A new grading system focusing on neurological outcomes for brain metastases treated with stereotactic radiosurgery: the modified Basic Score for Brain Metastases.

OBJECT: The Basic Score for Brain Metastases (BSBM) proposed by Lorenzoni and colleagues is one of the best grading systems for predicting survival periods after stereotactic radiosurgery (SRS) for brain metastases. However, it includes no brain factors and cannot predict neurological outcomes, such as preservation of neurological function and prevention of neurological death. Herein, the authors propose a modified BSBM, adding 4 brain factors to the original BSBM, enabling prediction of neurological outcomes, as well as of overall survival, in patients undergoing SRS.

METHODS: To serve as neurological prognostic scores (NPSs), the authors scored 4 significant brain factors for both preservation of neurological function (qualitative survival) and prevention of neurological death (neurological survival) as 0 or 1 as described in the following: > 10 brain tumors = 0 or ≤ 10 = 1, total tumor volume > 15 cm(3) = 0 or ≤ 15 cm(3) = 1, MRI findings of localized meningeal dissemination (yes = 0 or no = 1), and neurological symptoms (yes = 0 or no = 1). According to the sum of NPSs, patients were classified into 2 subgroups: Subgroup A with a total NPS of 3 or 4 and Subgroup B with an NPS of 0, 1, or 2. The authors defined the modified BSBM according to the NPS subgroup classification applied to the original BSBM groups. The validity of this modified BSBM in 2838 consecutive patients with brain metastases treated with SRS was verified.

RESULTS: Patients included 1868 with cancer of the lung (including 1604 with non-small cell lung cancer), 355 of the gastrointestinal tract, 305 of the breast, 176 of the urogenital tract, and 134 with other cancers. Subgroup A had 2089 patients and Subgroup B 749. Median overall survival times were 2.6 months in BSBM 0 (382 patients), 5.7 in BSBM 1 (1143), 11.4 in BSBM 2 (1011) and 21.7 in BSBM 3 (302), and pairwise differences between the BSBM groups were statistically significant (all p < 0.0001). One-year qualitative survival rates were 64.6% (modified BSBM 0A, 204 patients), 45.0% (0B, 178), 82.5% (1A, 825), 63.3% (1B, 318), 86.4% (2A, 792), 73.7% (2B, 219), 91.4% (3A, 268), and 73.5% (3B, 34). One-year neurological survival rates were 82.6% (0A), 52.4% (0B), 90.5% (1A), 78.1% (1B), 91.1% (2A), 83.2% (2B), 93.9% (3A), and 76.3% (3B), where A and B identify the subgroup. Statistically significant differences in both qualitative and neurological survivals between Subgroups A and B were detected in all BSBM groups.

CONCLUSIONS: The authors' new index, the modified BSBM, was found to be excellent for predicting neurological outcomes, independently of life expectancy, in SRS-treated patients with brain metastases.