S2424
Physics - Radiomics, functional and biological imaging, and outcome prediction
ESTRO 2026
Material/Methods: We evaluated doses to the large thoracic vessels (aorta, pulmonary artery, pulmonary veins, superior vena cava and inferior vena cava) in 202 patients treated with SBRT for 209 centrally and ultra-centrally located lung tumours included in the expanded HILUS- cohort. A Cox regression-based prediction model was developed, accounting for bronchial tumour compression, MIB near-maximum dose (D0.31cm3), maximum dose to each large vessel and time-to-event. We also performed a Spearman correlation analysis to address the risk of dosimetric cross-correlation between the MIB and the large vessels due to their anatomically close position. Results: In the univariable analysis, the maximum dose to the following structures predicted fatal bronchopulmonary bleeding: the pulmonary artery (p=0.011), the pulmonary veins (p=0.005) and the inferior vena cava (p=0.025). The dose to these three vessels also had the strongest correlation to the MIB dose, see Figure 1.
statistical characteristics of the cohort since this vessel had the largest standard deviation in dose (Figure 2). The model performance is known to depend on the variance in the cohort [2], and such a pattern is seen in Figure 2. Thus, it is unclear whether the pulmonary veins should be considered an important organ of interest for bronchopulmonary bleeding, or whether it predicts the endpoint through its cross-correlation with MIB, in combination with a greater variance in the cohort.
Conclusion: In the expanded HILUS-cohort, the dose to the aorta, superior vena cava and inferior vena cava does not contribute to the risk of fatal bronchopulmonary bleeding after SBRT of centrally and ultra-centrally located lung tumours. The role of the pulmonary artery and the pulmonary veins will be explored further, as their impact on the endpoint is difficult to separate from the adjacent structure of MIB. References: 1. Lindberg S, Grozman V, Karlsson K, Onjukka E, Lindbäck E, Al-Jirf K, et al. Expanded HILUS trial: a pooled analysis of risk factors for toxicity from stereotactic body radiation therapy of central and ultracentral lung tumors. Int J Radiation Oncol Biol Phys 2023;117(5):1222-31. 2. Rutkowska E, Baker C, Nahum A. Mechanistic simulation of normal-tissue damage in radiotherapy – implications for dose- volume analyses. Phys Med Biol 2010;55(8):2121-36. Keywords: SBRT, central lung, dose to large vessels
In the multivariable analysis including vessel dose, MIB D0.31cm3 and bronchial tumour compression, the only vessel significantly predicting bronchopulmonary bleeding was the dose to the pulmonary veins (p=0.040). This dependence may however result from
Made with FlippingBook - Share PDF online