Search

ImportanceIdentifying new prognostic features in colon cancer has the potential to refine histopathologic review and inform patient care. Although prognostic artificial intelligence systems have recently demonstrated significant risk stratification for several cancer types, studies have not yet shown that the machine learning–derived features associated with these prognostic artificial intelligence systems are both interpretable and usable by pathologists.ObjectiveTo evaluate whether pathologist scoring of a histopathologic feature previously identified by machine learning is associated with survival among patients with colon cancer.Design, Setting, and ParticipantsThis prognostic study used deidentified, archived colorectal cancer cases from January 2013 to December 2015 from the University of Milano-Bicocca. All available histologic slides from 258 consecutive colon adenocarcinoma cases were reviewed from December 2021 to February 2022 by 2 pathologists, who conducted semiquantitative scoring for tumor adipose feature (TAF), which was previously identified via a prognostic deep learning model developed with an independent colorectal cancer cohort.Main Outcomes and MeasuresPrognostic value of TAF for overall survival and disease-specific survival as measured by univariable and multivariable regression analyses. Interpathologist agreement in TAF scoring was also evaluated.ResultsA total of 258 colon adenocarcinoma histopathologic cases from 258 patients (138 men [53%]; median age, 67 years [IQR, 65-81 years]) with stage II (n = 119) or stage III (n = 139) cancer were included. Tumor adipose feature was identified in 120 cases (widespread in 63 cases, multifocal in 31, and unifocal in 26). For overall survival analysis after adjustment for tumor stage, TAF was independently prognostic in 2 ways: TAF as a binary feature (presence vs absence: hazard ratio [HR] for presence of TAF, 1.55 [95% CI, 1.07-2.25]; P = .02) and TAF as a semiquantitative categorical feature (HR for widespread TAF, 1.87 [95% CI, 1.23-2.85]; P = .004). Interpathologist agreement for widespread TAF vs lower categories (absent, unifocal, or multifocal) was 90%, corresponding to a κ metric at this threshold of 0.69 (95% CI, 0.58-0.80).Conclusions and RelevanceIn this prognostic study, pathologists were able to learn and reproducibly score for TAF, providing significant risk stratification on this independent data set. Although additional work is warranted to understand the biological significance of this feature and to establish broadly reproducible TAF scoring, this work represents the first validation to date of human expert learning from machine learning in pathology. Specifically, this validation demonstrates that a computationally identified histologic feature can represent a human-identifiable, prognostic feature with the potential for integration into pathology practice.

Various biological resources, such as biobanks and disease-specific registries, have become indispensable resources to better understand the epidemiology and biological mechanisms of disease and are fundamental for advancing medical research. Nevertheless, biobanks and similar resources still face significant challenges to become more findable and accessible by users on both national and global scales. One of the main challenges for users is to find relevant resources using cataloging and search services such as the BBMRI-ERIC Directory, operated by European Research Infrastructure on Biobanking and Biomolecular Resources (BBMRI-ERIC), as these often do not contain the information needed by the researchers to decide if the resource has relevant material/data; these resources are only weakly characterized. Hence, the researcher is typically left with too many resources to explore and investigate. In addition, resources often have complex procedures for accessing holdings, particularly for depletable biological materials. This article focuses on designing a system for effective negotiation of access to holdings, in which a researcher can approach many resources simultaneously, while giving each resource team the ability to implement their own mechanisms to check if the material/data are available and to decide if access should be provided. The BBMRI-ERIC has developed and implemented an access and negotiation tool called the BBMRI-ERIC Negotiator. The Negotiator enables access negotiation to more than 600 biobanks from the BBMRI-ERIC Directory and other discovery services such as GBA/BBMRI-ERIC Locator or RD-Connect Finder. This article summarizes the principles that guided the design of the tool, the terminology used and underlying data model, request workflows, authentication and authorization mechanism(s), and the mechanisms and monitoring processes to stimulate the desired behavior of the resources: to effectively deliver access to biological material and data.

During the COVID-19 pandemic, the European biobanking infrastructure is in a unique position to preserve valuable biological material complemented with detailed data for future research purposes. Biobanks can be either integrated into healthcare, where preservation of the biological material is a fork in clinical routine diagnostics and medical treatment processes or they can also host prospective cohorts or material related to clinical trials. The paper discussed objectives of BBMRI-ERIC, the European research infrastructure established to facilitate access to quality-defined biological materials and data for research purposes, with respect to the COVID-19 crisis: (a) to collect information on available European as well as non-European COVID-19-relevant biobanking resources in BBMRI-ERIC Directory and to facilitate access to these via BBMRI-ERIC Negotiator platform; (b) to help harmonizing guidelines on how data and biological material is to be collected to maximize utility for future research, including large-scale data processing in artificial intelligence, by participating in activities such as COVID-19 Host Genetics Initiative; (c) to minimize risks for all involved parties dealing with (potentially) infectious material by developing recommendations and guidelines; (d) to provide a European-wide platform of exchange in relation to ethical, legal, and societal issues (ELSI) specific to the collection of biological material and data during the COVID-19 pandemic.

In this paper an automatic classification system for pathological findings is presented. The starting point in our undertaking was a pathologic tissue collection with about 1.4 million tissue samples described by free text records over 23 years. Exploring knowledge out of this “big data” pool is a challenging task, especially when dealing with unstructured data spanning over many years. The classification is based on an ontology-based term extraction and decision tree build with a manually curated classification system. The information extracting system is based on regular expressions and a text substitution system. We describe the generation of the decision trees by medical experts using a visual editor. Also the evaluation of the classification process with a reference data set is described. We achieved an F-Score of 89,7% for ICD-10 and an F-Score of 94,7% for ICD-O classification. For the information extraction of the tumor staging and receptors we achieved am F-Score ranging from 81,8 to 96,8%.

// Sonja M. Kessler 1, 2, * , Eva Lederer 1, * , Stephan Laggai 2 , Nicole Golob-Schwarzl 1, 3 , Kevan Hosseini 2 , Johannes Petzold 1 , Caroline Schweiger 1 , Robert Reihs 1 , Marlen Keil 4 , Jens Hoffmann 4 , Christian Mayr 5 , Tobias Kiesslich 5 , Martin Pichler 6, 7, 8 , Kyung Sik Kim 9 , Hyungjin Rhee 10 , Young Nyun Park 11 , Sigurd Lax 12 , Peter Obrist 13 , Alexandra K. Kiemer 2 and Johannes Haybaeck 1, 14 1 Institute of Pathology, Medical University of Graz, Graz, Austria 2 Department of Pharmacy, Pharmaceutical Biology, Saarland University, Saarbrucken, Germany 3 Center for Biomarker Research in Medicine, Graz, Austria 4 Experimental Pharmacology and Oncology Berlin-Buch GmbH, Berlin, Germany 5 Laboratory for Tumor Biology and Experimental Therapies, Institute of Physiology and Pathophysiology, Department of Internal Medicine I, Paracelsus Medical University, Salzburg, Austria 6 Division of Oncology, Department of Internal Medicine, Medical University of Graz, Graz, Austria 7 Research Unit for Non-coding RNAs and Genome Editing in Cancer, Medical University of Graz, Graz, Austria 8 Department of Experimental Therapeutics, The University of Texas MD Anderson Cancer Center, Houston, USA 9 Department of Surgery, Yonsei University College of Medicine, Seoul, South Korea 10 Department of Radiology, Brain Korea 21 PLUS Project for Medical Science, Yonsei University College of Medicine, Seoul, South Korea 11 Department of Pathology, Brain Korea 21 PLUS Project for Medical Science, Yonsei University College of Medicine, Seoul, South Korea 12 Department of Pathology, General Hospital Graz Sued-West, Graz, Austria 13 Laboratory of Pathology, Dr. Obrist and Dr. Brunhuber OG, Zams, Austria 14 Department of Pathology, Medical Faculty, Otto-von-Guericke-University Magdeburg, Magdeburg, Germany * These authors contributed equally to this work Correspondence to: Johannes Haybaeck, email: johannes.haybaeck@med.ovgu.de Keywords: IGF2BP2/IMP2/p62, IGF2BP3/IMP3, IGF2BP1/IMP1, NADPH oxidase, gallbladder carcinoma Received: January 04, 2017 Accepted: September 04, 2017 Published: September 21, 2017 ABSTRACT Overexpression of the oncofetal insulin-like growth factor 2 mRNA-binding protein 2 (IMP2/IGF2BP2) has been described in different cancer types. Gallbladder carcinoma (GBC) is a rare but highly aggressive cancer entity with late clinical detection and poor prognosis. The aim of this study was to investigate the role of IMP2 in human GBC. Tissue microarrays (TMAs) of an international multi-center GBC sample collection from n = 483 patients were analyzed by immunohistochemistry. IMP2 immunoreactivity was found in 74.3% of the tumor samples on TMA, of which 14.0% showed strong and 86.0% low staining intensity. 72.4% of the tumor samples were IMP1 positive, but IMP1 showed lower expression in tumor tissue compared to control tissues. IMP3 immunoreactivity was observed in 92.7% of all tumors, of which 53.6% revealed strong IMP3 expression. Kaplan-Meier analysis linked high IMP2 expression to shorter survival time ( p = 0.033), whereas neither IMP1 nor IMP3 expression was linked to a decreased survival time. Eight different human biliary tract cancer (BTC) cell lines were evaluated for tumor growth kinetics in mouse xenografts. Cell lines with high IMP2 expression levels showed the fastest increase in tumor volumes in murine xenografts. Furthermore, IMP2 expression in these cells correlated with the generation of reactive oxygen species (ROS) and RAC1 expression in BTC cells, suggesting RAC1-induced ROS generation as a potential mechanism of IMP2-promoted progression of GBC. In conclusion, IMP2 is frequently overexpressed in GBC and significantly associated with poor prognosis and growth rates in vivo . IMP2 might therefore represent a new target for the treatment of advanced GBC.

The domain of biobanking has gone through many stages and as a result there are a wide range of commercial and open source software solutions available. The utilization of these software tools requires different levels of domain and technical skills for installation, configuration and ultimate us of these biobank software tools. To compound this complexity the biobanking community are required to work together in order to share knowledge and jointly build solutions to underpin the research infrastructure. We have evaluated the available tools, described them in a catalogue (BiobankApps) and made a selection of tools available to biobanks in a reference toolbox (BIBBOX) that are use-case driven. In the BiobankApps tool catalogue, both commercial and open source software solutions related to the biobanking domain are included, classified and evaluated. The evaluation covers: 1) "user review" by an authenticated user 2) domain expert: quick analysis by BBMRI members and 3) domain expert: detailed analysis and test installation with real world data. The evaluation is paired with a survey across the more "advanced" (from a technology perspective) biobanks to investigate what tools are currently used and summarises known benefits/drawbacks of the respective packages. In the second step we recommend tools for specific use cases, and install, configure and connect these in the BIBBOX framework. This service also builds on the existing work in the United Kingdom in seeking to establish the motivations for different stakeholders to become involved and therefore assisting in prioritising the use-cases based on the level of need and support within the research community. All tools associated to a use-case are available as BIBBOX applications (technically this is achieved by docker containers), which are integrated in the BIBBOX framework with central identification and user management. In future work we plan to share the acquired knowledge with other networks, develop an Application Programmable Interface (API) for the exchange of metadata with other tool catalogues and work on an ontology for the evaluation of biobank software.