Share：
Share this content in WeChat
X
[Chinese] [PDF] 189 8
• Clinical Article •
Construction and validation of a non-invasive differentiation model for glioblastoma and primary central nervous system lymphoma based on clinical-multimodal magnetic resonance imaging radiomics
SONG Tingting  HONG Shiqiang  ZHU He  ZHENG Lei  WU Changshun  FENG Hong 

DOI:10.12015/issn.1674-8034.2025.08.007.


[Abstract] Objective To overcome the limitations of conventional imaging in differentiating glioblastoma (GBM) from primary central nervous system lymphoma (PCNSL), we propose and validate a clinically integrated radiomics model for the preoperative, non-invasive stratification of these two oncological entities.Materials and Methods A retrospective cohort of 173 patients with intracranial masses (118 GBM, 55 PCNSL), confirmed by histopathology or diagnostic radiotherapy, was randomly divided into training (n = 121) and validation (n = 52) sets in a 7∶3 ratio. Preoperative clinical parameters (serological indices, imaging manifestations) and multimodal MRI sequences [CE-T1WI, T2-FLAIR, DWI (b=1000 s/mm²), and ADC] were acquired. Tumor core regions (excluding peritumoral edema) were delineated as regions of interest (ROIs). Following Z-score normalization, key features were selected using the Mann-Whitney U test, Spearman correlation analysis, and the least absolute shrinkage and selection operator (LASSO) algorithm. An XGBoost classifier with 10-fold cross-validation was employed for model construction. A comparative analysis of five models was performed: the clinical model, four single-modality radiomics models, the multimodal radiomics model, and the integrated clinical-radiomics model. The diagnostic performance was evaluated using receiver operating characteristic (ROC) curves, with the area under the curve (AUC), sensitivity, specificity, and accuracy calculated. The statistical validation included the DeLong test for AUC comparison, calibration curve assessment, and decision curve analysis (DCA) to quantify clinical utility.Results The clinical model demonstrated AUC values of 0.83 (95% CI: 0.76 to 0.90) in the training set and 0.74 (95% CI: 0.61 to 0.87) in the validation set. Among radiomics models, the multimodal radiomics model (T1+ADC+T2+DWI) achieved optimal performance with training/validation AUCs of 0.93 (95% CI: 0.88 to 0.98)/0.84 (95% CI: 0.72 to 0.96). The integrated clinical-radiomics model demonstrated superior diagnostic performance, achieving a training AUC of 0.94 (95% CI: 0.90 to 0.98) (accuracy 90.2%, sensitivity 96.7%) and a validation AUC of 0.85 (95% CI: 0.74 to 0.96) (accuracy 88.6%, sensitivity 83.3%). This combined model significantly outperformed individual models in predictive accuracy (DeLong test, P < 0.05) and clinical net benefit across threshold probability ranges (decision curve analysis).Conclusions The combined model, constructed by integrating clinical features and multimodal radiomics, can non-invasively and stably distinguish GBM from PCNSL, providing reliable references for the precise preoperative diagnosis of patients. It helps reduce the need for invasive tests and optimizes the clinical decision-making process.
[Keywords] glioblastoma；primary central nervous system lymphoma；clinical features；radiomics；multimodal magnetic resonance imaging；differential diagnosis

SONG Tingting1   HONG Shiqiang2   ZHU He1   ZHENG Lei1   WU Changshun3   FENG Hong2*  

1 Graduate Department, Shandong First Medical University (Shandong Academy of Medical Sciences), Jinan 250117, China

2 Tumor Center, Shandong Provincial Hospital Affiliated to Shandong First Medical University, Jinan 250021, China

3 Department of Orthopedics and Joint Surgery, Shandong Provincial Hospital Affiliated to Shandong First Medical University, Jinan 250021, China

Corresponding author: FENG H, E-mail: 18753157528@163.com

Conflicts of interest   None.

Received  2025-04-01
Accepted  2025-07-07
DOI: 10.12015/issn.1674-8034.2025.08.007
DOI:10.12015/issn.1674-8034.2025.08.007.

[1]
SCHAFF L R, MELLINGHOFF I K. Glioblastoma and Other Primary Brain Malignancies in Adults: A Review[J]. JAMA, 2023, 329(7): 574-587. DOI: 10.1001/jama.2023.0023.
[2]
RONG L, LI N, ZHANG Z. Emerging Therapies for Glioblastoma: Current State and Future Directions[J/OL]. J Exp Clin Cancer Res, 2022, 41(1): 142 [2025-04-01]. https://doi.org/10.1186/s13046-022-02349-7. DOI: 10.1186/s13046-022-02349-7.
[3]
LOUIS D, PERRY A, WESSELING P, et al. The 2021 WHO Classification of Tumors of the Central Nervous System: A Summary[J]. Neuro Oncol, 2021, 23(8): 1231-1251. DOI: 10.1093/neuonc/noab106.
[4]
SCHAFF L, GROMMES C. Primary Central Nervous System Lymphoma[J]. Blood, 2022, 140(9): 971-979. DOI: 10.1182/blood.2020008377.
[5]
CHEN C, SUN P, SUN X Q, et al. Primary Treatment and Recent Survival Trends in Patients with Primary Diffuse Large B-cell Lymphoma of Central Nervous System, 1995-2016: A population-based SEER analysis[J]. Hematol Oncol, 2023, 41(2): 248-256. DOI: 10.1002/hon.2918.
[6]
TAWEESOMBOONYAT C, TUNTHANATHIP T, SAE-HENG S, et al. Diagnostic Yield and Complication of Frameless Stereotactic Brain Biopsy[J]. J Neurosci Rural Pract, 2019, 10(1): 78-84. DOI: 10.4103/jnrp.jnrp_166_18.
[7]
TOSEFSKY K, REBCHUK A, MARTIN K, et al. Preoperative Corticosteroids Reduce Diagnostic Accuracy of Stereotactic Biopsies in Primary Central Nervous System Lymphoma: A Systematic Review and Meta-Analysis[J]. Neurosurgery, 2024, 95(4): 740-750. DOI: 10.1227/neu.0000000000002944.
[8]
BULLIS C, MALDONADO-PEREZ A, BOWDEN S, et al. Diagnostic Impact of Preoperative Corticosteroids in Primary Central Nervous System Lymphoma[J]. J Clin Neurosci, 2020, 72: 287-291. DOI: 10.1016/j.jocn.2019.10.010.
[9]
THERKELSEN K, OMURO A. Advances in Primary Central Nervous System Lymphoma[J/OL]. Curr Neurol Neurosci Rep, 2024, 25(1): 5 [2025-04-01]. https://doi.org/10.1007/s11910-024-01389-0. DOI: 10.1007/s11910-024-01389-0.
[10]
GUHA A, HALDER S, SHINDE S, et al. How Does Deep Learning/Machine Learning Perform in Comparison to Radiologists in Distinguishing Glioblastomas (or grade IV astrocytomas) From Primary CNS Lymphomas?: A Meta-analysis and Systematic Review[J]. Clin Radiol, 2024, 79(6): 460-472. DOI: 10.1016/j.crad.2024.03.007.
[11]
HAN Y, WANG Z, LI W, et al. Differentiation Between Primary Central Nervous System Lymphoma and Atypical Glioblastoma Based on MRI Morphological Feature and Signal Intensity Ratio: A Retrospective Multicenter Study[J/OL]. Front Oncol, 2022, 12: 811197 [2025-04-01]. https://doi.org/10.3389/fonc.2022.811197. DOI: 10.3389/fonc.2022.811197.
[12]
KUNIMATSU A, KUNIMATSU N, YASAKA K, et al. Machine Learning-based Texture Analysis of Contrast-enhanced MR Imaging to Differentiate between Glioblastoma and Primary Central Nervous System Lymphoma[J]. Magn Reson Med Sci, 2019, 18(1): 44-52. DOI: 10.2463/mrms.mp.2017-0178.
[13]
XIA W, HU B, LI H, et al. Deep Learning for Automatic Differential Diagnosis of Primary Central Nervous System Lymphoma and Glioblastoma: Multi-Parametric Magnetic Resonance Imaging Based Convolutional Neural Network Model[J]. J Magn Reson Imaging, 2021, 54(3): 880-887. DOI: 10.1002/jmri.27592.
[14]
MATSUDA R, MAEOKA R, MORIMOTO T, et al. Pre-treatment Systemic Inflammation Response Index and Systemic Immune Inflammation In Patients with Primary Central Nerve System Lymphoma as A Useful Prognostic Indicator[J]. J Neurooncol, 2024, 168(3): 487-494. DOI: 10.1007/s11060-024-04692-5.
[15]
FENG Y, LIU Y, ZHONG M, et al. Complete Blood Count Score Model Predicts Inferior Prognosis in Primary Central Nervous System Lymphoma[J/OL]. Front Oncol, 2021, 11: 618694 [2025-04-01]. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8596886. DOI: 10.3389/fonc.2021.618694.
[16]
WANG Y, SHI Q, SHI Z, et al. Biological Signatures of The International Prognostic Index in Diffuse Large B-cell Lymphoma[J]. Blood Adv, 2024, 8(7): 1587-1599. DOI: 10.1182/bloodadvances.2023011425.
[17]
JI H, NIU X, YIN L, et al. Ratio of Immune Response to Tumor Burden Predicts Survival Via Regulating Functions of Lymphocytes and Monocytes in Diffuse Large B-Cell Lymphoma[J]. Cell Physiol Biochem, 2018, 45(3): 951-961. DOI: 10.1159/000487288.
[18]
FERRERI A, ILLERHAUS G, DOORDUIJN J, et al. Primary Central Nervous System Lymphomas: EHA-ESMO Clinical Practice Guideline for Diagnosis, Treatment and Follow-up[J/OL]. Ann Oncol, 2024 [2025-04-01]. https://www.sciencedirect.com/science/article/pii/S0923753423012580. DOI: 10.1016/j.annonc.2023.11.010.
[19]
HOANG-XUAN K, DECKERT M, FERRERI A, et al. European Association of Neuro-Oncology (EANO) Guidelines for Treatment of Primary Central Nervous System Lymphoma (PCNSL)[J]. Neuro Oncol, 2023, 25(1): 37-53. DOI: 10.1093/neuonc/noac196.
[20]
YU X, HONG W, YE M, et al. Atypical Primary Central Nervous System Lymphoma and Glioblastoma: Multiparametric Differentiation Based on Non-enhancing Volume, Apparent Diffusion Coefficient, and Arterial Spin Labeling[J]. Eur Radiol, 2023, 33(8): 5357-5367. DOI: 10.1007/s00330-023-09681-2.
[21]
SCAPICCHIO C, GABELLONI M, BARUCCI A, et al. A Deep Look into Radiomics[J]. Radiol Med, 2021, 126(10): 1296-1311. DOI: 10.1007/s11547-021-01389-x.
[22]
COBO M, MENÉNDEZ FERNÁNDEZ-MIRANDA P, BASTARRIKA G, et al. Enhancing Radiomics and Deep Learning Systems Through the Standardization of Medical Imaging Workflows[J/OL]. Sci Data, 2023, 10(1): 732 [2025-04-01]. https://www.nature.com/articles/s41597-023-02641-x. DOI: 10.1038/s41597-023-02641-x.
[23]
SHUR J, DORAN S, KUMAR S, et al. Radiomics in Oncology: A Practical Guide[J]. Radiographics, 2021, 41(6): 1717-1732. DOI: 10.1148/rg.2021210037.
[24]
FENG A, LI L, HUANG T, et al. Differentiating Glioblastoma from Primary Central Nervous System Lymphoma of Atypical Manifestation Using Multiparametric Magnetic Resonance Imaging: A Comparative Study[J/OL]. Heliyon, 2023, 9(4): e15150 [2025-04-01]. https://www.sciencedirect.com/science/article/pii/S2405844023012852. DOI: 10.1016/j.heliyon.2023.e15150.
[25]
WANG S, YUAN Q, ZHANG G, et al. Preoperative Blood Testing for Glioblastoma, Brain Metastases, and Primary Central Nervous System Lymphoma Differentiation[J]. Transl Cancer Res, 2022, 11(1): 63-71. DOI: 10.21037/tcr-21-1957.
[26]
WU W, MIAO L, ZHAO L, et al. Prognostic Value of Lactate Dehydrogenase, Serum Albumin and The Lactate Dehydrogenase/Albumin Ratio in Patients with Diffuse Large B-cell Lymphoma[J/OL]. Hematology, 2024, 29(1): 2293514 [2025-04-01].https://www.tandfonline.com/doi/full/10.1080/16078454.2023.2293514. DOI: 10.1080/16078454.2023.2293514.
[27]
XI Y, KANG X, WANG N, et al. Differentiation of Primary Central Nervous System Lymphoma from High-grade Glioma and Brain Metastasis Using Arterial Spin Labeling and Dynamic Contrast-enhanced Magnetic Resonance Imaging[J]. Eur J Radiol, 2019, 112: 59-64. DOI: 10.1016/j.ejrad.2019.01.008.
[28]
GUPTA T, NAYAK P, BAVISKAR Y, et al. Systemic Inflammatory Biomarkers in Primary Central Nervous System Lymphoma Versus High-grade Glioma: Exploratory, Comparative and Correlative Analysis[J/OL]. CNS Oncol, 2022, 11(2): Cns83 [2025-04-01]. https://www.futuremedicine.com/doi/10.2217/cns-2022-0004. DOI: 10.2217/cns-2022-0004.
[29]
DHARMAJAYA R, SARI D. Role and Value of Inflammatory Markers in Brain Tumors: A Case Controlled Study[J/OL]. Ann Med Surg (Lond), 2021, 63: 102107 [2025-04-01]. https://doi.org/10.1016/j.amsu.2021.01.055. DOI: 10.1016/j.amsu.2021.01.055.
[30]
LI D, ZHANG Q, LI L, et al. β2-Microglobulin Maintains Glioblastoma Stem Cells and Induces M2-like Polarization of Tumor-Associated Macrophages[J]. Cancer Res, 2022, 82(18): 3321-3334. DOI: 10.1158/0008-5472.CAN-22-0507.
[31]
GARCIA-RUIZ A, PONS-ESCODA A, GRUSSU F, et al. An Accessible Deep Learning Tool for Voxel-wise Classification of Brain Malignancies from Perfusion MRI[J/OL]. Cell Rep Med, 2024, 5(3): 101464 [2025-04-01]. https://www.sciencedirect.com/science/article/pii/S2666379124001030. DOI: 10.1016/j.xcrm.2024.101464.
[32]
TATEISHI K, MIYAKE Y, NAKAMURA T, et al. Primary Central Nervous System Lymphoma: Clinicopathological and Genomic Insights for Therapeutic Development[J]. Brain Tumor Pathol, 2021, 38(3): 173-182. DOI: 10.1007/s10014-021-00408-z.
[33]
NOWACKA A, ŚNIEGOCKI M, SMUCZYŃSKI W, et al. Angiogenesis in Glioblastoma-Treatment Approaches[J/OL]. Cells, 2025, 14(6): 407 [2025-04-01]. https://www.mdpi.com/2073-4409/14/6/407. DOI: 10.3390/cells14060407.
[34]
CORTES BALLEN A, AMOSU M, RAVINDER S, et al. Metabolic Reprogramming in Glioblastoma Multiforme: A Review of Pathways and Therapeutic Targets[J/OL]. Cells, 2024, 13(18): 1574 [2025-04-01]. https://www.mdpi.com/2073-4409/13/18/1574. DOI: 10.3390/cells13181574.
[35]
FERRERI A, CALIMERI T, CWYNARSKI K, et al. Primary Central Nervous System Lymphoma[J/OL]. Nat Rev Dis Primers, 2023, 9(1): 29 [2025-04-01]. https://www.nature.com/articles/s41572-023-00439-0. DOI: 10.1038/s41572-023-00439-0.
[36]
FUSHIMI Y, NAKAJIMA S. Can We Distinguish Atypical Primary Brain Lymphoma from Glioblastoma Multiforme by Looking at The Non-enhancing Tumor?[J]. Eur Radiol, 2023, 33(8): 5355-5356. DOI: 10.1007/s00330-023-09750-6.
[37]
SHAO J, MA J, YU Y, et al. A Multimodal Integration Pipeline for Accurate Diagnosis, Pathogen Identification, and Prognosis Prediction of Pulmonary Infections[J/OL]. Innovation (Camb), 2024, 5(4): 100648 [2025-04-01]. https://www.sciencedirect.com/science/article/pii/S2666386424001422. DOI: 10.1016/j.xinn.2024.100648.
[38]
SCHMIDT D, PATEL R, KIRSCH D, et al. Metabolomics in Cancer Research and Emerging Applications in Clinical Oncology[J]. CA Cancer J Clin, 2021, 71(4): 333-358. DOI: 10.3322/caac.21670.
[39]
BATCHELOR T, GIRI S, RUPPERT A, et al. Myeloablative Vs Nonmyeloablative Consolidation for Primary Central Nervous System Lymphoma: Results of Alliance 51101[J]. Blood Adv, 2024, 8(12): 3189-3199. DOI: 10.1182/bloodadvances.2023011657.
[40]
FERRERI A, CWYNARSKI K, PULCZYNSKI E, et al. Long-term Efficacy, Safety and Neurotolerability of MATRix Regimen Followed by Autologous Transplant in Primary CNS Lymphoma: 7-year Results of The IELSG32 Randomized Trial[J]. Leukemia, 2022, 36(7): 1870-1878. DOI: 10.1038/s41375-022-01582-5.
[41]
GAJJAR A, MAHAJAN A, ABDELBAKI M, et al. Pediatric Central Nervous System Cancers, Version 2.2023, NCCN Clinical Practice Guidelines in Oncology[J]. J Natl Compr Canc Netw, 2022, 20(12): 1339-1362. DOI: 10.6004/jnccn.2022.0062.

PREV Value of integrated MRI radiomics and clinical factors for post-thrombolytic hemorrhagic transformation in acute ischemic stroke
NEXT Study on value of intra-tumoral and peri-tumoral features of multimodal MRI radiomics in distinguishing fibrous from nonfibrous meningiomas
  


LINK


Tel & Fax: +8610-67113815    E-mail: editor@cjmri.cn