MRI features can predict EGFR expression in lower grade gliomas: A voxel-based radiomic analysis

Multiparametric MRI texture analysis in prediction of glioma biomarker status: added value of MR diffusion

Background

Early identification of glioma molecular phenotypes can lead to understanding of patient prognosis and treatment guidance. We aimed to develop a multiparametric MRI texture analysis model using a combination of conventional and diffusion MRI to predict a wide range of biomarkers in patients with glioma.

Methods

In this retrospective study, patients were included if they (1) had diagnosis of gliomas with known IDH1, EGFR, MGMT, ATRX, TP53, and PTEN status from surgical pathology and (2) had preoperative MRI including FLAIR, T1c+ and diffusion for radiomic texture analysis. Statistical analysis included logistic regression and receiver-operating characteristic (ROC) curve analysis to determine the optimal model for predicting glioma biomarkers. A comparative analysis between ROCs (conventional only vs conventional + diffusion) was performed.

Results

From a total of 111 patients included, 91 (82%) were categorized to training and 20 (18%) to test datasets. Constructed cross-validated model using a combination of texture features from conventional and diffusion MRI resulted in overall AUC/accuracy of 1/79% for IDH1, 0.99/80% for ATRX, 0.79/67% for MGMT, and 0.77/66% for EGFR. The addition of diffusion data to conventional MRI features significantly (P < .05) increased predictive performance for IDH1, MGMT, and ATRX. The overall accuracy of the final model in predicting biomarkers in the test group was 80% (IDH1), 70% (ATRX), 70% (MGMT), and 75% (EGFR).

Conclusion

Addition of MR diffusion to conventional MRI features provides added diagnostic value in preoperative determination of IDH1, MGMT, and ATRX in patients with glioma.

Construction of a prognostic immune signature for lower grade glioma that can be recognized by MRI radiomics features to predict survival in LGG patients

BACKGROUND

This study aimed to identify a series of prognostically relevant immune features by immunophenoscore. Immune features were explored using MRI radiomics features to prediction the overall survival (OS) of lower-grade glioma (LGG) patients and their response to immune checkpoints.

METHOD

LGG data were retrieved from TCGA and categorized into training and internal validation datasets. Patients attending the First Affiliated Hospital of Harbin Medical University were included in an external validation cohort. An immunophenoscore-based signature was built to predict malignant potential and response to immune checkpoint inhibitors in LGG patients. In addition, a deep learning neural network prediction model was built for validation of the immunophenoscore-based signature.

RESULTS

Immunophenotype-associated mRNA signatures (IMriskScore) for outcome prediction and ICB therapeutic effects in LGG patients were constructed. Deep learning of neural networks based on radiomics showed that MRI radiomic features determined IMriskScore. Enrichment analysis and ssGSEA correlation analysis were performed. Mutations in CIC significantly improved the prognosis of patients in the high IMriskScore group. Therefore, CIC is a potential therapeutic target for patients in the high IMriskScore group. Moreover, IMriskScore is an independent risk factor that can be used clinically to predict LGG patient outcomes.

CONCLUSIONS

The IMriskScore model consisting of a sets of biomarkers, can independently predict the prognosis of LGG patients and provides a basis for the development of personalized immunotherapy strategies. In addition, IMriskScore features were predicted by MRI radiomics using a deep learning approach using neural networks. Therefore, they can be used for the prognosis of LGG patients.

Multiparametric MRI-Based Radiomics Approaches for Preoperative Prediction of EGFR Mutation Status in Spinal Bone Metastases in Patients with Lung Adenocarcinoma

BACKGROUND

Preoperative prediction of epidermal growth factor receptor (EGFR) mutation status in patients with spinal bone metastases (SBM) from primary lung adenocarcinoma is potentially important for treatment decisions.

PURPOSE

To develop and validate multiparametric magnetic resonance imaging (MRI)-based radiomics methods for preoperative prediction of EGFR mutation based on MRI of SBM.

STUDY TYPE

Retrospective.

POPULATION

A total of 97 preoperative patients with lumbar SBM from lung adenocarcinoma (77 in training set and 20 in validation set).

FIELD STRENGTH/SEQUENCE

T1-weighted, T2-weighted, and T2-weighted fat-suppressed fast spin echo sequences at 3.0 T.

ASSESSMENT

Radiomics handcrafted and deep learning-based features were extracted and selected from each MRI sequence. The abilities of the features to predict EGFR mutation status were analyzed and compared. A radiomics nomogram was constructed integrating the selected features.

STATISTICAL TESTS

The Mann-Whitney U test and χ² test were employed for evaluating associations between clinical characteristics and EGFR mutation status for continuous and discrete variables, respectively. Least absolute shrinkage and selection operator was used for selection of predictive features. Sensitivity (SEN), specificity (SPE), and area under the receiver operating characteristic curve (AUC) were used to evaluate the ability of radiomics models to predict the EGFR mutation. Calibration and decision curve analysis (DCA) were performed to assess and validate nomogram results.

RESULTS

The radiomics signature comprised five handcrafted and one deep learning-based features and achieved good performance for predicting EGFR mutation status, with AUCs of 0.891 (95% confidence interval [CI], 0.820-0.962, SEN = 0.913, SPE = 0.710) in the training group and 0.771 (95% CI, 0.551-0.991, SEN = 0.750, SPE = 0.875) in the validation group. DCA confirmed the potential clinical usefulness of the radiomics models.

DATA CONCLUSION

Multiparametric MRI-based radiomics is potentially clinical valuable for predicting EGFR mutation status in patients with SBM from lung adenocarcinoma.

LEVEL OF EVIDENCE

3 TECHNICAL EFFICACY: 2.

Radiomics Features Predict Telomerase Reverse Transcriptase Promoter Mutations in World Health Organization Grade II Gliomas via a Machine-Learning Approach

The detection of mutations in telomerase reverse transcriptase promoter (pTERT) is important since preoperative diagnosis of pTERT status helps with evaluating prognosis and determining the surgical strategy. Here, we aimed to establish a radiomics-based machine-learning algorithm and evaluated its performance with regard to the prediction of mutations in pTERT in patients with World Health Organization (WHO) grade II gliomas. In total, 164 patients with WHO grade II gliomas were enrolled in this retrospective study. We extracted a total of 1,293 radiomics features from multi-parametric magnetic resonance imaging scans. Elastic net (used for feature selection) and support vector machine with linear kernel were applied in nested 10-fold cross-validation loops. The predictive model was evaluated by receiver operating characteristic and precision-recall analyses. We performed an unpaired t-test to compare the posterior predictive probabilities among patients with differing pTERT statuses. We selected 12 valuable radiomics features using nested 10-fold cross-validation loops. The area under the curve (AUC) was 0.8446 (95% confidence interval [CI], 0.7735–0.9065) with an optimal summed value of sensitivity of 0.9355 (95% CI, 0.8802–0.9788) and specificity of 0.6197 (95% CI, 0.5071–0.7371). The overall accuracy was 0.7988 (95% CI, 0.7378–0.8598). The F1-score was 0.8406 (95% CI, 0.7684–0.902) with an optimal precision of 0.7632 (95% CI, 0.6818–0.8364) and recall of 0.9355 (95% CI, 0.8802–0.9788). Posterior probabilities of pTERT mutations were significantly different between patients with wild-type and mutant TERT promoters. Our findings suggest that a radiomics analysis with a machine-learning algorithm can be useful for predicting pTERT status in patients with WHO grade II glioma and may aid in glioma management.

Combined Spiral Transformation and Model-Driven Multi-Modal Deep Learning Scheme for Automatic Prediction of TP53 Mutation in Pancreatic Cancer

Pancreatic cancer is a malignant form of cancer with one of the worst prognoses. The poor prognosis and resistance to therapeutic modalities have been linked to TP53 mutation. Pathological examinations, such as biopsies, cannot be frequently performed in clinical practice; therefore, noninvasive and reproducible methods are desired. However, automatic prediction methods based on imaging have drawbacks such as poor 3D information utilization, small sample size, and ineffectiveness multi-modal fusion. In this study, we proposed a model-driven multi-modal deep learning scheme to overcome these challenges. A spiral transformation algorithm was developed to obtain 2D images from 3D data, with the transformed image inheriting and retaining the spatial correlation of the original texture and edge information. The spiral transformation could be used to effectively apply the 3D information with less computational resources and conveniently augment the data size with high quality. Moreover, model-driven items were designed to introduce prior knowledge in the deep learning framework for multi-modal fusion. The model-driven strategy and spiral transformation-based data augmentation can improve the performance of the small sample size. A bilinear pooling module was introduced to improve the performance of fine-grained prediction. The experimental results show that the proposed model gives the desired performance in predicting TP53 mutation in pancreatic cancer, providing a new approach for noninvasive gene prediction. The proposed methodologies of spiral transformation and model-driven deep learning can also be used for the artificial intelligence community dealing with oncological applications. Our source codes with a demon will be released at https://github.com/SJTUBME-QianLab/SpiralTransform.

MRI Features May Predict Molecular Features of Glioblastoma in Isocitrate Dehydrogenase Wild-Type Lower-Grade Gliomas

BACKGROUND AND PURPOSE

Isocitrate dehydrogenase (IDH) wild-type lower-grade gliomas (histologic grades II and III) with epidermal growth factor receptor (EGFR) amplification or telomerase reverse transcriptase (TERT) promoter mutation are reported to behave similar to glioblastoma. We aimed to evaluate whether MR imaging features could identify a subset of IDH wild-type lower-grade gliomas that carry molecular features of glioblastoma.

MATERIALS AND METHODS

In this multi-institutional retrospective study, pathologically confirmed IDH wild-type lower-grade gliomas from 2 tertiary institutions and The Cancer Genome Atlas constituted the training set (institution 1 and The Cancer Genome Atlas, 64 patients) and the independent test set (institution 2, 57 patients). Preoperative MRIs were analyzed using the Visually AcceSAble Rembrandt Images and radiomics. The molecular glioblastoma status was determined on the basis of the presence of EGFR amplification and TERT promoter mutation. Molecular glioblastoma was present in 73.4% and 56.1% in the training and test sets, respectively. Models using clinical, Visually AcceSAble Rembrandt Images, and radiomic features were built to predict the molecular glioblastoma status in the training set; then they were validated in the test set.

RESULTS

In the test set, a model using both Visually AcceSAble Rembrandt Images and radiomic features showed superior predictive performance (area under the curve = 0.854) than that with only clinical features or Visually AcceSAble Rembrandt Images (areas under the curve = 0.514 and 0.648, respectively; P < . 001, both). When both Visually AcceSAble Rembrandt Images and radiomics were added to clinical features, the predictive performance significantly increased (areas under the curve = 0.514 versus 0.863, P < .001).

CONCLUSIONS

MR imaging features integrated with machine learning classifiers may predict a subset of IDH wild-type lower-grade gliomas that carry molecular features of glioblastoma.

An MRI-based radiomics signature as a pretreatment noninvasive predictor of overall survival and chemotherapeutic benefits in lower-grade gliomas

OBJECTIVES

The aim of this study was to develop and validate a radiomics signature for predicting survival and chemotherapeutic benefits of patients with lower-grade gliomas (LGG).

METHODS

Radiomics features were extracted from precontrast axial fluid-attenuated inversion recovery (FLAIR) and contrast-enhanced axial T-1 weighted (CE-T1-w) sequence. Lasso Cox regression model was used for feature selection and radiomics signature building. The radiomics signature was developed in a primary cohort that consisted of 149 LGG patients and was then validated on an entirely new validation cohort that contained 66 LGG patients. A radiomics nomogram for the prediction of OS was established by adding the radiomics to clinicopathologic nomogram which developed with clinical data.

RESULTS

A radiomics signature derived from joint CE-T1-w and FLAIR images showed better prognostic performance (C-index, 0.798) than signatures derived from CE-T1-w (C-index, 0.744) or FLAIR (C-index, 0.736) sequences alone. Multivariable Cox regression revealed that the radiomics signature was an independent prognostic factor. One radiomics nomogram integrated the radiomics signature from joint CE-T1-w and FLAIR sequences with the clinicopathologic nomogram outperformed the clinicopathologic nomogram based on clinicopathologic data alone in predicting OS of LGG (C-index, 0.821 vs. 0.692; p < 0.001). Further analysis revealed that patients with higher radiomics signature were prone to benefit from chemotherapy.

CONCLUSIONS

The radiomics signature was independent with clinicopathologic data and was a noninvasive pretreatment predictor for LGG patients' survival. Moreover, it could predict which patients with LGG benefit from chemotherapy.

KEY POINTS

• A radiomics signature derived from joint CE-T1-w and FLAIR sequences showed better prognostic performance than signatures derived from either single imaging modality. • The radiomics signature is an independent prognostic factor and outperformed clinicopathologic features in predicting overall survival of LGG patients. • The radiomics signature could help preoperatively identify LGG patients who may benefit from chemotherapy.

Imaging genomics for accurate diagnosis and treatment of tumors: A cutting edge overview

Imaging genomics refers to the establishment of the connection between invasive gene expression features and non-invasive imaging features. Tumor imaging genomics can not only understand the macroscopic phenotype of tumor, but also can deeply analyze the cellular and molecular characteristics of tumor tissue. In recent years, tumor imaging genomics has been a key in the field of medicine. The incidence of cancer in China has increased significantly, which is the main reason of disease death of urban residents. With the rapid development of imaging medicine, depending on imaging genomics, many experts have made remarkable achievements in tumor screening and diagnosis, prognosis evaluation, new treatment targets and understanding of tumor biological mechanism. This review analyzes the relationship between tumor radiology and gene expression, which provides a favorable direction for clinical staging, prognosis evaluation and accurate treatment of tumors.

Systematic review of radiomic biomarkers for predicting immune checkpoint inhibitor treatment outcomes

BACKGROUND

Systemic therapy agents targeting immune checkpoint inhibitors have been approved for use since 2011. This type of therapy aims to trigger a patient's immune response to attack tumor cells, rather than acting against the tumor directly. Radiomics is an automated method of medical image analysis that is now being actively investigated for predictive markers of treatment response in immunotherapy.

OBJECTIVE

To conduct an early systematic review determining the current status of radiomic features as potential predictive markers of immunotherapy response. Provide a detailed critical appraisal of methodological quality of models, as this informs the degree of confidence about current reports of model performance. In addition, to offer some recommendations for future studies that could establish robust evidence for radiomic features as immunotherapy response markers.

METHOD

A PubMed citation search was conducted for publications up to and including April 2020, followed by full-text screening. A total of seven articles meeting the eligibility criteria were examined in detail for study characteristics, model information and methodological quality. The review was conducted in the Cochrane style but has not been prospectively registered. Results are reported following Preferred Reporting Items for Systematic Review and Meta-Analysis Protocols (PRISMA) guidelines.

RESULTS

A total of seven studies were examined in detail, comprising non-small cell lung cancer, metastatic melanoma and a diverse assortment of solid tumors. Methodological robustness of reviewed studies varied greatly. Principal shortcomings were lack of prospective registration, and deficiencies in feature selection and dimensionality reduction, model calibration, clinical utility and external validation. A few studies with overall moderate to good methodological quality were identified. These results suggest that current state-of-the-art performance of radiomics in regards to discrimination (area under the curve or concordance index) is in the vicinity of 0.7, but the very small number of studies to date prevents any conclusive remarks to be made. We recommended future improvements in regards to prospective study registration, clinical utility, methodological procedure and data sharing.

CONCLUSIONS

Radiomics has a potentially significant role for predicting immunotherapy response. Additional multi-institutional studies with robust methodological underpinning and repeated external validations are required to establish the (added) value of radiomics within the pantheon of clinical tools for decision-making in immunotherapy.

Deep learning algorithm to improve hypertrophic cardiomyopathy mutation prediction using cardiac cine images

OBJECTIVES

The high variability of hypertrophic cardiomyopathy (HCM) genetic phenotypes has prompted the establishment of risk-stratification systems that predict the risk of a positive genetic mutation based on clinical and echocardiographic profiles. This study aims to improve mutation-risk prediction by extracting cardiovascular magnetic resonance (CMR) morphological features using a deep learning algorithm.

METHODS

We recruited 198 HCM patients (48% men, aged 47 ± 13 years) and divided them into training (147 cases) and test (51 cases) sets based on different genetic testing institutions and CMR scan dates (2012, 2013, respectively). All patients underwent CMR examinations, HCM genetic testing, and an assessment of established genotype scores (Mayo Clinic score I, Mayo Clinic score II, and Toronto score). A deep learning (DL) model was developed to classify the HCM genotypes, based on a nonenhanced four-chamber view of cine images.

RESULTS

The areas under the curve (AUCs) for the test set were Mayo Clinic score I (AUC: 0.64, sensitivity: 64.29%, specificity: 47.83%), Mayo Clinic score II (AUC: 0.70, sensitivity: 64.29%, specificity: 65.22%), Toronto score (AUC: 0.74, sensitivity: 75.00%, specificity: 56.52%), and DL model (AUC: 0.80, sensitivity: 85.71%, specificity: 69.57%). The combination of the DL and the Toronto score resulted in a significantly higher predictive performance (AUC = 0.84, sensitivity: 83.33%, specificity: 78.26%), compared with Mayo I (p = 006), Mayo II (p = 022), and Toronto score (p = 0.029).

CONCLUSIONS

The combination of the DL model, based on nonenhanced cine CMR images and the Toronto score yielded significantly higher diagnostic performance in detecting HCM mutations.

KEY POINTS

• Deep learning method could enable the extraction of image features from cine images. • Deep learning method based on cine images performed better than established scores in identifying HCM patients with positive genotypes. • The combination of the deep learning method based on cine images and the Toronto score could further improve the performance of the identification of HCM patients with positive genotypes.

Longitudinal radiomics algorithm of posttreatment computed tomography images for early detecting recurrence of hepatocellular carcinoma after resection or ablation

OBJECTIVES

To develop a radiomics algorithm, improving the performance of detecting recurrence, based on posttreatment CT images within one month and at suspicious time during follow-up.

MATERIALS AND METHODS

A total of 114 patients with 228 images were randomly split (7:3) into training and validation cohort. Radiomics algorithm was trained using machine learning, based on difference-in-difference (DD) features extracted from tumor and liver regions of interest on posttreatment CTs within one month after resection or ablation and when suspected recurrent lesion was observed but cannot be confirmed as HCC during follow-up. The performance was evaluated by area under the receiver operating characteristic curve (AUC) and was compared among radiomics algorithm, change of alpha-fetoprotein (AFP) and combined model of both. Five-folded cross validation (CV) was used to present the training error.

RESULTS

A radiomics algorithm was established by 34 DD features selected by random forest and multivariable logistic models and showed a better AUC than that of change of AFP (0.89 [95% CI: 0.78, 1.00] vs 0.63 [95% CI: 0.42, 0.84], P = .04) and similar with the combined model in detecting recurrence in the validation set. Five-folded CV error in the validation cohort was 21% for the algorithm and 26% for the changes of AFP.

CONCLUSIONS

The algorithm integrated radiomic features of posttreatment CT showed superior performance to that of conventional AFP and may act as a potential marker in the early detecting recurrence of HCC.

Magnetic resonance imaging-based radiomic features for extrapolating infiltration levels of immune cells in lower-grade gliomas

PURPOSE

To extrapolate the infiltration levels of immune cells in patients with lower-grade gliomas (LGGs) using magnetic resonance imaging (MRI)-based radiomic features.

METHODS

A retrospective dataset of 516 patients with LGGs from The Cancer Genome Atlas (TCGA) database was analysed for the infiltration levels of six types of immune cells using Tumor IMmune Estimation Resource (TIMER) based on RNA sequencing data. Radiomic features were extracted from 107 patients whose pre-operative MRI data are available in The Cancer Imaging Archive; 85 and 22 of these patients were assigned to the training and testing cohort, respectively. The least absolute shrinkage and selection operator (LASSO) was applied to select optimal radiomic features to build the radiomic signatures for extrapolating the infiltration levels of immune cells in the training cohort. The developed radiomic signatures were examined in the testing cohort using Pearson's correlation.

RESULTS

The infiltration levels of B cells, CD4+ T cells, CD8+ T cells, macrophages, neutrophils and dendritic cells negatively correlated with overall survival in the 516 patient cohort when using univariate Cox's regression. Age, Karnofsky Performance Scale, WHO grade, isocitrate dehydrogenase mutant status and the infiltration of neutrophils correlated with survival using multivariate Cox's regression analysis. The infiltration levels of the 6 cell types could be estimated by radiomic features in the training cohort, and their corresponding radiomic signatures were built. The infiltration levels of B cells, CD8+ T cells, neutrophils and macrophages estimated by radiomics correlated with those estimated by TIMER in the testing cohort. Combining clinical/genomic features with the radiomic signatures only slightly improved the prediction of immune cell infiltrations.

CONCLUSION

We developed MRI-based radiomic models for extrapolating the infiltration levels of immune cells in LGGs. Our results may have implications for treatment planning.

Dynamic contrast-enhanced MRI differentiates hepatocellular carcinoma from hepatic metastasis of rectal cancer by extracting pharmacokinetic parameters and radiomic features

The aim of the present study was to explore how dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) may differentiate hepatocellular carcinoma (HCC) from hepatic metastasis of rectal cancer (HMRC) by extracting pharmacokinetic parameters and radiomic features. A total of 75 patients, including 41 cases with HCC and 34 cases with HMRC, underwent DCE-MRI examination. Dual-input two-compartment extended Tofts tracer kinetic model attached to a specialized image post-processing software package from OmniKinetics; GE Healthcare was used to calculate the values of the pharmacokinetic parameters and radiomic features, which were extracted from the lesions at the same region of interest. These values were evaluated using Student's t-test and receiver operating characteristic curves, and discriminant models were built to differentiate between HCC and HRMC. The results identified statistically significant differences in the values of the pharmacokinetic parameters hepatic perfusion index (HPI), endothelial transfer constant (K^trans), initial area under the gadolinium concentration curve during the first 60 sec (IAUC) between the HCC and HRMC groups. In addition, statistically significant differences in 17 radiomic features were observed between the two groups (P<0.05). The areas under the receiver operating characteristic (ROC) curves of the pharmacokinetic parameters K^trans, IAUC and HPI were 0.73, 0.77 and 0.67, respectively. The range of the areas under the ROC curves of the 17 radiomic features with statistical differences was 0.63-0.79. In addition, when pharmacokinetic parameters and radiomic features were incorporated, the area under the ROC curve was 0.86. The accuracy of Fisher's discriminant analysis model based on radiomic features was 89.3%, and the leave-one-out cross-validation accuracy was 80.0%. In conclusion, DCE-MRI was demonstrated to be useful in the differential diagnosis of HCC and HMRC by extracting pharmacokinetic parameters and radiomic features, and incorporation of the two paths improved the diagnostic efficacy. A discriminant model based on radiomic features further enhanced the identification of HCC and HMRC.

Financially effective test algorithm to identify an aggressive, EGFR-amplified variant of IDH-wildtype, lower-grade diffuse glioma

BACKGROUND

Update 3 of the Consortium to Inform Molecular and Practical Approaches to CNS Tumor Taxonomy (cIMPACT-NOW) recognizes amplification of epidermal growth factor receptor (EGFR) as one important aberration in diffuse gliomas (World Health Organization [WHO] grade II/III). While these recommendations endorse testing, a cost-effective, clinically relevant testing paradigm is currently lacking. Here, we use real-world clinical data to propose a financially effective diagnostic test algorithm in the context of new guidelines.

METHODS

To determine the prevalence, distribution, neuroradiographic features (Visually Accessible REMBRANDT Images [VASARI]), and prognostic relevance of EGFR amplification in lower-grade gliomas, we assembled a consecutive series of diffuse gliomas. For validation we included publicly available data from The Cancer Genome Atlas. For a cost-utility analysis we compared combined EGFR and isocitrate dehydrogenase (IDH) testing, EGFR testing based on IDH results, and no EGFR testing.

RESULTS

In n = 71 WHO grade II/III gliomas, we identified EGFR amplification in 28.2%. With one exception, all EGFR amplifications occurred in IDH-wildtype gliomas. Comparison of overall survival showed that EGFR amplification denotes a significantly more aggressive subset of tumors (P < 0.0001, log-rank). The radiologic phenotype in the EGFR-amplified tumors includes diffusion restriction (15%, P = 0.02), >5% tumor contrast enhancement (75%, P = 0.016), and mild (not avid) enhancement (P = 0.016). The proposed testing algorithm reserves EGFR fluorescence in situ hybridization (FISH) testing for IDH-wildtype cases. Implementation would result in ~37.9% cost reduction at our institution, or about $1.3-4 million nationally.

CONCLUSION

EGFR-amplified diffuse gliomas are "glioblastoma-like" in their behavior and may represent undersampled glioblastomas, or subsets of IDH-wildtype diffuse gliomas with inherently aggressive biology. EGFR FISH after IDH testing is a financially effective and clinically relevant test algorithm for routine clinical practice.

Integrative radiogenomics for virtual biopsy and treatment monitoring in ovarian cancer

BACKGROUND

Ovarian cancer survival rates have not changed in the last 20 years. The majority of cases are High-grade serous ovarian carcinomas (HGSOCs), which are typically diagnosed at an advanced stage with multiple metastatic lesions. Taking biopsies of all sites of disease is infeasible, which challenges the implementation of stratification tools based on molecular profiling.

MAIN BODY

In this review, we describe how these challenges might be overcome by integrating quantitative features extracted from medical imaging with the analysis of paired genomic profiles, a combined approach called radiogenomics, to generate virtual biopsies. Radiomic studies have been used to model different imaging phenotypes, and some radiomic signatures have been associated with paired molecular profiles to monitor spatiotemporal changes in the heterogeneity of tumours. We describe different strategies to integrate radiogenomic information in a global and local manner, the latter by targeted sampling of tumour habitats, defined as regions with distinct radiomic phenotypes.

CONCLUSION

Linking radiomics and biological correlates in a targeted manner could potentially improve the clinical management of ovarian cancer. Radiogenomic signatures could be used to monitor tumours during the course of therapy, offering additional information for clinical decision making. In summary, radiogenomics may pave the way to virtual biopsies and treatment monitoring tools for integrative tumour analysis.

Diffusion and perfusion MRI may predict EGFR amplification and the TERT promoter mutation status of IDH-wildtype lower-grade gliomas

OBJECTIVES

Epidermal growth factor receptor (EGFR) amplification and telomerase reverse transcriptase promoter (TERTp) mutation status of isocitrate dehydrogenase-wildtype (IDHwt) lower-grade gliomas (LGGs; grade II/III) are crucial for identifying IDHwt LGG with an aggressive clinical course. The purpose of this study was to assess whether parameters from diffusion tensor imaging, dynamic susceptibility contrast (DSC), and diffusion tensor imaging, dynamic contrast-enhanced imaging can predict the EGFR amplification and TERTp mutation status of IDHwt LGGs.

METHODS

A total of 49 patients with IDHwt LGGs with either known EGFR amplification (39 non-amplified, 10 amplified) or TERTp mutation (19 wildtype, 21 mutant) statuses underwent MRI. The mean ADC, fractional anisotropy (FA), normalized cerebral blood volume (nCBV), normalized cerebral blood flow (nCBF), volume transfer constant (K^trans), rate transfer coefficient (K_ep), extravascular extracellular volume fraction (V_e), and plasma volume fraction (V_p) values were assessed. Univariate and multivariate logistic regression models were constructed.

RESULTS

EGFR-amplified tumors showed lower mean ADC values than EGFR-non-amplified tumors (p = 0.019). Mean ADC was an independent predictor of EGFR amplification, with an AUC of 0.75. TERTp mutant tumors showed higher mean nCBV (p = 0.020), higher mean nCBF (p = 0.017), and higher mean V_p (p = 0.002) than TERTp wildtype tumors. With multivariate logistic regression, mean V_p was the independent predictor of TERTp mutation status, with an AUC of 0.85.

CONCLUSION

This exploratory pilot study shows that lower ADC values may be useful for prediction of EGFR amplification, whereas higher V_p values may be useful for prediction of the TERTp mutation status of IDHwt LGGs.

KEY POINTS

• EGFR amplification and TERTp mutation are key molecular markers that predict an aggressive clinical course of IDHwt LGGs. • EGFR-amplified tumors showed lower ADC values than EGFR-non-amplified tumors, suggesting higher cellularity. • TERTp mutant tumors showed a higher plasma volume fraction than TERTp wildtype tumors, suggesting higher vascular proliferation and tumor angiogenesis.

CT-based radiomics scores predict response to neoadjuvant chemotherapy and survival in patients with gastric cancer

BACKGROUND

Neoadjuvant chemotherapy is a promising treatment option for potential resectable gastric cancer, but patients' responses vary. We aimed to develop and validate a radiomics score (rad_score) to predict treatment response to neoadjuvant chemotherapy and to investigate its efficacy in survival stratification.

METHODS

A total of 106 patients with neoadjuvant chemotherapy before gastrectomy were included (training cohort: n = 74; validation cohort: n = 32). Radiomics features were extracted from the pre-treatment portal venous-phase CT. After feature reduction, a rad_score was established by Randomised Tree algorithm. A rad_clinical_score was constructed by integrating the rad_score with clinical variables, so was a clinical score by clinical variables only. The three scores were validated regarding their discrimination and clinical usefulness. The patients were stratified into two groups according to the score thresholds (updated with post-operative clinical variables), and their survivals were compared.

RESULTS

In the validation cohort, the rad_score demonstrated a good predicting performance in treatment response to the neoadjuvant chemotherapy (AUC [95% CI] =0.82 [0.67, 0.98]), which was better than the clinical score (based on pre-operative clinical variables) without significant difference (0.62 [0.42, 0.83], P = 0.09). The rad_clinical_score could not further improve the performance of the rad_score (0.70 [0.51, 0.88], P = 0.16). Based on the thresholds of these scores, the high-score groups all achieved better survivals than the low-score groups in the whole cohort (all P < 0.001).

CONCLUSION

The rad_score that we developed was effective in predicting treatment response to neoadjuvant chemotherapy and in stratifying patients with gastric cancer into different survival groups. Our proposed strategy is useful for individualised treatment planning.

Role of molecular biomarkers in glioma resection: a systematic review

New discoveries based on genetic and epigenetic evidence have significantly expanded the understanding of diffuse gliomas. Molecular biomarkers detected in diffuse gliomas are not only potential targets for radiotherapy, chemotherapy, and immunotherapy, but are also able to guide surgical treatment. Previous studies have suggested that the optimal extent of resection of diffuse gliomas varies according to the expression of specific molecular biomarkers. However, the specific guiding role of these biomarkers in the resection of diffuse gliomas has not been systemically analyzed. This review summarizes several critical molecular biomarkers of tumorigenesis and progression in diffuse gliomas and discusses different strategies of tumor resection in the context of varying genetic expression. With ongoing study and advances in technology, molecular biomarkers will play a more important role in glioma resection and maximize the survival benefit from surgery for diffuse gliomas.

Noninvasive Prediction of TERT Promoter Mutations in High-Grade Glioma by Radiomics Analysis Based on Multiparameter MRI

Objectives

To investigate the predictors of telomerase reverse transcriptase (TERT) promoter mutations in adults suffered from high-grade glioma (HGG) through radiomics analysis, develop a noninvasive approach to evaluate TERT promoter mutations.

Methods

126 adult patients with HGG (88 in the training cohort and 38 in the validation cohort) were retrospectively enrolled. Totally 5064 radiomics features were, respectively, extracted from three VOIs (necrosis, enhanced, and edema) in MRI. Firstly, an optimal radiomics signature (Radscore) was established based on LASSO regression. Secondly, univariate and multivariate logistic regression analyses were performed to investigate important potential variables as predictors of TERT promoter mutations. Besides, multiparameter models were established and evaluated. Eventually, an optimal model was visualized as radiomics nomogram for clinical evaluations.

Results

6 radiomics features were selected to build Radscore signature through LASSO regression. Among them, 5 were from necrotic VOIs and 1 was from enhanced ones. With univariate and multivariate analysis, necrotic volume percentages of core (CNV), Age, Cho/Cr, Lac, and Radscore were significantly higher in TERTm than in TERTw (p < 0.05). 4 models were built in our study. Compared with Model B (Age, Cho/Cr, Lac, and Radscore), Model A (Age, Cho/Cr, Lac, Radscore, and CNV) has a larger AUC in both training (0.955 vs. 0.917, p = 0.049) and validation (0.889 vs. 0.868, p = 0.039) cohorts. It also has higher performances in net reclassification improvement (NRI), integrated discrimination improvement (IDI), and decision curve analysis (DCA) evaluation. Conclusively, Model A was visualized as a radiomics nomogram. Calibration curve shows a good agreement between estimated and actual probabilities.

Conclusions

Age, Cho/Cr, Lac, CNV, and Radscore are important indicators for TERT promoter mutation predictions in HGG. Tumor necrosis seems to be closely related to TERT promoter mutations. Radiomics nomogram based on multiparameter MRI and CNV has higher prediction accuracies.

A systematic review reporting quality of radiomics research in neuro-oncology: toward clinical utility and quality improvement using high-dimensional imaging features

Background

To evaluate radiomics analysis in neuro-oncologic studies according to a radiomics quality score (RQS) system to find room for improvement in clinical use.

Methods

Pubmed and Embase were searched up the terms radiomics or radiogenomics and gliomas or glioblastomas until February 2019. From 189 articles, 51 original research articles reporting the diagnostic, prognostic, or predictive utility were selected. The quality of the methodology was evaluated according to the RQS. The adherence rates for the six key domains were evaluated: image protocol and reproducibility, feature reduction and validation, biologic/clinical utility, performance index, a high level of evidence, and open science. Subgroup analyses for journal type (imaging vs. clinical) and biomarker (diagnostic vs. prognostic/predictive) were performed.

Results

The median RQS was 11 out of 36 and adherence rate was 37.1%. Only 29.4% performed external validation. The adherence rate was high for reporting imaging protocol (100%), feature reduction (94.1%), and discrimination statistics (96.1%), but low for conducting test-retest analysis (2%), prospective study (3.9%), demonstrating potential clinical utility (2%), and open science (5.9%). None of the studies conducted a phantom study or cost-effectiveness analysis. Prognostic/predictive studies received higher score than diagnostic studies in comparison to gold standard (P < .001), use of calibration (P = .02), and cut-off analysis (P = .001).

Conclusions

The quality of reporting of radiomics studies in neuro-oncology is currently insufficient. Validation is necessary using external dataset, and improvements need to be made to feature reproducibility, demonstrating clinical utility, pursuits of a higher level of evidence, and open science.

Prediction of Core Signaling Pathway by Using Diffusion- and Perfusion-based MRI Radiomics and Next-generation Sequencing in Isocitrate Dehydrogenase Wild-type Glioblastoma

Background Next-generation sequencing (NGS) enables highly sensitive cancer genomics analysis, but its clinical implications for therapeutic options from imaging-based prediction have been limited. Purpose To predict core signaling pathways in isocitrate dehydrogenase (IDH) wild-type glioblastoma by using diffusion and perfusion MRI radiomics and NGS. Materials and Methods The radiogenomics model was developed by using retrospective patients with glioma who underwent NGS and anatomic, diffusion-, and perfusion-weighted imaging between March 2017 and February 2019. For testing model performance in predicting core signaling pathway, patients with IDH wild-type glioblastoma from a retrospective analysis from a registry (ClinicalTrials.gov NCT02619890) were evaluated. Radiogenomic feature selection was performed by using t tests, least absolute shrinkage and selection operator penalization, and random forest. Combining radiogenomic features, age, and location, the performance of predicting receptor tyrosine kinase (RTK), tumor protein p53 (P53), and retinoblastoma 1 pathways was evaluated by using the area under the receiver operating characteristic curve (AUC). Results There were 120 patients (52 years ± 13 [standard deviation]; 61 women) who were evaluated. Eighty-five patients (51 years ± 13; 43 men) were in the training set and 35 patients with IDH wild-type glioblastoma (56 years ± 12; 19 women) were in the validation set. Radiogenomics model identified 71 features in the RTK, 17 features in P53, and 35 features in the retinoblastoma pathway. The combined model showed better performance than anatomic imaging-based prediction in the RTK (P = .03) and retinoblastoma (P = .03) and perfusion imaging-based prediction in the P53 pathway (P = .04) in the training set. AUC values of the combined model for the prediction of core signaling pathways were 0.88 (95% confidence interval [CI]: 0.74, 1) for RTK, 0.76 (95% CI: 0.59, 0.92) for P53, and 0.81 (95% CI: 0.64, 0.97) for retinoblastoma in the validation set. Conclusion A diffusion- and perfusion-weighted MRI radiomics model can help characterize core signaling pathways and potentially guide targeted therapy for isocitrate dehydrogenase wild-type glioblastoma. © RSNA, 2019 Online supplemental material is available for this article.

A Comparison of the Efficiency of Using a Deep CNN Approach with Other Common Regression Methods for the Prediction of EGFR Expression in Glioblastoma Patients

To estimate epithermal growth factor receptor (EGFR) expression level in glioblastoma (GBM) patients using radiogenomic analysis of magnetic resonance images (MRI). A comparative study using a deep convolutional neural network (CNN)-based regression, deep neural network, least absolute shrinkage and selection operator (LASSO) regression, elastic net regression, and linear regression with no regularization was carried out to estimate EGFR expression of 166 GBM patients. Except for the deep CNN case, overfitting was prevented by using feature selection, and loss values for each method were compared. The loss values in the training phase for deep CNN, deep neural network, Elastic net, LASSO, and the linear regression with no regularization were 2.90, 8.69, 7.13, 14.63, and 21.76, respectively, while in the test phase, the loss values were 5.94, 10.28, 13.61, 17.32, and 24.19 respectively. These results illustrate that the efficiency of the deep CNN approach is better than that of the other methods, including Lasso regression, which is a regression method known for its advantage in high-dimension cases. A comparison between deep CNN, deep neural network, and three other common regression methods was carried out, and the efficiency of the CNN deep learning approach, in comparison with other regression models, was demonstrated.

Towards clinical application of image mining: a systematic review on artificial intelligence and radiomics

PURPOSE

The aim of this systematic review was to analyse literature on artificial intelligence (AI) and radiomics, including all medical imaging modalities, for oncological and non-oncological applications, in order to assess how far the image mining research stands from routine medical application. To do this, we applied a trial phases classification inspired from the drug development process.

METHODS

Among the articles we considered for inclusion from PubMed were multimodality AI and radiomics investigations, with a validation analysis aimed at relevant clinical objectives. Quality assessment of selected papers was performed according to the QUADAS-2 criteria. We developed the phases classification criteria for image mining studies.

RESULTS

Overall 34,626 articles were retrieved, 300 were selected applying the inclusion/exclusion criteria, and 171 high-quality papers (QUADAS-2 ≥ 7) were identified and analysed. In 27/171 (16%), 141/171 (82%), and 3/171 (2%) studies the development of an AI-based algorithm, radiomics model, and a combined radiomics/AI approach, respectively, was described. A total of 26/27(96%) and 1/27 (4%) AI studies were classified as phase II and III, respectively. Consequently, 13/141 (9%), 10/141 (7%), 111/141 (79%), and 7/141 (5%) radiomics studies were classified as phase 0, I, II, and III, respectively. All three radiomics/AI studies were categorised as phase II trials.

CONCLUSIONS

The results of the studies are promising but still not mature enough for image mining tools to be implemented in the clinical setting and be widely used. The transfer learning from the well-known drug development process, with some specific adaptations to the image mining discipline could represent the most effective way for radiomics and AI algorithms to become the standard of care tools.

Radiogenomic analysis of vascular endothelial growth factor in patients with diffuse gliomas

OBJECTIVE

To predict vascular endothelial growth factor (VEGF) expression in patients with diffuse gliomas using radiomic analysis.

MATERIALS AND METHODS

Preoperative magnetic resonance images were retrospectively obtained from 239 patients with diffuse gliomas (World Health Organization grades II-IV). The patients were randomly assigned to a training group (n = 160) or a validation group (n = 79) at a 2:1 ratio. For each patient, a total of 431 radiomic features were extracted. The minimum redundancy maximum relevance (mRMR) algorithm was used for feature selection. A machine-learning model for predicting VEGF status was then developed using the selected features and a support vector machine classifier. The predictive performance of the model was evaluated in both groups using receiver operating characteristic curve analysis, and correlations between selected features were assessed.

RESULTS

Nine radiomic features were selected to generate a VEGF-associated radiomic signature of diffuse gliomas based on the mRMR algorithm. This radiomic signature consisted of two first-order statistics or related wavelet features (Entropy and Minimum) and seven textural features or related wavelet features (including Cluster Tendency and Long Run Low Gray Level Emphasis). The predictive efficiencies measured by the area under the curve were 74.1% in the training group and 70.2% in the validation group. The overall correlations between the 9 radiomic features were low in both groups.

CONCLUSIONS

Radiomic analysis facilitated efficient prediction of VEGF status in diffuse gliomas, suggesting that using tumor-derived radiomic features for predicting genomic information is feasible.

Bringing radiomics into a multi-omics framework for a comprehensive genotype-phenotype characterization of oncological diseases

Genomic and radiomic data integration, namely radiogenomics, can provide meaningful knowledge in cancer diagnosis, prognosis and treatment. Despite several data structures based on multi-layer architecture proposed to combine multi-omic biological information, none of these has been designed and assessed to include radiomic data as well. To meet this need, we propose to use the MultiAssayExperiment (MAE), an R package that provides data structures and methods for manipulating and integrating multi-assay experiments, as a suitable tool to manage radiogenomic experiment data. To this aim, we first examine the role of radiogenomics in cancer phenotype definition, then the current state of radiogenomics data integration in public repository and, finally, challenges and limitations of including radiomics in MAE, designing an extended framework and showing its application on a case study from the TCGA-TCIA archives. Radiomic and genomic data from 91 patients have been successfully integrated in a single MAE object, demonstrating the suitability of the MAE data structure as container of radiogenomic data.

Preoperative prediction of microvascular invasion in hepatocellular cancer: a radiomics model using Gd-EOB-DTPA-enhanced MRI

OBJECTIVES

Preoperative prediction of microvascular invasion (MVI) in patients with hepatocellular cancer (HCC) is important for surgery strategy making. We aimed to develop and validate a combined intratumoural and peritumoural radiomics model based on gadolinium-ethoxybenzyl-diethylenetriamine (Gd-EOB-DTPA)-enhanced magnetic resonance imaging (MRI) for preoperative prediction of MVI in primary HCC patients.

METHODS

This study included a training cohort of 110 HCC patients and a validating cohort of 50 HCC patients. All the patients underwent preoperative Gd-EOB-DTPA-enhanced MRI examination and curative hepatectomy. The volumes of interest (VOIs) around the hepatic lesions including intratumoural and peritumoural regions were manually delineated in the hepatobiliary phase of MRI images, from which quantitative features were extracted and analysed. In the training cohort, machine-learning method was applied for dimensionality reduction and selection of the extracted features.

RESULTS

The proportion of MVI-positive patients was 38.2% and 40.0% in the training and validation cohort, respectively. Supervised machine learning selected ten features to establish a predictive model for MVI. The area under the receiver operating characteristic curve (AUC), sensitivity, specificity of the combined intratumoural and peritumoural radiomics model in the training and validation cohort were 0.85 (95% confidence interval (CI), 0.77-0.93), 88.2%, 76.2%, and 0.83 (95% CI, 0.71-0.95), 90.0%, 75.0%, respectively.

CONCLUSIONS

We evaluate quantitative Gd-EOB-DTPA-enhanced MRI image features of both intratumoural and peritumoural regions and provide an effective radiomics-based model for the prediction of MVI in HCC patients, and may therefore help clinicians make precise decisions regarding treatment before the surgery.

KEY POINTS

• An effective radiomics model for prediction of microvascular invasion in HCC patients is established. • The radiomics model is superior to the radiologist in prediction of MVI. • The radiomics model can help clinicians in pretreatment decision making.

Radiomics for the prediction of EGFR mutation subtypes in non-small cell lung cancer

PURPOSE

This retrospective study was designed to investigate the ability of radiomics to predict the mutation status of epidermal growth factor receptor (EGFR) subtypes (19Del and L858R) in patients with non-small cell lung cancer (NSCLC).

METHODS

In total, 312 patients with NSCLC were included, and 580 radiomic features were extracted from the computed tomography images of each patient. In the training set, univariate analysis was performed on the clinical and radiomic features; logistic regression models were established using a 5-fold cross validation strategy for the prediction of EGFR subtypes 19Del and L858R. Subsequently, the predictive ability of the joint models was evaluated using the test set.

RESULTS

The results revealed that the radiomic features specific for EGFR 19Del and L858R were Gabor's MTRVariance, Gabor's PTREntropy, and sphericity. Additionally, the respective areas under the receiver operating characteristic curves of the EGFR 19Del and L858R joint models were 0.7925 and 0.7750 for the test set.

CONCLUSIONS

Our study demonstrated the potential for radiomics to predict EGFR 19Del and L858R. Epidermal growth factor receptor 19Del and L858R exhibited distinct imaging phenotypes, which may help to guide the selection of more accurate and personalized treatment programs for patients with NSCLC.

Predicting Breast Cancer in Breast Imaging Reporting and Data System (BI-RADS) Ultrasound Category 4 or 5 Lesions: A Nomogram Combining Radiomics and BI-RADS

Radiomics reflects the texture and morphological features of tumours by quantitatively analysing the grey values of medical images. We aim to develop a nomogram incorporating radiomics and the Breast Imaging Reporting and Data System (BI-RADS) for predicting breast cancer in BI-RADS ultrasound (US) category 4 or 5 lesions. From January 2017 to August 2018, a total of 315 pathologically proven breast lesions were included. Patients from the study population were divided into a training group (n = 211) and a validation group (n = 104) according to a cut-off date of March 1^st, 2018. Each lesion was assigned a category (4A, 4B, 4C or 5) according to the second edition of the American College of Radiology (ACR) BI-RADS US. A radiomics score was generated from the US image. A nomogram was developed based on the results of multivariate regression analysis from the training group. Discrimination, calibration and clinical usefulness of the nomogram for predicting breast cancer were assessed in the validation group. The radiomics score included 9 selected radiomics features. The radiomics score and BI-RADS category were independently associated with breast malignancy. The nomogram incorporating the radiomics score and BI-RADS category showed better discrimination (area under the receiver operating characteristic curve [AUC]: 0.928; 95% confidence interval [CI]: 0.876, 0.980) between malignant and benign lesions than either the radiomics score (P = 0.029) or BI-RADS category (P = 0.011). The nomogram demonstrated good calibration and clinical usefulness. In conclusion, the nomogram combining the radiomics score and BI-RADS category is potentially useful for predicting breast malignancy in BI-RADS US category 4 or 5 lesions.

Computational quantitative MR image features - a potential useful tool in differentiating glioblastoma from solitary brain metastasis

PURPOSE

Glioblastomas (GBM) and metastases are the most frequent malignant brain tumors in the adult population. Their presentation on conventional MRI is quite similar, but treatment strategy and prognosis are substantially different. Even with advanced MR techniques, in some cases diagnostic uncertainty remains. The main objective of this study was to determine whether fractal, texture, or both MR image analyses could aid in differentiating glioblastoma from solitary brain metastasis.

METHOD

In a retrospective study of 55 patients (30 glioblastomas and 25 solitary metastases) who underwent T2W/SWI/CET1 MRI, quantitative parameters of fractal and texture analysis were estimated, using box-counting and gray level co-occurrence matrix (GLCM) methods.

RESULTS

All five GLCM parameters obtained from T2W images showed significant difference between glioblastomas and solitary metastases, as well as on CET1 images except correlation (S_COR), contrary to SWI images which showed different values of two parameters (angular second moment-S_ASM and contrast-S_CON). Only three fractal features (binary box dimension-D_bin, normalized box dimension-D_norm and lacunarity-λ) measured on T2W and D_norm measured on CET1 images significantly differed GBMs from solitary metastases. The highest sensitivity and specificity were obtained from inverse difference moment (S_IDM) on T2W and S_IDM on CET1 images, respectively. Combination of several GLCM parameters yielded better results. The processing of T2W images provided the most significantly different parameters between the groups, followed by CET1 and SWI images.

CONCLUSIONS

Computational-aided quantitative image analysis may potentially improve diagnostic accuracy. According to our results texture features are more significant than fractal-based features in differentiation glioblastoma from solitary metastasis.

Quality of science and reporting of radiomics in oncologic studies: room for improvement according to radiomics quality score and TRIPOD statement

OBJECTIVES

To evaluate radiomics studies according to radiomics quality score (RQS) and Transparent Reporting of a multivariable prediction model for Individual Prognosis Or Diagnosis (TRIPOD) to provide objective measurement of radiomics research.

MATERIALS AND METHODS

PubMed and Embase were searched for studies published in high clinical imaging journals until December 2018 using the terms "radiomics" and "radiogenomics." Studies were scored against the items in the RQS and TRIPOD guidelines. Subgroup analyses were performed for journal type (clinical vs. imaging), intended use (diagnostic vs. prognostic), and imaging modality (CT vs. MRI), and articles were compared using Fisher's exact test and Mann-Whitney analysis.

RESULTS

Seventy-seven articles were included. The mean RQS score was 26.1% of the maximum (9.4 out of 36). The RQS was low in demonstration of clinical utility (19.5%), test-retest analysis (6.5%), prospective study (3.9%), and open science (3.9%). None of the studies conducted a phantom or cost-effectiveness analysis. The adherence rate for TRIPOD was 57.8% (mean) and was particularly low in reporting title (2.6%), stating study objective in abstract and introduction (7.8% and 16.9%), blind assessment of outcome (14.3%), sample size (6.5%), and missing data (11.7%) categories. Studies in clinical journals scored higher and more frequently adopted external validation than imaging journals.

CONCLUSIONS

The overall scientific quality and reporting of radiomics studies is insufficient. Scientific improvements need to be made to feature reproducibility, analysis of clinical utility, and open science categories. Reporting of study objectives, blind assessment, sample size, and missing data is deemed to be necessary.

KEY POINTS

• The overall scientific quality and reporting of radiomics studies is insufficient. • The RQS was low in demonstration of clinical utility, test-retest analysis, prospective study, and open science. • Room for improvement was shown in TRIPOD in stating study objective in abstract and introduction, blind assessment of outcome, sample size, and missing data categories.

CT imaging-based histogram features for prediction of EGFR mutation status of bone metastases in patients with primary lung adenocarcinoma

OBJECTIVE

To identify imaging markers that reflect the epidermal growth factor receptor (EGFR) mutation status by comparing computed tomography (CT) imaging-based histogram features between bone metastases with and without EGFR mutation in patients with primary lung adenocarcinoma.

MATERIALS AND METHODS

This retrospective study included 57 patients, with pathologically confirmed bone metastasis of primary lung adenocarcinoma. EGFR mutation status of bone metastases was confirmed by gene detection. The CT imaging of the metastatic bone lesions which were obtained between June 2014 and December 2017 were collected and analyzed. A total of 42 CT imaging-based histogram features were automatically extracted. Feature selection was conducted using Student's t-test, Mann-Whitney U test, single-factor logistic regression analysis and Spearman correlation analysis. A receiver operating characteristic (ROC) curve was plotted to compare the effectiveness of features in distinguishing between EGFR(+) and EGFR(-) groups. DeLong's test was used to analyze the differences between the area under the curve (AUC) values.

RESULTS

Three histogram features, namely range, skewness, and quantile 0.975 were significantly associated with EGFR mutation status. After combining these three features and combining range and skewness, we obtained the same AUC values, sensitivity and specificity. Meanwhile, the highest AUC value was achieved (AUC 0.783), which also had a higher sensitivity (0.708) and specificity (0.788). The differences between AUC values of the three features and their various combinations were statistically insignificant.

CONCLUSION

CT imaging-based histogram features of bone metastases with and without EGFR mutation in patients with primary lung adenocarcinoma were identified, and they may contribute to diagnosis and prediction of EGFR mutation status.

Texture Analysis in Cerebral Gliomas: A Review of the Literature

Texture analysis is a continuously evolving, noninvasive radiomics technique to quantify macroscopic tissue heterogeneity indirectly linked to microscopic tissue heterogeneity beyond human visual perception. In recent years, systemic oncologic applications of texture analysis have been increasingly explored. Here we discuss the basic concepts and methodologies of texture analysis, along with a review of various MR imaging texture analysis applications in glioma imaging. We also discuss MR imaging texture analysis limitations and the technical challenges that impede its widespread clinical implementation. With continued advancement in computational processing, MR imaging texture analysis could potentially develop into a valuable clinical tool in routine oncologic imaging.

Radiogenomics: bridging imaging and genomics

From diagnostics to prognosis to response prediction, new applications for radiomics are rapidly being developed. One of the fastest evolving branches involves linking imaging phenotypes to the tumor genetic profile, a field commonly referred to as "radiogenomics." In this review, a general outline of radiogenomic literature concerning prominent mutations across different tumor sites will be provided. The field of radiogenomics originates from image processing techniques developed decades ago; however, many technical and clinical challenges still need to be addressed. Nevertheless, increasingly accurate and robust radiogenomic models are being presented and the future appears to be bright.

Toward automatic prediction of EGFR mutation status in pulmonary adenocarcinoma with 3D deep learning

To develop a deep learning system based on 3D convolutional neural networks (CNNs), and to automatically predict EGFR‐mutant pulmonary adenocarcinoma in CT images. A dataset of 579 nodules with EGFR mutation status labels of mutant (Mut) or wild‐type (WT) was retrospectively analyzed. A deep learning system, namely 3D DenseNets, was developed to process 3D patches of nodules from CT data, and learn strong representations with supervised end‐to‐end training. The 3D DenseNets were trained with a training subset of 348 nodules and tuned with a development subset of 116 nodules. A strong data augmentation technique, mixup, was used for better generalization. We evaluated our model on a holdout subset of 115 nodules. An independent public dataset of 37 nodules from the cancer imaging archive (TCIA) was also used to test the generalization of our method. Conventional radiomics analysis was also performed for comparison. Our method achieved promising performance on predicting EGFR mutation status, with AUCs of 75.8% and 75.0% for our holdout test set and public test set, respectively. Moreover, strong relations were found between deep learning feature and conventional radiomics, while deep learning worked through an enhanced radiomics manner, that is, deep learned radiomics (DLR), in terms of robustness, compactness and expressiveness. The proposed deep learning system predicts EGFR‐mutant of lung adenocarcinomas in CT images noninvasively and automatically, indicating its potential to help clinical decision‐making by identifying eligible patients of pulmonary adenocarcinoma for EGFR‐targeted therapy.

The Applications of Radiomics in Precision Diagnosis and Treatment of Oncology: Opportunities and Challenges

Medical imaging can assess the tumor and its environment in their entirety, which makes it suitable for monitoring the temporal and spatial characteristics of the tumor. Progress in computational methods, especially in artificial intelligence for medical image process and analysis, has converted these images into quantitative and minable data associated with clinical events in oncology management. This concept was first described as radiomics in 2012. Since then, computer scientists, radiologists, and oncologists have gravitated towards this new tool and exploited advanced methodologies to mine the information behind medical images. On the basis of a great quantity of radiographic images and novel computational technologies, researchers developed and validated radiomic models that may improve the accuracy of diagnoses and therapy response assessments. Here, we review the recent methodological developments in radiomics, including data acquisition, tumor segmentation, feature extraction, and modelling, as well as the rapidly developing deep learning technology. Moreover, we outline the main applications of radiomics in diagnosis, treatment planning and evaluations in the field of oncology with the aim of developing quantitative and personalized medicine. Finally, we discuss the challenges in the field of radiomics and the scope and clinical applicability of these methods.

Radiomics-based machine learning methods for isocitrate dehydrogenase genotype prediction of diffuse gliomas

Purpose

Reliable and accurate predictive models are necessary to drive the success of radiomics. Our aim was to identify the optimal radiomics-based machine learning method for isocitrate dehydrogenase (IDH) genotype prediction in diffuse gliomas.

Methods

Eight classical machine learning methods were evaluated in terms of their stability and performance for pre-operative IDH genotype prediction. A total of 126 patients were enrolled for analysis. Overall, 704 radiomic features extracted from the pre-operative MRI images were analyzed. The patients were randomly assigned to either the training set or the validation set at a ratio of 2:1. Feature selection and classification model training were done using the training set, whereas the predictive performance and stability of the model were independently assessed using the validation set.

Results

Random Forest (RF) showed high predictive performance (accuracy 0.885 ± 0.041, AUC 0.931 ± 0.036), whereas neural network (NN) (accuracy 0.829 ± 0.064, AUC 0.878 ± 0.052) and flexible discriminant analysis (FDA) (accuracy 0.851 ± 0.049, AUC 0.875 ± 0.057) displayed low predictive performance. With regard to stability, RF also showed high robustness against data perturbation (relative standard deviations, RSD 3.87%).

Conclusions

RF is a promising machine learning method in predicting IDH genotype. Development of an accurate and reliable model can assist in the initial diagnostic evaluation and treatment planning for diffuse glioma patients.

Electronic supplementary material

The online version of this article (10.1007/s00432-018-2787-1) contains supplementary material, which is available to authorized users.

Pretreatment prediction of immunoscore in hepatocellular cancer: a radiomics-based clinical model based on Gd-EOB-DTPA-enhanced MRI imaging

OBJECTIVES

Immunoscore evaluates the density of CD3+ and CD8+ T cells in both the tumor core and invasive margin. Pretreatment prediction of immunoscore in hepatocellular cancer (HCC) is important for precision immunotherapy. We aimed to develop a radiomics model based on gadolinium-ethoxybenzyl-diethylenetriamine (Gd-EOB-DTPA)-enhanced MRI for pretreatment prediction of immunoscore (0-2 vs. 3-4) in HCC.

MATERIALS AND METHODS

The study included 207 (training cohort: n = 150; validation cohort: n = 57) HCC patients with hepatectomy who underwent preoperative Gd-EOB-DTPA-enhanced MRI. The volumes of interest enclosing hepatic lesions including intratumoral and peritumoral regions were manually delineated in the hepatobiliary phase of MRI images, from which 1044 quantitative features were extracted and analyzed. Extremely randomized tree method was used to select radiomics features for building radiomics model. Predicting performance in immunoscore was compared among three models: (1) using only intratumoral radiomics features (intratumoral radiomics model); (2) using combined intratumoral and peritumoral radiomics features (combined radiomics model); (3) using clinical data and selected combined radiomics features (combined radiomics-based clinical model).

RESULTS

The combined radiomics model showed a better predicting performance in immunoscore than intratumoral radiomics model (AUC, 0.904 (95% CI 0.855-0.953) vs. 0.823 (95% CI 0.747-0.899)). The combined radiomics-based clinical model showed an improvement over the combined radiomics model in predicting immunoscore (AUC, 0·926 (95% CI 0·884-0·967) vs. 0·904 (95% CI 0·855-0·953)), although differences were not statistically significant. Results were confirmed in validation cohort and calibration curves showed good agreement.

CONCLUSION

The MRI-based combined radiomics nomogram is effective in predicting immunoscore in HCC and may help making treatment decisions.

KEY POINTS

• Radiomics obtained from Gd-EOB-DTPA-enhanced MRI help predicting immunoscore in hepatocellular carcinoma. • Combined intratumoral and peritumoral radiomics are superior to intratumoral radiomics only in predicting immunoscore. • We developed a combined clinical and radiomicsnomogram to predict immunoscore in hepatocellular carcinoma.

Radiomics based on multicontrast MRI can precisely differentiate among glioma subtypes and predict tumour-proliferative behaviour

PURPOSE

To explore the feasibility and diagnostic performance of radiomics based on anatomical, diffusion and perfusion MRI in differentiating among glioma subtypes and predicting tumour proliferation.

METHODS

220 pathology-confirmed gliomas and ten contrasts were included in the retrospective analysis. After being registered to T2FLAIR images and resampling to 1 mm³ isotropically, 431 radiomics features were extracted from each contrast map within a semi-automatic defined tumour volume. For single-contrast and the combination of all contrasts, correlations between the radiomics features and pathological biomarkers were revealed by partial correlation analysis, and multivariate models were built to identify the best predictive models with adjusted 0.632+ bootstrap AUC.

RESULTS

In univariate analysis, both non-wavelet and wavelet radiomics features were correlated significantly with tumour grade and the Ki-67 labelling index. The max R was 0.557 (p = 2.04E-14) in T₁C for tumour grade and 0.395 (p = 2.33E-07) in ADC for Ki-67. In the multivariate analysis, the combination of all-contrast radiomics features had the highest AUCs in both differentiating among glioma subtypes and predicting proliferation compared with those in single-contrast images. For low-/high-grade gliomas, the best AUC was 0.911. In differentiating among glioma subtypes, the best AUC was 0.896 for grades II-III, 0.997 for grades II-IV, and 0.881 for grades III-IV. In predicting proliferation levels, multicontrast features led to an AUC of 0.936.

CONCLUSION

Multicontrast radiomics supplies complementary information on both geometric characters and molecular biological traits, which correlated significantly with tumour grade and proliferation. Combining all-contrast radiomics models might precisely predict glioma biological behaviour, which may be attributed to presurgical personal diagnosis.

KEY POINTS

• Multicontrast MRI radiomics features are significantly correlated with tumour grade and Ki-67 LI. • Multimodality MRI provides independent but supplemental information in assessing glioma pathological behaviour. • Combined multicontrast MRI radiomics can precisely predict glioma subtypes and proliferation levels.

Radiomics Nomogram Building From Multiparametric MRI to Predict Grade in Patients With Glioma: A Cohort Study

BACKGROUND

Accurate classification of gliomas is crucial for prescribing therapy and assessing the prognosis of patients.

PURPOSE

To develop a radiomics nomogram using multiparametric MRI for predicting glioma grading.

STUDY TYPE

Retrospective.

POPULATION

This study involved 85 patients (training cohort: n = 56; validation cohort: n = 29) with pathologically confirmed gliomas.

FIELD STRENGTH/SEQUENCE

1.5T MR, containing contrast-enhanced T₁ -weighted (CET₁ WI), axial T₂ -weighted (T₂ WI), and apparent diffusion coefficient (ADC) sequences.

ASSESSMENT

A region of interest of the tumor was delineated. A total of 652 radiomics features were extracted and were reduced using least absolute shrinkage and selection operator regression.

STATISTICAL TESTING

Radiomic signature, participant's age, and gender were analyzed as potential predictors to perform logistic regression analysis and develop a prediction model of glioma grading, and a radiomics nomogram was used to represent this model. The performance of the nomogram was assessed in terms of discrimination, calibration, and clinical value in glioma grading.

RESULTS

The radiomic signature was significantly associated with glioma grade (P < 0.001) in both the training and validation cohorts. The performance of the radiomics nomogram derived from three MRI sequences (with C-index of 0.971 and 0.961 in the training and validation cohorts, respectively) was improved compared to those based on either CET₁ WI, T₂ WI, or ADC alone in glioma grading (with C-index of 0.914, 0.714, 0.842 in the training cohort, and 0.941, 0.500, 0.730 in the validation cohort). The nomogram derived from three sequences showed good calibration: the calibration curve showed good agreement between the estimated and the actual probability. The decision curve demonstrated that combining three sequences had more favorable clinical predictive value than single sequence imaging.

DATA CONCLUSION

We created and assessed a multiparametric MRI-based radiomics nomogram that may help clinicians classify gliomas more accurately.

LEVEL OF EVIDENCE

4 Technical Efficacy: Stage 2 J. Magn. Reson. Imaging 2019;49:825-833.

Predicting Malignancy in Thyroid Nodules: Radiomics Score Versus 2017 American College of Radiology Thyroid Imaging, Reporting and Data System

BACKGROUND

Visual interpretation of ultrasound (US) images alone may not be sensitive enough to detect important features of potentially malignant thyroid nodules. The aim of this study was to develop a radiomics score using US imaging to predict the probability for malignancy of thyroid nodules as compared with the Thyroid Imaging, Reporting, and Data System (TI-RADS) scoring criteria proposed by the American College of Radiology (ACR).

METHODS

One hundred thirty-seven pathologically proven thyroid nodules from hospital 1 were enrolled as a training cohort, while 95 nodules from hospital 2 served as the validation cohort. A radiomics score using US images was developed from the training cohort. Two junior and two senior radiologists reviewed all images and scored each nodule according to the 2017 updated ACR TI-RADS scoring criteria. Univariate logistic regression analysis was used to develop the prediction models based on the radiomics score and ACR scores. The performance of the models was evaluated and compared with respect to discrimination, calibration, and clinical application in the validation cohort.

RESULTS

Univariate regression indicated that the radiomics score and ACR scores were predictors for thyroid nodule malignancy (all p < 0.001). Five prediction models were built based on the above scores. The radiomics score showed good discrimination with an AUC of 0.921 in the training cohort and 0.931 in the validation cohort, which was significantly better than the ACR scores of junior radiologists in both cohorts. Although five models showed good calibration (all p > 0.05), the model based on the radiomics score presented the lowest errors (E max = 0.073 or E aver = 0.028) in predicting and calibrating probabilities. Decision curve analysis demonstrated that the model using the radiomics score added more benefit than using the ACR scores of junior radiologists.

CONCLUSION

Compared with ACR TI-RADS evaluation by junior radiologists, the radiomics score showed good performance in predicting malignancy of thyroid nodules in our set of histologically verified thyroid nodules from two tertiary hospitals.

Radiomics of high-grade serous ovarian cancer: association between quantitative CT features, residual tumour and disease progression within 12 months

OBJECTIVES

To determine if radiomic features, alone or combined with clinical data, are associated with residual tumour (RT) at surgery, and predict the risk of disease progression within 12 months (PD12) in ovarian cancer (OC) patients.

METHODS

This retrospective study enrolled 101 patients according to the following inclusion parameters: cytoreductive surgery performed at our institution (9 May 2007-23 February 2016), assessment of BRCA mutational status, preoperative CT available. Radiomic features of the ovarian masses were extracted from 3D structures drawn on CT images. A phantom experiment was performed to assess the reproducibility of radiomic features. The final radiomic features included in the analysis (n = 516) were grouped into clusters using a hierarchical clustering procedure. The association of each cluster's representative radiomic feature with RT and PD12 was assessed by chi-square test. Multivariate analysis was performed using logistic regression models. P values < 0.05 were considered significant.

RESULTS

Patients with values of F2-Shape/Compactness1 below the median, of F1- GrayLevelCooccurenceMatrix25/0-1InformationMeasureCorr2 below the median and of F1-GrayLevelCooccurenceMatrix25/-333-1InverseVariance above the median showed higher risk of RT (36%, 36% and 35%, respectively, as opposed to 18%, 18% and 18%). Patients with values of F4-GrayLevelRunLengthMatrix25/-333RunPercentage above the median, of F2 shape/Max3DDiameter below the median and F1-GrayLevelCooccurenceMatrix25/45-1InverseVariance above the median showed higher risk of PD12 (22%, 24% and 23%, respectively, as opposed to 6%, 5% and 6%). At multivariate analysis F2-Shape/Max3DDiameter remained significant (odds ratio (95% CI) = 11.86 (1.41-99.88)). To predict PD12, a clinical radiomics model performed better than a base clinical model.

CONCLUSION

This study demonstrated significant associations between radiomic features and prognostic factors such as RT and PD12.

KEY POINTS

• No residual tumour (RT) at surgery is the most important prognostic factor in OC. • Radiomic features related to mass size, randomness and homogeneity were associated with RT. • Progression of disease within 12 months (PD12) indicates worse prognosis in OC. • A model including clinical and radiomic features performed better than only-clinical model to predict PD12.

Re-irradiation of recurrent gliomas: pooled analysis and validation of an established prognostic score-report of the Radiation Oncology Group (ROG) of the German Cancer Consortium (DKTK)

The heterogeneity of high‐grade glioma recurrences remains an ongoing challenge for the interdisciplinary neurooncology team. Response to re‐irradiation (re‐RT) is heterogeneous, and survival data depend on prognostic factors such as tumor volume, primary histology, age, the possibility of reresection, or time between primary diagnosis and initial RT and re‐RT. In the present pooled analysis, we gathered data from radiooncology centers of the DKTK Consortium and used it to validate the established prognostic score by Combs et al. and its modification by Kessel et al. Data consisted of a large independent, multicenter cohort of 565 high‐grade glioma patients treated with re‐RT from 1997 to 2016 and a median dose of 36 Gy. Primary RT was between 1986 and 2015 with a median dose of 60 Gy. Median age was 54 years; median follow‐up was 7.1 months. Median OS after re‐RT was 7.5, 9.5, and 13.8 months for WHO IV, III, and I/II gliomas, respectively. All six prognostic factors were tested for their significance on OS. Aside from the time from primary RT to re‐RT (P = 0.074) and the reresection status (P = 0.101), all factors (primary histology, age, KPS, and tumor volume) were significant. Both the original and new score showed a highly significant influence on survival with P < 0.001. Both prognostic scores successfully predict survival after re‐RT and can easily be applied in the routine clinical workflow. Now, further prognostic features need to be found to even improve treatment decisions regarding neurooncological interventions for recurrent glioma patients.