Metabolic signatures derived from whole-brain MR-spectroscopy identify early tumor progression in high-grade gliomas using machine learning.
Binary classification
Glioblastoma
Machine learning
Multiclass classifier
Predictive models
Whole-brain magnetic resonance spectroscopy (WB-MRS)
Journal
Journal of neuro-oncology
ISSN: 1573-7373
Titre abrégé: J Neurooncol
Pays: United States
ID NLM: 8309335
Informations de publication
Date de publication:
24 Aug 2024
24 Aug 2024
Historique:
received:
03
03
2024
accepted:
19
08
2024
medline:
24
8
2024
pubmed:
24
8
2024
entrez:
24
8
2024
Statut:
aheadofprint
Résumé
Recurrence for high-grade gliomas is inevitable despite maximal safe resection and adjuvant chemoradiation, and current imaging techniques fall short in predicting future progression. However, we introduce a novel whole-brain magnetic resonance spectroscopy (WB-MRS) protocol that delves into the intricacies of tumor microenvironments, offering a comprehensive understanding of glioma progression to inform expectant surgical and adjuvant intervention. We investigated five locoregional tumor metabolites in a post-treatment population and applied machine learning (ML) techniques to analyze key relationships within seven regions of interest: contralateral normal-appearing white matter (NAWM), fluid-attenuated inversion recovery (FLAIR), contrast-enhancing tumor at time of WB-MRS (Tumor), areas of future recurrence (AFR), whole-brain healthy (WBH), non-progressive FLAIR (NPF), and progressive FLAIR (PF). Five supervised ML classification models and a neural network were developed, optimized, trained, tested, and validated. Lastly, a web application was developed to host our novel calculator, the Miami Glioma Prediction Map (MGPM), for open-source interaction. Sixteen patients with histopathological confirmation of high-grade glioma prior to WB-MRS were included in this study, totaling 118,922 whole-brain voxels. ML models successfully differentiated normal-appearing white matter from tumor and future progression. Notably, the highest performing ML model predicted glioma progression within fluid-attenuated inversion recovery (FLAIR) signal in the post-treatment setting (mean AUC = 0.86), with Cho/Cr as the most important feature. This study marks a significant milestone as the first of its kind to unveil radiographic occult glioma progression in post-treatment gliomas within 8 months of discovery. These findings underscore the utility of ML-based WB-MRS growth predictions, presenting a promising avenue for the guidance of early treatment decision-making. This research represents a crucial advancement in predicting the timing and location of glioblastoma recurrence, which can inform treatment decisions to improve patient outcomes.
Identifiants
pubmed: 39180640
doi: 10.1007/s11060-024-04812-1
pii: 10.1007/s11060-024-04812-1
doi:
Types de publication
Journal Article
Langues
eng
Sous-ensembles de citation
IM
Subventions
Organisme : NIH HHS
ID : R01CA172210, R01EB016064
Pays : United States
Organisme : NIH HHS
ID : R01CA172210, R01EB016064
Pays : United States
Organisme : NIH HHS
ID : R01CA172210, R01EB016064
Pays : United States
Organisme : NIH HHS
ID : R01CA172210, R01EB016064
Pays : United States
Organisme : NIH HHS
ID : R01CA172210, R01EB016064
Pays : United States
Organisme : NIH HHS
ID : R01CA172210, R01EB016064
Pays : United States
Organisme : NIH HHS
ID : R01CA172210, R01EB016064
Pays : United States
Organisme : NIH HHS
ID : R01CA172210, R01EB016064
Pays : United States
Organisme : NIH HHS
ID : R01CA172210, R01EB016064
Pays : United States
Organisme : NIH HHS
ID : R01CA172210, R01EB016064
Pays : United States
Organisme : NIH HHS
ID : R01CA172210, R01EB016064
Pays : United States
Organisme : NIH HHS
ID : R01CA172210, R01EB016064
Pays : United States
Organisme : NIH HHS
ID : R01CA172210, R01EB016064
Pays : United States
Organisme : NIH HHS
ID : R01CA172210, R01EB016064
Pays : United States
Organisme : NIH HHS
ID : R01CA172210, R01EB016064
Pays : United States
Organisme : NIH HHS
ID : R01CA172210, R01EB016064
Pays : United States
Organisme : NIH HHS
ID : R01CA172210, R01EB016064
Pays : United States
Organisme : NIH HHS
ID : R01CA172210, R01EB016064
Pays : United States
Organisme : NIH HHS
ID : R01CA172210, R01EB016064
Pays : United States
Organisme : NIH HHS
ID : R01CA172210, R01EB016064
Pays : United States
Informations de copyright
© 2024. The Author(s).
Références
Stupp R, Mason WP, van den Bent MJ et al (2005) Radiotherapy plus concomitant and adjuvant temozolomide for glioblastoma. N Engl J Med 352(10):987–996. https://doi.org/10.1056/NEJMoa043330
doi: 10.1056/NEJMoa043330
pubmed: 15758009
Shah AH, Mahavadi A, Di L et al (2020) Survival benefit of lobectomy for glioblastoma: moving towards radical supramaximal resection. J Neurooncol 148(3):501–508. https://doi.org/10.1007/s11060-020-03541-5
doi: 10.1007/s11060-020-03541-5
pubmed: 32627128
pmcid: 7442660
Louis DN (2006) Molecular pathology of malignant gliomas. Annu Rev Pathol 1:97–117. https://doi.org/10.1146/annurev.pathol.1.110304.100043
doi: 10.1146/annurev.pathol.1.110304.100043
pubmed: 18039109
Liu D, Liu Y, Hu X et al (2020) Alterations of white matter integrity associated with cognitive deficits in patients with glioma. Brain Behav 10(7):e01639. https://doi.org/10.1002/brb3.1639
doi: 10.1002/brb3.1639
pubmed: 32415731
pmcid: 7375068
Wen PY, Macdonald DR, Reardon DA et al (2010) Updated response assessment criteria for high-grade gliomas: response assessment in neuro-oncology working group. J Clin Oncol 28(11):1963–1972. https://doi.org/10.1200/jco.2009.26.3541
doi: 10.1200/jco.2009.26.3541
pubmed: 20231676
Maudsley AA, Domenig C, Govind V et al (2009) Mapping of brain metabolite distributions by volumetric proton MR spectroscopic imaging (MRSI). Magn Reson Med 61(3):548–559. https://doi.org/10.1002/mrm.21875
doi: 10.1002/mrm.21875
pubmed: 19111009
pmcid: 2724718
Gurbani S, Weinberg B, Cooper L et al (2019) The Brain Imaging Collaboration Suite (BrICS): a cloud platform for integrating whole-brain spectroscopic MRI into the radiation therapy planning workflow. Tomography 5(1):184–191. https://doi.org/10.18383/j.tom.2018.00028
doi: 10.18383/j.tom.2018.00028
pubmed: 30854456
pmcid: 6403040
Goryawala M, Saraf-Lavi E, Nagornaya N, Heros D, Komotar R, Maudsley AA (2020) The Association between whole-brain MR Spectroscopy and IDH mutation status in gliomas. J Neuroimaging 30(1):58–64. https://doi.org/10.1111/jon.12685
doi: 10.1111/jon.12685
pubmed: 31868291
Maudsley AA, Roy B, Gupta RK et al (2014) Association of metabolite concentrations and water diffusivity in normal appearing brain tissue with glioma grade. J Neuroimaging 24(6):585–589. https://doi.org/10.1111/jon.12063
doi: 10.1111/jon.12063
pubmed: 24251857
Dhermain FG, Hau P, Lanfermann H, Jacobs AH, van den Bent MJ (2010) Advanced MRI and PET imaging for assessment of treatment response in patients with gliomas. Lancet Neurol 9(9):906–920. https://doi.org/10.1016/S1474-4422(10)70181-2
doi: 10.1016/S1474-4422(10)70181-2
pubmed: 20705518
Bulik M, Jancalek R, Vanicek J, Skoch A, Mechl M (2013) Potential of MR spectroscopy for assessment of glioma grading. Clin Neurol Neurosurg 115(2):146–153. https://doi.org/10.1016/j.clineuro.2012.11.002
doi: 10.1016/j.clineuro.2012.11.002
pubmed: 23237636
Ekici S, Nye JA, Neill SG, Allen JW, Shu HK, Fleischer CC (2022) glutamine imaging: a new avenue for glioma management. AJNR Am J Neuroradiol 43(1):11–18. https://doi.org/10.3174/ajnr.A7333
doi: 10.3174/ajnr.A7333
pubmed: 34737183
pmcid: 8757564
El-Abtah ME, Wenke MR, Talati P et al (2022) Myo-inositol levels measured with MR spectroscopy can help predict failure of antiangiogenic treatment in recurrent glioblastoma. Radiology 302(2):410–418. https://doi.org/10.1148/radiol.2021210826
doi: 10.1148/radiol.2021210826
pubmed: 34751617
Papadopoulos MC, Saadoun S, Binder DK, Manley GT, Krishna S, Verkman AS (2004) Molecular mechanisms of brain tumor edema. Neuroscience 129(4):1011–1020. https://doi.org/10.1016/j.neuroscience.2004.05.044
doi: 10.1016/j.neuroscience.2004.05.044
pubmed: 15561416
Senders JT, Staples PC, Karhade AV et al (2018) Machine learning and neurosurgical outcome prediction: a systematic review. World Neurosurg 109:476-486.e1. https://doi.org/10.1016/j.wneu.2017.09.149
doi: 10.1016/j.wneu.2017.09.149
pubmed: 28986230
Maudsley AA, Darkazanli A, Alger JR et al (2006) Comprehensive processing, display and analysis for in vivo MR spectroscopic imaging. NMR Biomed 19(4):492–503. https://doi.org/10.1002/nbm.1025
doi: 10.1002/nbm.1025
pubmed: 16763967
pmcid: 2673915
Zhang X, Yan LF, Hu YC et al (2017) Optimizing a machine learning based glioma grading system using multi-parametric MRI histogram and texture features. Oncotarget 8(29):47816–47830. https://doi.org/10.18632/oncotarget.18001
doi: 10.18632/oncotarget.18001
pubmed: 28599282
pmcid: 5564607
Usinskiene J, Ulyte A, Bjørnerud A et al (2016) Optimal differentiation of high- and low-grade glioma and metastasis: a meta-analysis of perfusion, diffusion, and spectroscopy metrics. Neuroradiology 58(4):339–350. https://doi.org/10.1007/s00234-016-1642-9
doi: 10.1007/s00234-016-1642-9
pubmed: 26767528
Matsusue E, Fink JR, Rockhill JK, Ogawa T, Maravilla KR (2010) Distinction between glioma progression and post-radiation change by combined physiologic MR imaging. Neuroradiology 52(4):297–306. https://doi.org/10.1007/s00234-009-0613-9
doi: 10.1007/s00234-009-0613-9
pubmed: 19834699
Choi RY, Coyner AS, Kalpathy-Cramer J, Chiang MF, Campbell JP (2020) Introduction to machine learning, neural networks, and deep learning. Transl Vis Sci Technol 9(2):14. https://doi.org/10.1167/tvst.9.2.14
doi: 10.1167/tvst.9.2.14
pubmed: 32704420
pmcid: 7347027
Lasocki A, Gaillard F (2019) Non-contrast-enhancing tumor: a new frontier in glioblastoma research. AJNR Am J Neuroradiol 40(5):758–765. https://doi.org/10.3174/ajnr.A6025
doi: 10.3174/ajnr.A6025
pubmed: 30948373
pmcid: 7053910
Di L, Shah AH, Mahavadi A et al (2023) Radical supramaximal resection for newly diagnosed left-sided eloquent glioblastoma: safety and improved survival over gross-total resection. J Neurosurg 138(1):62–69. https://doi.org/10.3171/2022.3.JNS212399
doi: 10.3171/2022.3.JNS212399
pubmed: 35623362
Khalafallah AM, Rakovec M, Bettegowda C et al (2021) A crowdsourced consensus on supratotal resection versus gross total resection for anatomically distinct primary glioblastoma. Neurosurgery 89(4):712–719. https://doi.org/10.1093/neuros/nyab257
doi: 10.1093/neuros/nyab257
pubmed: 34320218
pmcid: 8440068
Zhong J, Huang V, Gurbani SS et al (2021) 3D whole-brain metabolite imaging to improve characterization of low-to-intermediate grade gliomas. J Neurooncol 153(2):303–311. https://doi.org/10.1007/s11060-021-03770-2
doi: 10.1007/s11060-021-03770-2
pubmed: 33983570
pmcid: 8237861
Morshed RA, Young JS, Hervey-Jumper SL, Berger MS (2019) The management of low-grade gliomas in adults. J Neurosurg Sci 63(4):450–457. https://doi.org/10.23736/S0390-5616.19.04701-5
doi: 10.23736/S0390-5616.19.04701-5
pubmed: 30916536
Shah AH, Madhavan K, Heros D et al (2011) The management of incidental low-grade gliomas using magnetic resonance imaging: systematic review and optimal treatment paradigm. Neurosurg Focus 31(6):E12. https://doi.org/10.3171/2011.9.FOCUS11219
doi: 10.3171/2011.9.FOCUS11219
pubmed: 22133169
Law M, Yang S, Wang H et al (2003) Glioma grading: sensitivity, specificity, and predictive values of perfusion MR imaging and proton MR spectroscopic imaging compared with conventional MR imaging. AJNR Am J Neuroradiol 24(10):1989–1998
pubmed: 14625221
pmcid: 8148904
Olman CA, Davachi L, Inati S (2009) Distortion and signal loss in medial temporal lobe. PloS one, 4(12):e8160. https://doi.org/10.1371/journal.pone.0008160