Transfer learning radiomic model predicts intratumoral tertiary lymphoid structures in hepatocellular carcinoma: a multicenter study

Background Intratumoral tertiary lymphoid structures (iTLS) in hepatocellular carcinoma (HCC) are associated with improved survival and may influence treatment decisions. However, their non-invasive detection remains challenging in HCC. We aim to develop a non-invasive model using baseline contrast-...

Full description

Saved in:
Bibliographic Details
Main Authors: Juan Chen, Xiong Chen, Kai Fu, Feng Yi, Shichao Long, Mengsi Li, Linhui Zhong, Wenguang Liu, Deng Pan, Ganmian Dai, Wenzheng Li, Yue Ruan, Bocheng Zou
Format: Article
Language:English
Published: BMJ Publishing Group 2025-03-01
Series:Journal for ImmunoTherapy of Cancer
Online Access:https://jitc.bmj.com/content/13/3/e011126.full
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849709468311355392
author Juan Chen
Xiong Chen
Kai Fu
Feng Yi
Shichao Long
Mengsi Li
Linhui Zhong
Wenguang Liu
Deng Pan
Ganmian Dai
Wenzheng Li
Yue Ruan
Bocheng Zou
author_facet Juan Chen
Xiong Chen
Kai Fu
Feng Yi
Shichao Long
Mengsi Li
Linhui Zhong
Wenguang Liu
Deng Pan
Ganmian Dai
Wenzheng Li
Yue Ruan
Bocheng Zou
author_sort Juan Chen
collection DOAJ
description Background Intratumoral tertiary lymphoid structures (iTLS) in hepatocellular carcinoma (HCC) are associated with improved survival and may influence treatment decisions. However, their non-invasive detection remains challenging in HCC. We aim to develop a non-invasive model using baseline contrast-enhanced MRI to predict the iTLS status.Methods A total of 660 patients with HCC who underwent surgery were retrospectively recruited from four centers between October 2015 and January 2023 and divided into training, internal test, and external validation sets. After features dimensionality and selection, corresponding features were used to construct transfer learning radiomic (TLR) models for diagnosing iTLS, and model interpretability was explored with pathway analysis in The Cancer Genome Atlas-Liver HCC. The performances of models were assessed using the area under the receiver operating characteristic curve (AUC). The log-rank test was used to evaluate the prognostic value of the TLR model. The combination therapy set of 101 patients with advanced HCC treated with first-line anti-programmed death 1 or ligand 1 plus antiangiogenic treatment between January 2021 and January 2024 was used to investigate the value of the TLR model for evaluating the treatment response.Results The presence of iTLS was identified in 46.0% (n=308) patients. The TLR model demonstrated excellent performance in predicting the presence of iTLS in training (AUC=0.91, 95% CI: 0.87, 0.94), internal test (AUC=0.85, 95% CI: 0.77, 0.93) and external validation set (AUC=0.85, 95% CI: 0.81, 0.90). The TLR model-predicted iTLS group has favorable overall survival (HR=0.66; 95% CI: 0.48, 0.90; p=0.007) and relapse-free survival (HR=0.64; 95% CI: 0.48, 0.85; p=0.001) in the external validation set. The model-predicted iTLS status was associated with inflammatory response and specific tumor-associated signaling activation (all p<0.001). The proportion of treatment responders was significantly higher in the model-predicted group with iTLS than in the group without iTLS (36% vs 13.73%, p=0.009).Conclusion The TLR model has indicated accurate prediction of iTLS status, which may assist in the risk stratification for patients with HCC in clinical practice.
format Article
id doaj-art-c22eb4f5191e4209b459b84ed99356c3
institution DOAJ
issn 2051-1426
language English
publishDate 2025-03-01
publisher BMJ Publishing Group
record_format Article
series Journal for ImmunoTherapy of Cancer
spelling doaj-art-c22eb4f5191e4209b459b84ed99356c32025-08-20T03:15:17ZengBMJ Publishing GroupJournal for ImmunoTherapy of Cancer2051-14262025-03-0113310.1136/jitc-2024-011126Transfer learning radiomic model predicts intratumoral tertiary lymphoid structures in hepatocellular carcinoma: a multicenter studyJuan Chen0Xiong Chen1Kai Fu2Feng Yi3Shichao Long4Mengsi Li5Linhui Zhong6Wenguang Liu7Deng Pan8Ganmian Dai9Wenzheng Li10Yue Ruan11Bocheng Zou12Department of Radiology, and National Clinical Research Center for Geriatric Disorders, Xiangya Hospital Central South University, Changsha, Hunan, ChinaDepartment of Oncology, Mengchao Hepatobiliary Hospital of Fujian Medical University, Fuzhou, Fujian, ChinaInstitute of Molecular Precision Medicine and Hunan Key Laboratory of Molecular Precision Medicine, Xiangya Hospital Central South University Department of General Surgery, Changsha, Hunan, ChinaInstitute of Molecular Precision Medicine and Hunan Key Laboratory of Molecular Precision Medicine, Xiangya Hospital Central South University Department of General Surgery, Changsha, Hunan, ChinaDepartment of Radiology, and National Clinical Research Center for Geriatric Disorders, Xiangya Hospital Central South University, Changsha, Hunan, ChinaDepartment of Radiology, and National Clinical Research Center for Geriatric Disorders, Xiangya Hospital Central South University, Changsha, Hunan, ChinaDepartment of Radiology, and National Clinical Research Center for Geriatric Disorders, Xiangya Hospital Central South University, Changsha, Hunan, ChinaDepartment of Radiology, and National Clinical Research Center for Geriatric Disorders, Xiangya Hospital Central South University, Changsha, Hunan, ChinaDepartment of Nuclear Medicine, Hainan Cancer Hospital, Haikou, Hainan, ChinaDepartment of Radiology, The Second Affiliated Hospital of Hainan Medical University, Haikou, Hainan, ChinaDepartment of Radiology, and National Clinical Research Center for Geriatric Disorders, Xiangya Hospital Central South University, Changsha, Hunan, ChinaDepartment of Radiology, and National Clinical Research Center for Geriatric Disorders, Xiangya Hospital Central South University, Changsha, Hunan, ChinaDepartment of Radiology, and National Clinical Research Center for Geriatric Disorders, Xiangya Hospital Central South University, Changsha, Hunan, ChinaBackground Intratumoral tertiary lymphoid structures (iTLS) in hepatocellular carcinoma (HCC) are associated with improved survival and may influence treatment decisions. However, their non-invasive detection remains challenging in HCC. We aim to develop a non-invasive model using baseline contrast-enhanced MRI to predict the iTLS status.Methods A total of 660 patients with HCC who underwent surgery were retrospectively recruited from four centers between October 2015 and January 2023 and divided into training, internal test, and external validation sets. After features dimensionality and selection, corresponding features were used to construct transfer learning radiomic (TLR) models for diagnosing iTLS, and model interpretability was explored with pathway analysis in The Cancer Genome Atlas-Liver HCC. The performances of models were assessed using the area under the receiver operating characteristic curve (AUC). The log-rank test was used to evaluate the prognostic value of the TLR model. The combination therapy set of 101 patients with advanced HCC treated with first-line anti-programmed death 1 or ligand 1 plus antiangiogenic treatment between January 2021 and January 2024 was used to investigate the value of the TLR model for evaluating the treatment response.Results The presence of iTLS was identified in 46.0% (n=308) patients. The TLR model demonstrated excellent performance in predicting the presence of iTLS in training (AUC=0.91, 95% CI: 0.87, 0.94), internal test (AUC=0.85, 95% CI: 0.77, 0.93) and external validation set (AUC=0.85, 95% CI: 0.81, 0.90). The TLR model-predicted iTLS group has favorable overall survival (HR=0.66; 95% CI: 0.48, 0.90; p=0.007) and relapse-free survival (HR=0.64; 95% CI: 0.48, 0.85; p=0.001) in the external validation set. The model-predicted iTLS status was associated with inflammatory response and specific tumor-associated signaling activation (all p<0.001). The proportion of treatment responders was significantly higher in the model-predicted group with iTLS than in the group without iTLS (36% vs 13.73%, p=0.009).Conclusion The TLR model has indicated accurate prediction of iTLS status, which may assist in the risk stratification for patients with HCC in clinical practice.https://jitc.bmj.com/content/13/3/e011126.full
spellingShingle Juan Chen
Xiong Chen
Kai Fu
Feng Yi
Shichao Long
Mengsi Li
Linhui Zhong
Wenguang Liu
Deng Pan
Ganmian Dai
Wenzheng Li
Yue Ruan
Bocheng Zou
Transfer learning radiomic model predicts intratumoral tertiary lymphoid structures in hepatocellular carcinoma: a multicenter study
Journal for ImmunoTherapy of Cancer
title Transfer learning radiomic model predicts intratumoral tertiary lymphoid structures in hepatocellular carcinoma: a multicenter study
title_full Transfer learning radiomic model predicts intratumoral tertiary lymphoid structures in hepatocellular carcinoma: a multicenter study
title_fullStr Transfer learning radiomic model predicts intratumoral tertiary lymphoid structures in hepatocellular carcinoma: a multicenter study
title_full_unstemmed Transfer learning radiomic model predicts intratumoral tertiary lymphoid structures in hepatocellular carcinoma: a multicenter study
title_short Transfer learning radiomic model predicts intratumoral tertiary lymphoid structures in hepatocellular carcinoma: a multicenter study
title_sort transfer learning radiomic model predicts intratumoral tertiary lymphoid structures in hepatocellular carcinoma a multicenter study
url https://jitc.bmj.com/content/13/3/e011126.full
work_keys_str_mv AT juanchen transferlearningradiomicmodelpredictsintratumoraltertiarylymphoidstructuresinhepatocellularcarcinomaamulticenterstudy
AT xiongchen transferlearningradiomicmodelpredictsintratumoraltertiarylymphoidstructuresinhepatocellularcarcinomaamulticenterstudy
AT kaifu transferlearningradiomicmodelpredictsintratumoraltertiarylymphoidstructuresinhepatocellularcarcinomaamulticenterstudy
AT fengyi transferlearningradiomicmodelpredictsintratumoraltertiarylymphoidstructuresinhepatocellularcarcinomaamulticenterstudy
AT shichaolong transferlearningradiomicmodelpredictsintratumoraltertiarylymphoidstructuresinhepatocellularcarcinomaamulticenterstudy
AT mengsili transferlearningradiomicmodelpredictsintratumoraltertiarylymphoidstructuresinhepatocellularcarcinomaamulticenterstudy
AT linhuizhong transferlearningradiomicmodelpredictsintratumoraltertiarylymphoidstructuresinhepatocellularcarcinomaamulticenterstudy
AT wenguangliu transferlearningradiomicmodelpredictsintratumoraltertiarylymphoidstructuresinhepatocellularcarcinomaamulticenterstudy
AT dengpan transferlearningradiomicmodelpredictsintratumoraltertiarylymphoidstructuresinhepatocellularcarcinomaamulticenterstudy
AT ganmiandai transferlearningradiomicmodelpredictsintratumoraltertiarylymphoidstructuresinhepatocellularcarcinomaamulticenterstudy
AT wenzhengli transferlearningradiomicmodelpredictsintratumoraltertiarylymphoidstructuresinhepatocellularcarcinomaamulticenterstudy
AT yueruan transferlearningradiomicmodelpredictsintratumoraltertiarylymphoidstructuresinhepatocellularcarcinomaamulticenterstudy
AT bochengzou transferlearningradiomicmodelpredictsintratumoraltertiarylymphoidstructuresinhepatocellularcarcinomaamulticenterstudy