BACKGROUND: Radiation pneumonitis (RP) is a major pulmonary adverse event of chest radiotherapy. The PACIFIC trial that identified durvalumab as an effective subsequent-line therapy after concurrent chemoradiotherapy (CCRT) found that patients with grade 2 or higher RP may have to be excluded from treatment under certain criteria. The purpose of this study was to investigate the relationship between grade ≥2 RP and the parameters of dose-volume histograms after CCRT with cisplatin/docetaxel for stage III non-small cell lung cancer and conduct a subset analysis of severe RP that can lead to the permanent discontinuation of treatment per the PACIFIC trial criteria to help determine treatment strategy. METHODS: We calculated the percentage of the lung volume received at least 5 Gy (V5) and 20 Gy (V20), the mean lung dose (MLD), and the lung volume spared from a 5 Gy dose (VS5) to the total lung volume. Factors affecting the incidence of grade ≥2 RP were identified; severe RP was defined as grade ≥3 as well as grade 2 RP that required ≥10 mg prednisolone for at least 12 weeks. RESULTS: This study included 45 patients. On univariate analysis, all parameters and total lung volume were found to be significant predictors of grade ≥2 RP (P = .001, .003, .03, .004, and .02, respectively). On multivariate analysis, V20 was a significant predictive factor of grade ≥2 RP (P = .007). Severe RP developed in 6 of 37 patients (16.2%) whose V20 values were 35% or lower. On univariate analysis, only V20 was a significant predictor of severe RP in these patients (P = .01). CONCLUSIONS: The best approach to reduce the rate of grade ≥2 RP is to maintain the V5, V20, MLD, and VS5 as low as possible during radiotherapy planning in patients receiving definitive CCRT with cisplatin/docetaxel.