Dose-Registered CT assessment of radiation-induced lung injury after chemoradiotherapy for stage III NSCLC
- Journal
- Annals of medicine (Q1)
- Published
- 10 September 2026
- Study design
- Retrospective cohort
- Evidence level
- Level 4, Very Low (CEBM 4)
- Authors
- Tomohiro Itonaga, Ryuji Mikami, Tatsuhiko Zama, Mitsuru Okubo, Shinji Sugahara, Norihiko Ikeda, et al.
- PMID
- 42723268
- DOI
- 10.1080/07853890.2026.2721850
Why clinicians should know about it
- Picked for Medical Physics (paper of the day, 12 September 2026): Dose‑registered CT analysis of radiation‑induced lung injury
Abstract
AIMS: To quantify radiation-induced lung injury (RILI) on deformably registered CT images after chemoradiotherapy (CCRT) for stage III NSCLC, and to evaluate spatial and dosimetric associations between RILI and planned dose distribution in a pre-durvalumab cohort. METHODS: This retrospective study included 58 patients with stage III NSCLC treated with CCRT using three-dimensional conformal radiotherapy (3D-CRT) prior to the era of consolidation immunotherapy. Follow-up CT images at 6 months (acute) and 12 months (late) were registered to treatment planning CT using deformable image registration (DIR). RILI was quantified by segmented volume (mL) and isodose-line overlap (%). Dosimetric parameters (V5Gy-V60Gy, mean lung dose) and symptomatic RILI (CTCAE grade ≥2) were analysed using correlation, ROC, and logistic regression methods. RESULTS: Symptomatic RILI occurred in 14 patients (24.1%). RILI volume decreased by 27.7% from the acute to the late phase, with moderate-to-strong dose-volume correlations in both phases (Spearman ρ = 0.69-0.76). Symptomatic patients showed broader spatial spread into lower-dose regions (median isodose-line overlap 20% vs 40%; p < 0.001). Acute-phase RILI reaching the ≤25% isodose line (≤16.5 Gy) was an exploratory imaging marker in this cohort (AUC 0.81) that remained associated with symptomatic RILI on multivariable analysis (p < 0.01). CONCLUSIONS: Dose-registered CT analysis demonstrated consistent spatial associations between planned dose and RILI, providing a hypothesis-generating dosimetric-radiographic reference dataset that requires external validation in IMRT/VMAT- and durvalumab-treated cohorts before clinical application.
Abstract as published, via PubMed.
For healthcare professionals. The summary is generated by AI from the published abstract, and the evidence level is assigned automatically from the study design on the Oxford CEBM hierarchy. Neither is medical advice. Read the full paper before changing practice.