Non-invasive genotype prediction of chromosome 1p/19q co-deletion by development and validation of an MRI-based radiomics signature in lower-grade gliomas

Abstract

Our study highlighted that an MRI-based radiomics signature can effectively identify the 1p/19q co-deletion in histopathologically diagnosed lower-grade gliomas, thereby offering the potential to facilitate non-invasive molecular subtype prediction of gliomas.

Thin-Slice Magnetic Resonance Imaging-Based Radiomics Signature Predicts Chromosomal 1p/19q Co-deletion Status in Grade II and III Gliomas

“…However, such imaging characteristics may be limited in clinical application due to the unbalanced sensitivity and specificity, and highthroughput quantitative features are intensively needed to better illustrate the radiological divergences and further predict the 1p/19q status non-invasively. Previous radiomics studies using conventional MRI or advanced MRI sequences to predict 1p/19q status reached AUCs ranging from 0.68 to 0.96 (if reported, without distinguishing the training and validation dataset) (18)(19)(20)(21)(22)(23)(24)(25)(26), and our study displayed a competent result, with AUCs around 0.90 for the whole population and further elevated in IDH-mutated tumors, suggesting the capability of our signature for non-invasive 1p/19q detection. In addition, the 3D signature also displayed a balanced sensitivity and specificity, which compensated for the disequilibrium of visual characteristics.…”

“…studies of glioma have investigated the association between selected radiomics features and WHO grading, molecular characteristics, clinical manifestations, and patient prognosis (16)(17)(18). A few studies have involved the non-invasive prediction of 1p/19q status through a radiomics approach but display only moderate prediction value (18)(19)(20)(21)(22)(23)(24)(25)(26), and further investigation is needed to establish a reliable radiomics signature. In addition, previous studies were conducted using MR images acquired with diverse spacing (ranging from 1 to 5-6 mm for contrastenhanced T1 [CE-T1]-weighted images), and whether such differences would influence the performance of the prediction model remains to be explored.…”

Domain Mapping and Deep Learning from Multiple MRI Clinical Datasets for Prediction of Molecular Subtypes in Low Grade Gliomas

“…Zhou et al [7] used hand-crafted features such as histograms, shape and texture from data that was collected from single institution combined with age information for a random forest classifier. Han et al [5] used an analysis to generate radiomics signature by extracting 647 MRI-based features from T2-MRIs and side information of patients. Van der Voort et al [11] used support vector machine classifier to extract features from T1 and T2-MRI along with age and sex information on 284 patients and validated results on 129 patients from TCIA.…”

“…Zhou et al [7] used histogram, shape and texture features combined with age information to a random forest algorithm for IDH mutation and 1p/19q codeletion prediction. Han et al [5] performed an analysis to generate radiomics signature by extracting 647 MRI based features for predicting 1p/19q codeletion status. Another radiomics based approach was studied by Yu et al [10] on IDH mutation prediction.…”

“…The molecular information would be of practical value since oligodendrogliomas harbor better prognosis than the other LGG subtypes and also seem to be more sensitive to oncological treatment [3,4]. This molecular information requires a tissue diagnosis, but recently several advanced machine learning techniques have been shown to predict molecular subtypes in gliomas based upon preoperative imaging [5][6][7][8]. Non-invasive diagnostic tools are attractive in identification since it may assist in prognostication and would significantly enhance patient counseling and shared decision making.…”