Hybrid 11C-MET PET/MRI Combined With “Machine Learning” in Glioma Diagnosis According to the Revised Glioma WHO Classification 2016

Abstract

Purpose With the advent of the revised WHO classification from 2016, molecular features, including isocitrate dehydrogenase (IDH) mutation have become important in glioma subtyping. This pilot trial analyzed the potential for 11C-methionine (MET) PET/MRI in classifying glioma according to the revised WHO classification using a machine learning model. Methods Patients with newly diagnosed WHO grade II–IV glioma underwent preoperative MET-PET/MRI imaging. Patients were retrospectively divided into four groups: IDH wild-type glioblastoma (GBM), IDH wild-type grade II/III glioma (GII/III-IDHwt), IDH mutant grade II/III glioma with codeletion of 1p19q (GII/III-IDHmut1p19qcod) or without 1p19q-codeletion (GII/III-IDHmut1p19qnc). Within each group, the maximum tumor-to-brain-ratio (TBRmax) of MET-uptake was calculated. To gain generalizable implications from our data, we made use of a machine learning algorithm based on a development and validation subcohort. A support vector machine model was fit to the development subcohort and evaluated on the validation subcohort. Receiver operating characteristic (ROC) analysis served as metric to assess model performance. Results Of a total of 259 patients, 39 patients met the inclusion criteria. TBRmax was highest in the GBM cohort (TBRmax 3.83 ± 1.30) and significantly higher (P = 0.004) compared to GII/III-IDHmut1p19qnc group, where TBRmax was lowest (TBRmax 2.05 ± 0.94). ROC analysis showed poor AUC for glioma subtyping (AUC 0.62) and high AUC of 0.79 for predicting IDH status. In the GII/III-IDHmut1p19qcod group, TBR values were slightly higher than in the IDHmut1p19qnc group. Conclusions MET-PET/MRI imaging in pre-operatively classifying glioma entities appears useful for the assessment of IDH status. However, a larger trial is needed prior to translation into the clinical routine.

A Nomogram Modeling 11C-MET PET/CT and Clinical Features in Glioma Helps Predict IDH Mutation

“…PET imaging was suggested to be informative for preoperatively differentiating gliomas according to 2016 WHO classification, particularly for differentiating IDH-wildtype and IDH-mutant tumors ( 19 ). A study of hybrid 11 C-MET PET/MRI imaging including 39 glioma patients described that ROC analysis of TNRmax had a high AUC of 0.79 for predicting IDH status ( 16 ). Another study retrospectively evaluated 109 patients with newly diagnosed glioma also indicated that 11 C-MET uptake was negatively correlated with IDH mutational status.…”

“…The results are inspiring since T2-weighted MR imaging is widely available and routinely performed in the assessment of gliomas. Some studies have explored the relationship between amino acid uptake characteristics of gliomas and IDH mutation status (15)(16)(17)(18)(19). We aimed to develop a novel and convenient statistical model that combines PET features and clinical factors for an IDH-predictive signature.…”

The diagnostic value of lower glucose consumption for IDH1 mutated gliomas on FDG-PET

“…PET images could reveal metabolic differences among gliomas, providing a possible non-invasive preoperative molecular diagnosis with FDG, 11 C-choline(CHO), or 11 C-methionine(MET) [21,22]. MET-PET/MRI has been reported to predict the IDH mutation of gliomas, by calculating the maximum tumor-brain ratio of MET uptake, and the receiver operating characteristic (ROC) analysis showed high AUC of 0.79 [23]. The biological tumor volume of early glioma differs from the standard 20-40 min FDG-PET to determine IDH status [24].…”

Die Wichtigkeit des „Machine Learning“ und der Texturanalyse in der Onkologie anhand der Fusionsbildgebung

“…Mittels der Texturanalyse konnte bereits die ärztliche Einschätzung und bildbasierte Charakterisierung des Tu-mors reproduziert werden. Dennoch sind größere Fallgruppen essenziell um diese Methodik zu verbessern sowie auf Basis des guten Kontrasts von Tumor-zu-Hintergrund der gängigen Radiotracer verlässliche semi-maschinelle oder maschinelle Segmentations-Modelle/-Programme zu entwickeln und zu etablieren [1,9].…”