Functional and biologic metrics for predicting radiation pneumonitis in locally advanced non-small cell lung cancer patients treated with chemoradiotherapy

To study the predictive value of functional and biologic metrics for predicting radiation pneumonitis (RP) in locally advanced non-small cell lung cancer (LANSCLC) patients treated with chemoradiotherapy. Between March 2006 and April 2010, 57 LANSCLC patients were enrolled in a prospective study. Fusion of SPECT and computed tomography scans provides perfusion-weighted functional dose-volume histogram (DVH) and associated functional dosimetric parameters. Blood for serum biomarkers-interleukin-6 (IL-6), transforming growth factor-beta1, and superoxide dismutase (SOD)-was drawn pre-RT and then 40 Gy/4 weeks during the treatment. The incidence of RP was related to the functional and biologic metrics. The predictability of predictors was calculated and compared based on the area under receiver-operating characteristic (ROC) curve (AUC). Relative volumes of functional lung receiving more than a threshold dose of 5-50 Gy at increments of 5 Gy and elevated levels of serum SOD after delivery of 40 Gy/4 weeks were associated with RP (p < 0.05). The best predictive efficacy of SOD was observed for a cutoff value of 56 U/ml, with a sensitivity of 0.80 (95 % CI 0.28-0.99) and a specificity of 0.67 (95 % CI 0.43-0.65) (p = 0.040). Functional DVH provided better predictive outcome (AUC 0.76-0.98) than standard DVH (AUC 0.62-0.86) for patients with poor baseline lung function. Functional metrics were identified to be better predictors for RP in patients with poor baseline lung function. SOD seemed to be a potential predictor for RP; however, it will need to be further verified using a larger sample size.