Presenter(s)
T. L. Vaziri1, I. C. Liu1, S. Downs2, T. Beckham2, M. T. Milano3, J. G. Roubil4, T. J. Harris4, H. R. Han5, E. L. Chang5, and K. J. Redmond6; 1Department of Radiation Oncology and Molecular Radiation Sciences, Johns Hopkins University School of Medicine, Baltimore, MD, 2Department of CNS Radiation Oncology, The University of Texas MD Anderson Cancer Center, Houston, TX, 3Department of Radiation Oncology, University of Rochester Medical Center, Rochester, NY, 4Virginia Commonwealth University, Richmond, VA, 5Department of Radiation Oncology, University of Southern California Keck School of Medicine, Los Angeles, CA, 6Department of Radiation Oncology and Molecular Radiation Sciences, Johns Hopkins University, Baltimore, MD
Purpose/Objective(s): Stereotactic radiosurgery (SRS) is the standard of care for patients with limited brain metastases (BMs) given favorable neurocognitive outcomes and survival compared to whole-brain radiotherapy (WBRT). The role of SRS in patients with extensive intracranial disease remains controversial. This retrospective, multi-institutional study evaluates outcomes and toxicity in patients treated with SRS to =15 BMs in a single course.
Materials/Methods: Patients who underwent SRS to =15 BMs in a single course between 2010-2023 were identified across five institutions. Primary endpoints included overall survival (OS), and freedom from WBRT (FFW). OS was estimated using Kaplan-Meier method and defined from the time of SRS. Cox proportional hazard models were used to determine prognostic factors associated with outcomes. FFW was estimated amongst patients without a history of prior WBRT using cumulative incidence with death as a competing risk, and Fine–Gray regression was used for competing-risks analysis.
Results: A total of 127 patients were included, with a median of 18 BMs treated (IQR, 16-24). Primary tumors included lung (43.3%), breast (23.6%), and melanoma (22.1%). Majority of patients had KPS >70 (89.8%). At the time of SRS, 23 patients (18.1%) had prior WBRT and 38 (29.9%) had prior SRS, most commonly a single prior course (68.4%). Treatment platforms included a non-invasive stereotactic radiosurgery instrument (35.4%), LINAC-based systems (37.8%), and a frameless robotic radiosurgery system (26.8%). Median total planning target volume (PTV) was 8.95 cm3. Thirty-two patients (25.2%) underwent at least one subsequent SRS course, with median interval to subsequent treatment of 7 months (IQR, 4-16). Local failure occurred in 7.9% of patients, with median time to failure of 5.8 months. Grade =3 radiation necrosis occurred in four patients (3.2%). Hospitalization within 30 days of SRS occurred in 19.7% of patients from any-cause and in 6.3% from neurologic causes. Median OS was 7.6 months (95% CI 5.8-12.6), with 12-month OS of 41.6% (95% CI 32.9%-50.1%). One-year FFW was 65.7%, and 84.6% ultimately avoided salvage WBRT. Leptomeningeal disease (LMD) developed in 4.7% of patients with median time to LMD of 6.8 months. On multivariate analysis, improved OS was independently associated with KPS >70 (HR 0.21; 95% CI 0.11-0.39, p <.0001); Worse OS was independently associated with previously known disease at the time of SRS (versus new diagnosis) (HR 1.86; 95% CI 1.21-2.87, p = .005). Total PTV, total number of treated BMs, and primary malignancy were not significantly associated with OS or FFW on univariate and multivariate analysis.
Conclusion: In this large, multi-institutional study evaluating SRS for patients with =15 BMs, SRS achieved high rates of local control, acceptable toxicity, and allowed majority of patients to avoid WBRT.