Multi-centre evaluation of variation in cumulative dose assessment in reirradiation scenarios

Nicholas Hardcastle,Eliana Vasquez Osorio,Andrew Jackson,Charles Mayo,Anja Einebærholm Aarberg,Myriam Ayadi,Francesca Belosi,Cemile Ceylan,Angela Davey,Pauline Dupuis,Julia-Claire Handley,Theresa Hemminger,Lone Hoffmann,Colin Kelly,Chrysanthi Michailidou,Sarah Muscat,Donna H Murrell,Jaime Pérez-Alija,Catherine Palmer,Lorenzo Placidi,Marija Popovic,Heidi S Rønde,Adam Selby,Theodora Skopidou,Natasa Solomou,Joep Stroom,Christopher Thompson,Nicholas S West,Ali Zaila,Ane L Appelt,Donna H. Murrell,Heidi S. Rønde
DOI: https://doi.org/10.1016/j.radonc.2024.110184
IF: 6.901
2024-03-08
Radiotherapy and Oncology
Abstract:Background and Purpose Safe reirradiation relies on assessment of cumulative doses to organs at risk (OARs) across multiple treatments. Different clinical pathways can result in inconsistent estimates. Here, we quantified the consistency of cumulative dose to OARs across multi-centre clinical pathways. Material and Methods We provided DICOM planning CT, structures and doses for two reirradiation cases: head & neck (HN) and lung. Participants followed their standard pathway to assess the cumulative physical and EQD2 doses (with provided α/β values), and submitted DVH metrics and a description of their pathways. Participants could also submit physical dose distributions from Course 1 mapped onto the CT of Course 2 using their best available tools. To assess isolated impact of image registrations, a single observer accumulated each submitted spatially mapped physical dose for every participating centre. Results Cumulative dose assessment was performed by 24 participants. Pathways included rigid (n = 15), or deformable (n = 5) image registration-based 3D dose summation, visual inspection of isodose line contours (n = 1), or summation of dose metrics extracted from each course (n = 3). Largest variations were observed in near-maximum cumulative doses (25.4–––41.8 Gy for HN, 2.4–––33.8 Gy for lung OARs), with lower variations in volume/dose metrics to large organs. A standardised process involving spatial mapping of the first course dose to the second course CT followed by summation improved consistency for most near-maximum dose metrics in both cases. Conclusion Large variations highlight the uncertainty in reporting cumulative doses in reirradiation scenarios, with implications for outcome analysis and understanding of published doses. Using a standardised workflow potentially including spatially mapped doses improves consistency in determination of accumulated dose in reirradiation scenarios.
oncology,radiology, nuclear medicine & medical imaging
What problem does this paper attempt to address?