Background: Thoracic reirradiation (re-RT) is increasingly administered. However, radiation pneumonitis (RP) remains to be the most common side effect from retreatment. This study aimed to determine the risk and predictors for severe RP in patients receiving thoracic re-RT.
Methods: Sixty seven patients with lung cancer received thoracic re-RT for recurrent or metastatic disease. Three-dimensional conformal radiotherapy (3D-CRT)/intensity modulated radiotherapy (IMRT) was used for 60 patients, and stereotactic body radiation therapy (SBRT) was used in 7 patients. Deformable image registration (DIR) was performed to create a composite plan. Severe (grade ≥ 3) RP was graded according to Common Terminology Criteria for Adverse Events version 4.0.
Results: Eighteen patients (26.9%) developed grade ≥ 3 RP (17 of grade 3, and 1 of grade 4). In univariate analyses, V5 and mean lung dose (MLD) of initial RT or re-RT plans, V5 and V20 of composite plans, and the overlap between V5 of initial RT and V5 of re-RT plans/V5 of re-RT plans (overlap-V5/re-V5) were significantly associated with grade ≥ 3 RP (P < 0.05 for each comparison). Multivariate analysis revealed that MLD of the initial RT plans (HR = 14.515, 95%CI:1.778-118.494, P = 0.013), V5 of the composite plans (HR = 7.398, 95%CI:1.319-41.495, P = 0.023), and overlap-V5/re-V5 (P = 0.041) were independent predictors for grade ≥ 3 RP. Out-of-field failures with medium overlap-V5/re-V5 of 0.4-0.8 was associated with higher risk of grade ≥ 3 RP compared with in-field failures (18.3% vs. 50%, P = 0.014).
Conclusions: The risk of grade ≥ 3 RP could be predicted not only by dose-volume variables from re-RT plan, but also by some from initial-RT and composite plans. Out-of-field failures was associated with higher risk of severe RP compared with in-field failures in some cases.
Keywords: Lung cancer; Predictors; Radiation pneumonitis; Reirradiation.