Predictive factors of late radiation fibrosis: a prospective study in non-small cell lung cancer.

PURPOSE: To determine predictive factors of late radiation fibrosis (RF) after conformal radiotherapy (3D-RT) in non-small cell lung cancer (NSCLC). METHODS AND MATERIALS: Ninety-six patients with Stage IA-IIIB NSCLC were included in a prospective trial. Clinical evaluation, chest X-ray, and pulmonary functional tests including diffusion parameters were performed before and 6 months after radiotherapy. An independent panel of experts prospectively analyzed RF, using Late Effects in Normal Tissues-Subjective, Objective, Management and Analytic scales classification. Logistic regression analysis was performed to identify relationships between clinical, functional, or treatment parameters and incidence of RF. Variations of circulating serum levels of pro-inflammatory (interleukin-6, tumor necrosis factor alpha, tumor growth factor beta1) and anti-inflammatory (interleukin-10) cytokines during 3D-RT were examined to identify correlations with RF.
RESULTS: Of the 96 patients included, 72 were evaluable for RF at 6 months. Thirty-seven (51.4%) developed RF (Grade >or=1), including six severe RF (Grades 2-3; 8.3%). In univariate analysis, only poor Karnofsky Performance Status and previous acute radiation pneumonitis were associated with RF (p < 0.05). Dosimetric factors (mean lung dose, percentage of lung volume receiving more than 10, 20, 30, 40, and 50 Gy) were highly correlated with RF (p < 0.001). In multivariate analysis, previous acute radiation pneumonitis and dosimetric parameters were significantly correlated with RF occurrence. It was not significantly correlated either with cytokines at baseline or with their variation during 3D-RT.
CONCLUSIONS: This study confirms the importance of dosimetric parameters to limit the risk of RF. Contrary to acute radiation pneumonitis, RF was not correlated to cytokine variations during 3D-RT.