Survival among patients with 10 or more brain metastases treated with stereotactic radiosurgery

Abstract

The goal of this study was to evaluate outcomes in patients with ≥ 10 CNS metastases treated with Gamma Knife stereotactic radiosurgery (GK-SRS). Patients with ≥ 10 brain metastases treated using GK-SRS during the period between 2004 and 2010 were identified. Overall survival and local and regional control as well as necrosis rates were determined. The influence of age, sex, histological type, extracranial metastases, whole-brain radiation therapy, and number of brain metastases was analyzed using the Kaplan-Meier method. Univariate (log-rank) analyses were performed, with a p value of < 0.05 considered significant. Fifty-three patients with ≥ 10 brain metastases were treated between 2004 and 2010. All had a Karnofsky Performance Status score of ≥ 70. Seventy-two percent had either non-small cell lung cancer (38%) or breast cancer (34%); melanoma, small cell lung cancer, renal cell carcinoma, and testicular, colon, and ovarian cancer contributed the remaining 28%. On average, 10.9 lesions were treated in a single session. Sixty-four percent of patients received prior whole-brain radiation therapy. The median survival was 6.5 months. One-year overall survival was 42% versus 14% when comparing breast cancer and other histological types, respectively (p = 0.074). Age, extracranial metastases, number of brain metastases, and previous CNS radiation therapy were not significant prognostic factors. Although the median time to local failure was not reached, the median time to regional failure was 3 months. Female sex was associated with longer time to regional failure (p = 0.004), as was breast cancer histological type (p = 0.089). No patient experienced symptomatic necrosis. Patients with ≥ 10 brain metastases who received prior CNS radiation can safely undergo repeat treatment with GK-SRS. With median survival exceeding 6 months, aggressive local treatment remains an option; however, rapid CNS failure is to be expected. Although numbers are limited, patients with breast cancer represent one group of individuals who would benefit most, with prolonged survival and extended time to CNS recurrence.