Radiomics model based on pretreatment PET-CT imaging to predict response to sequential combination therapy of induction chemotherapy and radiotherapy in lung cancer: a multicenter retrospective study.

Abstract

This study aimed to develop and validate a radiomics model based on pretreatment Fluorine-18 Fluorodeoxyglucose (18F-FDG) Positron Emission Tomography-Computed Tomography (PET-CT) imaging to predict response to sequential combination therapy (induction chemotherapy followed by radio-therapy) in patients with lung cancer. A total of 180 patients with lung cancer from multiple centers who underwent pretreatment PET-CT imaging followed by induction chemotherapy and subsequent radiotherapy were retrospectively en-rolled. Patients were randomly assigned to two cohorts: training (n = 126) and validation (n = 54). Radiomics features were extracted from PET-CT images, and feature selection was performed using the least absolute shrinkage and selection operator (LASSO) regression. Multiple machine learning algorithms, including support vector machines (SVMs), random forests (RFs), and logistic regression, were employed to construct predictive models. Model performance was evalu-ated using receiver operating characteristic (ROC) curves, area under the curve (AUC), accuracy, sensitivity, and specificity. The dataset comprised 129 male (71.67%) and 51 female (28.33%) patients, with a mean maximum standardized uptake value (SUVmax) of 9.10 ± 3.80. Feature selection identified 12 radiomics features significantly associated with treatment response. The combined PET-CT radiomics model achieved the best performance, with an AUC of 0.891 (95% CI: 0.835-0.947) in the training cohort and 0.851 (95% CI: 0.751-0.943) in the vali-dation cohort. The model demonstrated superior predictive capability compared to traditional clinical parameters alone (SUVmax AUC: 0.683). The pretreatment PET-CT-based radiomics model demonstrated robust performance in predicting response to sequential combination therapy (induction chemotherapy followed by radiotherapy) in patients with lung cancer.