Dose to High Ventilating Lung Predicts Radiation-Induced Pneumonitis in Lung Cancer Patients

Abstract

<h3>Purpose/Objective(s)</h3> Current lung cancer radiotherapy (RT) planning does not account for regional variation in pulmonary function. Functional avoidance RT is being studied as a technique to limit pulmonary toxicity. There is limited data on useful dose parameters for avoiding high ventilating lung. Our objective was to identify metrics of radiation dose delivered to high ventilating lung predictive of radiation-induced pneumonitis. <h3>Materials/Methods</h3> A combined prospective and retrospective cohort of 108 patients with locally advanced non-small cell lung cancer treated with standard fractionated RT of 60-66 Gy in 30-33 fractions were evaluated. Regional lung ventilation was determined from pre-RT 4-dimensional computed tomography (4DCT) using the Jacobian determinant of a B-spline deformable image registration to estimate lung tissue expansion during respiration. Lung voxels with greater expansion than the voxel-wise population median expansion were defined as high ventilating. Mean dose and volumes receiving dose over thresholds of 5-60 Gy were analyzed for both total lung-ITV (MLD, V5-V60) and high ventilating functional lung-ITV (fMLD, fV5-fV60). The primary endpoint was symptomatic grade 2+ (G2+) pneumonitis. Differences in dose metrics were compared and receiver operator curve (ROC) analyses were used to identify predictors of pneumonitis. <h3>Results</h3> G2+ pneumonitis occurred in 20.4% of patients. There were no differences between stage, smoking, COPD or chemo or immunotherapy use between G<2 or G2+ pneumonitis patients. Total lung median MLD, V5 Gy and V20 Gy were 15.4 Gy, 58.0% and 25.7%, respectively. Each metric was significantly different between patients with G<2 and G2+ pneumonitis (p<0.016). Optimal ROC points predicting pneumonitis from total lung dose were MLD < 17 Gy, V5 < 59% and V20 < 25%. Regional ventilation calculated from 4DCTs had a population-wide median of 18% voxel-level expansion. The median amount of high ventilating (≥ 18% expansion) lung per patient was 52.4% ± 23.0%. All functional dose metrics (fMLD, fV5-fV60) were significantly different between G<2 and G2+ patients (p<0.04). Optimal ROC points of fMLD < 15 Gy, fV5 < 53% and fV20 < 19% of functional lung were predictive of G2+ pneumonitis. Patients with fMLD ≥ 15 Gy had a 41% risk of developing G2+ pneumonitis. Similarly, pneumonitis risk increased from 9% to 33% for fV5 ≥ 53% and 11% to 33% for fV20 ≥ 19%. Spared functional lung volume receiving < 5 Gy was significantly different between G<2 and G2+ pneumonitis patients (671 cc vs 935 cc, p=0.04). <h3>Conclusion</h3> This work produced dosimetric values for ventilating lung associated with symptomatic pneumonitis. Predictive functional lung dose cut-offs were lower than total lung metrics; indicating planning could focus on limiting dose to functional regions. These findings provide important metrics to be used in functional lung avoidance RT planning and designing clinical trials.