# Evaluation of the accuracy of cone-beam CT–based dose calculation for target volumes and organs at risk in left-sided breast cancer radiotherapy

## TL;DR

This study shows that cone-beam CT with HU-RED correction can accurately calculate radiation doses in left-sided breast cancer radiotherapy, comparable to standard planning CT.

## Contribution

The study introduces a patient-specific HU-RED calibration method for CBCT dose calculation in breast cancer radiotherapy and validates its clinical feasibility.

## Key findings

- CBCT-based dose calculations showed minimal deviation (<1.3%) from planning CT for PTV dosimetric indices.
- Gamma analysis showed >90% agreement between CBCT and planning CT dose plans.
- Equivalence testing confirmed statistical equivalence for most PTV and OAR metrics.

## Abstract

Accurate dose calculation is essential in breast cancer radiotherapy. This study aimed to evaluate the feasibility and accuracy of cone-beam CT (CBCT) images with HU–relative electron density (HU-RED) correction for dose calculation in left-sided breast cancer radiotherapy.

Twenty postoperative patients receiving adjuvant radiotherapy at Subei People’s Hospital of Jiangsu Province (Yangzhou, China) from January 2022 to December 2024 were retrospectively enrolled. A patient-specific HU-RED calibration curve was generated using an improved density override method. Treatment plans created on planning CT (pCT) were transferred to CBCT for dose recalculation with identical optimization parameters. Dosimetric parameters of the planning target volume (PTV) and organs at risk (OARs) were compared. Two-dimensional gamma analysis was performed to assess dose consistency. Equivalence between CBCT- and pCT-based dose calculations was further evaluated using the two one-sided tests (TOST), Lin’s concordance correlation coefficient (CCC), intraclass correlation coefficient (ICC), and Bland–Altman analysis.

Differences in PTV dosimetric indices (D2, D50, D98, Dmean, HI, CI) between CBCT- and pCT-based plans were small, with mean deviations <1.3% and no statistically significant differences (P > 0.05). OAR parameters, including cardiac Dmean, V20, V30, V50 and lung Dmean, V20, V30, also showed minimal variation (<1.5%), with the largest deviation observed in cardiac V20 (2.1% in a single case). Gamma analysis revealed high agreement between both plans, with passing rates exceeding 90% for both the 3%/3 mm and 2%/2 mm criteria. Equivalence testing demonstrated statistical equivalence between CBCT- and pCT-based dose calculations for all PTV parameters and most OAR metrics. All 90% confidence intervals fell entirely within predefined equivalence margins (Δ), and all TOST P-values were <0.05. Lin’s CCC and ICC(A,1) exceeded 0.96 for all parameters, indicating excellent consistency, while Bland–Altman analyses showed minimal bias and narrow limits of agreement.

CBCT images corrected with HU-RED calibration achieved highly consistent dose calculation results compared with pCT in left-sided breast cancer radiotherapy. This method is feasible for clinical dose verification and may support future adaptive radiotherapy strategies, particularly with the integration of artificial intelligence techniques.