David Wishart (Presenter)
University of Alberta
Bio: Dr. David Wishart is a Professor in the Departments of Biological Sciences and Computing Science at the University of Alberta. For the past 10 years, he has led the “Human Metabolome Project” (HMP), a multi-university, multi-investigator project that is cataloguing all of the known metabolites in human tissues and biofluids. Using advanced methods in NMR, mass spectrometry and machine learning Dr. Wishart and his colleagues have identified or found evidence for more than 70,000 metabolites in the human body. This information has been archived on a freely accessible web-resource called the Human Metabolome Database (HMDB). He has published more than 330 research papers in many different areas including metabolomics, exposomics, bioinformatics, NMR, mass spectrometry, structural biology, and precision medicine. He is the director of Canada's national metabolomics laboratory (TMIC).
Authorship: David Wishart
University of Alberta
| Short Abstract In this presentation I will provide a number of specific examples of how quantitative metabolomics has been used to discover, validate and implement a number of clinically useful biomarkers and assays. These will include examples of predictive, prognostic and diagnostic multi-metabolite markers that exhibit substantially better performance than existing single biomarker assays. I will also discuss how quantitative metabolomics is beginning to have an impact in the fields of precision/personalized medicine by providing several interesting case studies where quantitative metabolomics has been used to assist in early disease diagnosis, disease prevention and “wellness” maintenance. |
Long Abstract
Introduction
Metabolomics is an emerging field of ‘omics science that is focused on the rapid, high-throughput characterization of hundreds to thousands of metabolites in biofluids, tissues or in organisms. Metabolomics exploits recent advances in chromatography, mass spectrometry, nuclear magnetic resonance (NMR) spectroscopy, big data and machine learning to identify and/or quantify a wide range of endogenous and exogenous metabolites. Because metabolite levels are highly sensitive to both an organism’s environment and its specific genotype, metabolomics offers a very powerful approach to molecular phenotyping. From a clinical perspective, metabolomics indicates what is happening while genomics indicates what might happen. With the recent development of highly reproducible, fully quantitative metabolomics techniques, the door has finally been opened to the concept of “clinical metabolomics”. Clinical metabolomics is an extension of clinical chemistry that employs quantitative metabolomics to aid in the discovery, validation and implementation of multiple, diagnostically useful biomarkers in a clinical setting. In this presentation I will provide a number of examples of how quantitative metabolomics has been used to discover, validate and implement a number of clinically useful biomarkers and assays.
Methods
A systematic review of more than 50 quantitative MS-based and NMR-based clinical metabolomics studies conducted in my laboratory over the past 7 years, as well as more than 200 MS-based quantitative clinical metabolomics studies published in the literature over the past 10 years was performed. These studies include biomarker studies focused on disease prediction, prognosis and diagnosis. Studies that provided validation cohorts or large sample numbers were analyzed in more detail. Using this data set, various machine learning and multivariate statistical analysis tools were employed to calculate receiver operating characteristic ROC curves to assess the sensitivity and specificity of the metabolite biomarkers. Optimization was performed to maximize the area under the ROC curve (AUROC) while minimizing the number of metabolite biomarkers. Additional literature research was done to determine the AUROC performance of existing clinical (single protein or metabolite) biomarkers to compare the overall performance of the metabolomics-derived biomarkers.
Results
This study revealed a large number of cases where quantitative metabolomic studies identified strongly predictive, prognostic or diagnostic mullti-metabolite biomarkers that far exceeded the performance of currently used (often single marker) clinical assays. Examples include multi-metabolite markers for predicting preeclampsia, birth defects, Alzheimer’s diseases and cancer cachexia as well as multi-metabolite markers for diagnosing kidney transplant rejection, heart failure categorization and colonic polyps. A smaller number of studies were identified that independently replicated and validated the identified biomarkers. Several of these metabolomic assays have recently moved into the clinic as laboratory developed tests. I will also present some recent data describing how quantitative metabolomics is having an impact in the fields of precision/personalized medicine. This will include several case studies where quantitative metabolomics has been used to assist in early disease diagnosis, disease prevention and “wellness” maintenance.
Conclusions & Discussion
These data provide compelling evidence that quantitative metabolomic techniques are now sufficiently robust that they can be used in the discovery, validation and implementation of multi-metabolite biomarker assays in clinical chemistry laboratories. The measurement of quantitative, multi-metabolite profiles via existing clinical MS infrastructure is now very simple and getting simpler. Therefore the use of quantitative metabolomics or the adoption of quantitative metabolomics techniques in clinical laboratories should be encouraged. Furthermore, rigorous, reproducible, absolute metabolite quantitation is now making metabolomics a vital technology for the advancement of in vitro diagnostics.
References & Acknowledgements:
| Description | Y/N | Source |
| Grants | no | |
| Salary | yes | Molecular You Incorporated |
| Board Member | yes | Molecular You Inc. |
| Stock | yes | Molecular You Inc. |
| Expenses | no |
IP Royalty: no
| Planning to mention or discuss specific products or technology of the company(ies) listed above: | no |