Prospective Validation of DVH Parameters as Objective Decision Criteria for Dose-Guided Adaptive Radiotherapy

Abstract

Anatomical changes during radiotherapy can lead to substantial dose differences. On line set-up verification is performed by two orthogonal kilo voltage images. If an anatomical difference of larger than 8 mm is detected ('code orange'), an additional Cone Beam Computed Tomography (CBCT) is performed. Based on these 3D images it is decided whether a new radiation plan has to be made. A retrospective study of Zegers et al. defined quantitative DVH threshold values for breast cancer (BC) radiotherapy as decision criteria for dose guided adaptive radiotherapy. Seroma was assumed to be a risk factor for adapting the original plan during treatment. This prospective study is designed to validate those DVH thresholds to enable a clinically applicable adaptation protocol. Between January 2018 and July 2018 32 BC patients with an indication for adjuvant breast or thoracic wall radiotherapy, but without regional nodal irradiation, were included in a single institution; 10 with seroma (defined as at least 3 cm in diameter on the planning CT), of whom 1 received a simultaneous integrated boost (SIB), and 22 patients without seroma, of whom 11 received a SIB. CBCTs were performed during the first and last treatment fraction in all included patients. Additional CBCTs were made in case of a ‘code orange’. The delineated structures on the planning CTs were copied to the CBCTs, manually corrected and a dose recalculation of the original plan was performed. The retrospectively defined DVH values were recorded and checked if the threshold value was exceeded. A total of 70 CBCTs were made in 31 patients (one patient was excluded because the target volume was not completely visible on the CBCTs); 31 CBCTs at the first fraction (including 4 'code orange'), 31 at the last fraction (including 1 'code orange') and 8 additional ‘code orange’ CBCTs. None of these 13 ‘code orange’ CBCTs resulted in an adaptive treatment, based on visual assessment by a radiation oncologist. The predefined DVH threshold values were exceeded in 60% of the CBCTs. Of these, 44% exceeded one DVH parameter, 15% two DVH parameters, 1% three DVH parameters and 2% even four DVH parameters. Only the threshold values of the Dmean CTV1, V95% CTV1 and V90% in patients with a SIB were between tolerance levels in all CBCTs. The previously proposed DVH threshold for adaptation could not be prospectively validated. The selected thresholds seem too strict and would result in too many false positives. To enable a clinically applicable adaptation protocol we suggest to use DVH parameters linked to TCP and NTCP differences.Abstract 3834; Table 1Percentage exceeding DVH parametersDmean CTV1 reduction > 2%0%Dmean CTV2 reduction > 2%7%V95% CTV1 reduction > 5%0%V95% CTV2 reduction > 5%15%Dmax increase > 2% Dmax (SIB) increase > 2%70% 19%V107% increase > 20cm3 V90% (SIB) increase > 20cm314% 0%MLD increase > 0.5Gy16%MHD increase > 0.5Gy11% Open table in a new tab