Radiation pneumonitis in lung cancer patients: a retrospective study of risk factors and the long-term prognosis.

Purpose: To retrospectively evaluate the risk factors for acute radiation pneumonitis (RP) and long-term prognosis of patients with lung cancer treated by thoracic radiotherapy.
Methods and Materials: Of the 256 lung cancer patients who underwent definitive thoracic radiotherapy between June 1988 and May 1998, the 191 patients who were capable of being evaluated were divided into three groups according to the grade of RP. RP was defined as "severe," when it caused severe clinical symptoms, such as intractable cough, dyspnea at rest, and the need for oxygen or steroid therapy. The definition was made by using a modification of the Radiation Therapy Oncology Group and the European Organization for Research and Treatment of Cancer acute radiation morbidity scoring criteria. Factors that influenced the incidence of severe RP were assessed by using the Mantel-Haenszel chi(2) test in the univariate analysis and the logistic regression test in the multivariate analysis. Survival rates was calculated by using the Kaplan-Meier method, and the p values indicating the significance of differences between the RP groups were calculated by the log-rank test.
Results: Of the 94 patients (49%) who experienced clinical RP in this study, the RP was mild in 69 (36%) and severe in 25 (13%) patients. The 3-year survival rates of the patients who experienced no, mild, and severe RP were 33.4%, 38.2%, and and 0%, respectively, and the survival rate of the patients who experienced severe RP was significantly poorer than the other two groups combined (p = 0.0028). The incidence of severe RP did not correlate with any of the baseline patient characteristics, radiotherapeutic factors, or chemotherapeutic variables. Two clinical risk factors were identified from medical records before radiotherapy: low PaO2 (< 80 torr) and high C-reactive protein (CRP) (> 1.0 ng/mL). Both of them were significantly related to the development of severe RP in the univariate analysis (p = 0.004 and 0.013, respectively), and low PaO2 remained a significant risk factor in the multivariate analysis (p = 0.034). Multivariate analysis also revealed the occurrence of severe RP to be the most important factor determining poor survival (p = 0.0065). There was no significant difference in survival rate according to whether the patients had been treated with corticosteroids.
Conclusion: Mild and severe RP occurred in 69 (36%) and 25 (13%), respectively, of 191 lung cancer patients who had undergone irradiation of the chest. Only severe RP was an adverse prognostic factor. Low PaO2 (< 80 torr) before radiotherapy was a significant risk factor predictive of severe RP. The role of corticosteroids in RP could not be accurately determined.