Microrna Profiles in Saliva of Children with and Without Severe Disease Due to Severe Acute Respiratory Syndrome Corona Virus 2 Infection

Background: The majority of children with exposure to SARS-CoV-2 virus have mild disease. However severe diseases such as Multisystem inflammatory syndrome (MISC) and pneumonia do occur in children. Currently, there are no established biomarkers that can predict progression to severe disease in children exposed to the virus. MicroRNAs (miRNAs) are non-coding RNAs that can be found in saliva and are thought to play a role in the regulation of inflammation following an infection. Our objective was to compare the miRNA profile in saliva of children with or without severe disease due to SARS-CoV-2 infection.

Methods: This prospective observational study was supported by the National Institutes of Health (NIH) RADx Program. Children ≤ 18 years of age presenting to two tertiary care children’s hospitals with symptoms of SARS-CoV-2 infection (confirmed by PCR test, serology or epidemiological link) were enrolled between 03/29/2021 and 04/ 30/2021. Severe infection was defined as any of the following within 30 days of testing: MISC or Kawasaki disease diagnosis, requirement for oxygen > 2L, inotropes, mechanical ventilation or ECMO, or the occurrence of death. Informed consent and a saliva swab were obtained at the time of SARS-CoV-2 diagnosis (DNA Genotek, Ottowa Canada), and RNA was extracted (Qiagen, Germantown, MD). Small RNA species ( <50 base pairs) were interrogated via shotgun sequencing (HiSeq 2500, Illumina, San Diego, CA) and miRNAs were quantified through alignment to the human genome (GRCh38). RNA features with sparse counts ( <10 in 90% of samples) were filtered, and the data was quantile normalized and mean-center scaled. Salivary miRNA levels were compared between those with severe and non-severe SARS-CoV-2 infection using Wilcoxon tests with Benjamini Hochberg multiple testing corrections. In addition, a logistic regression analysis was used to identify miRNA pairs that could best discriminate severe cases based on a Monte Carlo 100-fold cross-validated area under receiver operating characteristic curve (AUROC).

Results: Samples from 33 children were analyzed. Median age was 3 (3, 10) years and 54.5 % were males. Of the total, 29 were RT PCR positive, 4 had a positive serology and 6 children had severe infection. Seven miRNAs displayed significant differences (Fold change >2, FDR adjusted p < 0.1) among children with severe SARS-CoV-2 infection (Table). All seven miRNAs were up-regulated in severe SARS-CoV-2 cases. A logistic regression using a single ratio of miR-296-5p/miR-378j yielded 1.0 AUROC for differentiating children with severe infection (Figure).

Conclusion: In this interim analysis of salivary miRNA in childhood SARS-CoV-2 infection, we found a differential expression of 7 salivary miRNAs in children with severe infection. Ongoing work will seek to validate these findings and explore the role of miRNA in predicting severe SARS-CoV-2 infection in children.