128 - Fatal Pulmonary Hemorrhage in Patients Treated with Hypofractionated RT for Stage II-III NSCLC
Presenter(s)
S. Cooke1, J. Belderbos1, J. van Diessen1, C. Faivre-Finn2, B. Reymen3, M. Lambrecht4,5, E. M. Dieleman6, G. Fredberg Persson7,8, B. Stam9, D. de Ruysscher10, and J. J. Sonke11; 1The Netherlands Cancer Institute (NKI-AVL), Amsterdam, Netherlands, 2The Christie NHS Foundation Trust and The University of Manchester, Manchester, United Kingdom, 3Maastro Clinic - GROW School for Oncology and Developmental Biology - Maastricht University Medical Centre, Maastricht, Netherlands, 4KU Leuven – University of Leuven, Leuven, Belgium, 5Department of Radiation Oncology, University Hospitals Leuven, Gasthuisberg, Belgium, 6Department of Radiation Oncology, Amsterdam UMC, Amsterdam, Netherlands, 7Dept of Clinical Medicine, Faculty of Health Sciences, University of Copenhagen, Copenhagen, Denmark, 8Copenhagen University Hospital - Rigshospitalet, Copenhagen, Denmark, 9Department of Radiation Oncology, The Netherlands Cancer Institute-Antoni van Leeuwenhoek, Amsterdam, Netherlands, 10Maastro, Maastricht University Medical Center- School for Oncology and Developmental Biology Grow, Maastricht, Netherlands, 11Department of Radiation Oncology, The Netherlands Cancer Institute, Amsterdam, Netherlands
Purpose/Objective(s): For patients with stage II-III non-small cell lung cancer (NSCLC), fatal pulmonary hemorrhage (FPH) is rare (~1%) after conventional RT. However, the risk may increase with hypofractionated dose escalation, mirroring patterns observed in stereotactic radiotherapy for ultracentral tumours. This study aims to identify risk factors for FPH and investigates the potential dose-response relationship in patients receiving moderately hypofractionated (chemo)radiotherapy for stage II–III NSCLC.
Materials/Methods: To boost statistical power, data from 2 studies were combined. The randomized ARTFORCE PET-Boost trial (NCT01024829) (“PB cohort”) included patients who received personalized dose-escalation, targeting either the primary tumor as a whole or its FDG-avid subvolume. Median fraction dose was 3.3 Gy (IQR 3.2-3.4 Gy) and 3.5 Gy (IQR 3.4 -3.8 Gy), for 24 fractions, respectively. Involved lymph nodes received 24 x 2.75 Gy. In the retrospective “LN cohort” patients with node-positive NSCLC received 2.75 Gy to the primary tumor, while involved lymph nodes received either 2.75 Gy or 2.42 Gy over 24 fractions.
Hazard ratio’s (HR) and 95%CIs were calculated using Fine & Gray modelling, considering non-FPH death as a competing risk. Variables included histology, chemotherapy regimen (concurrent vs. sequential or none), and local recurrence. Pre-treatment cavitation, proximal bronchial tree (PBT) compression (narrowing), and =180° encasement of any large vessel were visually assessed. Dose metrics (in EQD2 a/b=3Gy) for the bronchi, PBT, vena cava superior, and pulmonary artery and vein included the maximum dose to 0.1 cm3 (D0.1) and 5 cm3 (D5), as well as Vx (cm3) metrics (50/60/70 Gy).Results: The current analysis included 398 patients (PB n=103; LN n=295). Median follow-up was 37 m (IQR 27-58). FPH occurred in 6 (5.8%) and 4 (1.4%) patients in the PB and LN cohorts, respectively. Of these 10, 3 had a local recurrence. Median time to FPH was 8 months (IQR 3-13). Squamous cell ca. (HR 6.88 CI 1.5-32), PBT compression (HR 8.9 CI 2.3-34.1), =180° vessel encasement (HR 5.0 CI 1.4-17.7), bronchi D5 (HR 1.06 CI 1.0-1.12), PBT V50 (HR 1.07 CI 1.01-1.14), and pulmonary vein V50 (HR 1.12 CI 1.03-1.23) and V60 (HR 1.14 CI 1.02 -1.28) were significantly associated with FPH (p<0.05). No statistically significant association was found for cavitation, chemotherapy, local recurrence or D0.1 for any of the assessed broncho-vascular structures.
Conclusion: Tumor pathology, PBT compression and vessel encasement may be more critical determinants of FPH than high maximum doses to broncho-vascular structures in patients with stage II-III NSCLC receiving moderately hypofractionated (chemo)radiotherapy. However, the association with bronchi D5, PBT V50, and pulmonary vein V50 and V60 suggest a relevant dose-volume effect. These findings may help refine patient selection for hypofractionated treatment.