Share：
Share this content in WeChat
X
[Chinese][PDF]311
• Clinical Article •
The radioscore based on pre-radiotherapy MRI for predicting poor outcome risk in long-term follow-up of glioblastoma patients
WANG Fei  QUAN Guanmin  YUAN Tao 

Cite this article as: WANG F, QUAN G M, YUAN T. The radioscore based on pre-radiotherapy MRI for predicting poor outcome risk in long-term follow-up of glioblastoma patients[J]. Chin J Magn Reson Imaging, 2025, 16(6): 78-84, 92. DOI:10.12015/issn.1674-8034.2025.06.012.


[Abstract] Objective To explore the value of radioscores which based on radiomics features extracted from pre-radiotherapy contrast enhanced T1WI (CE-T1WI) on the prediction of poor survival prognosis in long-term follow-up for glioblastoma (GBM) patients.Materials and Methods We retrospectively analyzed the pre-radiotherpy MRI and clinical data of 76 patients with GBM. Then we divided all cases into a training group and a validation group in a 7∶3 ratio, construct a model based on the training group, and conduct efficacy validation in the validation group. The radiomics features were extracted on CE-T1WI images. The overall survival (OS) was defined as poor prognosis if it was less than or equal to the median value (OS = 380 days), and good prognosis if it was greater than the median value. The patients were divided into two groups, with 38 cases in each groups. We compared the difference of clinical, conventional MRI and radiomics variables between poor and good prognosis groups. Univariate and multivariate analyses were employed to select the risk factors. Then the prognosis predictive models based on clinical factors, conventional MRI findings, radiomics factors were established separately. We compared area under the curve (AUC) of receiver operating characteristic (ROC) curve for subjects with shorter OS evaluated with different models.Results Compared to the patients with good prognosis, the patients with poor prognosis (OS ≤ 380 days) were older (P = 0.025), had shorter progression-free survival (PFS) (P < 0.001), had lower survival during follow-up (P < 0.001), tended to have coarse linear or nodular residual cavity wall enhancement (P = 0.018), had higher fluid attenuated inversion recovery (FLAIR) hyperintense orthogonal growth rate (rFLAIR) (P = 0.024) and growth rate orthogonal value of enhancement lesions (rCE) (P = 0.002). Multivariate analysis showed that coarse linear or nodular enhancement of residual cavity wall [hazard ratio (HR) = 2.127] were the independent risk predictors of shorter PFS of patients with GBM. Whereas, older age (HR = 1.046) and coarse linear or nodular enhancement in residual cavity wall (HR = 2.105) were independent risk predictors of shorter OS. In univariate and multivariate analyses, the HR values of radioscore for shorter OS were 2.392 (P = 0.003) and 1.129 (P = 0.054) separately. The AUCs of combination models in the training cohort and validation cohorts were 0.822 and 0.841 respectively, which indicated that the combined model including radioscores had predictive significance for poor prognosis.Conclusions The radioscore extracted from pre-radiotherpy MRI could be used as a predictive factor of poor survival of GBM patients. This radiomics feature could improve the predictive efficacy of the model which included conventional and clinical variables.
[Keywords] glioblastoma；magnetic resonance imaging；radiomics；contrast enhancement；prognosis；prediction

WANG Fei   QUAN Guanmin   YUAN Tao*  

Department of Medical Imaging, The Second Hospital of Hebei Medical University, Shijiazhuang 050000, China

Corresponding author: YUAN T, E-mail: yuantao1976@hebmu.edu.cn

Conflicts of interest   None.

Received  2025-03-03
Accepted  2025-05-13
DOI: 10.12015/issn.1674-8034.2025.06.012
Cite this article as: WANG F, QUAN G M, YUAN T. The radioscore based on pre-radiotherapy MRI for predicting poor outcome risk in long-term follow-up of glioblastoma patients[J]. Chin J Magn Reson Imaging, 2025, 16(6): 78-84, 92. DOI:10.12015/issn.1674-8034.2025.06.012.

[1]
MILLER K D, OSTROM Q T, KRUCHKO C, et al. Brain and other central nervous system tumor statistics, 2021[J]. CA Cancer J Clin, 2021, 71(5): 381-406. DOI: 10.3322/caac.21693.
[2]
JIANG T, MAO Y, MA W B, et al. CGCG clinical practice guidelines for the management of adult diffuse gliomas[J]. Cancer Lett, 2016, 375(2): 263-273. DOI: 10.1016/j.canlet.2016.01.024.
[3]
BUTLER M, PRASAD S, SRIVASTAVA S K. Targeting glioblastoma tumor microenvironment[J/OL]. Adv Exp Med Biol, 2020, 1296: 1-9 [2025-03-02]. https://pubmed.ncbi.nlm.nih.gov/34185283/. DOI: 10.1007/978-3-030-59038-3_1.
[4]
RODA D, VEIGA P, MELO J B, et al. Principles in the management of glioblastoma[J/OL]. Genes (Basel), 2024, 15(4): 501 [2025-03-02]. https://pubmed.ncbi.nlm.nih.gov/38674436/. DOI: 10.3390/genes15040501.
[5]
BOOTH T C, LUIS A, BRAZIL L, et al. Glioblastoma post-operative imaging in neuro-oncology: current UK practice (GIN CUP study)[J]. Eur Radiol, 2021, 31(5): 2933-2943. DOI: 10.1007/s00330-020-07387-3.
[6]
NEIL Z D, PIERZCHAJLO N, BOYETT C, et al. Assessing metabolic markers in glioblastoma using machine learning: a systematic review[J]. Metabolites, 2023, 13(2): 161. DOI: 10.3390/metabo13020161.
[7]
LU Y P, PATEL M, NATARAJAN K, et al. Machine learning-based radiomic, clinical and semantic feature analysis for predicting overall survival and MGMT promoter methylation status in patients with glioblastoma[J/OL]. Magn Reson Imaging, 2020, 74: 161-170 [2025-03-02]. https://pubmed.ncbi.nlm.nih.gov/32980505/. DOI: 10.1016/j.mri.2020.09.017.
[8]
SUH C H, KIM H S, JUNG S C, et al. Multiparametric MRI as a potential surrogate endpoint for decision-making in early treatment response following concurrent chemoradiotherapy in patients with newly diagnosed glioblastoma: a systematic review and meta-analysis[J]. Eur Radiol, 2018, 28(6): 2628-2638. DOI: 10.1007/s00330-017-5262-5.
[9]
SHI Y X, LIU D L, KONG Z R, et al. Prognostic value of choline and other metabolites measured using 1H-magnetic resonance spectroscopy in gliomas: a meta-analysis and systemic review[J/OL]. Metabolites, 2022, 12(12): 1219 [2025-03-02]. https://pubmed.ncbi.nlm.nih.gov/36557257/. DOI: 10.3390/metabo12121219.
[10]
LIANG Q, ZHANG H. Research progression of MRI radiomics in glioma[J]. Chin J Magn Reson Imag, 2024, 15(2): 192-197. DOI: 10.12015/issn.1674-8034.2024.02.031.
[11]
BERGER A, TZARFATI G G, SERAFIMOVA M, et al. Clinical and prognostic implications of rim restriction following glioma surgery[J/OL]. Sci Rep, 2022, 12(1): 12874 [2025-03-02]. https://pubmed.ncbi.nlm.nih.gov/35896589/. DOI: 10.1038/s41598-022-16717-y.
[12]
GUO J, FATHI KAZEROONI A, TOORENS E, et al. Integrating imaging and genomic data for the discovery of distinct glioblastoma subtypes: a joint learning approach[J/OL]. Sci Rep, 2024, 14(1): 4922 [2025-03-02]. https://pubmed.ncbi.nlm.nih.gov/38418494/. DOI: 10.1038/s41598-024-55072-y.
[13]
FAN H Q, LUO Y L, GU F, et al. Artificial intelligence-based MRI radiomics and radiogenomics in glioma[J/OL]. Cancer Imaging, 2024, 24(1): 36 [2025-03-02]. https://pubmed.ncbi.nlm.nih.gov/38486342/. DOI: 10.1186/s40644-024-00682-y.
[14]
GILLIES R J, SCHABATH M B. Radiomics improves cancer screening and early detection[J]. Cancer Epidemiol Biomarkers Prev, 2020, 29(12): 2556-2567. DOI: 10.1158/1055-9965.EPI-20-0075.
[15]
ZHAI X Y, REN J F, CHENG S J, et al. A preliminary study on predicting glioblastoma recurrence and postoperative survival time through MRI imaging radiomics[J]. Chin J Magn Reson Imag, 2023, 14(12): 26-32. DOI: 10.12015/issn.1674-8034.2023.12.005.
[16]
POURSAEED R, MOHAMMADZADEH M, SAFAEI A A. Survival prediction of glioblastoma patients using machine learning and deep learning: a systematic review[J/OL]. BMC Cancer, 2024, 24(1): 1581 [2025-03-02]. https://pubmed.ncbi.nlm.nih.gov/39731064/. DOI: 10.1186/s12885-024-13320-4.
[17]
GARRETT M D, YANAGIHARA T K, YEH R, et al. Monitoring radiation treatment effects in glioblastoma: flair volume as significant predictor of survival[J]. Tomography, 2017, 3(3): 131-137. DOI: 10.18383/j.tom.2017.00009.
[18]
WEN P Y, VAN DEN BENT M, YOUSSEF G, et al. RANO 2.0: update to the response assessment in neuro-oncology criteria for high- and low-grade gliomas in adults[J]. J Clin Oncol, 2023, 41(33): 5187-5199. DOI: 10.1200/JCO.23.01059.
[19]
KIM B R, CHOI S H, YUN T J, et al. MR imaging analysis of non-measurable enhancing lesions newly appearing after concomitant chemoradiotherapy in glioblastoma patients for prognosis prediction[J/OL]. PLoS One, 2016, 11(11): e0166096 [2025-03-02]. https://pubmed.ncbi.nlm.nih.gov/27835666/. DOI: 10.1371/journal.pone.0166096.
[20]
EISELE S C, WEN P Y, LEE E Q. Assessment of brain tumor response: RANO and its offspring[J/OL]. Curr Treat Options Oncol, 2016, 17(7): 35 [2025-03-02]. https://pubmed.ncbi.nlm.nih.gov/27262709/. DOI: 10.1007/s11864-016-0413-5.
[21]
CHADDAD A, DANIEL P, DESROSIERS C, et al. Novel radiomic features based on joint intensity matrices for predicting glioblastoma patient survival time[J]. IEEE J Biomed Health Inform, 2019, 23(2): 795-804. DOI: 10.1109/JBHI.2018.2825027.
[22]
CHIANG G C, PISAPIA D J, LIECHTY B, et al. The prognostic value of MRI subventricular zone involvement and tumor genetics in lower grade gliomas[J]. J Neuroimaging, 2020, 30(6): 901-909. DOI: 10.1111/jon.12763.
[23]
CHABAN A, WASCHULZIK B, BERNHARDT D, et al. Amino acid PET vs. RANO MRI for prediction of overall survival in patients with recurrent high grade glioma under bevacizumab therapy[J]. Eur J Nucl Med Mol Imaging, 2024, 51(6): 1698-1702. DOI: 10.1007/s00259-024-06601-4.
[24]
KIM J Y, PARK J E, JO Y, et al. Incorporating diffusion- and perfusion-weighted MRI into a radiomics model improves diagnostic performance for pseudoprogression in glioblastoma patients[J]. Neuro Oncol, 2019, 21(3): 404-414. DOI: 10.1093/neuonc/noy133.
[25]
ZEPPA P, NEITZERT L, MAMMI M, et al. How reliable are volumetric techniques for high-grade gliomas? a comparison study of different available tools[J/OL]. Neurosurgery, 2020, 87(6): E672-E679 [2025-03-02]. https://pubmed.ncbi.nlm.nih.gov/32629469/. DOI: 10.1093/neuros/nyaa282.
[26]
CHOI Y, JANG J, KIM B S, et al. Pretreatment MR-based radiomics in patients with glioblastoma: a systematic review and meta-analysis of prognostic endpoints[J/OL]. Eur J Radiol, 2023, 168: 111130 [2025-03-02]. https://pubmed.ncbi.nlm.nih.gov/37827087/. DOI: 10.1016/j.ejrad.2023.111130.
[27]
KU M C, WAICZIES S, NIENDORF T, et al. Assessment of blood brain barrier leakage with gadolinium-enhanced MRI[J/OL]. Methods Mol Biol, 2018, 1718: 395-408 [2025-03-02]. https://pubmed.ncbi.nlm.nih.gov/29341021/. DOI: 10.1007/978-1-4939-7531-0_23.
[28]
ALIZADEH M, BROOMAND LOMER N, AZAMI M, et al. Radiomics: the new promise for differentiating progression, recurrence, pseudoprogression, and radionecrosis in glioma and glioblastoma multiforme[J/OL]. Cancers (Basel), 2023, 15(18): 4429 [2025-03-02]. https://pubmed.ncbi.nlm.nih.gov/37760399/. DOI: 10.3390/cancers15184429.
[29]
ZHUANG Y K, CHEN J W, MAI Z H, et al. Signature construction and disulfidptosis-related molecular cluster identification for better prediction of prognosis in glioma[J/OL]. J Mol Neurosci, 2024, 74(2): 38 [2025-03-02]. https://pubmed.ncbi.nlm.nih.gov/38573391/. DOI: 10.1007/s12031-024-02216-4.
[30]
YOSHIMURA H, KAWAHARA D, SAITO A, et al. Prediction of prognosis in glioblastoma with radiomics features extracted by synthetic MRI images using cycle-consistent GAN[J]. Phys Eng Sci Med, 2024, 47(3): 1227-1243. DOI: 10.1007/s13246-024-01443-8.
[31]
HU M F, LI J G, LI Z Y, et al. Prediction of survival outcomes in patients with glioma using magnetic resonance imaging (MRI): a systematic review and meta-analysis[J/OL]. J Integr Neurosci, 2025, 24(1): 23389 [2025-03-02]. https://pubmed.ncbi.nlm.nih.gov/38573391/. DOI: 10.31083/JIN23389.
[32]
BAPURAJ J R, WANG N, SRINIVASAN A, et al. Advanced imaging and computational techniques for the diagnostic and prognostic assessment of malignant gliomas[J]. Cancer J, 2021, 27(5): 344-352. DOI: 10.1097/PPO.0000000000000545.
[33]
STYLIARA E I, ASTRAKAS L G, ALEXIOU G, et al. Survival outcome prediction in glioblastoma: insights from MRI radiomics[J]. Curr Oncol, 2024, 31(4): 2233-2243. DOI: 10.3390/curroncol31040165.
[34]
WISNU WARDHANA D P, MALIAWAN S, MAHADEWA T G B, et al. Radiomic features as artificial intelligence prognostic models in glioblastoma: a systematic review and meta-analysis[J/OL]. Diagnostics (Basel), 2024, 14(21): 2354 [2025-03-02]. https://pubmed.ncbi.nlm.nih.gov/39518322/. DOI: 10.3390/diagnostics14212354.
[35]
CHOI Y, NAM Y, JANG J, et al. Radiomics may increase the prognostic value for survival in glioblastoma patients when combined with conventional clinical and genetic prognostic models[J]. Eur Radiol, 2021, 31(4): 2084-2093. DOI: 10.1007/s00330-020-07335-1.

PREV MRI-based study of gray matter morphological in children with bilateral spastic cerebral palsy
NEXT Exploring the correlation between left atrial strain and left ventricular fibrosis in dilated cardiomyopathy based on cardiac magnetic resonance imaging
  


LINK


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