S833
Clinical - Lung
ESTRO 2026
no CTV and small PTV margins. However, the ITV approach generates motion margins that can exceed requirements for dosimetric coverage2. Here, we estimate the impact of the ITV planning approach on locoregional tumour control using a novel SDD model fitted to clinical data. We also conducted a ‘virtual trial’ replacing the ITV with a fixed low-dose CTV to see if this treatment approach would better cover SDDs, independent of tumour motion. Material/Methods: We analysed 257 patients with stage I-II NSCLC treated with 3-8 fractions to 54-60Gy using IMRT/VMAT, using PTV=ITV+5mm. We first virtually removed the ITV, and then added a virtual low-dose CTV that was defined as GTV+2cm, prescribed to 40% of maximum dose (Fig.1). To calculate Tumour Control Probability (TCP), delivered dose was estimated using planned dose and 4DCT. Our SDD model was fitted to locoregional recurrence data, and assumed a 3D distribution of 0.5mm radius deposits at random location, with a gradual decay in probability away from the GTV (Fig.2a). The TCP model3 incorporated realistic delineation uncertainty4, and its free parameters were fitted to observed locoregional failure data using the Brier score, varying average number of SDDs, probability decay and clonogen density.
V15Gy difference was 0.0% (IQR -17.4–1.45), per artery, LAD (-9.9%), RCA (0.0%), LCX (-0.1%). There was no significant difference between V15Gy differences for the ‘Acceptable’ and ‘Minor Edits’ groups (0.0Gy vs - 3.8Gy, p=0.089). There was no correlation between V15Gy absolute difference and time taken per artery for all segmentations (R2 0.019) or per clinician (R2 0.000–0.075). Conclusion: Efficient delineation of the CAs on 4DCT was achievable for radiation oncologists without previous experience or specific training. Dose estimates trended slightly lower than reference values, but were relatively robust to minor segmentation deviations. Manual segmentation of CAs should not be a barrier to their inclusion during optimisation of thoracic treatment plans. There is an imperative for clinicians to obtain CA delineation competencies in preparation for supervising auto-segmentation tools. References: 1. Atkins KM et al. Association of left anterior descending coronary artery radiation dose with major adverse cardiac events and mortality in patients with non-small cell lung cancer. JAMA Oncol 2021;7(2):206- 2192. Atkins KM et al. Cardiac substructure radiation dose and associations with tachyarrhythmia and bradyarrhythmia after lung cancer radiotherapy. JACC CardioOncol 2024;6(4):544-5563. Walls GM et al. Radiation oncology opinions and practice on cardiotoxicity in lung cancer: a cross-sectional study by the International Cardio-Oncology Society. Clin Oncol (R Coll Radiol) 2024;36(12):745-7564. Duane FK et al. A cardiac contouring atlas for radiotherapy. Radiother Oncol 2017;122(3):416-422 Keywords: Coronary artery, 4DCT, manual segmentation Is the Internal Target Volume important to control subclinical disease during Stereotactic Ablative Radiotherapy for lung cancer? Marcel van Herk 1 , Thomas Hahn 1 , Jamie Partridge 1 , Alan McWilliam 1 , Angela Davey 1 , Azadeh Abravan 2,1 , Ursula Nestle 3 , Corinne Faivre-Finn 1 1 Division of Cancer Sciences, The University of Manchester, Manchester, United Kingdom. 2 Institute of genetics and cancer, University of Edinburgh, Edinburgh, United Kingdom. 3 Department of Radiation Oncology, Kliniken Maria Hilf, Moenchengladbach, Germany Purpose/Objective: In early stage lung cancer, Subclinical Disease Deposits (SDDs) outside the GTV have been reported at 8-26 mm distance1. Despite this, conventional SABR uses Digital Poster Highlight 4167
Results: Observed and model-fitted locoregional failure rates were 12% (median follow-up 18 months). Most failures were due to SDDs. Removing the ITV reduced mean(SD) TCP by 2.2(2.3)% and decreased mean lung dose by 0.11(0.12)Gy. Replacing the ITV with the low-
Made with FlippingBook - Share PDF online