Related Experiment Video
Updated: Jul 13, 2025
Candidate Gene Testing in Clinical Cohort Studies with Multiplexed Genotyping and Mass Spectrometry
Published on: June 21, 2018
Multi-Omic Candidate Screening for Markers of Severe Clinical Courses of COVID-19
Alexander Dutsch1,2, Carsten Uhlig3, Matthias Bock1,2
1Department of Cardiology, German Heart Centre Munich, Technical University of Munich, Lazarettstraße 36, 80636 Munich, Germany.
Insights
A novel prediction model identifies protein signatures that precede severe COVID-19 complications like hyperinflammation and coagulopathy. These biomarkers may help predict disease progression and guide personalized treatment for coronavirus disease 2019 (COVID-19).
Area of Science:
- Biochemistry
- Immunology
- Proteomics
Background:
- Severe coronavirus disease 2019 (COVID-19) involves complex immuno-inflammatory, thrombotic, and parenchymal changes.
- Predicting individual COVID-19 trajectories for targeted prevention remains a significant clinical challenge.
- A distinct serologic signature preceding surges in Interleukin-6 (IL-6) and D-dimers in severe COVID-19 patients was hypothesized.
Purpose of the Study:
- To identify a serologic signature that predicts the onset of hyperinflammation and coagulopathy in severe COVID-19.
- To develop a predictive model for severe COVID-19 disease course.
- To explore candidate proteins and pathways associated with disease exacerbation.
Main Methods:
- Longitudinal plasma profiling (proteome, metabolome, biochemistry) of seven severe COVID-19 patients.
- Time-series analysis of patient sera to identify predictive biomarkers.
- Feature selection and dimensionality reduction using least absolute shrinkage and selection operator (LASSO) regression.
Main Results:
- A prediction model identified candidate proteins preceding elevated IL-6 (ΔIL-6) and D-dimers (ΔD-dimers) by 24-48 hours.
- Key pathways implicated include oxidative stress, inflammation, coagulation, immunoadhesion, and tissue repair.
- Proteins like P- and E-selectin, hepatocyte growth factor, and LDLR were identified as potential predictors.
Conclusions:
- A novel prediction model identified biological networks preceding COVID-19 hyperinflammation and coagulopathy.
- Proteins associated with D-dimer changes show particular promise as predictive biomarkers.
- Further validation is required, but these findings may enable personalized COVID-19 treatment strategies.
Background:
Severe coronavirus disease 2019 (COVID-19) disease courses are characterized by immuno-inflammatory, thrombotic, and parenchymal alterations. Prediction of individual COVID-19 disease courses to guide targeted prevention remains challenging. We hypothesized that a distinct serologic signature precedes surges of IL-6/D-dimers in severely affected COVID-19 patients.
Methods:
We performed longitudinal plasma profiling, including proteome, metabolome, and routine biochemistry, on seven seropositive, well-phenotyped patients with severe COVID-19 referred to the Intensive Care Unit at the German Heart Center. Patient characteristics were: 65 ± 8 years, 29% female, median CRP 285 ± 127 mg/dL, IL-6 367 ± 231 ng/L, D-dimers 7 ± 10 mg/L, and NT-proBNP 2616 ± 3465 ng/L.
Results:
Based on time-series analyses of patient sera, a prediction model employing feature selection and dimensionality reduction through least absolute shrinkage and selection operator (LASSO) revealed a number of candidate proteins preceding hyperinflammatory immune response (denoted ΔIL-6) and COVID-19 coagulopathy (denoted ΔD-dimers) by 24-48 h. These candidates are involved in biological pathways such as oxidative stress/inflammation (e.g., IL-1alpha, IL-13, MMP9, C-C motif chemokine 23), coagulation/thrombosis/immunoadhesion (e.g., P- and E-selectin), tissue repair (e.g., hepatocyte growth factor), and growth factor response/regulatory pathways (e.g., tyrosine-protein kinase receptor UFO and low-density lipoprotein receptor (LDLR)). The latter are host- or co-receptors that promote SARS-CoV-2 entry into cells in the absence of ACE2.
Conclusions:
Our novel prediction model identified biological and regulatory candidate networks preceding hyperinflammation and coagulopathy, with the most promising group being the proteins that explain changes in D-dimers. These biomarkers need validation. If causal, our work may help predict disease courses and guide personalized treatment for COVID-19.
