Clinical classifiers of COVID-19 infection from novel ultra-high-throughput proteomics
Christoph B. Messner, Vadim Demichev, Daniel Wendisch, Laura Michalick, Matthew White, Anja Freiwald, Kathrin Textoris-Taube, Spyros I. Vernardis, Anna-Sophia Egger, Marco Kreidl, Daniela Ludwig, Christiane Kilian, Federica Agostini, Aleksej Zelezniak, Charlotte Thibeault, Moritz Pfeiffer, Stefan Hippenstiel, Andreas Hocke, Christof von Kalle, Archie Campbell, Caroline Hayward, David J. Porteous, Riccardo E. Marioni, Claudia Langenberg, Kathryn S. Lilley, Wolfgang M. Kuebler, Michael Muelleder, Christian Drosten, Martin Witzenrath, Florian Kurth, Leif Erik Sander, View ORCID ProfileMarkus Ralser
doi: https://doi.org/10.1101/2020.04.27.20081810
Abstract
The COVID-19 pandemic is an unprecedented global challenge. Highly variable in its presentation, spread and clinical outcome, novel point-of-care diagnostic classifiers are urgently required. Here, we describe a set of COVID-19 clinical classifiers discovered using a newly designed low-cost high-throughput mass spectrometry-based platform. Introducing a new sample preparation pipeline coupled with short-gradient high-flow liquid chromatography and mass spectrometry, our methodology facilitates clinical implementation and increases sample throughput and quantification precision. Providing a rapid assessment of serum or plasma samples at scale, we report 27 biomarkers that distinguish mild and severe forms of COVID-19, of which some may have potential as therapeutic targets. These proteins highlight the role of complement factors, the coagulation system, inflammation modulators as well as pro-inflammatory signalling upstream and downstream of Interleukin 6. Application of novel methodologies hence transforms proteomics from a research tool into a rapid-response, clinically actionable technology adaptable to infectious outbreaks.
