INTRODUCTION: Small extracellular vesicles (sEVs) play a role in the pathophysiology of viral respiratory infections, and may be suitable biomarkers for COVID-19 and Influenza infections or targets for treatment. We investigated differences in the surface proteome of plasma sEVs in patients with COVID-19 and Influenza. METHODS: In a discovery cohort with 117 patients, we used a random forest (RF) classifier in order to discriminate COVID-19 and Influenza patients based on routine clinical parameters. Furthermore, plasma samples from these patients were analyzed with an EV Array containing 33 antibodies to capture sEVs, which were then visualized with a combination of CD9, CD63, and CD81 antibodies. We applied a RF classifier and a random depth-first search (RDFS) approach to extract markers with the best discriminatory potential. Data were then validated in an independent set of patient samples on a chip-based ExoView platform. RESULTS: In the initial cohort of 117 patients, leukocyte numbers, and heart rate discriminated best between COVID-19 and Influenza infection. In the plasma samples, 32 EV surface markers could be detected. Feature panels containing CD9, CD81, and CD141 allowed a discrimination between COVID-19 and Influenza. Consecutively, increased CD9 abundance was validated in a second, independent cohort, with the ExoView technology. The increased CD9 signal in Influenza patients was confirmed and shown to be mostly driven by CD9/CD41a double positive sEVs, hinting at a thrombocyte origin. DISCUSSION: We identified leukocyte numbers and heart rate, as well as CD9 as a sEV surface marker to differentiate COVID-19 from Influenza patients.
- Bertrams, W.
- Roessler, F. K.
- Bæk, R.
- Jung, A. L.
- Laakmann, K.
- Jørgensen, M. M.
- Lehmann, M.
- Weckler, B.
- Schulte, L. N.
- Rohde, G.
- Bar, N.
- Barten, G.
- Schmeck, B.
Keywords
- Covid-19
- Influenza
- extracellular vesicles
- machine learning
