MI-Common Data Model: Extending Observational Medical Outcomes Partnership-Common Data Model (OMOP-CDM) for Registering Medical Imaging Metadata and Subsequent Curation Processes.

PURPOSE: The explosion of big data and artificial intelligence has rapidly increased the need for integrated, homogenized, and harmonized health data. Many common data models (CDMs) and standard vocabularies have appeared in an attempt to offer harmonized access to the available information, with Observational Medical Outcomes Partnership (OMOP)-CDM being one of the most prominent ones, allowing the standardization and harmonization of health care information. However, despite its flexibility, still capturing imaging metadata along with the corresponding clinical data continues to pose a challenge. This challenge arises from the absence of a comprehensive standard representation for image-related information and subsequent image curation processes and their interlinkage with the respective clinical information. Successful resolution of this challenge holds the potential to enable imaging and clinical data to become harmonized, quality-checked, annotated, and ready to be used in conjunction, in the development of artificial intelligence models and other data-dependent use cases. METHODS: To address this challenge, we introduce medical imaging (MI)-CDM-an extension of the OMOP-CDM specifically designed for registering medical imaging data and curation-related processes. Our modeling choices were the result of iterative numerous discussions among clinical and AI experts to enable the integration of imaging and clinical data in the context of the ProCAncer-I project, for answering a set of clinical questions across the prostate cancer's continuum. RESULTS: Our MI-CDM extension has been successfully implemented for the use case of prostate cancer for integrating imaging and curation metadata along with clinical information by using the OMOP-CDM and its oncology extension. CONCLUSION: By using our proposed terminologies and standardized attributes, we demonstrate how diverse imaging modalities can be seamlessly integrated in the future.

Data infrastructures for AI in medical imaging: a report on the experiences of five EU projects.

Artificial intelligence (AI) is transforming the field of medical imaging and has the potential to bring medicine from the era of 'sick-care' to the era of healthcare and prevention. The development of AI requires access to large, complete, and harmonized real-world datasets, representative of the population, and disease diversity. However, to date, efforts are fragmented, based on single-institution, size-limited, and annotation-limited datasets. Available public datasets (e.g., The Cancer Imaging Archive, TCIA, USA) are limited in scope, making model generalizability really difficult. In this direction, five European Union projects are currently working on the development of big data infrastructures that will enable European, ethically and General Data Protection Regulation-compliant, quality-controlled, cancer-related, medical imaging platforms, in which both large-scale data and AI algorithms will coexist. The vision is to create sustainable AI cloud-based platforms for the development, implementation, verification, and validation of trustable, usable, and reliable AI models for addressing specific unmet needs regarding cancer care provision. In this paper, we present an overview of the development efforts highlighting challenges and approaches selected providing valuable feedback to future attempts in the area.Key points• Artificial intelligence models for health imaging require access to large amounts of harmonized imaging data and metadata.• Main infrastructures adopted either collect centrally anonymized data or enable access to pseudonymized distributed data.• Developing a common data model for storing all relevant information is a challenge.• Trust of data providers in data sharing initiatives is essential.• An online European Union meta-tool-repository is a necessity minimizing effort duplication for the various projects in the area.

Synergies of Radiomics and Transcriptomics in Lung Cancer Diagnosis: A Pilot Study.

Radiotranscriptomics is an emerging field that aims to investigate the relationships between the radiomic features extracted from medical images and gene expression profiles that contribute in the diagnosis, treatment planning, and prognosis of cancer. This study proposes a methodological framework for the investigation of these associations with application on non-small-cell lung cancer (NSCLC). Six publicly available NSCLC datasets with transcriptomics data were used to derive and validate a transcriptomic signature for its ability to differentiate between cancer and non-malignant lung tissue. A publicly available dataset of 24 NSCLC-diagnosed patients, with both transcriptomic and imaging data, was used for the joint radiotranscriptomic analysis. For each patient, 749 Computed Tomography (CT) radiomic features were extracted and the corresponding transcriptomics data were provided through DNA microarrays. The radiomic features were clustered using the iterative K-means algorithm resulting in 77 homogeneous clusters, represented by meta-radiomic features. The most significant differentially expressed genes (DEGs) were selected by performing Significance Analysis of Microarrays (SAM) and 2-fold change. The interactions among the CT imaging features and the selected DEGs were investigated using SAM and a Spearman rank correlation test with a False Discovery Rate (FDR) of 5%, leading to the extraction of 73 DEGs significantly correlated with radiomic features. These genes were used to produce predictive models of the meta-radiomics features, defined as p-metaomics features, by performing Lasso regression. Of the 77 meta-radiomic features, 51 can be modeled in terms of the transcriptomic signature. These significant radiotranscriptomics relationships form a reliable basis to biologically justify the radiomics features extracted from anatomic imaging modalities. Thus, the biological value of these radiomic features was justified via enrichment analysis on their transcriptomics-based regression models, revealing closely associated biological processes and pathways. Overall, the proposed methodological framework provides joint radiotranscriptomics markers and models to support the connection and complementarities between the transcriptome and the phenotype in cancer, as demonstrated in the case of NSCLC.

Data Ingestion for AI in Prostate Cancer.

Prostate cancer (PCa) is one of the most prevalent cancers in the male population. Current clinical practices lead to overdiagnosis and overtreatment necessitating more effective tools for improving diagnosis, thus the quality of life of patients. Recent advances in infrastructure, computing power and artificial intelligence enable the collection of tremendous amounts of clinical and imaging data that could assist towards this end. ProCAncer-I project aims to develop an AI platform integrating imaging data and models and hosting the largest collection of PCa (mp)MRI, anonymized image data worldwide. In this paper, we present an overview of the overall architecture focusing on the data ingestion part of the platform. We describe the workflow followed for uploading the data and the main repositories for storing imaging data, clinical data and their corresponding metadata.

CXCR4 and JUNB double-positive disseminated tumor cells are detected frequently in breast cancer patients at primary diagnosis.

BACKGROUND: The chemokine receptor CXCR4 and the transcription factor JUNB, expressed on a variety of tumor cells, seem to play an important role in the metastatic process. Since disseminated tumor cells (DTCs) in the bone marrow (BM) have been associated with worse outcomes, we evaluated the expression of CXCR4 and JUNB in DTCs of primary, nonmetastatic breast cancer (BC) patients before the onset of any systemic treatment. METHODS: Bilateral BM (10 ml) aspirations of 39 hormone receptor (HR)-positive, HER2-negative BC patients were assessed for the presence of DTCs using the following combination of antibodies: pan-cytokeratin (A45-B/B3)/CXCR4/JUNB. An expression pattern of the examined proteins was created using confocal laser scanning microscopy, Image J software and BC cell lines. RESULTS: CXCR4 was overexpressed in cancer cells and DTCs, with the following hierarchy of expression: SKBR3 > MCF7 > DTCs > MDA-MB231. Accordingly, the expression pattern of JUNB was: DTCs > MDA-MB231 > SKBR3 > MCF7. The mean intensity of CXCR4 (6411 ± 334) and JUNB (27725.64 ± 470) in DTCs was statistically higher compared with BM hematopoietic cells (2009 ± 456, p = 0.001; and 11112.89 ± 545, p = 0.001, respectively). The (CXCR4+JUNB+CK+) phenotype was the most frequently detected [90% (35/39)], followed by the (CXCR4-JUNB+CK+) phenotype [36% (14/39)]. However, (CXCR4+JUNB-CK+) tumor cells were found in only 5% (3/39) of patients. Those patients harboring DTCs with the (CXCR4+JUNB+CK+) phenotype revealed lower overall survival (Cox regression: p = 0.023). CONCLUSIONS: (CXCR4+JUNB+CK+)-expressing DTCs, detected frequently in the BM of BC patients, seem to identify a subgroup of patients at higher risk for relapse that may be considered for close follow up.

The prognostic value of JUNB-positive CTCs in metastatic breast cancer: from bioinformatics to phenotypic characterization.

BACKGROUND: Circulating tumor cells (CTCs) are important for metastatic dissemination of cancer. They can provide useful information, regarding biological features and tumor heterogeneity; however, their detection and characterization are difficult due to their limited number in the bloodstream and their mesenchymal characteristics. Therefore, new biomarkers are needed to address these questions. METHODS: Bioinformatics functional enrichment analysis revealed a subgroup of 24 genes, potentially overexpressed in CTCs. Among these genes, the chemokine receptor CXCR4 plays a central role. After prioritization according to the CXCR4 corresponding pathways, five molecules (JUNB, YWHAB, TYROBP, NFYA, and PRDX1) were selected for further analysis in biological samples. The SKBR3, MDA-MB231, and MCF7 cell lines, as well as PBMCs from normal (n = 10) blood donors, were used as controls to define the expression pattern of all the examined molecules. Consequently, 100 previously untreated metastatic breast cancer (mBC) patients (n = 100) were analyzed using the following combinations of antibodies: CK (cytokeratin)/CXCR4/JUNB, CK/NFYA/ΥWHΑΒ (14-3-3), and CK/TYROBP/PRDX1. A threshold value for every molecule was considered the mean expression in normal PBMCs. RESULTS: Quantification of CXCR4 revealed overexpression of the receptor in SKBR3 and in CTCs, following the subsequent scale (SKBR3>CTCs>Hela>MCF7>MDA-MB231). JUNB was also overexpressed in CTCs (SKBR3>CTCs>MCF7>MDA-MB231>Hela). According to the defined threshold for each molecule, CXCR4-positive CTCs were identified in 90% of the patients with detectable tumor cells in their blood. In addition, 65%, 75%, 14.3%, and 12.5% of the patients harbored JUNB-, TYROBP-, NFYA-, and PRDX-positive CTCs, respectively. Conversely, none of the patients revealed YWHAB-positive CTCs. Interestingly, JUNB expression in CTCs was phenotypically and statistically enhanced compared to patients' blood cells (p = 0.002) providing a possible new biomarker for CTCs. Furthermore, the detection of JUNB-positive CTCs in patients was associated with poorer PFS (p = 0.015) and OS (p = 0.002). Moreover, JUNB staining of 11 primary and 4 metastatic tumors from the same cohort of patients revealed a dramatic increase of JUNB expression in metastasis. CONCLUSIONS: CXCR4, JUNB, and TYROBP were overexpressed in CTCs, but only the expression of JUNB was associated with poor prognosis, providing a new biomarker and a potential therapeutic target for the elimination of CTCs.

Addressing drug-drug and drug-food interactions through personalized empowerment services for healthcare.

Personalized healthcare systems support the provision of timely and appropriate information regarding healthcare options and treatment alternatives. Especially for patients that receive multi-drug treatments a key issue is the minimization of the risk of adverse effects due to drug-drug interactions (DDIs). DDIs may be the result of doctor prescribed drugs but also due to self-medication of conventional drugs, alternative medicines, food habits, alcohol or smoking. It is therefore crucial for personalized health systems, apart from assisting physicians for optimal prescription practices, to also provide appropriate information for individual users for drug-drug interactions or similar information regarding risks for modulation of the ensuing treatment. In this manuscript we describe a DDI service including drug-food, drug-herb and other lifestyle-related factors, developed in the context of a personalized patient empowerment platform. The solution enables guidance to patients for their medication on how to reduce the risk of unwanted drug interactions and side effects in a seamless and transparent way. We present and analyze the implemented services and provide examples on using an alerting service to identify potential DDIs in two different chronic diseases, congestive heart failure and osteoarthritis.

Tamoxifen induces a pluripotency signature in breast cancer cells and human tumors.

Tamoxifen is the treatment of choice in estrogen receptor alpha breast cancer patients that are eligible for adjuvant endocrine therapy. However, ∼50% of ERα-positive tumors exhibit intrinsic or rapidly acquire resistance to endocrine treatment. Unfortunately, prediction of de novo resistance to endocrine therapy and/or assessment of relapse likelihood remain difficult. While several mechanisms regulating the acquisition and the maintenance of endocrine resistance have been reported, there are several aspects of this phenomenon that need to be further elucidated. Altered metabolic fate of tamoxifen within patients and emergence of tamoxifen-resistant clones, driven by evolution of the disease phenotype during treatment, appear as the most compelling hypotheses so far. In addition, tamoxifen was reported to induce pluripotency in breast cancer cell lines, in vitro. In this context, we have performed a whole transcriptome analysis of an ERα-positive (T47D) and a triple-negative breast cancer cell line (MDA-MB-231), exposed to tamoxifen for a short time frame (hours), in order to identify how early pluripotency-related effects of tamoxifen may occur. Our ultimate goal was to identify whether the transcriptional actions of tamoxifen related to induction of pluripotency are mediated through specific ER-dependent or independent mechanisms. We report that even as early as 3 hours after the exposure of breast cancer cells to tamoxifen, a subset of ERα-dependent genes associated with developmental processes and pluripotency are induced and this is accompanied by specific phenotypic changes (expression of pluripotency-related proteins). Furthermore we report an association between the increased expression of pluripotency-related genes in ERα-positive breast cancer tissues samples and disease relapse after tamoxifen therapy. Finally we describe that in a small group of ERα-positive breast cancer patients, with disease relapse after surgery and tamoxifen treatment, ALDH1A1 (a marker of pluripotency in epithelial cancers which is absent in normal breast tissue) is increased in relapsing tumors, with a concurrent modification of its intra-cellular localization. Our data could be of value in the discrimination of patients susceptible to develop tamoxifen resistance and in the selection of optimized patient-tailored therapies.

A network-based approach to enrich gene signatures for the prediction of breast cancer metastases.

Despite the multiplicity of the gene expression analysis studies for the identification of genomics based origins of cancerous diseases, the presented gene signatures have generally little overlap. The genes do not function in isolation and therefore a more holistic approach that takes into account the interactions among them is needed. In this study we present a stepwise refinement methodology where starting from some initial set of biomarkers we expand and enrich this set taking into account existing biological information. In particular, we start with a 27 gene signature previously identified as indicative of the presence of circulating tumor cells (CTCs) in peripheral blood of breast cancer patients. We use the manually curated HINT database of protein-protein interactions as the background biological network to locate the network-based similarity of the input genes and how they connect to each other. The result is an enriched connected set of genes that is subsequently expanded to form an even bigger network based on the ability of the surrounding genes to strongly correlate with the phenotypes of a training set of breast cancer patient cases. The induced network is then used as a new gene signature for the classification of breast brain metastases in an independent dataset. The results are encouraging for the validity of this method.

On the identification of circulating tumor cells in breast cancer.

Breast cancer is a highly heterogeneous disease and very common among western women. The main cause of death is not the primary tumor but its metastases at distant sites, such as lymph nodes and other organs (preferentially lung, liver, and bones). The study of circulating tumor cells (CTCs) in peripheral blood resulting from tumor cell invasion and intravascular filtration highlights their crucial role concerning tumor aggressiveness and metastasis. Genomic research regarding CTCs monitoring for breast cancer is limited due to the lack of indicative genes for their detection and isolation. Instead of direct CTC detection, in our study, we focus on the identification of factors in peripheral blood that can indirectly reveal the presence of such cells. Using selected publicly available breast cancer and peripheral blood microarray datasets, we follow a two-step elimination procedure for the identification of several discriminant factors. Our procedure facilitates the identification of major genes involved in breast cancer pathology, which are also indicative of CTCs presence.

Web-based workflow planning platform supporting the design and execution of complex multiscale cancer models.

Significant Virtual Physiological Human efforts and projects have been concerned with cancer modeling, especially in the European Commission Seventh Framework research program, with the ambitious goal to approach personalized cancer simulation based on patient-specific data and thereby optimize therapy decisions in the clinical setting. However, building realistic in silico predictive models targeting the clinical practice requires interactive, synergetic approaches to integrate the currently fragmented efforts emanating from the systems biology and computational oncology communities all around the globe. To further this goal, we propose an intelligent graphical workflow planning system that exploits the multiscale and modular nature of cancer and allows building complex cancer models by intuitively linking/interchanging highly specialized models. The system adopts and extends current standardization efforts, key tools, and infrastructure in view of building a pool of reliable and reproducible models capable of improving current therapies and demonstrating the potential for clinical translation of these technologies.

The technologically integrated oncosimulator: combining multiscale cancer modeling with information technology in the in silico oncology context.

This paper outlines the major components and function of the technologically integrated oncosimulator developed primarily within the Advancing Clinico Genomic Trials on Cancer (ACGT) project. The Oncosimulator is defined as an information technology system simulating in vivo tumor response to therapeutic modalities within the clinical trial context. Chemotherapy in the neoadjuvant setting, according to two real clinical trials concerning nephroblastoma and breast cancer, has been considered. The spatiotemporal simulation module embedded in the Oncosimulator is based on the multiscale, predominantly top-down, discrete entity-discrete event cancer simulation technique developed by the In Silico Oncology Group, National Technical University of Athens. The technology modules include multiscale data handling, image processing, invocation of code execution via a spreadsheet-inspired environment portal, execution of the code on the grid, and the visualization of the predictions. A refining scenario for the eventual coupling of the oncosimulator with immunological models is also presented. Parameter values have been adapted to multiscale clinical trial data in a consistent way, thus supporting the predictive potential of the oncosimulator. Indicative results demonstrating various aspects of the clinical adaptation and validation process are presented. Completion of these processes is expected to pave the way for the clinical translation of the system.

An architecture for integrating cancer model repositories.

The TUMOR project aims at developing a European clinically oriented semantic-layered cancer digital model repository from existing EU projects that will be interoperable with the US grid-enabled semantic-layered digital model repository platform at CViT.org (Center for the Development of a Virtual Tumor, Massachusetts General Hospital (MGH), Boston, USA) which is NIH/NCI-caGRID compatible. In this paper we describe the modular and federated architecture of TUMOR that effectively addresses model integration, interoperability, and security related issues.

ACGT: advancing clinico-genomic trials on cancer - four years of experience.

The challenges regarding seamless integration of distributed, heterogeneous and multilevel data arising in the context of contemporary, post-genomic clinical trials cannot be effectively addressed with current methodologies. An urgent need exists to access data in a uniform manner, to share information among different clinical and research centers, and to store data in secure repositories assuring the privacy of patients. Advancing Clinico-Genomic Trials (ACGT) was a European Commission funded Integrated Project that aimed at providing tools and methods to enhance the efficiency of clinical trials in the -omics era. The project, now completed after four years of work, involved the development of both a set of methodological approaches as well as tools and services and its testing in the context of real-world clinico-genomic scenarios. This paper describes the main experiences using the ACGT platform and its tools within one such scenario and highlights the very promising results obtained.

A disease annotation study of gene signatures in a breast cancer microarray dataset.

Breast cancer is a complex disease with heterogeneity between patients regarding prognosis and treatment response. Recent progress in advanced molecular biology techniques and the development of efficient methods for database mining lead to the discovery of promising novel biomarkers for prognosis and prediction of breast cancer. In this paper, we applied three computational algorithms (RFE-LNW, Lasso and FSMLP) to one microarray dataset for breast cancer and compared the obtained gene signatures with a recently described disease-agnostic tool, the Genotator. We identified a panel of 152 genes as a potential prognostic signature and the ERRFI1 gene as possible biomarker of breast cancer disease.

The ACGT Master Ontology and its applications--towards an ontology-driven cancer research and management system.

OBJECTIVE: This paper introduces the objectives, methods and results of ontology development in the EU co-funded project Advancing Clinico-genomic Trials on Cancer-Open Grid Services for Improving Medical Knowledge Discovery (ACGT). While the available data in the life sciences has recently grown both in amount and quality, the full exploitation of it is being hindered by the use of different underlying technologies, coding systems, category schemes and reporting methods on the part of different research groups. The goal of the ACGT project is to contribute to the resolution of these problems by developing an ontology-driven, semantic grid services infrastructure that will enable efficient execution of discovery-driven scientific workflows in the context of multi-centric, post-genomic clinical trials. The focus of the present paper is the ACGT Master Ontology (MO). METHODS: ACGT project researchers undertook a systematic review of existing domain and upper-level ontologies, as well as of existing ontology design software, implementation methods, and end-user interfaces. This included the careful study of best practices, design principles and evaluation methods for ontology design, maintenance, implementation, and versioning, as well as for use on the part of domain experts and clinicians. RESULTS: To date, the results of the ACGT project include (i) the development of a master ontology (the ACGT-MO) based on clearly defined principles of ontology development and evaluation; (ii) the development of a technical infrastructure (the ACGT Platform) that implements the ACGT-MO utilizing independent tools, components and resources that have been developed based on open architectural standards, and which includes an application updating and evolving the ontology efficiently in response to end-user needs; and (iii) the development of an Ontology-based Trial Management Application (ObTiMA) that integrates the ACGT-MO into the design process of clinical trials in order to guarantee automatic semantic integration without the need to perform a separate mapping process.

A semantic grid infrastructure enabling integrated access and analysis of multilevel biomedical data in support of postgenomic clinical trials on cancer.

This paper reports on original results of the Advancing Clinico-Genomic Trials on Cancer integrated project focusing on the design and development of a European biomedical grid infrastructure in support of multicentric, postgenomic clinical trials (CTs) on cancer. Postgenomic CTs use multilevel clinical and genomic data and advanced computational analysis and visualization tools to test hypothesis in trying to identify the molecular reasons for a disease and the stratification of patients in terms of treatment. This paper provides a presentation of the needs of users involved in postgenomic CTs, and presents such needs in the form of scenarios, which drive the requirements engineering phase of the project. Subsequently, the initial architecture specified by the project is presented, and its services are classified and discussed. A key set of such services are those used for wrapping heterogeneous clinical trial management systems and other public biological databases. Also, the main technological challenge, i.e. the design and development of semantically rich grid services is discussed. In achieving such an objective, extensive use of ontologies and metadata are required. The Master Ontology on Cancer, developed by the project, is presented, and our approach to develop the required metadata registries, which provide semantically rich information about available data and computational services, is provided. Finally, a short discussion of the work lying ahead is included.

Extending workflow management for knowledge discovery in clinico-genomic data.

Recent advances in research methods and technologies have resulted in an explosion of information and knowledge about cancers and their treatment. Knowledge Discovery (KD) is a key technique for dealing with this massive amount of data and the challenges of managing the steadily growing amount of available knowledge. In this paper, we present the ACGT integrated project, which is to contribute to the resolution of these problems by developing semantic grid services in support of multi-centric, post-genomic clinical trials. In particular, we describe the challenges of KD in clinico-genomic data in a collaborative Grid framework, and present our approach to overcome these difficulties by improving workflow management, construction and managing workflow results and provenance information. Our approach combines several techniques into a framework that is suitable to address the problems of interactivity and multiple dependencies between workflows, services, and data.