K. M. Atkins1, K. Silos1, O. Peony1, A. Singh1, C. E. Kehayias2, M. Bakhtiar3, S. C. Zhang1, J. He2, D. E. Kozono2, A. Nohria4, A. Nikolova5, C. V. Guthier2, and R. H. Mak2; 1Department of Radiation Oncology, Cedars-Sinai Medical Center, Los Angeles, CA, 2Department of Radiation Oncology, Dana-Farber Cancer Institute/Brigham and Women’s Hospital, Harvard Medical School, Boston, MA, 3Harvard Radiation Oncology Program, Boston, MA, 4Department of Cardiovascular Medicine, Dana-Farber Cancer Institute and Brigham and Women’s Hospital, Boston, MA, 5Department of Cardiology, Cedars-Sinai Medical Center, Los Angeles, CA
Purpose/Objective(s): Thoracic radiotherapy (RT) can result in pulmonary vascular remodeling and pulmonary fibrosis (PF) that may increase the risk of pulmonary hypertension (pHTN), a clinically significant and morbid condition. However, the incidence and RT dose volume predictors associated with pHTN are not well understood and are studied herein. Materials/
Methods: Multi-institutional retrospective analysis of 850 patients with locally advanced non-small cell lung cancer (LA-NSCLC) treated with RT from 2003-2014 (testing cohort, n=748) and 2005-2021 (validation cohort, n=102) with manually segmented The pulmonary arteries (PA) and more distal pulmonary vasculature were separately segmented using open-source deep learning-based algorithms and comprehensive osimetric parameters collected. Any grade pHTN was defined as = minimal dyspnea and elevated pulmonary artery systolic pressure on echocardiography (=35 mmHg) or mean pulmonary arterial pressure on right heart catheterization (=20 mmHg). We performed area under the receiver operating curve (AUC) and dosimetric cut-point analyses and Cox regression, adjusting for chronic obstructive pulmonary disease (COPD)/PF and other cardiopulmonary risk factors. Results: Of 748 patients in the testing cohort, the median age was 65 years (interquartile range [IQR], 57-73), 51% were men, 35.8% had coronary heart disease, 27.8% had COPD, and 1.4% PF. The median follow-up was 1.9 years (IQR, 0.8-3.8). The median time to pHTN was 20 months (IQR, 9-44) with a 2-year cumulative incidence of 13.1% (grade 1=101, grade 2=13, grade 3=9). The structure with the highest AUC predicting pHTN was the PA, for which the optimal cutpoint for the top dose variable was V10Gy =55 cc (0.58) overall, and V25Gy =46 cc (0.56) in those without COPD/PF. Lung and distal lung vessel parameters were not predictive (C-index <0.51). The 2-year pHTN incidences by V10Gy =55 cc vs <55 cc was 17.4% vs 9.4% (p=.0003). Adjusting for baseline COPD/PF and other factors, there was an increased risk of pHTN with PA V10Gy =55 cc (adjusted hazard ratio [aHR] 1.73; 95% confidence interval [CI] 1.15-2.61; p=.009). Similarly, among COPD/PF negative patients, there was an increased risk of pHTN with V25Gy =46 cc (aHR 1.70; 95%CI 1.07-2.70; p=.025). In the validation cohort, the 2-year pHTN incidence was 25.1%. Adjusting for baseline COPD/PF, PA V10Gy was associated with increased risk of pHTN (aHR 1.02/cc, 95%CI 1.00-1.04; p=.041). Conclusion: pHTN following thoracic RT was common, occurring in 13-25% of patients by 2-years, though most commonly low-grade. Pulmonary artery (but not lung or distal lung vessels) RT dose was an independent predictor of pHTN, revealing a distinct pathophysiological process from the previously presumed pathway of RT-induced pulmonary fibrosis leading to pHTN.