Related Experiment Video
Updated: Jan 11, 2026
Stereotactic Radiosurgery for Gynecologic Cancer
Published on: April 17, 2012
Prognostic scores for patients with multiple brain metastases treated with repeated stereotactic radiosurgery - a
Annabella Schiele1, Denise Bernhardt2, Thomas Welzel1
1University Hospital Heidelberg, Department of Radiation Oncology, INF 400, 69120 Heidelberg, Germany; Heidelberg Institute of Radiation Oncology (HIRO), Germany; National Center for Tumor diseases (NCT), Heidelberg, Germany.
Background:
Stereotactic radiosurgery (SRS) is frequently combined with modern systemic therapy for brain metastases (BM). Most established prognostic scores were developed from historical cohorts and may not reflect current treatment standards. The randomized phase II CYBER-SPACE trial investigated upfront and repeated SRS for up to ten simultaneous BMs during systemic therapy, aiming to avoid whole-brain radiotherapy (WBRT). This secondary analysis evaluates the validity of established scores in this contemporary cohort.
Methods:
Assessed scores included the Recursive Partitioning Analysis (RPA), modified RPA (M-RPA), Graded Prognostic Assessment (GPA), diagnosis-specific GPA (dsGPA), lung-mol GPA, Basic Score for Brain Metastases (BSBM), Score Index for Radiosurgery (SIR), and Brain Metastasis Velocity (BMV) score. Their predictive value for overall survival (OS) and WBRT indication (WBRTi; defined as > 10 simultaneous BMs or leptomeningeal disease) was analysed. The BMV score was calculated at first through fourth distant brain failure (DBF). Kaplan-Meier analyses and log-rank tests were used.
Results:
A total of 202 patients were enrolled. Common primary tumors were non-small cell lung cancer (63 %), melanoma (16 %), and breast cancer (10 %). Median OS after initial SRS was 13.1 months (Q1-Q3: 5.2-45.0). At 24 months, 40 patients (20.8 %) reached WBRTi. The dsGPA for melanoma (OS: p = 0.014; WBRTi: p = 0.042) and BMV (OS and WBRTi: p < 0.0001) were predictive for both endpoints; BSBM was predictive for OS only (p = 0.0011). The BMV remained prognostic through the fourth recurrence. Two endpoint-specific BMV cut-offs were identified, separating two prognostic groups for survival and WBRTi, respectively (p < 0.0001).
Conclusion:
BMV was the only score able to predict both OS and WBRTi across all tumor types. Two distinct endpoint-specific BMV cut-offs may support clinical decision-making between continued SRS and early WBRT, but warrant further external validation.
