Abstract
Introduction: Around 20-30% of the hospitalized patients with COVID-19 develop a severe phenotype that is associated with high mortality rate. There is a clinical demand of reliable tools to monitor the patient’s status. The non-coding transcriptome, and more specifically microRNAs (miRNAs), has emerged as a potential tool for clinical decision-making. The association between the circulating miRNA profile and COVID-19 has not been previously explored.
Objectives: To evaluate the alterations in the circulating miRNA signature linked to COVID-19 severity and its potential in decision-making.
Methods: Plasma samples were collected from 54 COVID-19 patients (23 patients hospitalized at clinical wards and 31 patients admitted to the ICU). A panel of 41 miRNAs associated with immune/inflammatory response, lung damage, myocardial damage and coagulation was analyzed using RT-qPCR. Quality control was performed using spike-ins and hemolysis tests. Differences between groups were evaluated by relative quantification using linear models for arrays. Dis-crimination was evaluated using the area under the ROC curve (AUC). A variable selection process based on random forest was used to con-struct a predictive model of ICU stay.
Results: Fifty-three samples passed the quality control. Four deregulated miRNAs; miR-16-5p (viral acute respiratory infection), miR-92-3p (immune response), miR-451a (immune response) and miR-486-5p (antiviral defense), were identified in ICU patients compared to ward patients. Both miR-451a and miR-486-5p showed a good discrimination value (AUC = 0.78 for miR-451 and 0.77 for miR-486-5p). Random forest analyses identified a signature of 3 miRNAs (miR-27a-3p, miR-92a-3p and miR-451a) that displayed an optimal discrimination accuracy (AUC = 0.84).
Conclusions: Our findings reveal for the first time a profile of miRNAs that is altered in plasma of patients with a severe phenotype of the disease. This circulating miRNA signature may be useful for the management of the COVID-19 patient.
Objectives: To evaluate the alterations in the circulating miRNA signature linked to COVID-19 severity and its potential in decision-making.
Methods: Plasma samples were collected from 54 COVID-19 patients (23 patients hospitalized at clinical wards and 31 patients admitted to the ICU). A panel of 41 miRNAs associated with immune/inflammatory response, lung damage, myocardial damage and coagulation was analyzed using RT-qPCR. Quality control was performed using spike-ins and hemolysis tests. Differences between groups were evaluated by relative quantification using linear models for arrays. Dis-crimination was evaluated using the area under the ROC curve (AUC). A variable selection process based on random forest was used to con-struct a predictive model of ICU stay.
Results: Fifty-three samples passed the quality control. Four deregulated miRNAs; miR-16-5p (viral acute respiratory infection), miR-92-3p (immune response), miR-451a (immune response) and miR-486-5p (antiviral defense), were identified in ICU patients compared to ward patients. Both miR-451a and miR-486-5p showed a good discrimination value (AUC = 0.78 for miR-451 and 0.77 for miR-486-5p). Random forest analyses identified a signature of 3 miRNAs (miR-27a-3p, miR-92a-3p and miR-451a) that displayed an optimal discrimination accuracy (AUC = 0.84).
Conclusions: Our findings reveal for the first time a profile of miRNAs that is altered in plasma of patients with a severe phenotype of the disease. This circulating miRNA signature may be useful for the management of the COVID-19 patient.
| Original language | English |
|---|---|
| Pages (from-to) | 5-6 |
| Number of pages | 2 |
| Journal | Archivos de Bronconeumologia |
| Volume | 56 |
| Issue number | SC2 |
| Publication status | Published - Nov 2020 |
| Externally published | Yes |
| Event | XIII Jornadas de Formación del Centro de Investigación Biomédica en Red de Enfermedades Respiratorias (CIBERES) - Virtual, Spain Duration: 19 Nov 2020 → 20 Nov 2020 |
Keywords
- MicroRNA
- coronavirus (COVID-19)
- plasma
