Assessing Radiation Pneumonitis Through Functional Lung Imaging: A Single Photon Emission Computed Tomography...

To develop models to assess the risk of symptomatic radiation pneumonitis (SRP) (Common Terminology Criteria for Adverse Events 4.03 grade ≥ 2) in lung cancer patients by utilizing single-photon emission computed tomography (SPECT) for functional lung volume identification and dosimetric analysis.

This retrospective study included 71 lung cancer patients who underwent SPECT before radiotherapy from 2018 to 2024. Perfusion and ventilation SPECT images were co-registered with planning CT to define functional and anatomical lung volumes. Functional lung was defined as voxels with ≥ 20% of the maximum intensity on SPECT. Models to assess the risk of SRP were constructed using Cox regression and evaluated using corrected Akaike Information Criterion (AICc) and time-dependent receiver operating characteristic analysis.

At a median follow-up of 16.8 months, 19 of 71 patients (26.8%) developed SRP. Factors significantly associated with SRP risk included planning target volume ≥ 150 mL, percentage of total perfusion-defined functional lung receiving ≥ 10 Gy (pVf10) exceeding that of total anatomical lung receiving ≥ 10 Gy (V10), percentage of total ventilation-defined lung receiving ≥ 10 Gy (vVf10) ≥ 45%, and ipsilateral vVf10 ≥ 60% (p=0.004, 0.004, 0.024, and 0.007, respectively). Among the three models, the model incorporating additional ventilation-based parameters demonstrated the best performance (AICc = 85.81, area under the curve = 0.819).

SPECT-based dosimetric parameters derived from perfusion and ventilation are significantly associated with the risk of SRP. Incorporating SPECT may improve risk stratification and enable lung-sparing strategies.