Miami Cancer Institute: Baptist Health South Florida Miami, Florida
O. Gal1, J. Yu1, M. P. Mehta1, M. D. Hall1, R. H. Press1, Y. Odia2, M. W. McDermott3, Z. Fellows1, A. Wroe1, A. Gutierrez1, and R. Kotecha1; 1Department of Radiation Oncology, Miami Cancer Institute, Baptist Health South Florida, Miami, FL, 2Department of Neuro-Oncology, Miami Cancer Institute, Baptist Health South Florida, Miami, FL, 3Department of Neurosurgery, Miami Neuroscience Institute, Baptist Health South Florida, Miami, FL
Purpose/Objective(s): Intensity modulated proton therapy (IMPT) is increasingly utilized in the treatment of primary central nervous system (CNS) tumors due to its favorable dose distribution and organ-at-risk (OAR) sparing characteristics. Lesions exhibiting radiation-induced contrast enhancement (RICE) have been subsequently observed; however, their exact incidence, spatial distribution patterns, kinetics, and clinical outcomes, as well as an understanding of predisposing risk factors remains poorly defined. Materials/
Methods: Imaging and clinical data were reviewed for consecutive patients treated at a single tertiary care institution between 2017-2023 with IMPT for primary CNS tumors with minimum of 1 year follow-up, at least 3 post-treatment MRI scans, receipt of at least 50 Gy RBE, and no prior radiotherapy. RICE distribution pattern was classified as overlapping with tumor/tumor bed (A), inside or marginal to high dose region (up to 5mm from the 95% isodose line) (B), distant (>5mm from 95% isodose line) (C) or restricted to cranial nerve involvement only (D). Results: The study cohort included 137 patients; the median age was 40 years (Range: 1-85) and 51% were female. The most common histologies were meningioma (32, 23%) and high-grade glioma (31, 23%). IMPT was delivered with pencil beam scanning technique to a median dose of 54 Gy (range 50.4-75 Gy RBE) in 30 fractions (range 28-41). At a median follow-up of 3 years, 77 discrete RICE lesions were observed in 36 (26%) patients, with a median of 1 lesion per patient (range 1-5), appearing at a median of 11 months after IMPT (range 1-39), with pathologic evaluation in 14%. The 1, 2, and 3-year actuarial rates of RICE were 15% (95% CI: 9-20%), 25% (95% CI: 18-32%), and 27% (95% 19-35%). Lesions were classified as group A, B, C, or D in 10 (7%), 22 (16%), 2 (1%) and 2 (1%) patients, respectively. The median time to RICE in group A was 8 months vs. 12 months for group B patients (p=0.07). RICE was most frequently localized in the periventricular space within 5mm of the ventricles (46%). Only 10 (7%) patients had symptomatic RICE and were treated with corticosteroids (n=2), bevacizumab (n=1), both (n=7), or surgery (n=2). At the last follow-up MRI, most lesions decreased in size (16, 44%) or resolved (12, 33%). The overall improvement rate was 75% for all RICE patients (90% and 64% for groups A and B, respectively, p=0.10) and the median time to improvement was 5 months (range: 1-12 months). Conclusion: This contemporary series demonstrates that RICE is a common radiographic finding following modern IMPT, primarily occurring in the periventricular region and approximately 1 year following treatment. Notably, four distinct distribution patterns of RICE are described with different frequencies and kinetics, but most with spontaneous resolution. Given that the majority occur in close proximity to the target volumes, additional analyses are underway to correlate dosimetric parameters to determine relation or causation.
