Quantitative evaluation of the effect of anatomical changes on the results of in vivo dosimetry during radiation therapy in the thoracic region

Background Anatomical changes during thoracic radiation therapy can alter dose delivery and compromise accuracy. Electronic portal imaging device (EPID)‐based in vivo dosimetry (IVD) enables noninvasive monitoring of these changes during treatment. Purpose This study quantitatively evaluated the impact of EPID‐based IVD in volumetric modulated arc therapy (VMAT) and explored the potential of IVD for guiding timely replanning. Methods Eleven patients undergoing thoracic VMAT were retrospectively analyzed: six required replanning as a result of anatomical changes (replanning group), and five did not (normal group). For each fraction, cumulative EPID transit images were acquired, and the global gamma passing rate (GPR) was calculated using six criteria. The difference between treatment GPR and patient‐specific QA GPR (ΔGPR) was used to reduce plan‐specific bias. For four patients in the replanning group (excluding two patients who were excluded due to changes in treatment machines or immobilization), the original plans were recalculated on post‐replanning CT images to estimate dose changes in the planning target volume (PTV) and organs at risks (OARs). Results Across all criteria, the normal group demonstrated higher GPR and ΔGPR values. The 3%/3 mm criterion best discriminated between groups, with a per‐fraction GPR of 98.2% ± 1.1% (normal) versus 90.5% ± 7.1% (replanning). GPR improved after replanning but remained below the levels observed in the normal group. Dose simulations indicated that continuing treatment without replanning increased the PTV dose (D50% up to + 3.5%) and OAR doses, with increases exceeding 5 Gy in some cases. Conclusions EPID‐based IVD detected reduced GPR values in replanning cases, identifying the 3%/3 mm criterion as optimal. A per‐fraction GPR of < 97% under this criterion may indicate notable anatomical changes. Continuing treatment without replanning tended to increase doses delivered to the target and normal tissues. These findings support the clinical utility of IVD and provide quantitative criteria for replanning decisions beyond CT or contour‐based assessments.