S804
Clinical - Lung
ESTRO 2026
before and after the CAA objective was implemented, comparing dose-survival associations between cohort 1 (pre-CAA) and cohort 2 (post-CAA). Material/Methods: Cohort 1 data included 559 patients (treated between 04/2021-03/2023), while cohort 2 data included 328 patients (04/2023-01/2024). RT planning data were spatially normalised via deformable registration of planning CTs and dose distributions to a common reference anatomy. Registration accuracy was assessed using heart ROI centroid offsets and visual inspection; patients with failed registrations were excluded. Patients were stratified by 12-month survival status. Voxel-wise dose comparisons between survivors and non-survivors used Student’s t-tests. Significance was assessed via permutation testing (1000 iterations with random survival label reassignment), allowing regions with significant dose- survival associations to be identified. Robustness of the identified regions was evaluated using multivariate Cox proportional hazards models including known clinical risk factors (age, sex, performance status, T- stage, N-stage, tumour volume, laterality, and mean lung dose). Results: After excluding failed image registrations (41 in cohort 1, 17 in cohort 2), 518 and 311 patients remained in cohorts 1 and 2, respectively. Registration accuracy was comparable between cohorts: heart centroid SD (LR/AP/CC) 3.1/3.6/5.8mm (cohort 1), 3.8/3.6/7.1mm (cohort 2). Figure 1 shows the difference in average dose between the cohorts with reduced dose to the CAA in cohort 2. Significant dose-survival associations were found in both cohorts (figure 2). In cohort 1, the most significant region overlapped the CAA (p < 0.01) and in cohort 2 it was located more posteriorly, overlapping the left bronchus and aorta, outside the CAA (p < 0.005). Multivariate Cox analysis confirmed that dose to the VBA-identified region in cohort 2 remained significantly associated with survival for cohort 2 patients after adjusting for clinical covariates (p = 0.01).
Conclusion: After implementing the CAA, reducing dose to this region removed the previously observed dose-survival association at the heart base. The region of significant association shifted outside of the heart more posteriorly, and remained significant after adjusting for clinical covariates. These findings suggest that other critical thoracic structures may also influence post-radiotherapy survival. References: [1] A McWilliam, J Kennedy, C Hodgson, E Vasquez Osorio, C Faivre-Finn, M van Herk (2017). Radiation dose to heart base linked with poorer survival in lung cancer patients. European Journal of Cancer, 85, 106– 113. https://doi.org/10.1016/J.EJCA.2017.07.053[2] I Fornacon-Wood, R Holley, H Crawford, K Banfill, T Marchant, C Morgan, H Turner-Uaandja, A Walker, E
Made with FlippingBook - Share PDF online