The ongoing pandemic with the new SARS-CoV2 virus shows the desperate and urgent need for better strategies to predict and treat Coronavirus disease 2019 (COVID-19). A subset of COVID-19 patients develop very severe respiratory symptoms, whereas others experience mild flu-like symptoms. Although it is evident that the host genetic and non-genetic factors, in interaction with new SARS-CoV2 virus, can determine variability in COVID-19 outcome, the underlying molecular mechanisms of patient-specific (COVID-19) outcome are unknown.
We have recently observed a striking time-dependent variability in immune response among COVID-19 patients, where 50% of the ICU patients showed immune response patterns similar to non-ICU patients. This suggests that instead of single layers of omics data measured cross-sectionally, we need longitudinal measurements of multi-omics data to predict severity and to obtain biological/molecular explanations to the clinical variability.
To determine how individual variation in molecular response (e.g. circulatory proteins and metabolites) affect COVID-19 severity and outcome we will use a unique and a largest cohort to date of COVID-19 patients in the Netherlands to profile longitudinal multi-omics data. We will then characterize: 1) the role of plasma metabolites, inflammatory markers and circulatory proteome variability in explaining COVID-19 outcome; 2) pinpoint causal molecular networks using dynamic changes in host multi-omics data; and 3) provide the genetic support for multi-omics variability that determine COVID-19 outcome in prospective independent cohorts. By conducting systematic longitudinal systems biology analyses, we will be able to establish causal relationships between omics-networks and COVID-19 clinical phenotypes. This will increase our understanding of the pathogenesis of COVID-19 and help to sub-group patients based on their response pattern so that treatment strategies can be adapted to individual patient categories.