Author: Katja M. Langen, Mark McDonald, Bill Stokes, Yinan Wang, Suk Whan (Paul) Yoon 👨🔬
Affiliation: Emory University, Department of Radiation Oncology and Winship Cancer Institute, Emory University 🌍
Purpose: Effective dose management in high-risk re-irradiation (ReRT) patients entails early, detailed special physics consultation (SPC) to mitigate risks from cumulative radiation dose. SPCs have traditionally relied on either conservative non-voxelized biological dose estimates (such as EQD2) applied to whole organs, or arduous iterative-planning processes to achieve an arbitrary coverage-to-cumulative-EQD2 balance, without prior agreement on this balance. We introduce a proactive preplanning SPC method leveraging voxelized EQD2 that: 1. Provides preplan voxel-level dose guidance and 2. Illustrates trade-offs between expected target coverage and cumulative EQD2 to organs-at-risk (OARs).
Methods: OARs were assigned consensus EQD2 limits and a/b ratios for ReRT. In-house RayStation script calculated EQD2remain distribution by subtracting discounted prior EQD2 from maximum allowed OAR EQD2. Physical dose (Dremain) was then derived from EQD2remain scaled to current fractionation. Isodose contours were created from Dremain, explicitly designating to planners volumes not to exceed XXGy. Simulated plans were created using rapid algorithms (Fast Pencil Beam), first prioritizing maximal target coverage then gradually increasing OAR constraints. EQD2 and coverage were recorded at each iteration for physician review.
Results: Preplanning consults were performed for a patient previously treated with 60Gy in 30 fractions to mediastinum, now planned for 40Gy in 5 fractions. Preplan SPC predicted coverage-priority plan to result in 156Gy3/163Gy3 for carotid arteries and great vessels respectively, while OAR-priority plan to achieve D99%=69.8% (27.9Gy) only. Both are well above/below institutional guidelines, indicating EQD2 and coverage goals cannot be met simultaneously. Dosimetrist-generated clinical plans verified preplan-SPC predictions, with coverage-priority plan achieving EQD2=146Gy3/162Gy3 with D99%=101.0% and OAR-priority plan achieving 114.2Gy3/115.9Gy3 with D99%=62.9%.
Conclusion: Our voxelized EQD2-based preplanning SPC method provides nuanced dosimetric guidance and informs physicians of coverage-to-cmulative-EQD2 trade-off early in the planning process, potentially enhancing workflow by reducing iterative planner-to-physician communication burden. The clinical plans reasonably aligned with preplanned predictions, supporting the method's utility.