Oncotarget

Research Papers:

Pre-surgical connectome features predict IDH status in diffuse gliomas

Shelli R. Kesler _, Rebecca A. Harrison, Melissa L. Petersen, Vikram Rao, Hannah Dyson, Kristin Alfaro-Munoz, Shiao-Pei Weathers and John de Groot

PDF  |  HTML  |  How to cite  |  Order a Reprint

Oncotarget. 2019; 10:6484-6493. https://doi.org/10.18632/oncotarget.27301

Metrics: PDF 13 views  |   HTML 26 views  |   ?  


Abstract

Shelli R. Kesler1,2,*, Rebecca A. Harrison3,*, Melissa L. Petersen3, Vikram Rao1,2, Hannah Dyson3, Kristin Alfaro-Munoz3, Shiao-Pei Weathers3 and John de Groot3

1 Cancer Neuroscience Laboratory, School of Nursing, The University of Texas at Austin, Austin, Texas, USA

2 Department of Diagnostic Medicine, Dell School of Medicine, The University of Texas at Austin, Austin, Texas, USA

3 Department of Neuro-Oncology, The University of Texas MD Anderson Cancer Center, Houston, Texas, USA

* These authors contributed equally to this work

Correspondence to:

Shelli R. Kesler,email: srkesler@austin.utexas.edu

Keywords: MRI; IDH; glioma; machine learning; connectomics

Received: July 04, 2019     Accepted: October 21, 2019     Published: November 05, 2019

ABSTRACT

Background

Gliomas are the most common type of malignant brain tumor. Clinical outcomes depend on many factors including tumor molecular characteristics. Mutation of the isocitrate dehydrogenase (IDH) gene confers significant benefits in terms of survival and quality of life. Preoperative determination of IDH genotype can facilitate surgical planning, allow for novel clinical trial designs, and assist clinical counseling surrounding the individual patient’s disease.

Methods

In this study, we aimed to evaluate a novel approach for non-invasively predicting IDH status from conventional MRI via connectomics, a whole-brain network-based technique. We retrospectively extracted 93 connectome features from the preoperative, T1-weighted MRI data of 234 adult patients (148 IDH mutated) and evaluated the performance of four common machine learning models to predict IDH genotype.

Results

Area under the curve (AUC) of the receiver operator characteristic were 0.76 to 0.94 with random forest (RF) showing significantly higher performance (p < 0.01) than other algorithms. Feature selection schemes and the addition of age and tumor location did not change RF performance.

Conclusions

Our findings suggest that connectomics is a feasible approach for preoperatively predicting IDH genotype in patients with gliomas. Our results support prior evidence that RF is an ideal machine learning method for this area of research. Additionally, connectomics provides unique insights regarding potential mechanisms of tumor genotype on large-scale brain network organization.


Creative Commons License All site content, except where otherwise noted, is licensed under a Creative Commons Attribution 3.0 License.
PII: 27301