Prognostic and Predictive Value of Integrated Qualitative and Quantitative Magnetic Resonance Imaging Analysis in Glioblastoma

Abstract

Glioblastoma (GBM) is the most malignant primary brain tumor for which no curative treatment options exist. Non-invasive qualitative (Visually Accessible Rembrandt Images (VASARI)) and quantitative (radiomics) imaging features to predict prognosis and clinically relevant markers for GBM patients are needed to guide clinicians. A retrospective analysis of GBM patients in two neuro-oncology centers was conducted. The multimodal Cox-regression model to predict overall survival (OS) was developed using clinical features with VASARI and radiomics features in isocitrate dehydrogenase (IDH)-wild type GBM. Predictive models for IDH-mutation, 06-methylguanine-DNA-methyltransferase (MGMT)-methylation and epidermal growth factor receptor (EGFR) amplification using imaging features were developed using machine learning. The performance of the prognostic model improved upon addition of clinical, VASARI and radiomics features, for which the combined model performed best. This could be reproduced after external validation (C-index 0.711 95% CI 0.64-0.78) and used to stratify Kaplan-Meijer curves in two survival groups (p-value < 0.001). The predictive models performed significantly in the external validation for EGFR amplification (area-under-the-curve (AUC) 0.707, 95% CI 0.582-8.25) and MGMT-methylation (AUC 0.667, 95% CI 0.522-0.82) but not for IDH-mutation (AUC 0.695, 95% CI 0.436-0.927). The integrated clinical and imaging prognostic model was shown to be robust and of potential clinical relevance. The prediction of molecular markers showed promising results in the training set but could not be validated after external validation in a clinically relevant manner. Overall, these results show the potential of combining clinical features with imaging features for prognostic and predictive models in GBM, but further optimization and larger prospective studies are warranted.

Keywords: MRI; glioblastoma; machine learning; prediction; prognosis; radiomics; survival.

Figure 1

Performance of prognostic models: (A) visualization of C-index for all prognostic models (including 95% CI) in the training (n = 95) and validation cohort (n = 38); (B) Kaplan–Meier curve of integrated radiomics, Visually Accessible Rembrandt Images (VASARI) and clinical model (Model 7) in the training cohort and (C) validation cohort. Low- and high-risk groups (blue and red lines, respectively) cut-off values were determined by set cut-off (75th percentile) in the training cohort. The solid lines represent the observed survival curves, the dashed the corresponding predicted survival curves.

Figure 2

Performance of predictive models: (A) Area-under-the-curve (AUC) values and corresponding 95% confidence intervals of different predictive models in the testing cohort and (B) in the validation cohort; (C) receiver operating characteristic (ROC)-curves of combined VASARI and radiomics model predictive performance in external validation set.

Figure 3

Quantitative and qualitative imaging analysis pipeline.