Interpretable MRI diversity index imaging for noninvasive evaluation of ferritin light chain expression and its clinical prognostic value in glioblastoma

Noninvasive biomarkers are urgently needed for glioblastoma (GBM) precision management. Ferritin light chain (FTL) not only regulates ferroptosis but is also a crucial factor in modulating the tumor immune microenvironment. Therefore, this study aimed to develop an MRI diversity index model to estimate FTL expression and evaluate its prognostic value. GBM RNA-seq and clinical data were obtained from The Cancer Genome Atlas (TCGA), along with corresponding MRI data from The Cancer Imaging Archive (TCIA). The 87 patients with matched imaging, RNA-seq, and clinical data were randomly divided 1:1 into a training set ( n = 43) and an internal test set ( n = 44). Intratumoral heterogeneity (ITH) features were extracted via diversity index imaging to establish radiomics and ITH model for estimating FTL expression. An independent external validation cohort of 35 GBM patients from the Chinese Glioma Genome Atlas (CGGA) database was included. Model efficacy in prognostic stratification was evaluated using Kaplan-Meier analysis, Cox regression, and nomogram. Subsequently, GSEA explored biological pathways, CIBERSORT assessed immune infiltration, and correlations between imaging features and immune checkpoint genes were analyzed. The FTL estimation model based on diversity index imaging achieved an AUC of 0.896 in the training set, 0.791 in the internal test set, and 0.688 in the external validation cohort. Survival analysis confirmed that the imaging-derived score was an independent risk factor for overall survival (HR = 1.79, P = 0.044). High-score tumors exhibited significant enrichment of immune-related pathways, accompanied by reduced follicular helper T cell infiltration. Additionally, the score showed significant correlations with immune checkpoint genes LAG-3 and CD47, suggesting a potential association with immunotherapy-related molecular features that warrants further investigation. MRI diversity index imaging enables noninvasive estimation of FTL expression in GBM and shows correlative associations with ferroptosis status and immune microenvironment features, including follicular helper T cell infiltration and immune checkpoint molecule expression patterns. The model retained discriminative ability in an independent external validation cohort, suggesting that the diversity index may serve as a potential imaging biomarker for further exploration in personalized GBM management.

Authors

Institutions

Publication Details

Journal
BMC Medical Imaging
Published
2026-11-21
DOI
https://doi.org/10.1186/s12880-026-03043-2
Primary Topic
Radiomics and Machine Learning in Medical Imaging
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Interpretable MRI diversity index imaging for noninvasive evaluation of ferritin light chain expression and its clinical prognostic value in glioblastoma

Wei Chen, P Chen, Meng Zeng, Fuwen Yao et al.
BMC Medical Imaging
Radiomics and Machine Learning in Medical Imaging
article

Interpretable MRI diversity index imaging for noninvasive evaluation of ferritin light chain expression and its clinical prognostic value in glioblastoma

Wei Chen, P Chen, Meng Zeng, Fuwen Yao, Fengwei Yu, Jiafei Chen, Hailong Jiang, Taotao Yang, Jun Chen, Zilong Li, Jing Yang
article en

Abstract

Noninvasive biomarkers are urgently needed for glioblastoma (GBM) precision management. Ferritin light chain (FTL) not only regulates ferroptosis but is also a crucial factor in modulating the tumor immune microenvironment. Therefore, this study aimed to develop an MRI diversity index model to estimate FTL expression and evaluate its prognostic value. GBM RNA-seq and clinical data were obtained from The Cancer Genome Atlas (TCGA), along with corresponding MRI data from The Cancer Imaging Archive (TCIA). The 87 patients with matched imaging, RNA-seq, and clinical data were randomly divided 1:1 into a training set ( n = 43) and an internal test set ( n = 44). Intratumoral heterogeneity (ITH) features were extracted via diversity index imaging to establish radiomics and ITH model for estimating FTL expression. An independent external validation cohort of 35 GBM patients from the Chinese Glioma Genome Atlas (CGGA) database was included. Model efficacy in prognostic stratification was evaluated using Kaplan-Meier analysis, Cox regression, and nomogram. Subsequently, GSEA explored biological pathways, CIBERSORT assessed immune infiltration, and correlations between imaging features and immune checkpoint genes were analyzed. The FTL estimation model based on diversity index imaging achieved an AUC of 0.896 in the training set, 0.791 in the internal test set, and 0.688 in the external validation cohort. Survival analysis confirmed that the imaging-derived score was an independent risk factor for overall survival (HR = 1.79, P = 0.044). High-score tumors exhibited significant enrichment of immune-related pathways, accompanied by reduced follicular helper T cell infiltration. Additionally, the score showed significant correlations with immune checkpoint genes LAG-3 and CD47, suggesting a potential association with immunotherapy-related molecular features that warrants further investigation. MRI diversity index imaging enables noninvasive estimation of FTL expression in GBM and shows correlative associations with ferroptosis status and immune microenvironment features, including follicular helper T cell infiltration and immune checkpoint molecule expression patterns. The model retained discriminative ability in an independent external validation cohort, suggesting that the diversity index may serve as a potential imaging biomarker for further exploration in personalized GBM management.

BMC Medical Imaging
Bayer (United States) (US), Army Medical University (CN), Southwest Hospital (CN)
Radiomics and Machine Learning in Medical Imaging
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon