Open AccessEarly prediction of COVID‐19 severity using extracellular vesicle COPB2
Author(s) -
Fujita Yu,
Hoshina Tokio,
Matsuzaki Juntaro,
Yoshioka Yusuke,
Kadota Tsukasa,
Hosaka Yusuke,
Fujimoto Shota,
Kawamoto Hironori,
Watanabe Naoaki,
Sawaki Kenji,
Sakamoto Yohei,
Miyajima Makiko,
Lee Kwangyole,
Nakaharai Kazuhiko,
Horino Tetsuya,
Nakagawa Ryo,
Araya Jun,
Miyato Mitsuru,
Yoshida Masaki,
Kuwano Kazuyoshi,
Ochiya Takahiro
Publication year - 2021
Publication title -
journal of extracellular vesicles
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 5.94
H-Index - 68
ISSN - 2001-3078
DOI - 10.1002/jev2.12092
Subject(s) - asymptomatic , cohort , medicine , extracellular vesicles , covid-19 , pathological , extracellular , severity of illness , extracellular vesicle , biomarker , immunology , microvesicles , biology , disease , infectious disease (medical specialty) , microrna , biochemistry , gene , microbiology and biotechnology
Abstract The clinical manifestations of COVID‐19 vary broadly, ranging from asymptomatic infection to acute respiratory failure and death. But the predictive biomarkers for characterizing the variability are still lacking. Since emerging evidence indicates that extracellular vesicles (EVs) and extracellular RNAs (exRNAs) are functionally involved in a number of pathological processes, we hypothesize that these extracellular components may be key determinants and/or predictors of COVID‐19 severity. To test our hypothesis, we collected serum samples from 31 patients with mild COVID‐19 symptoms at the time of their admission for discovery cohort. After symptomatic treatment without corticosteroids, 9 of the 31 patients developed severe/critical COVID‐19 symptoms. We analyzed EV protein and exRNA profiles to look for correlations between these profiles and COVID‐19 severity. Strikingly, we identified three distinct groups of markers (antiviral response‐related EV proteins, coagulation‐related markers, and liver damage‐related exRNAs) with the potential to serve as early predictive biomarkers for COVID‐19 severity. As the best predictive marker, EV COPB2 protein, a subunit of the Golgi coatomer complex, exhibited significantly higher abundance in patients remained mild than developed severe/critical COVID‐19 and healthy controls in discovery cohort (AUC 1.00 (95% CI: 1.00‐1.00)). The validation set included 40 COVID‐19 patients and 39 healthy controls, and showed exactly the same trend between the three groups with excellent predictive value (AUC 0.85 (95% CI: 0.73‐0.97)). These findings highlight the potential of EV COPB2 expression for patient stratification and for making early clinical decisions about strategies for COVID‐19 therapy.