Clinical outcomes of gastrointestinal brain metastases treated with radiotherapy.

Background: Brain metastases of gastrointestinal origin are a rare occurrence. Radiation therapy (RT) in the form of stereotactic radiosurgery (SRS) or whole brain radiation therapy (WBRT) is an effective established treatment modality in either the definitive or adjuvant setting. The aim of this study is to assess the long-term clinical outcomes of patients with gastrointestinal (GI) brain metastases treated with SRS or WBRT.
Methods: In this single institutional retrospective review, we detail the outcomes of patients diagnosed with metastatic brain tumors from an adenocarcinoma gastrointestinal primary. Patients were treated using stereotactic radiosurgery or whole brain radiation therapy. Initial site control (defined as lesions visualized on imaging at time of treatment), new site control (defined as new intracranial lesions visualized on follow-up imaging), and overall survival were calculated using the Kaplan-Meier method.
Results: Thirty-three patients were treated from August 2008 to December 2015. Primary malignancy locations were as follows: 18 colon, 6 esophagus, 4 rectum, 5 other. Median total dose delivered was 25 Gy (18-35 Gy) in a median of 4 fractions for SRS and 30 Gy (10.8-40 Gy) in 10 fractions for WBRT. Crude initial site control at last radiographic follow-up was 64.3% after SRS and 41.7% after WBRT. Eleven of the 28 brain lesions (39.3%) treated with SRS had resection of the SRS-treated lesion prior to radiation therapy. Five of the twelve patients (41.7%) undergoing WBRT underwent cranial resection prior to radiation therapy. Crude new site control at last radiographic follow-up was 46.4% after SRS and 83.3% after WBRT. Kaplan-Meier analysis of overall survival did not show any statistically significant difference between WBRT and SRS (p = 0.424). Median overall survival for SRS patients was 5.2 months (0.5-57.5) and for WBRT patients 4.4 months (0-15). Kaplan-Meier analysis of new site control was significantly improved with WBRT versus SRS (p = 0.017). Total dose, treatment with WBRT, and active extracranial disease were statistically significant on multivariate analysis for new site control (p < 0.05).
Conclusions: Survival and intracranial disease control are poor following RT for brain metastases from GI primaries. In this small series, outcomes are worse than published series for other primary malignancies metastatic to the brain and further research into methods of local control improvement is warranted. Future studies should explore the utility of dose escalation or radiosensitization in this patient population.