Variability of Target Volumes and Organs at Risk Delineation in Breast Cancer Radiation Therapy: Quality Assurance Results of the Pretrial Benchmark Case for the POTENTIAL Trial.

ABSTRACT

PURPOSE: To estimate the variations in clinical target volumes (CTVs) and organs at risk delineation within the quality assurance (QA) program of the POTENTIAL trial, which is a multicenter, randomized phase 3 trial evaluating postmastectomy radiation therapy (RT), with or without internal mammary nodal irradiation, for patients with high-risk breast cancer. METHODS AND MATERIALS The simulating computed tomography scan data set of a benchmark case was sent to the participating centers, and the delineation of CTVs and organs at risk was required to be completed by the investigators following protocol guidelines. All submitted contours were reviewed and compared with the reference contours created by the QA team, using quantitative geometric analysis regarding volume and the Jaccard Index (JCI), Dice similarity coefficient, Geographic Miss Index, Discordance Index, and mean distance to agreement. In addition to the whole-volume analysis of all structures, the combination contour of the supraclavicular fossa and level III and II axilla (CTVsc + axIII + axII) was further analyzed on a slice-by-slice basis. RESULTS: The contours from 26 centers were reviewed and variations were observed between submission and reference. The variations of the CTV of the chest wall, contralateral breast, and heart were small, for which the mean JCI values were 0.62, 0.68, and 0.87, respectively. However, the mean JCI values of the CTV of the internal mammary nodal region, ipsilateral brachial plexus, left anterior descending coronary artery, and right coronary artery were 0.38, 0.21, 0.29, and 0.18, respectively, suggesting marked variations. In addition, marked under- and overoutlining variations were identified on 4 slices of CTVsc + axIII + axII on slice-by-slice analysis. CONCLUSIONS: There were residual contouring variations despite a detailed protocol being provided, confirming the importance of pretrial QA in RT and highlighting the need for education and consideration of a real-time central review of the target delineation before the trial participants begin RT.