Radiomic features from multiparametric magnetic resonance imaging predict molecular subgroups of pediatric low-grade gliomas.

BACKGROUND: We aimed to develop machine learning models for prediction of molecular subgroups (low-risk group and intermediate/high-risk group) and molecular marker (KIAA1549-BRAF fusion) of pediatric low-grade gliomas (PLGGs) based on radiomic features extracted from multiparametric MRI. METHODS: 61 patients with PLGGs were included in this retrospective study, which were divided into a training set and an internal validation set at a ratio of 2:1 based on the molecular subgroups or the molecular marker. The patients were classified into low-risk and intermediate/high-risk groups, BRAF fusion positive and negative groups, respectively. We extracted 5929 radiomic features from multiparametric MRI. Thereafter, we removed redundant features, trained random forest models on the training set for predicting the molecular subgroups or the molecular marker, and validated their performance on the internal validation set. The performance of the prediction model was verified by 3-fold cross-validation. RESULTS: We constructed the classification model differentiating low-risk PLGGs from intermediate/high-risk PLGGs using 4 relevant features, with an AUC of 0.833 and an accuracy of 76.2% in the internal validation set. In the prediction model for predicting KIAA1549-BRAF fusion using 4 relevant features, an AUC of 0.818 and an accuracy of 81.0% were achieved in the internal validation set. CONCLUSIONS: The current study demonstrates that MRI radiomics is able to predict molecular subgroups of PLGGs and KIAA1549-BRAF fusion with satisfying sensitivity. TRIAL REGISTRATION: This study was retrospectively registered at clinicaltrials.gov (NCT04217018).

Image-based deep learning identifies glioblastoma risk groups with genomic and transcriptomic heterogeneity: a multi-center study.

OBJECTIVES: To develop and validate a deep learning imaging signature (DLIS) for risk stratification in patients with multiforme (GBM), and to investigate the biological pathways and genetic alterations underlying the DLIS. METHODS: The DLIS was developed from multi-parametric MRI based on a training set (n = 600) and validated on an internal validation set (n = 164), an external test set 1 (n = 100), an external test set 2 (n = 161), and a public TCIA set (n = 88). A co-profiling framework based on a radiogenomics analysis dataset (n = 127) using multiscale high-dimensional data, including imaging, transcriptome, and genome, was established to uncover the biological pathways and genetic alterations underpinning the DLIS. RESULTS: The DLIS was associated with survival (log-rank p < 0.001) and was an independent predictor (p < 0.001). The integrated nomogram incorporating the DLIS achieved improved C indices than the clinicomolecular nomogram (net reclassification improvement 0.39, p < 0.001). DLIS significantly correlated with core pathways of GBM (apoptosis and cell cycle-related P53 and RB pathways, and cell proliferation-related RTK pathway), as well as key genetic alterations (del_CDNK2A). The prognostic value of DLIS-correlated genes was externally confirmed on TCGA/CGGA sets (p < 0.01). CONCLUSIONS: Our study offers a biologically interpretable deep learning predictor of survival outcomes in patients with GBM, which is crucial for better understanding GBM patient's prognosis and guiding individualized treatment. KEY POINTS: • MRI-based deep learning imaging signature (DLIS) stratifies GBM into risk groups with distinct molecular characteristics. • DLIS is associated with P53, RB, and RTK pathways and del_CDNK2A mutation. • The prognostic value of DLIS-correlated pathway genes is externally demonstrated.

Predicting 1p/19q co-deletion status from magnetic resonance imaging using deep learning in adult-type diffuse lower-grade gliomas: a discovery and validation study.

Determination of 1p/19q co-deletion status is important for the classification, prognostication, and personalized therapy in diffuse lower-grade gliomas (LGG). We developed and validated a deep learning imaging signature (DLIS) from preoperative magnetic resonance imaging (MRI) for predicting the 1p/19q status in patients with LGG. The DLIS was constructed on a training dataset (n = 330) and validated on both an internal validation dataset (n = 123) and a public TCIA dataset (n = 102). The receiver operating characteristic (ROC) analysis and precision recall curves (PRC) were used to measure the classification performance. The area under ROC curves (AUC) of the DLIS was 0.999 for training dataset, 0.986 for validation dataset, and 0.983 for testing dataset. The F1-score of the prediction model was 0.992 for training dataset, 0.940 for validation dataset, and 0.925 for testing dataset. Our data suggests that DLIS could be used to predict the 1p/19q status from preoperative imaging in patients with LGG. The imaging-based deep learning has the potential to be a noninvasive tool predictive of molecular markers in adult diffuse gliomas.

Biologic Pathways Underlying Prognostic Radiomics Phenotypes from Paired MRI and RNA Sequencing in Glioblastoma.

Background The biologic meaning of prognostic radiomics phenotypes remains poorly understood, hampered in part by lack of multicenter reproducible evidence. Purpose To uncover the biologic meaning of individual prognostic radiomics phenotypes in glioblastomas using paired MRI and RNA sequencing data and to validate the reproducibility of the identified radiogenomics linkages externally. Materials and Methods This retrospective multicenter study included four data sets gathered between January 2015 and December 2016. From a radiomics analysis set, a 13-feature radiomics signature was built using preoperative MRI for overall survival prediction. Using a radiogenomics training set with both MRI and RNA sequencing, biologic pathways were enriched and correlated with each of the 13 radiomics phenotypes. Radiomics-correlated key genes were identified to derive a prognostic radiomics gene expression (RadGene) score. The reproducibility of identified pathways and genes was validated with an external test set and a public data set (The Cancer Genome Atlas [TCGA]). A log-rank test was performed to assess prognostic significance. Results A total of 435 patients (mean age, 55 years ± 15 [standard deviation]; 263 men) were enrolled. The radiomics signature was associated with overall survival (hazard ratio [HR], 3.68; 95% CI: 2.08, 6.52; P < .001) in the radiomics validation subset. Four types of prognostic radiomics phenotypes were correlated with distinct pathways: immune, proliferative, treatment responsive, and cellular functions (false-discovery rate < 0.10). Thirty radiomics-correlated genes were identified. The prognostic significance of the RadGene score was confirmed in an external test set (HR, 2.02; 95% CI: 1.19, 3.41; P = .01) and a TCGA test set (HR, 1.43; 95% CI: 1.001, 2.04; P = .048). The radiomics-associated pathways and key genes can be replicated in an external test set. Conclusion Individual radiomics phenotypes on MRI scans predictive of overall survival were driven by distinct key pathways involved in immune regulation, tumor proliferation, treatment responses, and cellular functions in glioblastoma, which could be reproduced externally. © RSNA, 2021 Online supplemental material is available for this article.

Validation of CT radiomics for prediction of distant metastasis after surgical resection in patients with clear cell renal cell carcinoma: exploring the underlying signaling pathways.

OBJECTIVES: To develop a radiomics model using preoperative multiphasic CT for predicting distant metastasis after surgical resection in patients with localized clear cell renal cell carcinoma (ccRCC) and to identify key biological pathways underlying the predictive radiomics features using RNA sequencing data. METHODS: In this multi-institutional retrospective study, a CT radiomics metastasis score (RMS) was developed from a radiomics analysis cohort (n = 184) for distant metastasis prediction. Using a gene expression analysis cohort (n = 326), radiomics-associated gene modules were identified. Based on a radiogenomics discovery cohort (n = 42), key biological pathways were enriched from the gene modules. Furthermore, a multigene signature associated with RMS was constructed and validated on an independent radiogenomics validation cohort (n = 37). RESULTS: The 9-feature-based RMS predicted distant metastasis with an AUC of 0.861 in validation set and was independent with clinical factors (p < 0.001). A gene module comprising 114 genes was identified to be associated with all nine radiomics features (p < 0.05). Four enriched pathways were identified, including ECM-receptor interaction, focal adhesion, protein digestion and absorption, and PI3K-Akt pathways. Most of them play important roles in tumor progression and metastasis. A 19-gene signature was constructed from the radiomics-associated gene module and predicted metastasis with an AUC of 0.843 in the radiogenomics validation cohort. CONCLUSIONS: CT radiomics features can predict distant metastasis after surgical resection of localized ccRCC while the predictive radiomics phenotypes may be driven by key biological pathways related to cancer progression and metastasis. KEY POINTS: • Radiomics features from primary tumor in preoperative CT predicted distant metastasis after surgical resection in patients with localized ccRCC. • CT radiomics features predictive of distant metastasis were associated with key signaling pathways related to tumor progression and metastasis. • Gene signature associated with radiomics metastasis score predicted distant metastasis in localized ccRCC.

Multiregional radiomics features from multiparametric MRI for prediction of MGMT methylation status in glioblastoma multiforme: A multicentre study.

OBJECTIVES: To build a reliable radiomics model from multiregional and multiparametric magnetic resonance imaging (MRI) for pretreatment prediction of O6-methylguanine-DNA methyltransferase (MGMT) promotor methylation status in glioblastoma multiforme (GBM). METHODS: In this retrospective multicentre study, 1,705 multiregional radiomics features were automatically extracted from multiparametric MRI. A radiomics model with a minimal set of all-relevant features and a radiomics model with univariately-predictive and non-redundant features were built for MGMT methylation prediction from a primary cohort (133 patients) and tested on an independent validation cohort (60 patients). Predictive models combing clinical factors were built and evaluated. Both radiomics models were assessed on subgroups stratified by clinical factors. RESULTS: The radiomics model with six all-relevant features allowed pretreatment prediction of MGMT methylation (AUC=0.88, accuracy=80 %), which significantly outperformed the model with eight univariately-predictive and non-redundant features (AUC=0.76, accuracy=70 %). Combing clinical factors with radiomics features did not benefit the prediction performance. The all-relevant model achieved significantly better performance in stratified analysis. CONCLUSIONS: Radiomics model built from multiregional and multiparameter MRI may serve as a potential imaging biomarker for pretreatment prediction of MGMT methylation in GBM. The all-relevant features have the potential of offering better predictive power than the univariately-predictive and non-redundant features. KEY POINTS: • Multiregional and multiparametric MRI features reliably predicted MGMT methylation in multicentre cohorts. • All-relevant imaging features predicted MGMT methylation better than univariately-predictive and non-redundant features. • Combing clinical factors with radiomics features did not benefit the prediction performance.

Imaging-proteomic analysis for prediction of neoadjuvant chemotherapy responses in patients with breast cancer.

BACKGROUND: Optimizing patient selection for neoadjuvant chemotherapy in patients with breast cancer remains an unmet clinical need. Quantitative features from medical imaging were reported to be predictive of treatment responses. However, the biologic meaning of these latent features is poorly understood, preventing the clinical use of such noninvasive imaging markers. The study aimed to develop a deep learning signature (DLS) from pretreatment magnetic resonance imaging (MRI) for predicting responses to neoadjuvant chemotherapy in patients with breast cancer and to further investigate the biologic meaning of the DLS by identifying its underlying pathways using paired MRI and proteomic sequencing data. METHODS: MRI-based DLS was constructed (radiogenomic training dataset, n = 105) and validated (radiogenomic validation dataset, n = 26) for the prediction of pathologic complete response (pCR) to neoadjuvant chemotherapy. Proteomic sequencing revealed biological functions facilitating pCR (n = 139). Their associations with DLS were uncovered by radiogenomic analysis. RESULTS: The DLS achieved a prediction accuracy of 0.923 with an AUC of 0.958, higher than the performance of the model trained by transfer learning. Cellular membrane formation, endocytosis, insulin-like growth factor binding, protein localization to membranes, and cytoskeleton-dependent trafficking were differentially regulated in patients showing pCR. Oncogenic signaling pathways, features correlated with human phenotypes, and features correlated with general biological processes were significantly correlated with DLS in both training and validation dataset (p.adj < 0.05). CONCLUSIONS: Our study offers a biologically interpretable DLS for the prediction of pCR to neoadjuvant chemotherapy in patients with breast cancer, which may guide personalized medication.

A radio-pathologic integrated model for prediction of lymph node metastasis stage in patients with gastric cancer.

BACKGROUND: Accurate prediction of lymph node metastasis stage (LNMs) facilitates precision therapy for gastric cancer. We aimed to develop and validate a deep learning-based radio-pathologic model to predict the LNM stage in patients with gastric cancer by integrating CT images and histopathological whole-slide images (WSIs). METHODS: A total of 252 patients were enrolled and randomly divided into a training set (n = 202) and a testing set (n = 50). Both pretreatment contrast-enhanced abdominal CT and WSI of biopsy specimens were collected for each patient. The deep radiologic and pathologic features were extracted from CT and WSI using ResNet-50 and Vision Transformer (ViT) network, respectively. By fusing both radiologic and pathologic features, a radio-pathologic integrated model was constructed to predict the five LNM stages. For comparison, four single-modality models using CT images or WSIs were also constructed, respectively. All models were trained on the training set and validated on the testing set. RESULTS: The radio-pathologic integrated mode achieved an overall accuracy of 84.0% and a kappa coefficient of 0.795 on the testing set. The areas under the curves (AUCs) of the integrated model in predicting the five LNM stages were 0.978 (95% Confidence Interval (CI 0.917-1.000), 0.946 (95% CI 0.867-1.000), 0.890 (95% CI 0.718-1.000), 0.971 (95% CI 0.920-1.000), and 0.982 (95% CI 0.911-1.000), respectively. Moreover, the integrated model achieved an AUC of 0.978 (95% CI 0.912-1.000) in predicting the binary status of nodal metastasis. CONCLUSION: Our study suggests that radio-pathologic integrated model that combined both macroscale radiologic image and microscale pathologic image can better predict lymph node metastasis stage in patients with gastric cancer.

Imaging phenotypes from MRI for the prediction of glioma immune subtypes from RNA sequencing: A multicenter study.

Tumor subtyping based on its immune landscape may guide precision immunotherapy. The aims of this study were to identify immune subtypes of adult diffuse gliomas with RNA sequencing data, and to noninvasively predict this subtype using a biologically interpretable radiomic signature from MRI. A subtype discovery dataset (n = 210) from a public database and two radiogenomic datasets (n = 130 and 55, respectively) from two local hospitals were included. Brain tumor microenvironment-specific signatures were constructed from RNA sequencing to identify the immune types. A radiomic signature was built from MRI to predict the identified immune subtypes. The pathways underlying the radiomic signature were identified to annotate their biological meanings. The reproducibility of the findings was verified externally in multicenter datasets. Three distinctive immune subtypes were identified, including an inflamed subtype marked by elevated hypoxia-induced immunosuppression, a "cold" subtype that exhibited scarce immune infiltration with downregulated antigen presentation, and an intermediate subtype that showed medium immune infiltration. A 10-feature radiomic signature was developed to predict immune subtypes, achieving an AUC of 0.924 in the validation dataset. The radiomic features correlated with biological functions underpinning immune suppression, which substantiated the hypothesis that molecular changes can be reflected by radiomic features. The immune subtypes, predictive radiomic signature, and radiomics-correlated biological pathways were validated externally. Our data suggest that adult-type diffuse gliomas harbor three distinctive immune subtypes that can be predicted by MRI radiomic features with clear biological significance. The immune subtypes, radiomic signature, and radiogenomic links can be replicated externally.

Neuropathologist-level integrated classification of adult-type diffuse gliomas using deep learning from whole-slide pathological images.

Current diagnosis of glioma types requires combining both histological features and molecular characteristics, which is an expensive and time-consuming procedure. Determining the tumor types directly from whole-slide images (WSIs) is of great value for glioma diagnosis. This study presents an integrated diagnosis model for automatic classification of diffuse gliomas from annotation-free standard WSIs. Our model is developed on a training cohort (n = 1362) and a validation cohort (n = 340), and tested on an internal testing cohort (n = 289) and two external cohorts (n = 305 and 328, respectively). The model can learn imaging features containing both pathological morphology and underlying biological clues to achieve the integrated diagnosis. Our model achieves high performance with area under receiver operator curve all above 0.90 in classifying major tumor types, in identifying tumor grades within type, and especially in distinguishing tumor genotypes with shared histological features. This integrated diagnosis model has the potential to be used in clinical scenarios for automated and unbiased classification of adult-type diffuse gliomas.

Deep learning features from diffusion tensor imaging improve glioma stratification and identify risk groups with distinct molecular pathway activities.

BACKGROUND: To develop and validate a deep learning signature (DLS) from diffusion tensor imaging (DTI) for predicting overall survival in patients with infiltrative gliomas, and to investigate the biological pathways underlying the developed DLS. METHODS: The DLS was developed based on a deep learning cohort (n = 688). The key pathways underlying the DLS were identified on a radiogenomics cohort with paired DTI and RNA-seq data (n=78), where the prognostic value of the pathway genes was validated in public databases (TCGA, n = 663; CGGA, n = 657). FINDINGS: The DLS was associated with survival (log-rank P < 0.001) and was an independent predictor (P < 0.001). Incorporating the DLS into existing risk system resulted in a deep learning nomogram predicting survival better than either the DLS or the clinicomolecular nomogram alone, with a better calibration and classification accuracy (net reclassification improvement 0.646, P < 0.001). Five kinds of pathways (synaptic transmission, calcium signaling, glutamate secretion, axon guidance, and glioma pathways) were significantly correlated with the DLS. Average expression value of pathway genes showed prognostic significance in our radiogenomics cohort and TCGA/CGGA cohorts (log-rank P < 0.05). INTERPRETATION: DTI-derived DLS can improve glioma stratification by identifying risk groups with dysregulated biological pathways that contributed to survival outcomes. Therapies inhibiting neuron-to-brain tumor synaptic communication may be more effective in high-risk glioma defined by DTI-derived DLS. FUNDING: A full list of funding bodies that contributed to this study can be found in the Acknowledgements section.

Radiomics nomogram for preoperative prediction of progression-free survival using diffusion-weighted imaging in patients with muscle-invasive bladder cancer.

PURPOSE: To develop a radiomics signature using diffusion-weighted imaging (DWI) for predicting progression-free survival (PFS) in muscle-invasive bladder cancer (MIBC) patients and to assess its incremental value over traditional staging system. METHOD: 210 MIBC patients undergoing preoperative DWI were enrolled. A radiomics signature was built using LASSO model. A radiomics nomogram was generated to assess the incremental value of the radiomics signature over existing risk factors in PFS estimation in terms of calibration, discrimination, reclassification and clinical usefulness. Kaplan-Meier analysis was used to assess the association of the radiomics signature with PFS. C-index was used as a discrimination measure. Net reclassification improvement (NRI) was calculated to evaluate the usefulness improvement added by the radiomics signature. Decision curve analysis was performed to evaluate the clinical usefulness of the nomograms. RESULTS: The radiomics signature was significantly associated with PFS (log-rank P = 0.0073) and was independent with clinicopathological factors (P = 0.0004). The radiomics nomogram achieved better performance in PFS prediction (C-index: 0.702, 95 % confidence interval [CI]: 0.602, 0.802) than either clinicopathological nomogram (C-index: 0.682, 95 % CI: 0.575, 0.788) or radiomics signature (C-index: 0.612, 95 % CI: 0.493, 0.731), and achieved better calibration and classification (NRI: 0.226, 95 % CI: 0.016, 0.415, P = 0.038). Decision curve analysis demonstrated the better clinical usefulness of the radiomics nomogram. CONCLUSIONS: The DWI-based radiomics signature was an independent predictor of PFS in MIBC patients. Combining the radiomics signature, clinical staging and other clinicopathological factors achieved better performance in individual PFS prediction.

Deep Learning vs. Radiomics for Predicting Axillary Lymph Node Metastasis of Breast Cancer Using Ultrasound Images: Don't Forget the Peritumoral Region.

Objective: Axillary lymph node (ALN) metastasis status is important in guiding treatment in breast cancer. The aims were to assess how deep convolutional neural network (CNN) performed compared with radiomics analysis in predicting ALN metastasis using breast ultrasound, and to investigate the value of both intratumoral and peritumoral regions in ALN metastasis prediction. Methods: We retrospectively enrolled 479 breast cancer patients with 2,395 breast ultrasound images. Based on the intratumoral, peritumoral, and combined intra- and peritumoral regions, three CNNs were built using DenseNet, and three radiomics models were built using random forest, respectively. By combining the molecular subtype, another three CNNs and three radiomics models were built. All models were built on training cohort (343 patients 1,715 images) and evaluated on testing cohort (136 patients 680 images) with ROC analysis. Another prospective cohort of 16 patients was enrolled to further test the models. Results: AUCs of image-only CNNs in both training/testing cohorts were 0.957/0.912 for combined region, 0.944/0.775 for peritumoral region, and 0.937/0.748 for intratumoral region, which were numerically higher than their corresponding radiomics models with AUCs of 0.940/0.886, 0.920/0.724, and 0.913/0.693. The overall performance of image-molecular CNNs in terms of AUCs on training/testing cohorts slightly increased to 0.962/0.933, 0.951/0.813, and 0.931/0.794, respectively. AUCs of both CNNs and radiomics models built on combined region were significantly better than those on either intratumoral or peritumoral region on the testing cohort (p < 0.05). In the prospective study, the CNN model built on combined region achieved the highest AUC of 0.95 among all image-only models. Conclusions: CNNs showed numerically better overall performance compared with radiomics models in predicting ALN metastasis in breast cancer. For both CNNs and radiomics models, combining intratumoral, and peritumoral regions achieved significantly better performance.

Multiregional radiomics profiling from multiparametric MRI: Identifying an imaging predictor of IDH1 mutation status in glioblastoma.

PURPOSE: Isocitrate dehydrogenase 1 (IDH1) has been proven as a prognostic and predictive marker in glioblastoma (GBM) patients. The purpose was to preoperatively predict IDH mutation status in GBM using multiregional radiomics features from multiparametric magnetic resonance imaging (MRI). METHODS: In this retrospective multicenter study, 225 patients were included. A total of 1614 multiregional features were extracted from enhancement area, non-enhancement area, necrosis, edema, tumor core, and whole tumor in multiparametric MRI. Three multiregional radiomics models were built from tumor core, whole tumor, and all regions using an all-relevant feature selection and a random forest classification for predicting IDH1. Four single-region models and a model combining all-region features with clinical factors (age, sex, and Karnofsky performance status) were also built. All models were built from a training cohort (118 patients) and tested on an independent validation cohort (107 patients). RESULTS: Among the four single-region radiomics models, the edema model achieved the best accuracy of 96% and the best F1-score of 0.75 while the non-enhancement model achieved the best area under the receiver operating characteristic curve (AUC) of 0.88 in the validation cohort. The overall performance of the tumor-core model (accuracy 0.96, AUC 0.86 and F1-score 0.75) and the whole-tumor model (accuracy 0.96, AUC 0.88 and F1-score 0.75) was slightly better than the single-regional models. The 8-feature all-region radiomics model achieved an improved overall performance of an accuracy 96%, an AUC 0.90, and an F1-score 0.78. Among all models, the model combining all-region imaging features with age achieved the best performance of an accuracy 97%, an AUC 0.96, and an F1-score 0.84. CONCLUSIONS: The radiomics model built with multiregional features from multiparametric MRI has the potential to preoperatively detect the IDH1 mutation status in GBM patients. The multiregional model built with all-region features performed better than the single-region models, while combining age with all-region features achieved the best performance.

A Fully-Automatic Multiparametric Radiomics Model: Towards Reproducible and Prognostic Imaging Signature for Prediction of Overall Survival in Glioblastoma Multiforme.

In fully-automatic radiomics model for predicting overall survival (OS) of glioblastoma multiforme (GBM) patients, the effect of image standardization parameters such as voxel size, quantization method and gray level on model reproducibility and prognostic performance are still unclear. In this study, 45792 multiregional radiomics features were automatically extracted from multi-modality MR images with different voxel sizes, quantization methods, and gray levels. The feature reproducibility and prognostic performance were assessed. Multiparametric and fixed-parameter radiomics signatures were constructed based on a training cohort (60 patients). In an independent validation cohort (32 patients), the multiparametric signature achieved better performance for OS prediction (C-Index = 0.705, 95% CI: 0.672, 0.738) and significant stratification of patients into high- and low-risk groups (P = 0.0040, HR = 3.29, 95% CI: 1.40, 7.70), which outperformed the fixed-parameter signatures and conventional factors such as age, Karnofsky Performance Score and tumor volume. This study demonstrated that voxel size, quantization method and gray level had influence on reproducibility and prognosis of radiomics features for GBM OS prediction. An automatic method to determine the optimal parameter settings was provided. It indicated that multiparametric radiomics signature had the potential of offering better prognostic performance than fixed-parameter signatures.