EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Metabolic signatures derived from whole-brain MR-spectroscopy identify early tumor progression in high-grade gliomas using machine learning

Cameron A Rivera, Shovan Bhatia, Alexis A Morell, Lekhaj C Daggubati, Martin A Merenzon, Sulaiman A Sheriff, Evan Luther, Jay Chandar, Adam S Levy, Ashley R Metzler, Chandler N Berke, Mohammed Goryawala und 8 weitere

Journal of Neuro-Oncology · 2024-12 · 170 (3) · 579-589

DOI: 10.1007/s11060-024-04812-1 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Neuro-Oncology, online erschienen am 24.08.2024. Verfasst von Cameron A Rivera und 19 weiteren Autor:innen. Beteiligte Einrichtung: Department of Neurosurgery, University of Miami Miller School of Medicine, Miami, FL, USA. car909@med.miami.edu. Hauptthemen laut MEDLINE: Glioma, Brain Neoplasms, Machine Learning, Disease Progression. Die Arbeit wurde bisher 18-mal zitiert (Crossref) und nennt 28 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: NCI NIH HHS, NIH HHS, NIH.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Purpose

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.

Methods

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.

Results

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.

Conclusions

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.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

GliomaBrain NeoplasmsMachine LearningDisease ProgressionMagnetic Resonance SpectroscopyHumansMaleFemaleMiddle AgedAdultAgedBrainNeoplasm Grading

Stichwörter der Autor:innen

Binary classificationGlioblastomaMachine learningMulticlass classifierPredictive modelsWhole-brain magnetic resonance spectroscopy (WB-MRS)

Autor:innen und Institutionen

  • Department of Neurosurgery, University of Miami Miller School of Medicine, Miami, FL, USA. car909@med.miami.edu.
  • Department of Neurosurgery, University of Miami Miller School of Medicine, Miami, FL, USA.
  • Department of Neurosurgery, University of Pittsburgh School of Medicine, Pittsburgh, PA, USA.
  • Department of Critical Care, University of Pittsburgh School of Medicine, Pittsburgh, PA, USA.
  • Surgical Neuro-Oncology, District of Columbia, George Washington Medical Faculty Associates, Washington, USA.
  • Department of Neurosurgery, Yale University School of Medicine, New Haven, CT, USA.

Förderung

  • NCI NIH HHS
  • NIH HHS
  • NIH

Zitieren

APA

Rivera, C. A., Bhatia, S., Morell, A. A., Daggubati, L. C., Merenzon, M. A., Sheriff, S. A., Luther, E., Chandar, J., S Levy, A., Metzler, A. R., Berke, C. N., Goryawala, M., Mellon, E. A., Bhatia, R. G., Nagornaya, N., Saigal, G., I de la Fuente, M., Komotar, R. J., Ivan, M. E., & Shah, A. H. (2024). Metabolic signatures derived from whole-brain MR-spectroscopy identify early tumor progression in high-grade gliomas using machine learning. Journal of Neuro-Oncology, 170(3), 579-589. https://doi.org/10.1007/s11060-024-04812-1

Vancouver

Rivera CA, Bhatia S, Morell AA, Daggubati LC, Merenzon MA, Sheriff SA, et al. Metabolic signatures derived from whole-brain MR-spectroscopy identify early tumor progression in high-grade gliomas using machine learning. Journal of Neuro-Oncology. 2024;170(3):579-589. doi:10.1007/s11060-024-04812-1

BibTeX
@article{Rivera2024,
  author = {Rivera, Cameron A and Bhatia, Shovan and Morell, Alexis A and Daggubati, Lekhaj C and Merenzon, Martin A and Sheriff, Sulaiman A and Luther, Evan and Chandar, Jay and S Levy, Adam and Metzler, Ashley R and Berke, Chandler N and Goryawala, Mohammed and Mellon, Eric A and Bhatia, Rita G and Nagornaya, Natalya and Saigal, Gaurav and I de la Fuente, Macarena and Komotar, Ricardo J and Ivan, Michael E and Shah, Ashish H},
  title = {Metabolic signatures derived from whole-brain MR-spectroscopy identify early tumor progression in high-grade gliomas using machine learning},
  journal = {Journal of Neuro-Oncology},
  year = {2024},
  volume = {170},
  number = {3},
  pages = {579--589},
  doi = {10.1007/s11060-024-04812-1}
}
Kostenlos starten →