Dosimetric impact of esophageal motion in esophagus-sparing radiotherapy of spinal metastases

The phase III trial (NCT05109819) investigates whether esophagus-sparing radiotherapy (RT) reduces dysphagia in patients with spinal metastases. This retrospective simulation study evaluates the dosimetric impact of inter-fractional esophageal motion in this setting. Patient/materials and methods: Patients receiving daily image-guided RT on high-quality cone beam computed tomography (CBCT)-equipped units between September 2023 and December 2024 were screened. Inclusion required five consecutive CBCTs with vertebrae T1-T10 in view and a visible esophagus. The esophagus was contoured on the planning CT (pCT) and five CBCTs. Standard and esophagus-sparing 25 Gy/5 fraction volumetric-modulated arc therapy plans were simulated for five thoracic targets. In esophagus-sparing plans, the esophageal D0.027cc was limited to 8 Gy equivalent dose in 2 Gy fractions (α/β = 3 Gy), consistent with the ESO-SPARE constraint. CBCT-based esophagus contours were rigidly transferred to the pCT after bony matching. Esophageal metrics (Mean, D0.027cc-D5cc, V8.5 Gy) were extracted from the original plan and compared across CBCT and pCT contours and between plan types. Inter-fraction motion was assessed using 95% Hausdorff Distance (HD). CBCT quality was evaluated in two patients (five targets) using Dice similarity coefficients and interobserver 95% HDs. Twelve patients with 23 targets were simulated. Inter-fraction motion led to dose constraint violations in 20 of 23 plans, though only small esophageal volumes entered the high-dose region. Dice scores > 0.8 confirmed good esophageal visibility on CBCT. Most 95% HDs fell within interobserver variability, indicating motion was comparable to contouring uncertainty. Despite constraint violations, high-dose exposure was limited to small esophageal volumes, and overall sparing was preserved.