.
WEBWIRE Friday, December 13, 2020
An article in American Journal of Pathology ( ) details a promising way to classify subtypes of pancreatic-ductal-adenocarcinoma, allowing for quicker and more accessible diagnoses. This will lead to better patient outcomes by providing timely, tailored treatment.
Scientists have successfully developed a model that uses histopathology images to classify pancreatic-ductal-adenocarcino This method is highly accurate and provides a cost-effective, rapid alternative to the current methods which rely on expensive molecular tests. The new study, which opens in a new tab/window The American Journal of Pathologyopened in a new tab/window published by Elsevier holds promise for improving patient outcomes and advancing personalized treatment strategies. PDACs are now the third leading cause for cancer death in Canada and the United States. If detected early, surgery can cure about one-fifth PDAC cases. Despite the fact that these patients receive surgical intervention, their five-year survival rate is still 20%. About 80% of patients already have metastatic disease when they are diagnosed, and the majority of these patients die within a year.
PDAC’s aggressiveness is a challenge when using sequencing technology to determine a treatment plan for patients. The rapid clinical deterioration of the disease requires swift action to identify individuals eligible for targeted therapies and inclusion into clinical trials. Despite this, the current turnaround times of molecular profiling are between 19 and 52 days after biopsy. This is not enough to meet these time-sensitive needs.
David Schaeffer MD, Department of Pathology and Laboratory Medicine at the University of British Columbia and Vancouver General Hospital, and Pancreas Centre BC explains: [i”More and more potentially actionable subtypes to personalize treatment for pancreatic cancer patients are being discovered. However, the subtyping is still entirely based on genomic methodology based on DNA and RNA extracted from tissue. This methodology is outstanding if sufficient tissue is present, which is not always the case for PDAC tumors given the difficult anatomical location of this organ. Our study provides a promising method to cost-effectively and rapidly classify PDAC molecular subtypes based on routine hematoxylin-eosinstained slides, potentially leading to more effective clinical management of this disease”[/i]
This study involved training deep-learning AI models on whole slide pathology images in order to identify molecular subtypes of PDACbasal and classical using hematoxylin and Eosin (H&E)-stained slides. H&E staining, a widely used and cost-effective technique for diagnostics and prediction in pathology labs, is performed routinely with fast turnaround times. The models were tested on 110 slides of 44 patients from a local cohort. They were trained on 97 slide from The Cancer Genome Atlas. The best-performing model was able to identify the classical and basal types with an accuracy of 96.19% in the TCGA dataset, and 83.03 % in the local cohort. This demonstrates its robustness when used across different datasets.
Ali Bashashati PhD, co-lead researcher, School of Biomedical Engineering and Department of Pathology and Laboratory Medicine at the University of British Columbia notes: [i”The sensitivity and specificity of the model was 85% and 100%, respectively, making this AI tool a highly applicable tool for triaging patients for molecular testing. Also, the main achievement of this study is the fact that the AI model was able to detect the subtypes from biopsy images, making it a highly useful tool that can be deployed at the time of diagnosis”[/i]
As Dr. Bashashati concludes: “The AI-based approach is an exciting advancement for pancreatic cancer diagnosis, enabling us identify key molecular types quickly and cost-effectively.
Notes.
This article is A Deep-Learning Approach for the Identification Molecular Subtypes in Pancreatic Ductal Adenocarcinoma Using Whole Slide Pathology Images by Pouya Ahmedvand, Hossein Farahani David Farnell Amirali Darbandsari James Topham Joanna Karasinska Jessica Nelson Julia Naso Steven J.M. Jones, Daniel Renouf, David F. Schaeffer, and Ali Bashashati (https://doi.org/10.1016/j.ajpath.2024.08.006opens in new tab/window). The article appears in The American Journal of Pathology (), volume 194, issue 12, December 2024, published by Elsevier.
The article is openly available at https://ajp.amjpathol.org/article/S0002-9440(24)00325-0/fulltextopens in new tab/window. The full text of the article can be obtained by credentialed reporters upon request. Contact Eileen Leahy at +1 732 406 1313 or ajpmedia@elsevier.com to request a PDF of the article or more information. To reach the studys authors contact David Schaeffer, MD, at david.schaeffer@vch.ca, or Ali Bashashati, PhD, at ali.bashashati@ubc.ca.
This study was supported by Canadian Institutes of Health Research, number 418734, the Natural Sciences and Engineering Research Council of Canada (RGPIN-2019-04896), the Michael Smith Foundation for Health Research, the BC (British Columbia Cancer Foundation), and the VGH and UBC Hospital Foundations (University of British Columbia).
About the American Journal of Pathology (
) The American Journal of Pathology () opens in a new tab/window is the official journal of the American Society for Investigational PathologyOpens in a new tab/window and published by Elsevier. It seeks high quality original research reports, review articles, and comments related to the cellular and molecular basis of disease. The editors will take into consideration basic, translational and clinical investigations which directly address mechanisms of disease or provide a basis for future mechanistic inquires. These foundational investigations can include data mining, biomarker identification, molecular pathology and discovery research. The priority is placed on studies of human diseases and relevant experimental models, using molecular and cellular approaches. https://ajp.amjpathol.orgopens in new tab/window
About Elsevier
As a global leader in scientific information and analytics, Elsevier helps researchers and healthcare professionals advance science and improve health outcomes for the benefit of society. We achieve this by facilitating insights, critical decision-making and innovative solutions based upon trusted, evidence-based digital technologies and advanced AI-enabled content. Since 19659017, we have supported our research and healthcare communities. Our 9,500 employees, including 2,500 technologist, work around the globe to support researchers, librarians and academic leaders, as well as funders, governments, R&D intensive companies, doctors, nurses and future healthcare professionals, in their vital work. Cell Press, The Lancet, and Grays Anatomy are just a few of the 2,900 iconic reference books and scientific journals that we publish. Together with the Elsevier Foundationopens a new tab/window we work with the communities we support to advance inclusion and diversification in science, healthcare and research in developing countries.
Elsevier, a part of RELXopens a new tab/window is a global provider for information-based analytics and business decision tools. Visit www.elsevier.com for more information about our work, digital content and solutions.
( Press Release Image: https://photos.webwire.com/prmedia/6/331186/331186-1.png )
WebWireID331186 (19659021) This news content has been configured by WebWire’s editorial staff. Linking is allowed. WebWire offers News Release Distribution Services and Press Release Distribution Services.
