MRI-based radiomics for prognosis of pediatric diffuse intrinsic pontine glioma: an international study. Issue 1 (5th March 2021)

• Record Type: Journal Article
• Authors: Tam, Lydia T; Yeom, Kristen W; Wright, Jason N; Jaju, Alok; Radmanesh, Alireza; Han, Michelle; Toescu, Sebastian; Maleki, Maryam; Chen, Eric; Campion, Andrew; Lai, Hollie A; Eghbal, Azam A; Oztekin, Ozgur; Mankad, Kshitij; Hargrave, Darren; Jacques, Thomas S; Goetti, Robert; Lober, Robert M; Cheshier, Samuel H; Napel, Sandy; Said, Mourad; Aquilina, Kristian; Ho, Chang Y; Monje, Michelle; Vitanza, Nicholas A; Mattonen, Sarah A
• Is Part Of: Neuro-oncology advances. Volume 3:Issue 1(2021)
• Publication Date: 2021-03-05
• Subjects: diffuse intrinsic pontine gliomas -- diffuse midline glioma -- H3K27M-mutant -- machine learning -- magnetic resonance imaging -- radiomics; 616.99481
• DOI: 10.1093/noajnl/vdab042
• Languages: English
• ISSNs: 2632-2498

Abstract

Abstract: Background: Diffuse intrinsic pontine gliomas (DIPGs) are lethal pediatric brain tumors. Presently, MRI is the mainstay of disease diagnosis and surveillance. We identify clinically significant computational features from MRI and create a prognostic machine learning model. Methods: We isolated tumor volumes of T1-post-contrast (T1) and T2-weighted (T2) MRIs from 177 treatment-naïve DIPG patients from an international cohort for model training and testing. The Quantitative Image Feature Pipeline and PyRadiomics was used for feature extraction. Ten-fold cross-validation of least absolute shrinkage and selection operator Cox regression selected optimal features to predict overall survival in the training dataset and tested in the independent testing dataset. We analyzed model performance using clinical variables (age at diagnosis and sex) only, radiomics only, and radiomics plus clinical variables. Results: All selected features were intensity and texture-based on the wavelet-filtered images (3 T1 gray-level co-occurrence matrix (GLCM) texture features, T2 GLCM texture feature, and T2 first-order mean). This multivariable Cox model demonstrated a concordance of 0.68 (95% CI: 0.61–0.74) in the training dataset, significantly outperforming the clinical-only model ( C = 0.57 [95% CI: 0.49–0.64]). Adding clinical features to radiomics slightly improved performance ( C = 0.70 [95% CI: 0.64–0.77]). The combined radiomics and clinical model was validated in the independent testing dataset ( C = 0.59 [95% CI: 0.51–0.67], Noether's test P = .02). Conclusions: In this international study, we demonstrate the use of radiomic signatures to create a machine learning model for DIPG prognostication. Standardized, quantitative approaches that objectively measure DIPG changes, including computational MRI evaluation, could offer new approaches to assessing tumor phenotype and serve a future role for optimizing clinical trial eligibility and tumor surveillance.