IDH mutation status trumps the Pignatti risk score as a prognostic marker in low-grade gliomas

Abstract Purpose Many low-grade gliomas (LGG) harbor isocitrate dehydrogenase (IDH) mutations. Although IDH mutation is known to be epileptogenic, the rate of refractory seizures in LGG with IDH mutation vs wild-type had not been previously compared. We therefore compared seizure pharmacoresistance in IDH-mutated and wild-type LGGs. Methods Single-institution retrospective study of patients with histologic proven LGG, known IDH mutation status, seizures, and ≥ 2 neurology clinic encounters. Seizure history was followed until histological high-grade transformation or death. Seizures requiring ≥ 2 changes in anti-epileptic drugs were considered pharmacoresistant. Incidence rates of pharmacoresistant seizures were estimated using competing risks methodology. Results Of 135 patients, 25 patients (19%) had LGGs classified as IDH wild-type. Of those with IDH mutation, 104 (94.5%) were IDH1 R132H; only six were IDH2 R172K. 120 patients (89%) had tumor resection and 14 (10%) had biopsy. Initial post-surgical management included observation (64%), concurrent chemoradiation (23%), chemotherapy alone (9%), and radiotherapy alone (4%). Seizures became pharmacoresistant in 24 IDH-mutated patients (22%) and in 3 IDH wild-type patients (12%). The 4-year cumulative incidence of intractable seizures was 17.6% (95% CI: 10.6%-25.9%) in IDH-mutated and 11% (95% CI: 1.3%-32.6%) in IDH wild-type LGG (Gray’s P-value= 0.26). Conclusions 22% of the IDH-mutated patients developed pharmacoresistant seizures, compared to 12% of the IDH wild-type tumors.The likelihood of developing pharmacoresistant seizures in patients with LGG-related epilepsy is independent to IDH mutation status, however, IDH-mutated tumors were approximately twice as likely to experience LGG-related pharmacoresistant seizures.

Download Full-text

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

Brain Sciences ◽

10.3390/brainsci10070463 ◽

2020 ◽

Vol 10 (7) ◽

pp. 463 ◽

Cited By ~ 1

Author(s):

Muhaddisa Barat Ali ◽

Irene Yu-Hua Gu ◽

Mitchel S. Berger ◽

Johan Pallud ◽

Derek Southwell ◽

...

Keyword(s):

Molecular Subtypes ◽

Negative Impact ◽

Molecular Classification ◽

Tumor Segmentation ◽

Test Accuracy ◽

Low Grade ◽

Idh Mutation ◽

Molecular Type ◽

Low Grade Gliomas ◽

Domain Mapping

Brain tumors, such as low grade gliomas (LGG), are molecularly classified which require the surgical collection of tissue samples. The pre-surgical or non-operative identification of LGG molecular type could improve patient counseling and treatment decisions. However, radiographic approaches to LGG molecular classification are currently lacking, as clinicians are unable to reliably predict LGG molecular type using magnetic resonance imaging (MRI) studies. Machine learning approaches may improve the prediction of LGG molecular classification through MRI, however, the development of these techniques requires large annotated data sets. Merging clinical data from different hospitals to increase case numbers is needed, but the use of different scanners and settings can affect the results and simply combining them into a large dataset often have a significant negative impact on performance. This calls for efficient domain adaption methods. Despite some previous studies on domain adaptations, mapping MR images from different datasets to a common domain without affecting subtitle molecular-biomarker information has not been reported yet. In this paper, we propose an effective domain adaptation method based on Cycle Generative Adversarial Network (CycleGAN). The dataset is further enlarged by augmenting more MRIs using another GAN approach. Further, to tackle the issue of brain tumor segmentation that requires time and anatomical expertise to put exact boundary around the tumor, we have used a tight bounding box as a strategy. Finally, an efficient deep feature learning method, multi-stream convolutional autoencoder (CAE) and feature fusion, is proposed for the prediction of molecular subtypes (1p/19q-codeletion and IDH mutation). The experiments were conducted on a total of 161 patients consisting of FLAIR and T1 weighted with contrast enhanced (T1ce) MRIs from two different institutions in the USA and France. The proposed scheme is shown to achieve the test accuracy of 74 . 81 % on 1p/19q codeletion and 81 . 19 % on IDH mutation, with marked improvement over the results obtained without domain mapping. This approach is also shown to have comparable performance to several state-of-the-art methods.

Download Full-text

Classification of Brain Tumor IDH Status using MRI and Deep Learning

10.1101/757344 ◽

2019 ◽

Author(s):

Sahil Nalawade ◽

Gowtham Murugesan ◽

Maryam Vejdani-Jahromi ◽

Ryan A. Fisicaro ◽

Chandan Ganesh Bangalore Yogananda ◽

...

Keyword(s):

Deep Learning ◽

Classification Accuracy ◽

Test Accuracy ◽

Low Grade ◽

Idh Mutation ◽

Validation Data ◽

Mutation Status ◽

Radiologic Imaging ◽

Address Data

AbstractIsocitrate dehydrogenase (IDH) mutation status is an important marker in glioma diagnosis and therapy. We propose a novel automated pipeline for predicting IDH status noninvasively using deep learning and T2-weighted (T2w) MR images with minimal preprocessing (N4 bias correction and normalization to zero mean and unit variance). T2w MRI and genomic data were obtained from The Cancer Imaging Archive dataset (TCIA) for 260 subjects (120 High grade and 140 Low grade gliomas). A fully automated 2D densely connected model was trained to classify IDH mutation status on 208 subjects and tested on another held-out set of 52 subjects, using 5-fold cross validation. Data leakage was avoided by ensuring subject separation during the slice-wise randomization. Mean classification accuracy of 90.5% was achieved for each axial slice in predicting the three classes of no tumor, IDH mutated and IDH wild-type. Test accuracy of 83.8% was achieved in predicting IDH mutation status for individual subjects on the test dataset of 52 subjects. We demonstrate a deep learning method to predict IDH mutation status using T2w MRI alone. Radiologic imaging studies using deep learning methods must address data leakage (subject duplication) in the randomization process to avoid upward bias in the reported classification accuracy.

Download Full-text

P10.21 2-hydroxyglutarate as a biomarker for IDH mutation in low grade gliomas

Neuro-Oncology ◽

10.1093/neuonc/nox036.339 ◽

2017 ◽

Vol 19 (suppl_3) ◽

pp. iii89-iii90

Author(s):

R. Nejad ◽

H. Sim ◽

K. Aldape ◽

W. Mason ◽

M. Bernstein ◽

...

Keyword(s):

Low Grade ◽

Idh Mutation ◽

Low Grade Gliomas

Download Full-text

IDH mutation and 1p19q codeletion distinguish two radiological patterns of diffuse low-grade gliomas

Journal of Neuro-Oncology ◽

10.1007/s11060-017-2421-0 ◽

2017 ◽

Vol 133 (1) ◽

pp. 37-45 ◽

Cited By ~ 29

Author(s):

Amélie Darlix ◽

Jérémy Deverdun ◽

Nicolas Menjot de Champfleur ◽

Florence Castan ◽

Sonia Zouaoui ◽

...

Keyword(s):

Low Grade ◽

Idh Mutation ◽

Low Grade Gliomas ◽

1P19q Codeletion ◽

Radiological Patterns

Download Full-text

Identification of Autophagy-Related Prognostic Signature and Analysis of Immune Cell Infiltration in Low-Grade Gliomas

BioMed Research International ◽

10.1155/2021/7918693 ◽

2021 ◽

Vol 2021 ◽

pp. 1-24

Author(s):

Qingli Quan ◽

Xinxin Xiong ◽

Shanyun Wu ◽

Meixing Yu

Keyword(s):

Risk Score ◽

Immune Cells ◽

Roc Analysis ◽

Immune Cell ◽

Time Dependent ◽

Low Grade ◽

Immune Cell Infiltration ◽

Prognostic Signature ◽

Low Grade Gliomas ◽

Key Genes

Autophagy plays an important role in cancer. Many studies have demonstrated that autophagy-related genes (ARGs) can act as a prognostic signature for some cancers, but little has been known in low-grade gliomas (LGG). In our study, we aimed to establish a prognostical model based on ARGs and find prognostic risk-related key genes in LGG. In the present study, a prognostic signature was constructed based on a total of 8 ARGs (MAPK8IP1, EEF2, GRID2, BIRC5, DLC1, NAMPT, GRID1, and TP73). It was revealed that the higher the risk score, the worse was the prognosis. Time-dependent ROC analysis showed that the risk score could precisely predict the prognosis of LGG patients. Additionally, four key genes (TGFβ2, SERPING1, SERPINE1, and TIMP1) were identified and found significantly associated with OS of LGG patients. Besides, they were also discovered to be strongly related to six types of immune cells which infiltrated in LGG tumor. Taken together, the present study demonstrated the promising potential of the ARG risk score formula as an independent factor for LGG prediction. It also provided the autophagy-related signature of prognosis and potential therapeutic targets for the treatment of LGG.

Download Full-text