Severity Predictors Using Immunology and Transcriptomics in Saliva Using Multi Neural Network Intelligence in SARS-CoV2 Infection in Children (SPITS MISC)

Children have been disproportionately less impacted by the Corona Virus Disease 2019 (COVID-19) caused by the Severe Acute Respiratory Syndrome Corona Virus 2 (SARS-CoV-2) compared to adults. However, severe illnesses including Multisystem Inflammatory Syndrome (MIS-C) and respiratory failure have occurred in a small proportion of children with SARS-CoV-2 infection. Nearly 80% of children with MIS-C are critically ill with a 2-4% mortality rate. Currently, there are no modalities to characterize the spectrum of disease severity and predict which child with SARS-CoV-2 exposure will likely develop severe illness including MIS-C. Thus, there is an urgent need to develop a diagnostic modality to distinguish the varying phenotypes of disease and risk stratify disease. The epigenetic changes in microRNA (miRNA) profiles that occur due to an infection can impact disease severity by altering immune response and cytokine regulation which may be detected in body fluids including saliva. The long-term goal of this study was to improve outcomes of children with SARS-CoV-2 by early identification and treatment of those at risk for severe illness. The central hypothesis was that a model that integrates salivary biomarkers with social and clinical determinants of health would predict disease severity in children with SARS-CoV-2 infection. This central hypothesis was pursued through four specific, phased aims. The first two aims from the R61 phase were: 1) Defining and comparing the salivary molecular host response in children with varying phenotypes (severe and non-severe) SARS-CoV-2 infections, and 2) Developing and validating a sensitive and specific model to predict severe SARS-CoV-2 illness in children. The R33 phase pursued the following two aims: 3) Development of a portable, rapid device that quantifies salivary miRNAs with comparable accuracy to predicate technology (qRT-PCR), and 4) Development of an artificial intelligence (AI) assisted cloud and mobile system for early recognition of severe SARS-CoV-2 infection in children. The above aims were pursued using an innovative combination of salivaomics and bioinformatics, analytic techniques of AI and clinical informatics. This research is significant because development of a sensitive model to risk stratify disease is expected to improve outcomes of children with severe SARS-CoV-2 infection via early recognition and timely intervention. The outcome of this study is a better understanding of the epigenetic regulation of host immune response to the viral infection which is expected to lead to personalized therapy in the future. The results have immediately had a positive impact and will lead to the creation of patient profiles based on individual risk factors which can enable early identification of severe disease and appropriate resource allocation during the pandemic.