Evaluation of Organs-at-Risk Sparing for Glioblastoma Patients Using a Small-Margin Weekly Adaptive Protocol on a 1.5T MR-Linac

Thomas Mann; James Stewart; Amir H Safavi; Jay Detsky; Chia-Lin Tseng; Deepak Dinakaran; Hany Soliman; Sten Myrehaug; Hanbo Chen; John Hudson; Arjun Sahgal; Mark Ruschin

doi:10.1016/j.ijrobp.2026.02.217

Evaluation of Organs-at-Risk Sparing for Glioblastoma Patients Using a Small-Margin Weekly Adaptive Protocol on a 1.5T MR-Linac

Int J Radiat Oncol Biol Phys. 2026 Feb 18:S0360-3016(26)00404-9. doi: 10.1016/j.ijrobp.2026.02.217. Online ahead of print.

Authors

Affiliations

¹ Department of Radiation Oncology, Sunnybrook Odette Cancer Center, University of Toronto, Toronto, Canada.
² Department of Radiation Oncology, Sunnybrook Odette Cancer Center, University of Toronto, Toronto, Canada. Electronic address: mark.ruschin@sunnybrook.ca.

PMID: 41720169
DOI: 10.1016/j.ijrobp.2026.02.217

Abstract

Purpose: To assess the dosimetric impact on organs-at-risk (OARs) and target coverage of a small-margin weekly adaptive treatment for glioblastoma delivered on an MR-Linac.

Methods and materials: The UNIty-Based MR-Linac Guided AdapTive Radiation thErapy for High GraDe Glioma (UNITED) phase 2 prospective trial used a 5 mm clinical target volume (CTV) margin compared to conventional 1.5 to 3 cm CTV margins. Margin reduction was supported by weekly administration of gadolinium contrast and adaptation to gross tumor volume changes using onboard MR imaging. In the present study, dosimetric data from 29 patients with UNITED plans (P_UNITED) treated with 60 Gy in 30 fractions were retrospectively replanned with conventional margins (P_Conv) and compared. For each patient, 6 weekly adapted and 1 conventional margin nonadapted plan were included in the analysis. Relevant OAR dose metrics were compared using Friedman tests followed by Nemenyi post hoc analysis (α = 0.005, corrected for multiple comparisons).

Results: A substantial and statistically significant reduction in irradiated normal brain volume was achieved with the median V60Gy lowered from 146.3-157.3 cm³ (range of median weekly values) with P_Convto 84.2-93.4 cm³ for P_UNITED (P < .0001). The median max dose to the brainstem and optic chiasm was reduced from 56.1 Gy to 49.0 Gy (P = .0001) and from 41.5 Gy to 38.4 Gy (P < .0001), respectively. In addition to OAR sparing, 10 of 29 patients experienced suboptimal target coverage when conventional margin dose distributions were applied to weekly adapted contours, with 5 patients showing consistent underdosing and CTV D98% below 50 Gy, highlighting the importance of adaptation for maintaining target coverage.

Conclusions: The UNITED small-margin weekly adaptive treatment protocol for glioblastoma improves OAR sparing and reduces the irradiated normal brain volume while maintaining consistent target coverage.