Central Nervous System
PV QA 2 - Poster Viewing Q&A 2
MO_7_2640 - Improved Local Control with Internal Dose-Escalation for Melanoma Brain Metastases Treated with Stereotactic Radiosurgery
Monday, October 22
10:45 AM - 12:15 PM
Location: Innovation Hub, Exhibit Hall 3
Improved Local Control with Internal Dose-Escalation for Melanoma Brain Metastases Treated with Stereotactic Radiosurgery
W. R. Kennedy1, S. Acharya2, M. Mahmood1, T. A. DeWees1, S. M. Perkins1, J. Huang1, C. Tsien1, C. G. Robinson1, and C. D. Abraham Jr1; 1Washington University School of Medicine, Department of Radiation Oncology, St. Louis, MO, 2St. Jude Children's Research Hospital, Memphis, TN
Purpose/Objective(s): The wide variability in local control (LC) rates of brain metastases (BM) after stereotactic radiosurgery (SRS) is due to a variety of patient, tumor and treatment-related factors. For a given prescription dose and tumor size, the high-dose-volume may vary widely. Increasing high-dose volume has been demonstrated to improve local control (LC) without increasing toxicity in non-small cell lung cancer BM treated with SRS. The purpose of this study is to identify potentially actionable dosimetric predictors of LC after SRS for melanoma BM.
Materials/Methods: Between 2006 and 2017, patients with melanoma BM treated with single-fraction SRS were reviewed. Patients who had a prior metastatectomy or re-SRS to the same lesion were excluded. Clinical and neuroimaging follow-up were reviewed from the time of SRS to either last follow-up or death. For each treated lesion, local control was assessed at each post-SRS MRI and defined as stability or decrease in size of the BM. Outcome-oriented approaches were utilized to determine optimal cutpoints for dosimetric variables relative to LC. Statistical comparisons utilized univariate (UVA) and multivariate (MVA) Cox regression to evaluate the impact of collected parameters on LC. Lesions were stratified by V30 Gy greater than or less than 25%.
Results: There were 287 melanoma brain metastases identified in 79 patients. Median age was 56 years (range, 31 – 86). DS-GPA was ≤3 in 90% of patients. At a median follow-up of 7.6 months, median overall survival was 9.3 months. Median tumor volume was 377 mm3 (range, 5 – 23700). For the entire cohort, 1-year LC was 83% versus 66% for V30 above and below 25%, respectively (p=0.001). Stratifying by maximal axial dimension-equivalent volume, lesions 2cm or less (n=215) had a 1 year LC of 82% versus 70% (p=0.01) for V30 above and below 25%, respectively. Lesions >2-3cm (n=32) had a 1 year LC of 100% versus 43% (p=0.214) for V30 above and below 25%, respectively. The overall rate of radiation necrosis was 2.8%, and was not associated with V30.
Conclusion: Measures of high dose within the target volume are predictive of local control after SRS for melanoma BM. Optimizing dosimetric measures such as V30>25% represents actionable parameters in SRS treatment planning. Given the relatively low frequency of lesions larger than 2cm in our current series, conclusions regarding the benefit and toxicity in this subset are unclear and warrant further investigation.
Author Disclosure: W.R. Kennedy: None. S. Acharya: None. T.A. DeWees: None. J. Huang: Speaker's Bureau; Viewray Inc. Travel Expenses; Viewray Inc. C. Tsien: None. C.G. Robinson: Research Grant; Varian Medical Systems, Elekta. Speaker's Bureau; Varian Medical Systems, DFINE. Advisory Board; Radialogica. Stock Options; Radialogica. C.D. Abraham: Employee; Phillips Medical Systems.