Dataset Information

Magnetic resonance radiomics features and prognosticators in different molecular subtypes of pediatric Medulloblastoma.

ABSTRACT:

Purpose

Medulloblastoma (MB) is a highly malignant pediatric brain tumor. In the latest classification, medulloblastoma is divided into four distinct groups: wingless (WNT), sonic hedgehog (SHH), Group 3, and Group 4. We analyzed the magnetic resonance imaging radiomics features to find the imaging surrogates of the 4 molecular subgroups of MB.

Material and methods

Frozen tissue, imaging data, and clinical data of 38 patients with medulloblastoma were included from Taipei Medical University Hospital and Taipei Veterans General Hospital. Molecular clustering was performed based on the gene expression level of 22 subgroup-specific signature genes. A total 253 magnetic resonance imaging radiomic features were generated from each subject for comparison between different molecular subgroups.

Results

Our cohort consisted of 7 (18.4%) patients with WNT medulloblastoma, 12 (31.6%) with SHH tumor, 8 (21.1%) with Group 3 tumor, and 11 (28.9%) with Group 4 tumor. 8 radiomics gray-level co-occurrence matrix texture (GLCM) features were significantly different between 4 molecular subgroups of MB. In addition, for tumors with higher values in a gray-level run length matrix feature-Short Run Low Gray-Level Emphasis, patients have shorter survival times than patients with low values of this feature (p = 0.04). The receiver operating characteristic analysis revealed optimal performance of the preliminary prediction model based on GLCM features for predicting WNT, Group 3, and Group 4 MB (area under the curve = 0.82, 0.72, and 0.78, respectively).

Conclusion

The preliminary result revealed that 8 contrast-enhanced T1-weighted imaging texture features were significantly different between 4 molecular subgroups of MB. Together with the prediction models, the radiomics features may provide suggestions for stratifying patients with MB into different risk groups.

SUBMITTER: Chang FC

PROVIDER: S-EPMC8321137 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:Purpose To construct a machine learning model based on radiomics of multiparametric magnetic resonance imaging (MRI) combined with clinical parameters for predicting Sonic Hedgehog (SHH) and Group 4 (G4) molecular subtypes of pediatric medulloblastoma (MB). Methods The preoperative MRI images and clinical data of 95 patients with MB were retrospectively analyzed, including 47 cases of SHH subtype and 48 cases of G4 subtype. Radiomic features were extracted from T1-weighted imaging (T1), contrast-enhanced T1 weighted imaging (T1c), T2-weighted imaging (T2), T2 fluid-attenuated inversion recovery imaging (T2FLAIR), and apparent diffusion coefficient (ADC) maps, using variance thresholding, SelectKBest, and Least Absolute Shrinkage and Selection Operator (LASSO) regression algorithms. The optimal features were filtered using LASSO regression, and a logistic regression (LR) algorithm was used to build a machine learning model. The receiver operator characteristic (ROC) curve was plotted to evaluate the prediction accuracy, and verified by its calibration, decision and nomogram. The Delong test was used to compare the differences between different models. Results A total of 17 optimal features, with non-redundancy and high correlation, were selected from 7,045 radiomics features, and used to build an LR model. The model showed a classification accuracy with an under the curve (AUC) of 0.960 (95% CI: 0.871−1.000) in the training cohort and 0.751 (95% CI: 0.587−0.915) in the testing cohort, respectively. The location of the tumor, pathological type, and hydrocephalus status of the two subtypes of patients differed significantly (p < 0.05). When combining radiomics features and clinical parameters to construct the combined prediction model, the AUC improved to 0.965 (95% CI: 0.898−1.000) in the training cohort and 0.849 (95% CI: 0.695−1.000) in the testing cohort, respectively. There was a significant difference in the prediction accuracy, as measured by AUC, between the testing cohorts of the two prediction models, which was confirmed by Delong’s test (p = 0.0144). Decision curves and nomogram further validate that the combined model can achieve net benefits in clinical work. Conclusion The combined prediction model, constructed based on radiomics of multiparametric MRI and clinical parameters can potentially provide a non-invasive clinical approach to predict SHH and G4 molecular subtypes of MB preoperatively.

Project description:BackgroundColorectal adenoma develops into cancer with the accumulation of genetic and epigenetic changes. We studied the underlying molecular and clinicopathological features to better understand the heterogeneity of colorectal neoplasms (CRNs).MethodsWe evaluated both genetic (mutations of KRAS, BRAF, TP53, and PIK3CA, and microsatellite instability [MSI]) and epigenetic (methylation status of nine genes or sequences, including the CpG island methylator phenotype [CIMP] markers) alterations in 158 CRNs including 56 polypoid neoplasms (PNs), 25 granular type laterally spreading tumors (LST-Gs), 48 non-granular type LSTs (LST-NGs), 19 depressed neoplasms (DNs) and 10 small flat-elevated neoplasms (S-FNs) on the basis of macroscopic appearance.ResultsS-FNs showed few molecular changes except SFRP1 methylation. Significant differences in the frequency of KRAS mutations were observed among subtypes (68% for LST-Gs, 36% for PNs, 16% for DNs and 6% for LST-NGs) (P<0.001). By contrast, the frequency of TP53 mutation was higher in DNs than PNs or LST-Gs (32% vs. 5% or 0%, respectively) (P<0.007). We also observed significant differences in the frequency of CIMP between LST-Gs and LST-NGs or PNs (32% vs. 6% or 5%, respectively) (P<0.005). Moreover, the methylation level of LINE-1 was significantly lower in DNs or LST-Gs than in PNs (58.3% or 60.5% vs. 63.2%, P<0.05). PIK3CA mutations were detected only in LSTs. Finally, multivariate analyses showed that macroscopic morphologies were significantly associated with an increased risk of molecular changes (PN or LST-G for KRAS mutation, odds ratio [OR] 9.11; LST-NG or DN for TP53 mutation, OR 5.30; LST-G for PIK3CA mutation, OR 26.53; LST-G or DN for LINE-1 hypomethylation, OR 3.41).ConclusionWe demonstrated that CRNs could be classified into five macroscopic subtypes according to clinicopathological and molecular differences, suggesting that different mechanisms are involved in the pathogenesis of colorectal tumorigenesis.

Project description:BackgroundLymphoma is a common malignant tumor in children. The pathologic subtyping of lymphoma is high complex, and the treatment options vary. The different pathologic subtypes of lymphomas have no significant differences on computed tomography (CT) images. As it is a hematologic disease, patients with lymphoma often show abnormalities in the spleen, and so the aim of this study was to construct a model for differentiating Burkitt lymphoma (BL) from lymphoblastic lymphoma through the extraction of radiomic features of the spleen from CT images. This could provide an efficient, noninvasive method that can differentiate the common pathological subtypes in patients with pediatric lymphoma.MethodsThe clinical data and imaging data of 48 patients with lymphoblastic lymphoma and 61 patients with BL were retrospectively analyzed. The dataset was divided into a training set (n=76) and a test set (n=33) through complete randomization. Radiomics features of the spleen were separately extracted from CT images in the noncontrast enhanced, arterial, and venous phases. These phase-specific features were integrated to construct fusion models. Three classifiers, quadratic discriminant analysis (QDA), logistic regression (LR), and support vector machine (SVM), were employed to build the models.ResultsThe fusion model exhibited superior performance compared to individual models. There was no significant difference between the fusion models constructed by QDA and LR in either the training set or the test set. Among the four fusion models constructed with the SVM classifier, SVM_4 emerged as the best performing model. The area under the curve, sensitivity, specificity, and F1-score of the SVM_4 model were 0.967 [95% confidence interval (CI): 0.935-0.998], 0.86, 0.97, and 0.913 in the training set, respectively, and 0.754 (95% CI: 0.584-0.924), 0.611, 0.867, and 0.71 in the test set, respectively.ConclusionsThe radiomics features of the spleen demonstrated the capability to distinguish between the two most common lymphoma subtypes in pediatric patients. This noninvasive approach holds promise for efficient and accurate discrimination.