Unsupervised Clustering of Quantitative Image Phenotypes Reveals Breast Cancer Subtypes with Distinct Prognoses and Molecular Pathways

Radiomics-based discriminant analysis of principal components to stratify the treatment response of lung metastases following stereotactic body radiation therapy

PURPOSE

Discriminant analysis of principal components (DAPC) was introduced to describe the clusters of genetically related individuals focusing on the variation between the groups of individuals. Borrowing this approach, we evaluated the potential of DAPC for the evaluation of clusters in terms of treatment response to SBRT of lung lesions using radiomics analysis on pre-treatment CT images.

MATERIALS AND METHODS

80 pulmonary metastases from 56 patients treated with SBRT were analyzed. Treatment response was stratified as complete, incomplete and null responses. For each lesion, 107 radiomics features were extracted using the PyRadiomics software. The concordance correlation coefficients (CCC) between the radiomics features obtained by two segmentations were calculated. DAPC analysis was performed to infer the structure of "radiomically" related lesions for treatment response assessment. The DAPC was performed using the "adegenet" package for the R software.

RESULTS

The overall mean CCC was 0.97 ± 0.14. The analysis yields 14 dimensions in order to explain 95 % of the variance. DAPC was able to group the 80 lesions into the 3 different clusters based on treatment response depending on the radiomics features characteristics. The first Linear Discriminant achieved the best discrimination of individuals into the three pre-defined groups. The greater radiomics loadings who contributed the most to the treatment response differentiation were associated with the "sphericity", "correlation" and "maximal correlation coefficient" features.

CONCLUSION

This study demonstrates that a DAPC analysis based on radiomics features obtained from pretreatment CT is able to provide a reliable stratification of complete, incomplete or null response of lung metastases following SBRT.

Enhancing NSCLC recurrence prediction with PET/CT habitat imaging, ctDNA, and integrative radiogenomics-blood insights

While we recognize the prognostic importance of clinicopathological measures and circulating tumor DNA (ctDNA), the independent contribution of quantitative image markers to prognosis in non-small cell lung cancer (NSCLC) remains underexplored. In our multi-institutional study of 394 NSCLC patients, we utilize pre-treatment computed tomography (CT) and 18F-fluorodeoxyglucose positron emission tomography (FDG-PET) to establish a habitat imaging framework for assessing regional heterogeneity within individual tumors. This framework identifies three PET/CT subtypes, which maintain prognostic value after adjusting for clinicopathologic risk factors including tumor volume. Additionally, these subtypes complement ctDNA in predicting disease recurrence. Radiogenomics analysis unveil the molecular underpinnings of these imaging subtypes, highlighting downregulation in interferon alpha and gamma pathways in the high-risk subtype. In summary, our study demonstrates that these habitat imaging subtypes effectively stratify NSCLC patients based on their risk levels for disease recurrence after initial curative surgery or radiotherapy, providing valuable insights for personalized treatment approaches.

Clustering on hierarchical heterogeneous data with prior pairwise relationships

BACKGROUND

Clustering is a fundamental problem in statistics and has broad applications in various areas. Traditional clustering methods treat features equally and ignore the potential structure brought by the characteristic difference of features. Especially in cancer diagnosis and treatment, several types of biological features are collected and analyzed together. Treating these features equally fails to identify the heterogeneity of both data structure and cancer itself, which leads to incompleteness and inefficacy of current anti-cancer therapies.

OBJECTIVES

In this paper, we propose a clustering framework based on hierarchical heterogeneous data with prior pairwise relationships. The proposed clustering method fully characterizes the difference of features and identifies potential hierarchical structure by rough and refined clusters.

RESULTS

The refined clustering further divides the clusters obtained by the rough clustering into different subtypes. Thus it provides a deeper insight of cancer that can not be detected by existing clustering methods. The proposed method is also flexible with prior information, additional pairwise relationships of samples can be incorporated to help to improve clustering performance. Finally, well-grounded statistical consistency properties of our proposed method are rigorously established, including the accurate estimation of parameters and determination of clustering structures.

CONCLUSIONS

Our proposed method achieves better clustering performance than other methods in simulation studies, and the clustering accuracy increases with prior information incorporated. Meaningful biological findings are obtained in the analysis of lung adenocarcinoma with clinical imaging data and omics data, showing that hierarchical structure produced by rough and refined clustering is necessary and reasonable.

Application of radiomics-based multiomics combinations in the tumor microenvironment and cancer prognosis

The advent of immunotherapy, a groundbreaking advancement in cancer treatment, has given rise to the prominence of the tumor microenvironment (TME) as a critical area of research. The clinical implications of an improved understanding of the TME are significant and far-reaching. Radiomics has been increasingly utilized in the comprehensive assessment of the TME and cancer prognosis. Similarly, the advancement of pathomics, which is based on pathological images, can offer additional insights into the panoramic view and microscopic information of tumors. The combination of pathomics and radiomics has revolutionized the concept of a "digital biopsy". As genomics and transcriptomics continue to evolve, integrating radiomics with genomic and transcriptomic datasets can offer further insights into tumor and microenvironment heterogeneity and establish correlations with biological significance. Therefore, the synergistic analysis of digital image features (radiomics, pathomics) and genetic phenotypes (genomics) can comprehensively decode and characterize the heterogeneity of the TME as well as predict cancer prognosis. This review presents a comprehensive summary of the research on important radiomics biomarkers for predicting the TME, emphasizing the interplay between radiomics, genomics, transcriptomics, and pathomics, as well as the application of multiomics in decoding the TME and predicting cancer prognosis. Finally, we discuss the challenges and opportunities in multiomics research. In conclusion, this review highlights the crucial role of radiomics and multiomics associations in the assessment of the TME and cancer prognosis. The combined analysis of radiomics, pathomics, genomics, and transcriptomics is a promising research direction with substantial research significance and value for comprehensive TME evaluation and cancer prognosis assessment.

Multiparametric MRI: From Simultaneous Rapid Acquisition Methods and Analysis Techniques Using Scoring, Machine Learning, Radiomics, and Deep Learning to the Generation of Novel Metrics

ABSTRACT

With the recent advancements in rapid imaging methods, higher numbers of contrasts and quantitative parameters can be acquired in less and less time. Some acquisition models simultaneously obtain multiparametric images and quantitative maps to reduce scan times and avoid potential issues associated with the registration of different images. Multiparametric magnetic resonance imaging (MRI) has the potential to provide complementary information on a target lesion and thus overcome the limitations of individual techniques. In this review, we introduce methods to acquire multiparametric MRI data in a clinically feasible scan time with a particular focus on simultaneous acquisition techniques, and we discuss how multiparametric MRI data can be analyzed as a whole rather than each parameter separately. Such data analysis approaches include clinical scoring systems, machine learning, radiomics, and deep learning. Other techniques combine multiple images to create new quantitative maps associated with meaningful aspects of human biology. They include the magnetic resonance g-ratio, the inner to the outer diameter of a nerve fiber, and the aerobic glycolytic index, which captures the metabolic status of tumor tissues.

Multi-View Radiomics Feature Fusion Reveals Distinct Immuno-Oncological Characteristics and Clinical Prognoses in Hepatocellular Carcinoma

Hepatocellular carcinoma (HCC) is one of the most prevalent malignancies worldwide, and the pronounced intra- and inter-tumor heterogeneity restricts clinical benefits. Dissecting molecular heterogeneity in HCC is commonly explored by endoscopic biopsy or surgical forceps, but invasive tissue sampling and possible complications limit the broadeer adoption. The radiomics framework is a promising non-invasive strategy for tumor heterogeneity decoding, and the linkage between radiomics and immuno-oncological characteristics is worth further in-depth study. In this study, we extracted multi-view imaging features from contrast-enhanced CT (CE-CT) scans of HCC patients, followed by developing a fused imaging feature subtyping (FIFS) model to identify two distinct radiomics subtypes. We observed two subtypes of patients with distinct texture-dominated radiomics profiles and prognostic outcomes, and the radiomics subtype identified by FIFS model was an independent prognostic factor. The heterogeneity was mainly attributed to inflammatory pathway activity and the tumor immune microenvironment. The predominant radiogenomics association was identified between texture-related features and immune-related pathways by integrating network analysis, and was validated in two independent cohorts. Collectively, this work described the close connections between multi-view radiomics features and immuno-oncological characteristics in HCC, and our integrative radiogenomics analysis strategy may provide clues to non-invasive inflammation-based risk stratification.

Weakly supervised label propagation algorithm classifies lung cancer imaging subtypes

Aiming at the problems of long time, high cost, invasive sampling damage, and easy emergence of drug resistance in lung cancer gene detection, a reliable and non-invasive prognostic method is proposed. Under the guidance of weakly supervised learning, deep metric learning and graph clustering methods are used to learn higher-level abstract features in CT imaging features. The unlabeled data is dynamically updated through the k-nearest label update strategy, and the unlabeled data is transformed into weak label data and continue to update the process of strong label data to optimize the clustering results and establish a classification model for predicting new subtypes of lung cancer imaging. Five imaging subtypes are confirmed on the lung cancer dataset containing CT, clinical and genetic information downloaded from the TCIA lung cancer database. The successful establishment of the new model has a significant accuracy rate for subtype classification (ACC = 0.9793), and the use of CT sequence images, gene expression, DNA methylation and gene mutation data from the cooperative hospital in Shanxi Province proves the biomedical value of this method. The proposed method also can comprehensively evaluate intratumoral heterogeneity based on the correlation between the final lung CT imaging features and specific molecular subtypes.

Explainable domain transfer of distant supervised cancer subtyping model via imaging-based rules extraction

Image texture analysis has for decades represented a promising opportunity for cancer assessment and disease progression evaluation, evolving in a discipline, i.e., radiomics. However, the road to a complete translation into clinical practice is still hampered by intrinsic limitations. As purely supervised classification models fail in devising robust imaging-based biomarkers for prognosis, cancer subtyping approaches would benefit from the employment of distant supervision, for instance exploiting survival/recurrence information. In this work, we assessed, tested, and validated the domain-generality of our previously proposed Distant Supervised Cancer Subtyping model on Hodgkin Lymphoma. We evaluate the model performance on two independent datasets coming from two hospitals, comparing and analyzing the results. Although successful and consistent, the comparison confirmed the instability of radiomics due to an across-center lack of reproducibility, leading to explainable results in one center and poor interpretability in the other. We thus propose a Random Forest-based Explainable Transfer Model for testing the domain-invariance of imaging biomarkers extracted from retrospective cancer subtyping. In doing so, we tested the predictive ability of cancer subtyping in a validation and perspective setting, which led to successful results and supported the domain-generality of the proposed approach. On the other hand, the extraction of decision rules enables to draw of risk factors and robust biomarkers to inform clinical decisions. This work shows the potentialities of the Distant Supervised Cancer Subtyping model to be further evaluated in larger multi-center datasets, to reliably translate radiomics into medical practice. The code is available at this GitHub repository.

Multiparametric Quantitative Imaging Biomarkers for Phenotype Classification: A Framework for Development and Validation

This manuscript is the third in a five-part series related to statistical assessment methodology for technical performance of multi-parametric quantitative imaging biomarkers (mp-QIBs). We outline approaches and statistical methodologies for developing and evaluating a phenotype classification model from a set of multiparametric QIBs. We then describe validation studies of the classifier for precision, diagnostic accuracy, and interchangeability with a comparator classifier. We follow with an end-to-end real-world example of development and validation of a classifier for atherosclerotic plaque phenotypes. We consider diagnostic accuracy and interchangeability to be clinically meaningful claims for a phenotype classification model informed by mp-QIB inputs, aiming to provide tools to demonstrate agreement between imaging-derived characteristics and clinically established phenotypes. Understanding that we are working in an evolving field, we close our manuscript with an acknowledgement of existing challenges and a discussion of where additional work is needed. In particular, we discuss the challenges involved with technical performance and analytical validation of mp-QIBs. We intend for this manuscript to further advance the robust and promising science of multiparametric biomarker development.

CT-based radiogenomic analysis dissects intratumor heterogeneity and predicts prognosis of colorectal cancer: a multi-institutional retrospective study

BACKGROUND

This study aimed to develop a radiogenomic prognostic prediction model for colorectal cancer (CRC) by investigating the biological and clinical relevance of intratumoural heterogeneity.

METHODS

This retrospective multi-cohort study was conducted in three steps. First, we identified genomic subclones using unsupervised deconvolution analysis. Second, we established radiogenomic signatures to link radiomic features with prognostic subclone compositions in an independent radiogenomic dataset containing matched imaging and gene expression data. Finally, the prognostic value of the identified radiogenomic signatures was validated using two testing datasets containing imaging and survival information collected from separate medical centres.

RESULTS

This multi-institutional retrospective study included 1601 patients (714 females and 887 males; mean age, 65 years ± 14 [standard deviation]) with CRC from 5 datasets. Molecular heterogeneity was identified using unsupervised deconvolution analysis of gene expression data. The relative prevalence of the two subclones associated with cell cycle and extracellular matrix pathways identified patients with significantly different survival outcomes. A radiogenomic signature-based predictive model significantly stratified patients into high- and low-risk groups with disparate disease-free survival (HR = 1.74, P = 0.003). Radiogenomic signatures were revealed as an independent predictive factor for CRC by multivariable analysis (HR = 1.59, 95% CI:1.03-2.45, P = 0.034). Functional analysis demonstrated that the 11 radiogenomic signatures were predominantly associated with extracellular matrix and immune-related pathways.

CONCLUSIONS

The identified radiogenomic signatures might be a surrogate for genomic signatures and could complement the current prognostic strategies.

Unsupervised Analysis Based on DCE-MRI Radiomics Features Revealed Three Novel Breast Cancer Subtypes with Distinct Clinical Outcomes and Biological Characteristics

Background: This study aimed to reveal the heterogeneity of dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) of breast cancer (BC) and identify its prognosis values and molecular characteristics. Methods: Two radiogenomics cohorts (n = 246) were collected and tumor regions were segmented semi-automatically. A total of 174 radiomics features were extracted, and the imaging subtypes were identified and validated by unsupervised analysis. A gene-profile-based classifier was developed to predict the imaging subtypes. The prognostic differences and the biological and microenvironment characteristics of subtypes were uncovered by bioinformatics analysis. Results: Three imaging subtypes were identified and showed high reproducibility. The subtypes differed remarkably in tumor sizes and enhancement patterns, exhibiting significantly different disease-free survival (DFS) or overall survival (OS) in the discovery cohort (p = 0.024) and prognosis datasets (p ranged from <0.0001 to 0.0071). Large sizes and rapidly enhanced tumors usually had the worst outcomes. Associations were found between imaging subtypes and the established subtypes or clinical stages (p ranged from <0.001 to 0.011). Imaging subtypes were distinct in cell cycle and extracellular matrix (ECM)-receptor interaction pathways (false discovery rate, FDR < 0.25) and different in cellular fractions, such as cancer-associated fibroblasts (p < 0.05). Conclusions: The imaging subtypes had different clinical outcomes and biological characteristics, which may serve as potential biomarkers.

Predicting hormone receptors and PAM50 subtypes of breast cancer from multi-scale lesion images of DCE-MRI with transfer learning technique

BACKGROUND

The recent development of artificial intelligence (AI) technologies coupled with medical imaging data has gained considerable attention, and offers a non-invasive approach for cancer diagnosis and prognosis. In this context, improved breast cancer (BC) molecular characteristics assessment models are foreseen to enable personalized strategies with better clinical outcomes compared to existing screening strategies. And it is a promising approach to developing models for hormone receptors (HR) and subtypes of BC patients from dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) data.

METHODS

In this institutional review board-approved study, 174 BC patients with both DCE-MRI and RNA-seq data in the local database were analyzed. Slice images from tumor lesions and multi-scale peri-tumor regions were used as model inputs, and five representative pre-trained transfer learning (TF) networks, such as Inception-v3 and Xception, were employed to establish prediction models. A comprehensive analysis was performed using five-fold cross-validation to avoid overfitting, and accuracy (ACC) and area under the receiver operating characteristic curve (AUROC) to evaluate model performance.

RESULTS

Xception achieved the superior results when using solely tumor regions, with highest AUROCs of 0.844 (95% CI: [0.841, 0.847]) and 0.784 (95% CI: [0.781, 0.788]) for estrogen receptor (ER) and progesterone receptor (PR), respectively, and best ACC of 0.467 (95% CI: [0.462, 0.470]) for PAM50 subtypes. A significant improvement in the model performance was observed when images of the peri-tumor region were included, with optimal results achieved using images of the tumor and the 10 mm peri-tumor regions. Xception-based TF models performed most effectively in predicting ER and PR statuses, with the AUROCs were 0.942 (95% CI: [0.940, 0.944]) and 0.920 (95% CI: [0.917, 0.922]), respectively, whereas for PAM50 subtypes, the Inception-v3-based network yielded the highest ACC as 0.742 (95% CI: [0.738, 0.746]).

CONCLUSIONS

Transfer learning analysis based on DCE-MRI data of tumor and peri-tumor regions was helpful to the non-invasive assessment of molecular characteristics of BC.

Integrated imaging and molecular analysis to decipher tumor microenvironment in the era of immunotherapy

Radiological imaging is an integral component of cancer care, including diagnosis, staging, and treatment response monitoring. It contains rich information about tumor phenotypes that are governed not only by cancer cellintrinsic biological processes but also by the tumor microenvironment, such as the composition and function of tumor-infiltrating immune cells. By analyzing the radiological scans using a quantitative radiomics approach, robust relations between specific imaging and molecular phenotypes can be established. Indeed, a number of studies have demonstrated the feasibility of radiogenomics for predicting intrinsic molecular subtypes and gene expression signatures in breast cancer based on MRI. In parallel, promising results have been shown for inferring the amount of tumor-infiltrating lymphocytes, a key factor for the efficacy of cancer immunotherapy, from standard-of-care radiological images. Compared with the biopsy-based approach, radiogenomics offers a unique avenue to profile the molecular makeup of the tumor and immune microenvironment as well as its evolution in a noninvasive and holistic manner through longitudinal imaging scans. Here, we provide a systematic review of the state of the art radiogenomics studies in the era of immunotherapy and discuss emerging paradigms and opportunities in AI and deep learning approaches. These technical advances are expected to transform the radiogenomics field, leading to the discovery of reliable imaging biomarkers. This will pave the way for their clinical translation to guide precision cancer therapy.

Construction of Prognostic Risk Model of Patients with Skin Cutaneous Melanoma Based on TCGA-SKCM Methylation Cohort

Skin cutaneous melanoma (SKCM) is a common malignant skin cancer. Early diagnosis could effectively reduce SKCM patient's mortality to a large extent. We managed to construct a model to examine the prognosis of SKCM patients. The methylation-related data and clinical data of The Cancer Gene Atlas- (TCGA-) SKCM were downloaded from TCGA database. After preprocessing the methylation data, 21,861 prognosis-related methylated sites potentially associated with prognosis were obtained using the univariate Cox regression analysis and multivariate Cox regression analysis. Afterward, unsupervised clustering was used to divide the patients into 4 clusters, and weighted correlation network analysis (WGCNA) was applied to construct coexpression modules. By overlapping the CpG sites between the clusters and turquoise model, a prognostic model was established by LASSO Cox regression and multivariate Cox regression. It was found that 9 methylated sites included cg01447831, cg14845689, cg20895058, cg06506470, cg09558315, cg06373660, cg17737409, cg21577036, and cg22337438. After constructing the prognostic model, the performance of the model was validated by survival analysis and receiver operating characteristic (ROC) curve, and the independence of the model was verified by univariate and multivariate regression. It was represented that the prognostic model was reliable, and riskscore could be used as an independent prognostic factor in SKCM patients. At last, we combined clinical data and patient's riskscore to establish and testify the nomogram that could determine patient's prognosis. The results found that the reliability of the nomogram was relatively good. All in all, we constructed a prognostic model that could determine the prognosis of SKCM patients and screened 9 key methylated sites through analyzing data in TCGA-SKCM dataset. Finally, a prognostic nomogram was established combined with clinical diagnosed information and riskscore. The results are significant for improving the prognosis of SKCM patients in the future.

MRI Radiogenomics in Precision Oncology: New Diagnosis and Treatment Method

Precision medicine for cancer affords a new way for the most accurate and effective treatment to each individual cancer. Given the high time-evolving intertumor and intratumor heterogeneity features of personal medicine, there are still several obstacles hindering its diagnosis and treatment in clinical practice regardless of extensive exploration on it over the past years. This paper is to investigate radiogenomics methods in the literature for precision medicine for cancer focusing on the heterogeneity analysis of tumors. Based on integrative analysis of multimodal (parametric) imaging and molecular data in bulk tumors, a comprehensive analysis and discussion involving the characterization of tumor heterogeneity in imaging and molecular expression are conducted. These investigations are intended to (i) fully excavate the multidimensional spatial, temporal, and semantic related information regarding high-dimensional breast magnetic resonance imaging data, with integration of the highly specific structured data of genomics and combination of the diagnosis and cognitive process of doctors, and (ii) establish a radiogenomics data representation model based on multidimensional consistency analysis with multilevel spatial-temporal correlations.

Quantifying Tumor Heterogeneity via MRI Habitats to Characterize Microenvironmental Alterations in HER2+ Breast Cancer

This study identifies physiological habitats using quantitative magnetic resonance imaging (MRI) to elucidate intertumoral differences and characterize microenvironmental response to targeted and cytotoxic therapy. BT-474 human epidermal growth factor receptor 2 (HER2+) breast tumors were imaged before and during treatment (trastuzumab, paclitaxel) with diffusion-weighted MRI and dynamic contrast-enhanced MRI to measure tumor cellularity and vascularity, respectively. Tumors were stained for anti-CD31, anti-ɑSMA, anti-CD45, anti-F4/80, anti-pimonidazole, and H&E. MRI data was clustered to identify and label each habitat in terms of vascularity and cellularity. Pre-treatment habitat composition was used stratify tumors into two "tumor imaging phenotypes" (Type 1, Type 2). Type 1 tumors showed significantly higher percent tumor volume of the high-vascularity high-cellularity (HV-HC) habitat compared to Type 2 tumors, and significantly lower volume of low-vascularity high-cellularity (LV-HC) and low-vascularity low-cellularity (LV-LC) habitats. Tumor phenotypes showed significant differences in treatment response, in both changes in tumor volume and physiological composition. Significant positive correlations were found between histological stains and tumor habitats. These findings suggest that the differential baseline imaging phenotypes can predict response to therapy. Specifically, the Type 1 phenotype indicates increased sensitivity to targeted or cytotoxic therapy compared to Type 2 tumors.

Radiomics predicts the prognosis of patients with locally advanced breast cancer by reflecting the heterogeneity of tumor cells and the tumor microenvironment

Background

This study investigated the efficacy of radiomics to predict survival outcome for locally advanced breast cancer (LABC) patients and the association of radiomics with tumor heterogeneity and microenvironment.

Methods

Patients with LABC from 2010 to 2015 were retrospectively reviewed. Radiomics features were extracted from enhanced MRI. We constructed the radiomics score using lasso and assessed its prognostic value. An external validation cohort from The Cancer Imaging Archive was used to assess phenotype reproducibility. Sequencing data from TCGA and our center were applied to reveal genomic landscape of different radiomics score groups. Tumor infiltrating lymphocytes map and bioinformatics methods were applied to evaluate the heterogeneity of tumor microenvironment. Computational histopathology was also applied.

Results

A total of 278 patients were divided into training cohort and validation cohort. Radiomics score was constructed and significantly associated with disease-free survival (DFS) of the patients in training cohort, validation cohort and external validation cohort (p < 0.001, p = 0.014 and p = 0.041, respectively). The radiomics-based nomogram showed better predictive performance of DFS compared with TNM model. Distinct gene expression patterns were identified. Immunophenotype and immune cell composition was different in each radiomics score group. The link between radiomics and computational histopathology was revealed.

Conclusions

The radiomics score could effectively predict prognosis of LABC after neoadjuvant chemotherapy and radiotherapy. Radiomics revealed heterogeneity of tumor cell and tumor microenvironment and holds great potential to facilitate individualized DFS estimation and guide personalized care.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13058-022-01516-0.

Image retrieval-based parenchymal analysis for breast cancer risk assessment

PURPOSE

This research on breast cancer risk assessment aims to develop models that predict the likelihood of breast cancer. In recent years, the computerized analysis of visual texture patterns in mammograms, namely parenchymal analysis, has shown great potential for risk assessment. However, the visual complexity and heterogeneity of visual patterns limit the performance of parenchymal analysis in large populations. In this work, we propose a method to create individualized risk assessment models based on the radiological visual appearance (radiomic phenotypes) of the mammograms.

METHODS

We developed a content-based image retrieval system to stratify mammographic analysis according to the similarities of their radiomic phenotypes. We collected 1144 mammograms from 286 women following a case-control study design. We compared the classical parenchymal analysis with the proposed approach using the area under the ROC curve (AUC) with 95% confidence intervals (CI). Statistical significance was assessed using DeLong's test ( p < 0.05).

RESULTS

At a patient level, AUC values of 0.504 (95% CI: 0.398-0.611) with classical parenchymal analysis increased to 0.813 (95% CI: 0.734-0.892) when the radiomic phenotypes are incorporated with the proposed method. In risk estimation from individual, standard mammographic views, the highest performance was obtained with the mediolateral oblique view of the right breast (RMLO), with an AUC value of 0.727 (95% CI: 0.634-0.820). Differences in performance among views were statistically significant ( p < 0.05 ) CONCLUSIONS: These results indicate that the utilization of radiomic phenotypes increases the performance of computerized risk assessment based on parenchymal analysis of mammographic images.

SIGNIFICANCE

The creation of individualized risk assessment models may be leveraged to target personalized screening and prevention recommendations according to the person's risk.

Contralateral parenchymal enhancement on MRI is associated with tumor proteasome pathway gene expression and overall survival of early ER+/HER2-breast cancer patients

Purpose

To assess whether contralateral parenchymal enhancement (CPE) on MRI is associated with gene expression pathways in ER+/HER2-breast cancer, and if so, whether such pathways are related to survival.

Methods

Preoperative breast MRIs were analyzed of early ER+/HER2-breast cancer patients eligible for breast-conserving surgery included in a prospective observational cohort study (MARGINS). The contralateral parenchyma was segmented and CPE was calculated as the average of the top-10% delayed enhancement. Total tumor RNA sequencing was performed and gene set enrichment analysis was used to reveal gene expression pathways associated with CPE (N = 226) and related to overall survival (OS) and invasive disease-free survival (IDFS) in multivariable survival analysis. The latter was also done for the METABRIC cohort (N = 1355).

Results

CPE was most strongly correlated with proteasome pathways (normalized enrichment statistic = 2.04, false discovery rate = .11). Patients with high CPE showed lower tumor proteasome gene expression. Proteasome gene expression had a hazard ratio (HR) of 1.40 (95% CI = 0.89, 2.16; P = .143) for OS in the MARGINS cohort and 1.53 (95% CI = 1.08, 2.14; P = .017) for IDFS, in METABRIC proteasome gene expression had an HR of 1.09 (95% CI = 1.01, 1.18; P = .020) for OS and 1.10 (95% CI = 1.02, 1.18; P = .012) for IDFS.

Conclusion

CPE was negatively correlated with tumor proteasome gene expression in early ER+/HER2-breast cancer patients. Low tumor proteasome gene expression was associated with improved survival in the METABRIC data.

•

Contralateral parenchymal enhancement on MRI was associated with tumor proteasome gene expression in ER+/HER2-breast cancer.

•

A high contralateral parenchymal enhancement was associated with a low proteasome gene expression in the breast cancer.

•

Low proteasome tumor gene expression was associated with improved survival in an independent patient cohort.

Machine Learning: Applications and Advanced Progresses of Radiomics in Endocrine Neoplasms

Endocrine neoplasms remain a great threat to human health. It is extremely important to make a clear diagnosis and timely treatment of endocrine tumors. Machine learning includes radiomics, which has long been utilized in clinical cancer research. Radiomics refers to the extraction of valuable information by analyzing a large amount of standard data with high-throughput medical images mainly including computed tomography, positron emission tomography, magnetic resonance imaging, and ultrasound. With the quantitative imaging analysis and model building, radiomics can reflect specific underlying characteristics of a disease that otherwise could not be evaluated visually. More and more promising results of radiomics in oncological practice have been seen in recent years. Radiomics may have the potential to supplement traditional imaging analysis and assist in providing precision medicine for patients. Radiomics had developed rapidly in endocrine neoplasms practice in the past decade. In this review, we would introduce the general workflow of radiomics and summarize the applications and developments of radiomics in endocrine neoplasms in recent years. The limitations of current radiomic research studies and future development directions would also be discussed.

Predicting cancer outcomes with radiomics and artificial intelligence in radiology

The successful use of artificial intelligence (AI) for diagnostic purposes has prompted the application of AI-based cancer imaging analysis to address other, more complex, clinical needs. In this Perspective, we discuss the next generation of challenges in clinical decision-making that AI tools can solve using radiology images, such as prognostication of outcome across multiple cancers, prediction of response to various treatment modalities, discrimination of benign treatment confounders from true progression, identification of unusual response patterns and prediction of the mutational and molecular profile of tumours. We describe the evolution of and opportunities for AI in oncology imaging, focusing on hand-crafted radiomic approaches and deep learning-derived representations, with examples of their application for decision support. We also address the challenges faced on the path to clinical adoption, including data curation and annotation, interpretability, and regulatory and reimbursement issues. We hope to demystify AI in radiology for clinicians by helping them to understand its limitations and challenges, as well as the opportunities it provides as a decision-support tool in cancer management.

Quantitative Multiparametric MRI as an Imaging Biomarker for the Prediction of Breast Cancer Receptor Status and Molecular Subtypes

Objectives

To assess breast cancer receptor status and molecular subtypes by using the CAIPIRINHA-Dixon-TWIST-VIBE and readout-segmented echo-planar diffusion weighted imaging techniques.

Methods

A total of 165 breast cancer patients were retrospectively recruited. Patient age, estrogen receptor, progesterone receptor, human epidermal growth factorreceptor-2 (HER-2) status, and the Ki-67 proliferation index were collected for analysis. Quantitative parameters (K^trans, V_e, K_ep), semiquantitative parameters (W_-in, W_-out, TTP), and apparent diffusion coefficient (ADC) values were compared in relation to breast cancer receptor status and molecular subtypes. Statistical analysis were performed to compare the parameters in the receptor status and molecular subtype groups.Multivariate analysis was performed to explore confounder-adjusted associations, and receiver operating characteristic curve analysis was used to assess the classification performance and calculate thresholds.

Results

Younger age (<49.5 years, odds ratio (OR) =0.95, P=0.004), lower K_ep (<0.704,OR=0.14, P=0.044),and higher TTP (>0.629 min, OR=24.65, P=0.011) were independently associated with progesterone receptor positivity. A higher TTP (>0.585 min, OR=28.19, P=0.01) was independently associated with estrogen receptor positivity. Higher K_ep (>0.892, OR=11.6, P=0.047), lower TTP (<0.582 min, OR<0.001, P=0.004), and lower ADC (<0.719 ×10^-3 mm²/s, OR<0.001, P=0.048) had stronger independent associations with triple-negative breast cancer (TNBC) compared to luminal A, and those parameters could differentiate TNBC from luminal A with the highest AUC of 0.811.

Conclusions

K_ep and TTP were independently associated with hormone receptor status. In addition, the K_ep, TTP, and ADC values had stronger independent associations with TNBC than with luminal A and could be used as imaging biomarkers for differentiate TNBC from Luminal A.

A costimulatory molecule-related signature in regard to evaluation of prognosis and immune features for clear cell renal cell carcinoma

Costimulatory molecules have been proven to enhance antitumor immune responses, but their roles in clear cell renal cell carcinoma (ccRCC) remain unexplored. In this study, we aimed to explore the gene expression profiles of costimulatory molecule genes in ccRCC and construct a prognostic signature to improve treatment decision-making and clinical outcomes. We performed the first comprehensive analysis of costimulatory molecules in patients with ccRCC and identified 13 costimulatory molecule genes with prognostic values and diagnostic values. Consensus clustering analysis based on these 13 costimulatory molecular genes showed different distribution patterns and prognostic differences for the two clusters identified. Then, a costimulatory molecule-related signature was constructed based on these 13 costimulatory molecular genes, and validated in an external dataset, showing good performance for predicting a patient’s prognosis. The signature was an independent risk factor for ccRCC patients and was significantly correlated with patients’ clinical factors, which could be used as a complement for clinical factors. In addition, the signature was associated with the tumor immune microenvironment and the response to immunotherapy. Patients identified as high-risk based on our signature exhibited a high mutation frequency, a high level of immune cell infiltration, and an immunosuppressive microenvironment. High-risk patients tended to have high cytolytic activity scores and immunophenoscore of CTLA4 and PD1/PD-L1/PD-L2 blocker than low-risk patients, suggesting these patients may be more suitable for immunotherapy. Therefore, our signature could provide clinicians with prognosis predictions and help guide treatment for ccRCC patients.

Radiogenomic Analysis of Locally Advanced Lung Cancer Based on CT Imaging and Intratreatment Changes in Cell-Free DNA

The radiologic appearance of locally advanced lung cancer may be linked to molecular changes of the disease during treatment, but characteristics of this phenomenon are poorly understood. Radiomics, liquid biopsy of cell-free DNA (cfDNA), and next-generation sequencing of circulating tumor DNA (ctDNA) encode tumor-specific radiogenomic expression patterns that can be probed to study this problem. Preliminary findings are reported from a radiogenomic analysis of CT imaging, cfDNA, and ctDNA in 24 patients (median age, 64 years; range, 49-74 years) with stage III lung cancer undergoing chemoradiation on a prospective pilot study (NCT00921739) between September 2009 and September 2014. Unsupervised clustering of radiomic signatures resulted in two clusters that were associated with ctDNA TP53 mutations (P = .03) and changes in cfDNA concentration after 2 weeks of chemoradiation (P = .02). The radiomic features dissimilarity (hazard ratio [HR] = 0.56; P = .05), joint entropy (HR = 0.56; P = .04), sum entropy (HR = 0.53; P = .02), and normalized inverse difference (HR = 1.77; P = .05) were associated with overall survival. These results suggest heterogeneous and low-attenuating disease without a detectable ctDNA TP53 mutation was associated with early surges of cfDNA concentration in response to therapy and a generally better prognosis. Keywords: CT-Quantitative, Radiation Therapy, Lung, Computer Applications-3D, Oncology, Tumor Response, Outcomes Analysis Clinical trial registration no. NCT00921739 Supplemental material is available for this article. © RSNA, 2021.

Deep learning radiomics-based prediction of distant metastasis in patients with locally advanced rectal cancer after neoadjuvant chemoradiotherapy: A multicentre study

BACKGROUND

Accurate predictions of distant metastasis (DM) in locally advanced rectal cancer (LARC) patients receiving neoadjuvant chemoradiotherapy (nCRT) are helpful in developing appropriate treatment plans. This study aimed to perform DM prediction through deep learning radiomics.

METHODS

We retrospectively sampled 235 patients receiving nCRT with the minimum 36 months' postoperative follow-up from three hospitals. Through transfer learning, a deep learning radiomic signature (DLRS) based on multiparametric magnetic resonance imaging (MRI) was constructed. A nomogram was established integrating deep MRI information and clinicopathologic factors for better prediction. Harrell's concordance index (C-index) and time-dependent receiver operating characteristic (ROC) were used as performance metrics. Furthermore, the risk of DM in patients with different response to nCRT was evaluated with the nomogram.

FINDINGS

DLRS performed well in DM prediction, with a C-index of 0·747 and an area under curve (AUC) at three years of 0·894 in the validation cohort. The performance of nomogram was better, with a C-index of 0·775. In addition, the nomogram could stratify patients with different responses to nCRT into high- and low-risk groups of DM (P < 0·05).

INTERPRETATION

MRI-based deep learning radiomics had potential in predicting the DM of LARC patients receiving nCRT and could help evaluate the risk of DM in patients who have different responses to nCRT.

FUNDING

The funding bodies that contributed to this study are listed in the Acknowledgements section.

Radiomic and radiogenomic modeling for radiotherapy: strategies, pitfalls, and challenges

The power of predictive modeling for radiotherapy outcomes has historically been limited by an inability to adequately capture patient-specific variabilities; however, next-generation platforms together with imaging technologies and powerful bioinformatic tools have facilitated strategies and provided optimism. Integrating clinical, biological, imaging, and treatment-specific data for more accurate prediction of tumor control probabilities or risk of radiation-induced side effects are high-dimensional problems whose solutions could have widespread benefits to a diverse patient population-we discuss technical approaches toward this objective. Increasing interest in the above is specifically reflected by the emergence of two nascent fields, which are distinct but complementary: radiogenomics, which broadly seeks to integrate biological risk factors together with treatment and diagnostic information to generate individualized patient risk profiles, and radiomics, which further leverages large-scale imaging correlates and extracted features for the same purpose. We review classical analytical and data-driven approaches for outcomes prediction that serve as antecedents to both radiomic and radiogenomic strategies. Discussion then focuses on uses of conventional and deep machine learning in radiomics. We further consider promising strategies for the harmonization of high-dimensional, heterogeneous multiomics datasets (panomics) and techniques for nonparametric validation of best-fit models. Strategies to overcome common pitfalls that are unique to data-intensive radiomics are also discussed.

Radiomic profiling of clear cell renal cell carcinoma reveals subtypes with distinct prognoses and molecular pathways

•

Radiomics profile of clear cell renal cell carcinoma is heterogeneity.

•

Multi-scale Radiogenomics could link molecular features and images.

•

Radiomic subtypes could be used for risk stratification.

Background

To identify radiomic subtypes of clear cell renal cell carcinoma (ccRCC) patients with distinct clinical significance and molecular characteristics reflective of the heterogeneity of ccRCC.

Methods

Quantitative radiomic features of ccRCC were extracted from preoperative CT images of 160 ccRCC patients. Unsupervised consensus cluster analysis was performed to identify robust radiomic subtypes based on these features. The Kaplan–Meier method and chi-square test were used to assess the different clinicopathological characteristics and gene mutations among the radiomic subtypes. Subtype-specific marker genes were identified, and gene set enrichment analyses were performed to reveal the specific molecular characteristics of each subtype. Moreover, a gene expression-based classifier of radiomic subtypes was developed using the random forest algorithm and tested in another independent cohort (n = 101).

Results

Radiomic profiling revealed three ccRCC subtypes with distinct clinicopathological features and prognoses. VHL, MUC16, FBN2, and FLG were found to have different mutation frequencies in these radiomic subtypes. In addition, transcriptome analysis revealed that the dysregulation of cell cycle-related pathways may be responsible for the distinct clinical significance of the obtained subtypes. The prognostic value of the radiomic subtypes was further validated in another independent cohort (log-rank P = 0.015).

Conclusion

In the present multi-scale radiogenomic analysis of ccRCC, radiomics played a central role. Radiomic subtypes could help discern genomic alterations and non-invasively stratify ccRCC patients.

Tuberculosis endotypes to guide stratified host-directed therapy

There is hope that host-directed therapy (HDT) for Tuberculosis (TB) can either shorten treatment duration, help cure drug resistant disease or limit the immunopathology. Many candidate HDT drugs have been proposed, however solid evidence only exists for a few select patient groups. The clinical presentation of TB is variable, with differences in severity, tissue pathology, and bacillary burden. TB clinical phenotypes likely determine the potential benefit of HDT. Underlying TB clinical phenotypes, there are TB "endotypes," defined as distinct molecular profiles, with specific metabolic, epigenetic, transcriptional, and immune phenotypes. TB endotypes can be characterized by either immunodeficiency or pathologic excessive inflammation. Additional factors, like comorbidities (HIV, diabetes, helminth infection), structural lung disease or Mycobacterial virulence also drive TB endotypes. Precise disease phenotyping, combined with in-depth immunologic and molecular profiling and multimodal omics integration, can identify TB endotypes, guide endotype-specific HDT, and improve TB outcomes, similar to advances in cancer medicine.

The Biological Meaning of Radiomic Features

An earlier incorrect version appeared online. This article was corrected on February 10, 2021.

Machine and deep learning methods for radiomics

Radiomics is an emerging area in quantitative image analysis that aims to relate large-scale extracted imaging information to clinical and biological endpoints. The development of quantitative imaging methods along with machine learning has enabled the opportunity to move data science research towards translation for more personalized cancer treatments. Accumulating evidence has indeed demonstrated that noninvasive advanced imaging analytics, that is, radiomics, can reveal key components of tumor phenotype for multiple three-dimensional lesions at multiple time points over and beyond the course of treatment. These developments in the use of CT, PET, US, and MR imaging could augment patient stratification and prognostication buttressing emerging targeted therapeutic approaches. In recent years, deep learning architectures have demonstrated their tremendous potential for image segmentation, reconstruction, recognition, and classification. Many powerful open-source and commercial platforms are currently available to embark in new research areas of radiomics. Quantitative imaging research, however, is complex and key statistical principles should be followed to realize its full potential. The field of radiomics, in particular, requires a renewed focus on optimal study design/reporting practices and standardization of image acquisition, feature calculation, and rigorous statistical analysis for the field to move forward. In this article, the role of machine and deep learning as a major computational vehicle for advanced model building of radiomics-based signatures or classifiers, and diverse clinical applications, working principles, research opportunities, and available computational platforms for radiomics will be reviewed with examples drawn primarily from oncology. We also address issues related to common applications in medical physics, such as standardization, feature extraction, model building, and validation.

Integrated analysis of the functions of RNA binding proteins in clear cell renal cell carcinoma

RNA binding proteins (RBPs) dysregulation is involved in the processes of various tumors. However, the roles of RBPs in clear cell renal cell carcinoma (ccRCC) remain poorly understand. In present study, we first performed consensus clustering and identified two clusters, of which cluster 2 was closely correlated with the malignancy of ccRCC. Differentially expressed RBPs between normal and tumor tissues were obtained, comprising 71 up-regulated and 44 down-regulated ones. Then, ten hub genes were selected and validated using The Human Protein Atlas database and receiver operating characteristic curves, showing good diagnostic value for cancers. Besides, we identified ten RBPs with the most useful prognostic values, and were used to construct a risk score model. The model could be used to stratify patients with different prognosis and phenotype distributions. The model showed good performance and can be used as a complementation for clinical factors to guide clinical practice in the future.

Effects of RNA Binding Proteins on the Prognosis and Malignant Progression in Prostate Cancer

Prostate cancer (PCa) is a common lethal malignancy in men. RNA binding proteins (RBPs) have been proven to regulate the biological processes of various tumors, but their roles in PCa remain less defined. In the present study, we used bioinformatics analysis to identify RBP genes with prognostic and diagnostic values. A total of 59 differentially expressed RBPs in PCa were obtained, comprising 28 upregulated and 31 downregulated RBP genes, which may play important roles in PCa. Functional enrichment analyses showed that these RBPs were mainly involved in mRNA processing, RNA splicing, and regulation of RNA splicing. Additionally, we identified nine RBP genes (EXO1, PABPC1L, REXO2, MBNL2, MSI1, CTU1, MAEL, YBX2, and ESRP2) and their prognostic values by a protein-protein interaction network and Cox regression analyses. The expression of these nine RBPs was validated using immunohistochemical staining between the tumor and normal samples. Further, the associations between the expression of these nine RBPs and pathological T staging, Gleason score, and lymph node metastasis were evaluated. Moreover, these nine RBP genes showed good diagnostic values and could categorize the PCa patients into two clusters with different malignant phenotypes. Finally, we constructed a prognostic model based on these nine RBP genes and validated them using three external datasets. The model showed good efficiency in predicting patient survival and was independent of other clinical factors. Therefore, our model could be used as a supplement for clinical factors to predict patient prognosis and thereby improve patient survival.

Radiogenomic signatures reveal multiscale intratumour heterogeneity associated with biological functions and survival in breast cancer

Advanced tumours are often heterogeneous, consisting of subclones with various genetic alterations and functional roles. The precise molecular features that characterize the contributions of multiscale intratumour heterogeneity to malignant progression, metastasis, and poor survival are largely unknown. Here, we address these challenges in breast cancer by defining the landscape of heterogeneous tumour subclones and their biological functions using radiogenomic signatures. Molecular heterogeneity is identified by a fully unsupervised deconvolution of gene expression data. Relative prevalence of two subclones associated with cell cycle and primary immunodeficiency pathways identifies patients with significantly different survival outcomes. Radiogenomic signatures of imaging scale heterogeneity are extracted and used to classify patients into groups with distinct subclone compositions. Prognostic value is confirmed by survival analysis accounting for clinical variables. These findings provide insight into how a radiogenomic analysis can identify the biological activities of specific subclones that predict prognosis in a noninvasive and clinically relevant manner.

Tumours are made up of heterogeneous subclones. Here, the authors show using breast cancer imaging and gene expression datasets that these subclones can be inferred by the deconvolution of gene expression data, mapped to MRI derived radiogenomic signatures and used to estimate prognosis.

Radiomics at a Glance: A Few Lessons Learned from Learning Approaches

Simple Summary

Radiomics has become a prominent component of medical imaging research and many studies show its specific value as a support tool for clinical decision-making processes. Radiomic data are typically analyzed with statistical and machine learning methods, which change depending on the disease context and the imaging modality. We found a certain bias in the literature towards the use of such methods and believe that this limitation may influence the capacity of producing accurate and reliable decisions. Therefore, in view of the relevance of various types of learning methods, we report their significance and discuss their unrevealed potential.

Abstract

Processing and modeling medical images have traditionally represented complex tasks requiring multidisciplinary collaboration. The advent of radiomics has assigned a central role to quantitative data analytics targeting medical image features algorithmically extracted from large volumes of images. Apart from the ultimate goal of supporting diagnostic, prognostic, and therapeutic decisions, radiomics is computationally attractive due to specific strengths: scalability, efficiency, and precision. Optimization is achieved by highly sophisticated statistical and machine learning algorithms, but it is especially deep learning that stands out as the leading inference approach. Various types of hybrid learning can be considered when building complex integrative approaches aimed to deliver gains in accuracy for both classification and prediction tasks. This perspective reviews some selected learning methods by focusing on both their significance for radiomics and their unveiled potential.

Dual-mode ultrasound radiomics and intrinsic imaging phenotypes for diagnosis of lymph node lesions

Background

The ultrasonic diagnosis of lymph node lesions is usually based on a small number of subjective visual features from a single ultrasonic modality, which limits diagnostic accuracy. Therefore, our study aimed to propose a computerized method for using dual-mode ultrasound radiomics and the intrinsic imaging phenotypes for accurately differentiating benign, lymphomatous, and metastatic lymph nodes.

Methods

A total of 543 lymph nodes from 538 patients were examined with both B-mode ultrasonography and elastography. The data set was randomly divided into a training set of 407 nodes and a validation set of 136 nodes. First, we extracted 430 radiomic features from dual-mode images. Then, we combined the least absolute shrinkage and selection operator with the analysis of variance to select several typical features. We retrieved the intrinsic imaging phenotypes by using a hierarchical clustering of all radiomics features, and we integrated the phenotypes with the selected features for the classification of benign, lymphomatous, and metastatic nodes.

Results

The areas under the receiver operating characteristic curves (AUCs) on the validation set were 0.960 for benign vs. lymphomatous, 0.716 for benign vs. metastatic, 0.933 for lymphomatous vs. metastatic, and 0.856 for benign vs. malignant.

Conclusions

The radiomics features and intrinsic imaging phenotypes derived from the dual-mode ultrasound can capture the distinctions between benign, lymphomatous, and metastatic nodes and are valuable in node differentiation.

AUNet: attention-guided dense-upsampling networks for breast mass segmentation in whole mammograms

Mammography is one of the most commonly applied tools for early breast cancer screening. Automatic segmentation of breast masses in mammograms is essential but challenging due to the low signal-to-noise ratio and the wide variety of mass shapes and sizes. Existing methods deal with these challenges mainly by extracting mass-centered image patches manually or automatically. However, manual patch extraction is time-consuming and automatic patch extraction brings errors that could not be compensated in the following segmentation step. In this study, we propose a novel attention-guided dense-upsampling network (AUNet) for accurate breast mass segmentation in whole mammograms directly. In AUNet, we employ an asymmetrical encoder-decoder structure and propose an effective upsampling block, attention-guided dense-upsampling block (AU block). Especially, the AU block is designed to have three merits. Firstly, it compensates the information loss of bilinear upsampling by dense upsampling. Secondly, it designs a more effective method to fuse high- and low-level features. Thirdly, it includes a channel-attention function to highlight rich-information channels. We evaluated the proposed method on two publicly available datasets, CBIS-DDSM and INbreast. Compared to three state-of-the-art fully convolutional networks, AUNet achieved the best performances with an average Dice similarity coefficient of 81.8% for CBIS-DDSM and 79.1% for INbreast.

The Impact of Normalization Approaches to Automatically Detect Radiogenomic Phenotypes Characterizing Breast Cancer Receptors Status

In breast cancer studies, combining quantitative radiomic with genomic signatures can help identifying and characterizing radiogenomic phenotypes, in function of molecular receptor status. Biomedical imaging processing lacks standards in radiomic feature normalization methods and neglecting feature normalization can highly bias the overall analysis. This study evaluates the effect of several normalization techniques to predict four clinical phenotypes such as estrogen receptor (ER), progesterone receptor (PR), human epidermal growth factor receptor 2 (HER2), and triple negative (TN) status, by quantitative features. The Cancer Imaging Archive (TCIA) radiomic features from 91 T1-weighted Dynamic Contrast Enhancement MRI of invasive breast cancers were investigated in association with breast invasive carcinoma miRNA expression profiling from the Cancer Genome Atlas (TCGA). Three advanced machine learning techniques (Support Vector Machine, Random Forest, and Naïve Bayesian) were investigated to distinguish between molecular prognostic indicators and achieved an area under the ROC curve (AUC) values of 86%, 93%, 91%, and 91% for the prediction of ER+ versus ER-, PR+ versus PR-, HER2+ versus HER2-, and triple-negative, respectively. In conclusion, radiomic features enable to discriminate major breast cancer molecular subtypes and may yield a potential imaging biomarker for advancing precision medicine.

Development and Validation of a Deep Learning CT Signature to Predict Survival and Chemotherapy Benefit in Gastric Cancer: A Multicenter, Retrospective Study

OBJECTIVE

We aimed to develop a deep learning-based signature to predict prognosis and benefit from adjuvant chemotherapy using preoperative computed tomography (CT) images.

BACKGROUND

Current staging methods do not accurately predict the risk of disease relapse for patients with gastric cancer.

METHODS

We proposed a novel deep neural network (S-net) to construct a CT signature for predicting disease-free survival (DFS) and overall survival in a training cohort of 457 patients, and independently tested it in an external validation cohort of 1158 patients. An integrated nomogram was constructed to demonstrate the added value of the imaging signature to established clinicopathologic factors for individualized survival prediction. Prediction performance was assessed with respect to discrimination, calibration, and clinical usefulness.

RESULTS

The DeLIS was associated with DFS and overall survival in the overall validation cohort and among subgroups defined by clinicopathologic variables, and remained an independent prognostic factor in multivariable analysis (P< 0.001). Integrating the imaging signature and clinicopathologic factors improved prediction performance, with C-indices: 0.792-0.802 versus 0.719-0.724, and net reclassification improvement 10.1%-28.3%. Adjuvant chemotherapy was associated with improved DFS in stage II patients with high-DeLIS [hazard ratio = 0.362 (95% confidence interval 0.149-0.882)] and stage III patients with high- and intermediate-DeLIS [hazard ratio = 0.611 (0.442-0.843); 0.633 (0.433-0.925)]. On the other hand, adjuvant chemotherapy did not affect survival for patients with low-DeLIS, suggesting a predictive effect (Pinteraction = 0.048, 0.016 for DFS in stage II and III disease).

CONCLUSIONS

The proposed imaging signature improved prognostic prediction and could help identify patients most likely to benefit from adjuvant chemotherapy in gastric cancer.

TCGA-TCIA Impact on Radiogenomics Cancer Research: A Systematic Review

In the last decade, the development of radiogenomics research has produced a significant amount of papers describing relations between imaging features and several molecular 'omic signatures arising from next-generation sequencing technology and their potential role in the integrated diagnostic field. The most vulnerable point of many of these studies lies in the poor number of involved patients. In this scenario, a leading role is played by The Cancer Genome Atlas (TCGA) and The Cancer Imaging Archive (TCIA), which make available, respectively, molecular 'omic data and linked imaging data. In this review, we systematically collected and analyzed radiogenomic studies based on TCGA-TCIA data. We organized literature per tumor type and molecular 'omic data in order to discuss salient imaging genomic associations and limitations of each study. Finally, we outlined the potential clinical impact of radiogenomics to improve the accuracy of diagnosis and the prediction of patient outcomes in oncology.

Tumour heterogeneity revealed by unsupervised decomposition of dynamic contrast-enhanced magnetic resonance imaging is associated with underlying gene expression patterns and poor survival in breast cancer patients

BACKGROUND

Heterogeneity is a common finding within tumours. We evaluated the imaging features of tumours based on the decomposition of tumoural dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) data to identify their prognostic value for breast cancer survival and to explore their biological importance.

METHODS

Imaging features (n = 14), such as texture, histogram distribution and morphological features, were extracted to determine their associations with recurrence-free survival (RFS) in patients in the training cohort (n = 61) from The Cancer Imaging Archive (TCIA). The prognostic value of the features was evaluated in an independent dataset of 173 patients (i.e. the reproducibility cohort) from the TCIA I-SPY 1 TRIAL dataset. Radiogenomic analysis was performed in an additional cohort, the radiogenomic cohort (n = 87), using DCE-MRI from TCGA-BRCA and corresponding gene expression data from The Cancer Genome Atlas (TCGA). The MRI tumour area was decomposed by convex analysis of mixtures (CAM), resulting in 3 components that represent plasma input, fast-flow kinetics and slow-flow kinetics. The prognostic MRI features were associated with the gene expression module in which the pathway was analysed. Furthermore, a multigene signature for each prognostic imaging feature was built, and the prognostic value for RFS and overall survival (OS) was confirmed in an additional cohort from TCGA.

RESULTS

Three image features (i.e. the maximum probability from the precontrast MR series, the median value from the second postcontrast series and the overall tumour volume) were independently correlated with RFS (p values of 0.0018, 0.0036 and 0.0032, respectively). The maximum probability feature from the fast-flow kinetics subregion was also significantly associated with RFS and OS in the reproducibility cohort. Additionally, this feature had a high correlation with the gene expression module (r = 0.59), and the pathway analysis showed that Ras signalling, a breast cancer-related pathway, was significantly enriched (corrected p value = 0.0044). Gene signatures (n = 43) associated with the maximum probability feature were assessed for associations with RFS (p = 0.035) and OS (p = 0.027) in an independent dataset containing 1010 gene expression samples. Among the 43 gene signatures, Ras signalling was also significantly enriched.

CONCLUSIONS

Dynamic pattern deconvolution revealed that tumour heterogeneity was associated with poor survival and cancer-related pathways in breast cancer.

Radiomics Signature: A Biomarker for the Preoperative Distant Metastatic Prediction of Stage I Nonsmall Cell Lung Cancer

OBJECTIVES

To evaluate the predictive value of radiomics features on the distant metastasis (DM) of stage I nonsmall cell lung cancer (NSCLC) preoperatively, by comparing with clinical characteristics and CT morphological features, and to screen the important prognostic predictors.

METHODS

One hundred ninety-four stage I NSCLC patients were retrospectively enrolled, DM free survival (DMFS) was evaluated. The consensus clustering analysis was used to build the radiomics signatures in the primary cohort and validated in the validation cohort. The univariate survival analysis was performed in clinical characteristics, CT morphological features and radiomics signatures, respectively. Cox model was performed and C-index was calculated.

RESULTS

There were 25 patients (12.9%) with DM. The median DMFS was 15 months. Three hundred thirteen radiomics features were selected, then classified into five groups, two subtypes (I and II) with each group. The RS1 showed the best prognostic ability with C-index of 0.355(95% confidence interval [CI], 0.269-0.442; p < 0.001). The histological type exhibited a good prognostic ability with C-index of 0.123 (95% CI, 0.000-0.305; p < 0.001) for DMFS. Cox model showed RS1(hazard ratio [HR] 18.025, 95% CI 2.366-137.340), pleural indentation sign (HR 2.623, 95% CI 1.070-6.426) and histological type (HR 4.461, 95% CI 1.783-11.162) were the independent prognostic factors (p < 0.05).

CONCLUSION

Radiomics provided a new modality for the distant metastatic prediction of stage I NSCLC. Patients with type II of RS1, pleural indentation sign and nonadenocarcinoma indicated the high probability of postsurgical DM.

Machine learning in breast MRI

Machine-learning techniques have led to remarkable advances in data extraction and analysis of medical imaging. Applications of machine learning to breast MRI continue to expand rapidly as increasingly accurate 3D breast and lesion segmentation allows the combination of radiologist-level interpretation (eg, BI-RADS lexicon), data from advanced multiparametric imaging techniques, and patient-level data such as genetic risk markers. Advances in breast MRI feature extraction have led to rapid dataset analysis, which offers promise in large pooled multiinstitutional data analysis. The object of this review is to provide an overview of machine-learning and deep-learning techniques for breast MRI, including supervised and unsupervised methods, anatomic breast segmentation, and lesion segmentation. Finally, it explores the role of machine learning, current limitations, and future applications to texture analysis, radiomics, and radiogenomics. Level of Evidence: 3 Technical Efficacy Stage: 2 J. Magn. Reson. Imaging 2019. J. Magn. Reson. Imaging 2020;52:998-1018.

Decoding the Interdependence of Multiparametric Magnetic Resonance Imaging to Reveal Patient Subgroups Correlated with Survivals

Glioblastoma is highly heterogeneous in microstructure and vasculature, creating various tumor microenvironments among patients, which may lead to different phenotypes. The purpose was to interrogate the interdependence of microstructure and vasculature using perfusion and diffusion imaging and to investigate the utility of this approach in tumor invasiveness assessment. A total of 115 primary glioblastoma patients were prospectively recruited for preoperative magnetic resonance imaging (MRI) and surgery. Apparent diffusion coefficient (ADC) was calculated from diffusion imaging, and relative cerebral blood volume (rCBV) was calculated from perfusion imaging. The empirical copula transform was applied to ADC and rCBV voxels in the contrast-enhancing tumor region to obtain their joint distribution, which was discretized to extract second-order features for an unsupervised hierarchical clustering. The lactate levels of patient subgroups, measured by MR spectroscopy, were compared. Survivals were analyzed using Kaplan-Meier and multivariate Cox regression analyses. The results showed that three patient subgroups were identified by the unsupervised clustering. These subtypes showed no significant differences in clinical characteristics but were significantly different in lactate level and patient survivals. Specifically, the subtype demonstrating high interdependence of ADC and rCBV displayed a higher lactate level than the other two subtypes (P = .016 and P = .044, respectively). Both subtypes of low and high interdependence showed worse progression-free survival than the intermediate (P = .046 and P = .009 respectively). Our results suggest that the interdependence between perfusion and diffusion imaging may be useful in stratifying patients and evaluating tumor invasiveness, providing overall measure of tumor microenvironment using multiparametric MRI.

Evolutionary Multiobjective Clustering and Its Applications to Patient Stratification

Patient stratification has a major role in enabling efficient and personalized medicine. An important task in patient stratification is to discover disease subtypes for effective treatment. To achieve this goal, the research on clustering algorithms for patient stratification has brought attention from both academia and medical community over the past decades. However, existing clustering algorithms suffer from realistic restrictions such as experimental noises, high dimensionality, and poor interpretability. In particular, the existing clustering algorithms usually determine clustering quality using only one internal evaluation function. Unfortunately, it is obvious that one internal evaluation function is hard to be fitted and robust for all datasets. Therefore, in this paper, a novel multiobjective framework called multiobjective clustering algorithm by fast search and find of density peaks is proposed to address those limitations altogether. In the proposed framework, a parameter candidate population is evolved under multiple objectives to select features and evaluate clustering densities automatically. To guide the multiobjective evolution, five cluster validity indices including compactness, separation, Calinski-Harabasz index, Davies-Bouldin index, and Dunn index, are chosen as the objective functions, capturing multiple characteristics of the evolving clusters. Multiobjective differential evolution algorithm based on decomposition is adopted to optimize those five objective functions simultaneously. To demonstrate its effectiveness, extensive experiments have been conducted, comparing the proposed algorithm with 45 algorithms including nine state-of-the-art clustering algorithms, five multiobjective evolutionary algorithms, and 31 baseline algorithms under different objective subsets on 94 datasets featuring 35 real patient stratification datasets, 55 synthetic datasets based on a real human transcription regulation network model, and four other medical datasets. The numerical results reveal that the proposed algorithm can achieve better or competitive solutions than the others. Besides, time complexity analysis, convergence analysis, and parameter analysis are conducted to demonstrate the robustness of the proposed algorithm from different perspectives.

Association of Peritumoral Radiomics With Tumor Biology and Pathologic Response to Preoperative Targeted Therapy for HER2 (ERBB2)-Positive Breast Cancer

IMPORTANCE

There has been significant recent interest in understanding the utility of quantitative imaging to delineate breast cancer intrinsic biological factors and therapeutic response. No clinically accepted biomarkers are as yet available for estimation of response to human epidermal growth factor receptor 2 (currently known as ERBB2, but referred to as HER2 in this study)-targeted therapy in breast cancer.

OBJECTIVE

To determine whether imaging signatures on clinical breast magnetic resonance imaging (MRI) could noninvasively characterize HER2-positive tumor biological factors and estimate response to HER2-targeted neoadjuvant therapy.

DESIGN, SETTING, AND PARTICIPANTS

In a retrospective diagnostic study encompassing 209 patients with breast cancer, textural imaging features extracted within the tumor and annular peritumoral tissue regions on MRI were examined as a means to identify increasingly granular breast cancer subgroups relevant to therapeutic approach and response. First, among a cohort of 117 patients who received an MRI prior to neoadjuvant chemotherapy (NAC) at a single institution from April 27, 2012, through September 4, 2015, imaging features that distinguished HER2+ tumors from other receptor subtypes were identified. Next, among a cohort of 42 patients with HER2+ breast cancers with available MRI and RNaseq data accumulated from a multicenter, preoperative clinical trial (BrUOG 211B), a signature of the response-associated HER2-enriched (HER2-E) molecular subtype within HER2+ tumors (n = 42) was identified. The association of this signature with pathologic complete response was explored in 2 patient cohorts from different institutions, where all patients received HER2-targeted NAC (n = 28, n = 50). Finally, the association between significant peritumoral features and lymphocyte distribution was explored in patients within the BrUOG 211B trial who had corresponding biopsy hematoxylin-eosin-stained slide images. Data analysis was conducted from January 15, 2017, to February 14, 2019.

MAIN OUTCOMES AND MEASURES

Evaluation of imaging signatures by the area under the receiver operating characteristic curve (AUC) in identifying HER2+ molecular subtypes and distinguishing pathologic complete response (ypT0/is) to NAC with HER2-targeting.

RESULTS

In the 209 patients included (mean [SD] age, 51.1 [11.7] years), features from the peritumoral regions better discriminated HER2-E tumors (maximum AUC, 0.85; 95% CI, 0.79-0.90; 9-12 mm from the tumor) compared with intratumoral features (AUC, 0.76; 95% CI, 0.69-0.84). A classifier combining peritumoral and intratumoral features identified the HER2-E subtype (AUC, 0.89; 95% CI, 0.84-0.93) and was significantly associated with response to HER2-targeted therapy in both validation cohorts (AUC, 0.80; 95% CI, 0.61-0.98 and AUC, 0.69; 95% CI, 0.53-0.84). Features from the 0- to 3-mm peritumoral region were significantly associated with the density of tumor-infiltrating lymphocytes (R2 = 0.57; 95% CI, 0.39-0.75; P = .002).

CONCLUSIONS AND RELEVANCE

A combination of peritumoral and intratumoral characteristics appears to identify intrinsic molecular subtypes of HER2+ breast cancers from imaging, offering insights into immune response within the peritumoral environment and suggesting potential benefit for treatment guidance.

Radiomics signature: A potential and incremental predictor for EGFR mutation status in NSCLC patients, comparison with CT morphology

OBJECTIVES

To compare the predictive performance of radiomics signature and CT morphological features for epidermal growth factor receptor (EGFR) mutation status; then further to develop and compare the different predictive models for EGFR mutation in non-small cell lung cancer (NSCLC) patients.

MATERIALS AND METHODS

This retrospective study involved 404 patients with NSCLC (243 cases in the training cohort and 161 cases in the validation cohort). Radiomics features were extracted from preoperative non-contrast CT images of the entire tumor. Correlations between the EGFR mutation status and candidate predictors were assessed using Mann-Whitney U test or Chi-square test. Unsupervised consensus clustering was used to analyze the representativeness and reduce the redundancy of radiomics features. Multivariable logistic regression analysis was performed to build radiomics signature and develop predictive models of EGFR mutation. ROC curve analysis and Delong test were used to compare the predictive performance among individual features and models.

RESULTS

Of the 234 radiomics features, 93 radiomics features with high repeatability and high predictive significance were selected. The radiomics signature, which was built with one histogram and two textural features, showed the best predictive performance (AUC = 0.762 and 0.775 in the training and validation cohort) in comparison with all the clinical characteristics and conventional CT morphological features to differentiate EGFR mutation status (P < 0.05). The integrated model was developed with maximum diameter, location, sex and radiomics signature. In the training and validation cohort, the integrated model showed the most optimal predictive performance (AUC = 0.798, 0.818 in the training and validation cohort) compared with the clinical models.

CONCLUSION

The radiomics signature showed better performance for predicting EGFR mutant than all the clinical and morphological features. Moreover, the integrated model built with radiomics signature, clinical and morphological features outperformed the clinical models, which is helpful for physicians to determine the targeted therapy.