Predicting severe acute radiation pneumonitis in patients with non-small cell lung cancer receiving postoperative radiotherapy: Development and internal validation of a nomogram based on the clinical and dose-volume histogram parameters

Background and purpose: Postoperative radiotherapy (PORT) can potentially lead to radiation pneumonitis. We aim to develop a nomogram predicting the severe acute radiation pneumonitis (SARP, grade >= 3) in patients with non-small cell lung cancer (NSCLC) receiving PORT. Materials and methods: Clinical and dose-volume histogram (DVH) factors were collected from 109 patients between 2006 and 2017. The endpoint was the development of SARP within 3 months after PORT. Logistic regression was used to evaluate the prognostic value of each factor in predicting SARP. Nomogram was generated based on multivariate regression coefficients. Area under the ROC curve (AUC), calibration curves, and decision curve analyses (DCA) were conducted to validate the model. Results: Univariate and multivariate analysis indicated that total lung mean dose (tlMD) (OR: 1.003, 95% CI: 1.001-1.006, p = 0.013), percentage of ipsilateral lung volume receiving >= 5 Gy (ilV5) (OR: 1.084, 95% CI: 1.020-1.151, p = 0.009), and concurrent chemoradiotherapy (CCRT) (OR: 4.091, 95% CI: 1.331-12.572, p = 0.014) were independent prognosticators of SARP and were included in the nomogram. ROC curves revealed the AUC of the nomogram was 0.842, which was much higher than any factor alone (tlMD: 0.769; ilV5: 0.744; CCRT: 0.661). Calibration curves showed favorable consistency between the predicted SARP and the actual observation. DCA showed satisfactory positive net benefits of the model among most of the threshold probabilities, indicating great clinical effect. Conclusion: We identified that the tlMD (>10.8 Gy), ilV5 (>64.9%), and CCRT could predict SARP among patients with NSCLC receiving PORT. Combining clinical and DVH parameters, a nomogram was first built and validated, showing its potential value in practice. (C) 2018 Elsevier B.V. All rights reserved.