Development of a Phantom-Based End-to-End QA Workflow for Online Adaptive Radiotherapy Incorporating High-Resolution 2D Detector Measurements
Abstract
Purpose
The development and assessment of an end-to-end quality assurance (QA) procedure for Ethos online adaptive radiotherapy (oART), integrating high-resolution detector-based dose verification with interchangeable anthropomorphic phantom inserts.
Methods
A new QA workflow was developed using the myQA® SRS phantom with a 2D CMOS detector and a custom 3D-printed anthropomorphic pelvis insert. A synthetically deformed planning CT was created to mimic anatomical changes, serving as the basis for the adaptive workflow. The procedure spans CBCT acquisition, manual contour adaptation, adaptive plan optimization on pelvis insert, and dose delivery. Measured and calculated dose distributions were compared using Gamma analysis. Workflow reproducibility was assessed through repeated measurements, and the procedure’s sensitivity to errors was examined by modifying gantry and collimator angles, monitor units (MU), phantom positioning, and CBCT presets. Following AAPM TG-119 methodology, Gamma Pass Rates (GPR) were evaluated using Confidence Limits as plan-specific thresholds.
Results
Repeated workflow executions yielded consistently high GPR values across all criteria (3 %/2 mm, 2 %/2 mm, 3 %/1 mm, 2 %/1 mm). Confidence Limits served effectively as thresholds for detection of plan deviations. Most simulated errors, including gantry angle deviations >3°, collimator angle deviations ≥3°, MU changes ≥3 %, incorrect CBCT presets, and 3 mm setup shifts, were detectable. Minor variations, such as 1° angular changes or 1 % MU alterations, were not detected, with GPRs remaining above the respective Confidence Limit. Manual contouring was necessary, as Ethos auto-contouring did not generate accurate contours on the phantom geometry.
Conclusion
The developed workflow is practical, reproducible, and capable of identifying plan-specific deviations in Ethos oART. High-resolution dose measurements and Gamma analysis provided robust verification of adaptive dose calculation and delivery. Although constrained by manual contouring requirements and data-handling limitations of the Ethos system, this approach offers a viable solution for implementing clinical end-to-end QA in oART.