Related Experiment Video
Updated: Jun 30, 2026
Intramucosal Inoculation of Squamous Cell Carcinoma Cells in Mice for Tumor Immune Profiling and Treatment Response Assessment
Published on: April 22, 2019
Integration of Delta-Radiomics Features and Inflammatory Biomarkers for Neoadjuvant Chemo-Immunotherapy Efficacy in
Benjie Xu1, Jiahao Zhu1, Tiantian Fan2
1Department of Outpatient Chemotherapy, Harbin Medical University Cancer Hospital, Harbin, Heilongjiang, 150000, People's Republic of China.
Purpose:
Neoadjuvant chemo-immunotherapy (NACI) significantly increases the pathological complete response (pCR) rate in locally advanced esophageal squamous cell carcinoma (LA-ESCC). However, not all patients benefit from this approach. This study aims to construct a multi-omics model for predicting the efficacy of NACI in LA-ESCC patients by integrating delta-radiomics features (delta-RFs) and the dynamic changes in inflammatory biomarkers.
Patients And Methods:
Our study was the first to develop a multi-omics model based on dynamic CT imaging features and inflammatory indices in patients with LA-ESCC. A total of 217 patients were divided into training (n = 152) and validation (n = 65) cohorts. Following the completion of the data standardization process, delta-RFs were extracted using the PyRadiomics, representing the relative imaging changes between pre- and post-treatment. The algorithm known as the least absolute shrinkage and selection operator (LASSO) was selected to pinpoint the most significant RFs for predicting pCR. The imaging signatures and clinical characteristics were integrated by the logistic regression analysis. Finally, a nomogram was constructed based on the identified independent predictors.
Results:
The pCR rates were approximately 30% among the enrolled patients. A total of 1834 delta-RFs were extracted. The 12 delta-RFs signature was determined by the optimal regularization parameter using the LASSO algorithm. Furthermore, multivariate regression analyses confirmed that tumor length, delta-RFs signature, and delta-systemic immune-inflammation index (SII) served as independent predictors for pCR prediction. Finally, a nomogram was constructed and the multi-omics model exhibited impressive predictive performance, with area under curve (AUC) values of 0.853 and 0.796 for the training and validation cohorts, respectively. This method aims to support informed treatment decisions and facilitate the development of personalized treatment strategies.
Conclusion:
The multi-omics model, which is based on the delta-RFs signature and delta-SII indicators prior to surgery, effectively predicts the NACI treatment response in patients with LA-ESCC.
