Habitat radiomics combined with clinical features for predicting tertiary lymphoid structure status in pancreatic ductal adenocarcinoma: development and validation of a preoperative CECT-based model.
- 1Department of Radiology, Southwest Hospital, Army Medical University (Third Military Medical University), Chongqing, China.
- 2Department of Medical Imaging, Sichuan Corps Hospital of Chinese People's Armed Police Force, Leshan, China.
- 3Institute of Hepatobiliary Surgery, Southwest Hospital, Army Medical University (Third Military Medical University), Chongqing, China.
Abstract
Tertiary lymphoid structures (TLS) are emerging prognostic biomarkers in pancreatic ductal adenocarcinoma (PDAC), yet no reliable non-invasive tool exists for their preoperative assessment. This study aimed to develop and validate a contrast-enhanced computed tomography (CECT)-based habitat radiomics model for preoperative prediction of intratumoral TLS status in PDAC. In this retrospective study, 204 surgically resected PDAC patients were enrolled and allocated into training (n=142) and validation (n=62) cohorts. TLS status was determined by fluorescent multiplex immunohistochemistry (mIHC), with positive cases subclassified as primary or secondary follicle-like TLS. K-means clustering partitioned the tumor volume of interest (VOI) into spatially distinct subregions; radiomics features were extracted from each subregion and selected via least absolute shrinkage and selection operator (LASSO) regression. Five machine learning classifiers were evaluated. Independent clinical predictors were identified by multivariate logistic regression (LR) and integrated with the optimal imaging model to construct a combined model. TLS positivity was observed in 32.84% of patients. The LR-based habitat model achieved area under the curve (AUC) of 0.939 and 0.873 in the training and validation cohorts, substantially outperforming the conventional radiomics model (AUCs: 0.815 and 0.701). Tumor differentiation and neural invasion were identified as independent clinical predictors. The combined model yielded the highest AUCs of 0.956 and 0.881, with favorable calibration and superior net clinical benefit on decision curve analysis (DCA). CECT-based habitat radiomics enables robust, non-invasive preoperative prediction of intratumoral TLS in PDAC, offering a practical tool for individualized risk stratification and treatment planning; incorporation of clinical variables yielded numerically higher AUCs but without statistically significant improvement.
Related
- Development and validation of machine learning models for predicting pancreatic injury observed after tigecycline initiation: a retrospective real-world study
- Development and internal validation of a clinical-MRI nomogram for predicting lymph node metastasis in rectal adenocarcinoma.
- Development and Internal Validation of the BRICS Score: A Preoperative Clinical and Hematological Model for Predicting High-Grade and Muscle-Invasive Bladder Cancer
- Development and validation of an interpretable clinical-radiomics model for predicting 3-year postoperative recurrence in clinical stage I lung adenocarcinoma: a multicentre study.
- Radiologic–pathologic discrepancy in the tumor size assessment of pancreatic ductal adenocarcinoma in the era of preoperative therapy
BibTeX
@article{li2026habitat,
title = {{Habitat radiomics combined with clinical features for predicting tertiary lymphoid structure status in pancreatic ductal adenocarcinoma: development and validation of a preoperative CECT-based model.}},
author = {Ke Li and Rui Guo and Xianqi Wang and Lian Li and Fengwei Yu and Zilong Li and Taotao Yang and Jiafei Chen and Wei Chen},
journal = {Gland surgery},
year = {2026},
month = sep,
doi = {10.21037/gs-2026-0439},
url = {https://europepmc.org/article/MED/42996878},
note = {CC BY-NC-ND},
}
RIS
TY - JOUR
TI - Habitat radiomics combined with clinical features for predicting tertiary lymphoid structure status in pancreatic ductal adenocarcinoma: development and validation of a preoperative CECT-based model.
AU - Ke Li
AU - Rui Guo
AU - Xianqi Wang
AU - Lian Li
AU - Fengwei Yu
AU - Zilong Li
AU - Taotao Yang
AU - Jiafei Chen
AU - Wei Chen
PY - 2026
DA - 2026/09/28
JO - Gland surgery
DO - 10.21037/gs-2026-0439
UR - https://europepmc.org/article/MED/42996878
AB - Tertiary lymphoid structures (TLS) are emerging prognostic biomarkers in pancreatic ductal adenocarcinoma (PDAC), yet no reliable non-invasive tool exists for their preoperative assessment. This study aimed to develop and validate a contrast-enhanced computed tomography (CECT)-based habitat radiomics model for preoperative prediction of intratumoral TLS status in PDAC. In this retrospective study, 204 surgically resected PDAC patients were enrolled and allocated into training (n=142) and validation (n=62) cohorts. TLS status was determined by fluorescent multiplex immunohistochemistry (mIHC), with positive cases subclassified as primary or secondary follicle-like TLS. K-means clustering partitioned the tumor volume of interest (VOI) into spatially distinct subregions; radiomics features were extracted from each subregion and selected via least absolute shrinkage and selection operator (LASSO) regression. Five machine learning classifiers were evaluated. Independent clinical predictors were identified by multivariate logistic regression (LR) and integrated with the optimal imaging model to construct a combined model. TLS positivity was observed in 32.84% of patients. The LR-based habitat model achieved area under the curve (AUC) of 0.939 and 0.873 in the training and validation cohorts, substantially outperforming the conventional radiomics model (AUCs: 0.815 and 0.701). Tumor differentiation and neural invasion were identified as independent clinical predictors. The combined model yielded the highest AUCs of 0.956 and 0.881, with favorable calibration and superior net clinical benefit on decision curve analysis (DCA). CECT-based habitat radiomics enables robust, non-invasive preoperative prediction of intratumoral TLS in PDAC, offering a practical tool for individualized risk stratification and treatment planning; incorporation of clinical variables yielded numerically higher AUCs but without statistically significant improvement.
SN - 2227-684X
SN - 2227-8575
ER -