2586 - Tumor Dynamics and Dosimetric Impact during Multi-Day Adaptive Stereotactic Radiosurgery (SRS) for Multiple Brain Metastases
Presenter(s)
I. Durowoju1, J. Stewart2, M. E. Ruschin2, H. Soliman2, S. D. Myrehaug2, J. Detsky2, H. Chen2, D. Dinakaran2, E. Atenafu3, A. Sahgal2, L. Holden4, and C. L. Tseng2; 1University of Toronto, Toronto, ON, Canada, 2Department of Radiation Oncology, Odette Cancer Centre, Sunnybrook Health Sciences Centre, University of Toronto, Toronto, ON, Canada, 3Department of Biostatistics, University Health Network, University of Toronto, Toronto, ON, Canada, 4Sunnybrook Health Sciences Center, Toronto, ON, Canada
Purpose/Objective(s): Our frameless Spatially Partitioned Adaptive RadiosurgEry (SPARE) approach using the non-invasive stereotactic radiosurgery icon (GKI-Spr) consists of multi-day single-fraction and/or hypofractionated SRS for the treatment of multiple brain metastases. As treatment may span several days, our practice has been to perform an interim MRI (iMRI) typically 7 to 14 days from the reference treatment planning MRI (rMRI) or last iMRI, on which the plan is adapted for untreated lesions. The goal of this study was to assess target dynamics and consequent dosimetric impact of the re-planned metastases during GKI-Spr.
Materials/Methods: Consecutive patients with multiple brain metastases treated with GKI-Spr between 2020 and 2024, with at least one iMRI, were retrospectively reviewed. For each lesion, geometric metrics including volumetric changes and migration distance (dmigration; the linear displacement of the volume relative to reference planning MRI, rMRI), were assessed relative to the rMRI. Potential dosimetric impact including V100 (volume receiving 100% of prescription dose) and the conformity index was evaluated on the iMRI with the reference plan. Linear mixed model analysis was used to assess factors that may predict geometric changes.
Results: A total of 732 brain metastases were identified in 42 patients. Of these, 313 metastases (42.8%) were re-contoured and re-planned in 27 patients. The median number of brain metastases was 26 (range, 3-73) at rMRI with 10 re-contoured lesions (range, 2-30) at iMRI per patient. Lesions were treated to a dose of 14 – 27.5 Gy in 1-5 fractions. Four patients (9%) underwent multiple iMRIs (range, 2-3) and the median time from the rMRI to the first iMRI was 11 days (range, 4-23 days). Across all metastases, the median lesion volume at rMRI was 0.04 cm3 (range, 0.003 – 18.66 cm3). In the re-contoured lesions, the median change in volume was 0.01 cm3 (range, -14.79-5.15 cm3) and the median migration distance was 1.22 mm (range, -0.5-4.64 mm). The absolute change in volume and migration distance were greater than 0.10 cm3 and 1 mm in 25% and 67% of lesions at iMRI, respectively. The mean gross tumor volume (GTV) V100 decreased from 99.9% on the rMRI to 93.7% on the iMRI (p < 0.0001). Without adaptation, GTV V100 would have been less than 95 % on the iMRI in 26.3% of the re-contoured lesions. The mean (SD) planning target volume (PTV) conformity index decreased from 0.53 (0.21) at rMRI to 0.48 (0.20) at iMRI (p-value < 0.0001). On multivariable analysis, systemic therapy administered within 1 month prior to or concurrently with SRS (p=0.0196) and prior whole brain radiation (p<0.0001) predicted for volumetric changes.
Conclusion: Clinically meaningful brain metastasis dynamics with resultant dosimetric impact were observed during multi-day SRS, which supports the role of iMRI to guide mid-treatment adaptive re-planning in selected patients.