Association of pulmonary artery radiation with pulmonary hypertension after lung cancer radiotherapy
Katelyn M Atkins, Samuel C Zhang, Christopher E Kehayias, Christian Guthier, Katrina D Silos, Olivia Peony, Andrew Warrington, Mina Bakhtiar, Maria Oorloff, Cosmin Ciausu, Asneh Singh, John He, David E Kozono, David C Christiani, Cody Ramin, Anju Nohria, Andriana P Nikolova, Aaron B Waxman, Raymond H MakBackground
Thoracic radiotherapy (RT) results in pulmonary fibrosis that has been hypothesised to increase the risk of pulmonary hypertension (PH); however, dose effects to cardiopulmonary structures predicting PH have not been systematically studied.
Methods
Multi-institutional retrospective cohort analysis of patients with non-small cell lung cancer treated with thoracic RT. The pulmonary arteries (PA), distal pulmonary vasculature, lungs and heart were segmented. Radiotherapy dose parameters (mean, maximum and volume (V) receiving X Gy in 5 Gy increments) were extracted. Area under the receiver operating curve (AUROC) analyses estimating PH and Fine-Gray regressions were performed.
Results
Of 848 patients included (n=746 discovery; n=102 validation), 49.9% were women and the median age was 65 years (IQR, 58–73 years). The 2-year cumulative incidence of PH was 11.6%. PA V10 Gy demonstrated the highest AUROC for predicting PH (0.58) and was significantly associated with the risk of PH on multivariable regression adjusting for cardiopulmonary risk factors in the discovery (subdistribution HR (sHR) 1.01; 95% CI 1.00 to 1.02; p=0.013) and validation (sHR 1.03, 95% CI 1.00 to 1.05; p=0.019) cohorts. In those who developed PH after RT (vs PH-negative), moderate or greater tricuspid disease and right ventricular systolic dysfunction (RVSD) were more common (18.7% vs 0.2% (p<0.001) and 3.3% vs 0.2% (p=0.003), respectively).
Conclusions
PH was common following thoracic RT and associated with PA radiation dose. Patients with PH were more likely to harbour tricuspid disease and RVSD. These hypothesis-generating findings are suggestive that PH following thoracic RT may occur through a pathophysiological process distinct from RT-induced pulmonary fibrosis.