Related Experiment Video
Updated: Jun 18, 2026
Development of Compendium for Esophageal Squamous Cell Carcinoma
Published on: April 12, 2024
Development and validation of a collagen score-based prediction model for neoadjuvant chemoimmunotherapy efficacy in
Xiao-Feng Chen1,2,3, Huai-Yuan Zhang1,2,3, Wei-Jie Chen1,2,3
1Department of Thoracic Oncology Surgery, Clinical Oncology School of Fujian Medical University, Fujian Cancer Hospital, Fuzhou, China.
Background:
The relationship between collagen features (CFs) in the tumor microenvironment (TME) of esophageal cancer (EC) and the therapeutic response to neoadjuvant chemoimmunotherapy (nCIT) remains unclear. This study aimed to develop and validate a novel model based on CFs to predict the therapeutic response of patients with locally advanced EC to nCIT.
Methods:
In this retrospective study, 79 newly diagnosed patients with resectable locally advanced EC were enrolled and randomly divided into a training cohort (54 patients) and a validation cohort (25 patients) at a 7:3 ratio. Multiphoton imaging (MPI) was used to quantitatively analyze collagen in pretreatment biopsy tissues. Least absolute shrinkage and selection operator (LASSO) regression was applied to select significant CFs and calculate the collagen score. Univariate and multivariate analyses were performed to verify the predictive value of the collagen score for a good pathological response (GR).
Results:
LASSO regression identified collagen-related parameters in EC biopsy specimens: F1 (collagen percentage area), F2 (fiber density), F8 (mean of histograms), F9 (mean of histograms), F102 (Gabor_variance_90°_1 scale), F117 (Gabor_mean_150°_2 scale), and F138 (Gabor_variance_90°_4 scale), on the basis of which a formula was established to calculate the collagen score. Univariate analysis revealed a significant difference in the collagen score between the GR and non-GR groups (P<0.001), and multivariate analysis confirmed that the collagen score was an independent risk factor (P=0.001). These results were validated in the validation cohort (P=0.009 for univariate analysis, P=0.02 for multivariate analysis). For predicting the GR, the area under the curve (AUC) values of the collagen score in the experimental and validation cohorts were 0.819 [95% confidence interval (CI): 0.704-0.935] and 0.820 (95% CI: 0.645-0.995), respectively.
Conclusions:
MPI can be used to quantify the TME of EC on the basis of a collagen score, which has a certain ability to predict the efficacy of nCIT.
More Related Videos
13:34A Combined 3D Tissue Engineered In Vitro/In Silico Lung Tumor Model for Predicting Drug Effectiveness in Specific Mutational Backgrounds
Published on: April 6, 2016
07:29Intramucosal Inoculation of Squamous Cell Carcinoma Cells in Mice for Tumor Immune Profiling and Treatment Response Assessment
Published on: April 22, 2019