Circulating proteins to predict adverse COVID-19 outcomes (preprint)

Predicting COVID-19 severity is difficult, and the biological pathways involved are not fully understood. To approach this problem, we measured 4,701 circulating human protein abundances in two independent cohorts totaling 986 individuals. We then trained prediction models including protein abundances and clinical risk factors to predict adverse COVID-19 outcomes in 417 subjects and tested these models in a separate cohort of 569 individuals. For severe COVID-19, a baseline model including age and sex provided an area under the receiver operator curve (AUC) of 65% in the test cohort. Selecting 92 proteins from the 4,701 unique protein abundances improved the AUC to 88% in the training cohort, which remained relatively stable in the testing cohort at 86%, suggesting good generalizability. Proteins selected from different adverse COVID-19 outcomes were enriched for cytokine and cytokine receptors, but more than half of the enriched pathways were not immune-related. Taken together, these findings suggest that circulating proteins measured at early stages of disease progression are reasonably accurate predictors of adverse COVID-19 outcomes. Further research is needed to understand how to incorporate protein measurement into clinical care.

Application of a 27-protein candidate cardiovascular surrogate endpoint to track risk ascendancy and resolution in COVID-19. (preprint)

Background: There is an urgent need for tools allowing the early prognosis and subsequent monitoring of individuals with heterogeneous COVID-19 disease trajectories. Pre-existing cardiovascular (CV) disease is a leading risk factor for COVID-19 susceptibility and poor outcomes, and cardiac involvement is prevalent in COVID-19 patients both during the acute phase as well as in convalescence. The utility of traditional CV risk biomarkers in mild COVID-19 disease or across disease course is poorly understood. We sought to determine if a previously validated 27-protein predictor of CV outcomes served a purpose in COVID-19. Methods: The 27-protein test of residual CV (RCV) risk was applied without modification to n=860 plasma samples from hospitalized and non-hospitalized SARS-CoV-2 infected individuals at disease presentation from three independent cohorts to predict COVID-19 severity and mortality. The same test was applied to an additional n=991 longitudinal samples to assess sensitivity to change in CV risk throughout the course of infection into convalescence. Results: In each independent cohort, RCV predictions were significantly related to maximal subsequent COVID-19 severity and to mortality. At the baseline blood draw, the mean protein-predicted likelihood of an event in subjects who died during the study period ranged from 88-99% while it ranged from 8-36% in subjects who were not admitted to hospital. Additionally, the test outperformed existing risk predictors based on commonly used laboratory chemistry values or presence of comorbidities. Application of the RCV test to sequential samples showed dramatic increases in risk during the first few days of infection followed by risk reduction in the survivors; a period of catastrophically high cardiovascular risk (above 50%) typically lasted 8-12 days and had not resolved to normal levels in most people within that timescale. Conclusions: The finding that a 27-protein candidate CV surrogate endpoint developed in multi-morbid patients prior to the pandemic is both prognostic and acutely sensitive to the adverse effects of COVID-19 suggests that this disease activates the same biologic risk-related mechanisms. The test may be useful for monitoring recovery and drug response.

Circulating Proteins Influencing COVID-19 Susceptibility and Severity: a Mendelian Randomization Study (preprint)

Proteins detectable in peripheral blood may influence COVID-19 susceptibility or severity. However, understanding which circulating proteins are etiologically involved is difficult because their levels may be influenced by COVID-19 itself and also subject to confounding factors. To identify circulating proteins influencing COVID-19 susceptibility and severity we undertook a large-scale two-sample Mendelian randomization (MR) study, since this study design can rapidly scan hundreds of circulating proteins and reduces bias due to confounding and reverse causation. We began by identifying the genetic determinants of 955 circulating proteins in up to 10,708 SARS-CoV-2 uninfected individuals, retaining only single nucleotide polymorphisms near the gene encoded by the circulating protein. We then undertook an MR study to estimate the effect of these proteins on COVID-19 susceptibility and severity using the Host Genetics Initiative. We found that a standard deviation increase in OAS1 levels was associated with reduced COVID-19 death or ventilation (N = 2,972 cases / 284,472 controls; OR = 0.48, P = 7x10-8), COVID-19 hospitalization (N = 6,492 / 1,012,809; OR = 0.60, P = 2x10-7) and COVID-19 susceptibility (N = 17,607 / 1,345,334; OR = 0.81, P = 6x10-5). Results were consistent despite multiple sensitivity analyses probing MR assumptions. OAS1 is an interferon-stimulated gene that promotes viral RNA degradation. Other potentially implicated proteins included IL10RB. Available medicines, such as interferon-beta-1b, increase OAS1 and could be explored for their effect on COVID-19 susceptibility and severity.