Daily online contour adaptation by radiation therapists 43 Discussion In this study, we evaluated the CTV contours, adapted and approved by RTTs, which were used for clinical treatment of prostate cancer patients on an MR-Linac. The contours were evaluated using objective parameters and subjective clinical judgement. Interobserver DSC analysis showed high agreement and little to no impact of the adaptations performed. Contours were judged clinically acceptable without any adaptations in 94.2% of the fractions. DVHs analysis of the remaining cases showed that for only one case (one fraction), a significant impact on the CTV coverage occurred. To our knowledge, no other published studies have evaluated the clinical feasibility and acceptability of online contour adaptation by RTTs on an MR-Linac. Several reports were published on first clinical experiences of prostate cancer treatment using MR-Linac systems.11–13 Most of these used a similar ATS workflow, with only Alongi et al.11 specifically reporting that physicians performed the manual contour adaptation. Also, no other studies have evaluated contour adaptation variability in similar clinical circumstances. Pathmanathan et al.14 reported on interobserver variability in prostate cancer contouring for MRI-guided radiotherapy. Five physicians delineated the prostate in Figure 4 – Examples of outlier cases. Transversal (left) and sagittal (right) images with Clinical Target Volume (CTV) contours by radiation therapists (RTTs) (blue) for the seven fractions that were judged to need larger, potentially clinically relevant, adaptations, as judged by Observer 3, next to the adapted contours by Observer 1 (yellow) and/or Observer 2 (red). For Patient 19, fraction 3 and 5 are not displayed separately, since all contours were almost identical (shown here: fraction 3). For Patient 2, for fraction 2, 4, and 5, the CTV contour by the RTTs included the neurovascular bundle at the apex. For Patient 4, fraction 4, the seminal vesicles were missed in the CTV contour by the RTTs and the CTV contour was too wide towards the base and ventrally at the mid-prostate. For Patient 5 and 19, a larger part of the base of the prostate was left out. 2
RkJQdWJsaXNoZXIy MTk4NDMw