MU variability in CBCT‐guided online adaptive radiation therapy

• DOI: 10.1002/acm2.14440
• Publication Date: 2024-06-19T20:44:17Z
• Authors: Sean Tanny; Olga M Dona Lemus; Joshua Wancura; Nicholas Sperling; Matthew Webster; Hyunuk Jung; Yuwei Zhou; Fiona Li; Jihyung Yoon; Alexander Podgorsak; Dandan Zheng
• Keywords: 03 medical and health sciences; 0302 clinical medicine; Radiation Oncology Physics

Abstract Purpose CBCT‐guided online‐adaptive radiotherapy (oART) systems have been made possible by using artificial intelligence and automation to substantially reduce treatment planning time during on‐couch adaptive sessions. Evaluating plans generated an session presents significant challenges the clinical team as process gets compressed into a shorter window than offline planning. We identified MU variations up 30% difference between plan reference in several oART sessions that caused question accuracy of dose calculation. investigated cause variation overall delivered when appear unnecessarily large. Methods Dosimetric data from 604 19 patients undergoing were collected. The analysis included total per fraction, target volume (PTV) organs at risk (OAR) volumes, changes PTV‐OAR overlap, DVH curves. Sessions with greater two standard deviations mean reoptimized offline, verified independent calculation system, measured detector array. Results relative normally distributed −1.0% deviation 11.0%. No correlation was found anatomic changes. Offline reoptimization did not reliably reproduce either or MUs, suggesting stochastic effects within optimizer are likely causing variations. Independent array measurements resulted acceptable agreement planned dose. Conclusions observed any errors workflow. Providers should refrain variability way express their confidence accuracy. Clinical decisions rely on validated secondary calculations ensure optimal selection.