Prognostic value of interim CT-based peritumoral and intratumoral radiomics in laryngeal and hypopharyngeal cancer patients undergoing definitive radiotherapy

Background and purpose We aimed to investigate the prognostic value of peritumoral and intratumoral computed tomography (CT)-based radiomics during the course of radiotherapy (RT) in patients with laryngeal and hypopharyngeal cancer (LHC). Materials and methods A total of 92 eligible patients were 1:1 randomly assigned into training and validation cohorts. Pre-RT and mid-RT radiomic features were extracted from pre-treatment and interim CT. LASSO–Cox regression was used for feature selection and model construction. Time-dependent area under the receiver operating curve (AUC) analysis was applied to evaluate the models’ prognostic performances. Risk stratification ability on overall survival (OS) and progression-free survival (PFS) were assessed using the Kaplan–Meier method and Cox regression. The associations between radiomics and clinical parameters as well as circulating lymphocyte counts were also evaluated. Results The mid-RT peritumoral (AUC: 0.77) and intratumoral (AUC: 0.79) radiomic models yielded better performance for predicting OS than the pre-RT intratumoral model (AUC: 0.62) in validation cohort. This was confirmed by Kaplan-Meier analysis, in which risk stratification depended on the mid-RT peritumoral (p = 0.009) and intratumoral (p = 0.003) radiomics could be improved for OS, in comparison to the pre-RT intratumoral radiomics (p = 0.199). Multivariate analysis identified mid-RT peritumoral and intratumoral radiomic models as independent prognostic factors for both OS and PFS. Mid-RT peritumoral and intratumoral radiomics were correlated with treatment-related lymphopenia. Conclusion Mid-RT peritumoral and intratumoral radiomic models are promising image biomarkers that could have clinical utility for predicting OS and PFS in patients with LHC treated with RT.