Functional dosimetric metrics for predicting radiation-induced lung injury in non-small cell lung cancer patients treated with chemoradiotherapy

BACKGROUND Radiation-induced lung injury (RILI) is an important dose-limiting toxicity during thoracic radiotherapy. The purpose of this study is to investigate single photon emission computed tomography (SPECT) perfusion-weighted functional dose-volume histogram (FDVH) for predicting RILI in non-small cell lung cancer (NSCLC) patients treated with definitive chemoradiotherapy. METHODS Fifty-seven locally advanced NSCLC patients receiving chemoradiotherapy were enrolled prospectively. Patients had treatment scans and dose calculations to provide a standard dose-volume histogram (DVH). Fusion of SPECT and computed tomography scans provided perfusion-weighted FDVH and associated functional dosimetric parameters (relative volumes of functional lung receiving more than a threshold dose of 5 - 60 Gy at increments of 5 Gy [FV5 - FV60]). The predictive abilities of FDVH and DVH were calculated and compared based on the area under receiver operating characteristic (ROC) curve (AUC). RESULTS The accumulative incidence of  ≥  2 grade RILI was 19.3% with a median follow-up of 12 months. Univariate analysis showed that the functional (FV5 - FV60) and standard (V5 - V40) parameters were associated with RILI (all value of p < 0.05). Close correlations between a variety of functional and standard parameters were found. By ROC curve analysis, functional metrics (AUCs were 0.784 - 0.869) provided similarly (p value 0.233 - 1.000) predictive outcome to standard metrics (AUCs were 0.695 - 0.902) in lower - median dose level parameters (FV5 - FV40). However, FDVH seemed to add some predictive value in higher dose level, the best statistical significance for comparing FV60 with V60 was 0.693 vs. 0.511 (p = 0.055). CONCLUSIONS Functional metrics are identified as reliable predictors for RILI, however, this observation still needs to be further verified using a larger sample size.