively. The median survival grouped under SIR classification from 1 to 3, 4 to 7 and 8 to 10 was 2.16 months, 3.1 months, 7.96 months, respectively. Median survival for the BSBM was class 0: 3.53 m., class I: 8.8m., class II: 10.5, class III: 14.6. Correlation of prognostic indexes with survival were: SIR (rho = 0.2903 , p = 0.0127) ; RPA (rho = -0.1014, p = 0.3932); GPA (rho 0.1696, 0.1513); BSBM (rho = 0.3370, p 0.0041).ConclusionsOur findings suggest that SIR and BSBM were the most accurate in predicting survival. We recommend using these indexes to guide clinical decisions in patients with the same characteristics as our population. Purpose/Objective(s)To define the most accurate prognostic index for patients with brain metastases in a Brazilian Public Hospital. To define the most accurate prognostic index for patients with brain metastases in a Brazilian Public Hospital. Materials/MethodsWe retrospectively analyzed all patients with brain metastases treated with radiosurgery at Brazil’s largest public institution. Data was gathered from the first case treated (September 2008) until December 2010. In this period, 78 patients received radiosurgery for brain metastases. 4 patients were excluded due to having no follow-up data (missing completely at random). 74 patients entered our final analysis. Patients were treated with a 6 MV LINAC Varian 6EX with brainLAb micro-MLC or cones. Data on the prognostic indexes SIR (Score Index for Radiosurgery), GPA (Graded Prognostic Assessment), RPA (Recursive Partitioning Analysis) and BSBM (Basic Score for Brain Metastases) were collected on baseline and compared. We retrospectively analyzed all patients with brain metastases treated with radiosurgery at Brazil’s largest public institution. Data was gathered from the first case treated (September 2008) until December 2010. In this period, 78 patients received radiosurgery for brain metastases. 4 patients were excluded due to having no follow-up data (missing completely at random). 74 patients entered our final analysis. Patients were treated with a 6 MV LINAC Varian 6EX with brainLAb micro-MLC or cones. Data on the prognostic indexes SIR (Score Index for Radiosurgery), GPA (Graded Prognostic Assessment), RPA (Recursive Partitioning Analysis) and BSBM (Basic Score for Brain Metastases) were collected on baseline and compared. ResultsThe median age was 56 years (range 20 - 82 years). Median dose was 1800cGy (1000 - 2200cGy). 44 patients died during follow-up, with a median follow-up of 6.23 months (1 - 25.6 months). The majority of primary tumors were lung (40.54 %) and breast (20.27%). The median survival was 9.53 months (1 - 25.6months). The median survival by RPA classes I, II, and III was 10.3 months, 9.53 months, 3.53 months respectively. The median survival grouped under SIR classification from 1 to 3, 4 to 7 and 8 to 10 was 2.16 months, 3.1 months, 7.96 months, respectively. Median survival for the BSBM was class 0: 3.53 m., class I: 8.8m., class II: 10.5, class III: 14.6. Correlation of prognostic indexes with survival were: SIR (rho = 0.2903 , p = 0.0127) ; RPA (rho = -0.1014, p = 0.3932); GPA (rho 0.1696, 0.1513); BSBM (rho = 0.3370, p 0.0041). The median age was 56 years (range 20 - 82 years). Median dose was 1800cGy (1000 - 2200cGy). 44 patients died during follow-up, with a median follow-up of 6.23 months (1 - 25.6 months). The majority of primary tumors were lung (40.54 %) and breast (20.27%). The median survival was 9.53 months (1 - 25.6months). The median survival by RPA classes I, II, and III was 10.3 months, 9.53 months, 3.53 months respectively. The median survival grouped under SIR classification from 1 to 3, 4 to 7 and 8 to 10 was 2.16 months, 3.1 months, 7.96 months, respectively. Median survival for the BSBM was class 0: 3.53 m., class I: 8.8m., class II: 10.5, class III: 14.6. Correlation of prognostic indexes with survival were: SIR (rho = 0.2903 , p = 0.0127) ; RPA (rho = -0.1014, p = 0.3932); GPA (rho 0.1696, 0.1513); BSBM (rho = 0.3370, p 0.0041). ConclusionsOur findings suggest that SIR and BSBM were the most accurate in predicting survival. We recommend using these indexes to guide clinical decisions in patients with the same characteristics as our population. Our findings suggest that SIR and BSBM were the most accurate in predicting survival. We recommend using these indexes to guide clinical decisions in patients with the same characteristics as our population.