Association of SARS-CoV-2 nucleocapsid viral antigen and the receptor for advanced glycation end products with development of severe disease in patients presenting to the emergency department with COVID-19

Introduction There remains a need to better identify patients at highest risk for developing severe Coronavirus Disease 2019 (COVID-19) as additional waves of the pandemic continue to impact hospital systems. We sought to characterize the association of receptor for advanced glycation end products (RAGE), SARS-CoV-2 nucleocapsid viral antigen, and a panel of thromboinflammatory biomarkers with development of severe disease in patients presenting to the emergency department with symptomatic COVID-19. Methods Blood samples were collected on arrival from 77 patients with symptomatic COVID-19, and plasma levels of thromboinflammatory biomarkers were measured. Results Differences in biomarkers between those who did and did not develop severe disease or death 7 days after presentation were analyzed. After adjustment for multiple comparisons, RAGE, SARS-CoV-2 nucleocapsid viral antigen, interleukin (IL)-6, IL-10 and tumor necrosis factor receptor (TNFR)-1 were significantly elevated in the group who developed severe disease (all p<0.05). In a multivariable regression model, RAGE and SARS-CoV-2 nucleocapsid viral antigen remained significant risk factors for development of severe disease (both p<0.05), and each had sensitivity and specificity >80% on cut-point analysis. Discussion Elevated RAGE and SARS-CoV-2 nucleocapsid viral antigen on emergency department presentation are strongly associated with development of severe disease at 7 days. These findings are of clinical relevance for patient prognostication and triage as hospital systems continue to be overwhelmed. Further studies are warranted to determine the feasibility and utility of point-of care measurements of these biomarkers in the emergency department setting to improve patient prognostication and triage.

In a multivariable regression model, RAGE and SARS-CoV-2 nucleocapsid viral antigen remained significant risk factors for development of severe disease (both p<0.05), and each had sensitivity and specificity >80% on cut-point analysis.
Discussion: Elevated RAGE and SARS-CoV-2 nucleocapsid viral antigen on emergency department presentation are strongly associated with development of severe disease at 7 days. These findings are of clinical relevance for patient prognostication and triage as hospital systems continue to be overwhelmed. Further studies are warranted to determine the feasibility and utility of point-of care measurements of these biomarkers in the emergency department setting to improve patient prognostication and triage. KEYWORDS COVID-19, biomarkers, thromboinflammation, receptor for advanced glycation end products, triage

Introduction
The COVID-19 pandemic continues to present major risks to public health as additional waves and variants lead to new spikes in infections, at times overwhelming health care systems (1,2). Although progress has been made in the prevention and treatment of COVID-19, a better understanding of early biologic markers predictive of ultimate disease severity may further facilitate patient prognostication and triage, and our understanding of the disease's pathophysiology. Previous studies have identified a range o f p l a s m a b i o m a r k e r s i n c l u d i n g t h o s e r e l a t e d t o thromboinflammation, lung injury, and endothelial damage, that may be useful for predicting outcomes and understanding COVID-19 pathophysiology (3)(4)(5)(6)(7). However, most studies have focused on patients who have already been admitted to the hospital or intensive care unit, with a time lag between hospital presentation and sample collection. Data identifying biomarkers on patient presentation in the emergency department predictive of development of severe disease remains limited (8,9). Common laboratory assays such as ddimer, c-reactive protein, and fibrinogen may be of value, but may lack sensitivity and specificity to accurately predict COVID-19 outcomes (3,10,11).
The soluble receptor for advanced glycation end-products (RAGE) and SARS-CoV-2 nucleocapsid viral antigen are two biomarkers recently shown to have prognostic value in hospitalized patients with COVID-19 (12)(13)(14). However, few studies to date have examined these biomarkers in patients presenting to the emergency department. RAGE is an established biomarker of alveolar epithelial injury and is prognostic in patients with acute respiratory distress syndrome (ARDS) (15)(16)(17). Recent data from hospitalized patients with COVID-19 suggest that rises in biomarkers of alveolar epithelial injury including RAGE occur early in the disease course, whereas markers of endothelial injury manifest later (18). We therefore hypothesized that RAGE would be a useful early biomarker of disease severity in patients with COVID-19. Additionally, viral load measured by plasma SARS-CoV-2 ribonucleic acid (RNA) or by plasma SARS-CoV-2 nucleocapsid viral antigen is known to be associated with disease severity (19-22). However, to what extent plasma SARS-CoV-2 nucleocapsid viral antigen correlates with RAGE and with ultimate disease severity early in patients' disease course is not well described. Further, understanding the association of SARS-CoV-2 nucleocapsid viral antigen plasma concentrations and associated thromboinflammatory biomarkers on arrival with patient outcome may inform clinical decision making in patients with COVID-19, and support future studies examining the role of point-of-care testing for these biomarkers.
The aims of this study were to test the prognostic value of RAGE and SARS-CoV-2 nucleocapsid viral antigen, as well as a panel of common thromboinflammatory biomarkers, measured on arrival to the emergency department in patients presenting with symptomatic COVID-19. We additionally aimed to examine the correlation between SARS-CoV-2 nucleocapsid antigen levels with a panel of thromboinflammatory biomarkers in this cohort.

Patient enrollment, data collection, inclusion/exclusion criteria
Participants included in this study were enrolled as part of the Covid-19 Associated Coagulopathy, Inflammation, and Thrombosis (Co-ACIT) Study (23). All adults presenting to our institution's emergency department who were undergoing evaluation for COVID-19 were eligible. Study participants (n=125) were enrolled under an initial waiver of consent due to their respiratory compromise, and informed consent was obtained after initial blood collection. In addition to laboratory testing performed at the discretion of the treating clinician, whole blood was collected in sodium citrate (3.2%) on all patients for investigational assays. Exclusion criteria included patients who declined to participate (n=9), and those who were pregnant (n=1). After these exclusions, 115 patients remained in the study, and of those 77 tested positive for SARS-CoV-2 by reverse transcriptase PCR nasal swab testing and were included in this analysis (June 2020-March 2021). All patients were followed longitudinally to hospital discharge with comprehensive collection of clinical data including demographics, physiological and laboratory parameters, and clinical outcomes. The study was approved by the Committee on Human Research at the University of California, San Francisco (study number 20-30895).

Statistical analysis
The primary outcome for this study was disease severity 7 days after presentation to the emergency department, classified by the World Health Organization's (WHO) Ordinal Scale (24). This outcome was dichotomized as mild disease (WHO Scale 1-4; ambulatory/discharged or hospitalized but receiving no supplemental oxygen or only supplemental oxygen by mask/nasal cannula) versus severe disease or death (WHO scale 5-7; hospitalized and receiving non-invasive ventilation, high-flow nasal cannula, invasive mechanical ventilation, or death). Patient characteristics and biomarkers were compared between the mild and severe disease groups using Student's t tests for normally distributed continuous variables, Wilcoxon-Rank Sum tests for non-normally distributed continuous variables, and Fisher's exact test for binary variables. Biomarkers with non-normal distributions were log-transformed to reduce skew.
The association of biomarkers with development of severe disease at day 7 was assessed with bivariate analysis. After adjustment for multiple comparisons (Sidak-Holm adjusted P-Values (25)), the biomarkers showing significant association with development of severe disease or death were included in a multivariable model. The model additional included Acute Physiology and Chronic Health Evaluation III score (APACHE-III) to account physiologic and demographic predictors of poor outcome, and the final model was fit using backward stepwise logistic regression with significance set at 0.1 for retention. The APACHE-III score is a well validated tool to estimate mortality across a range of disease states including sepsis and acute respiratory distress syndrome (26,27). Cut-point analysis using Youden's index was used to determine optimal cutoffs for each of the statistically significant biomarkers for discriminating development of severe disease or death (28). The association of the SARS-CoV-2 nucleocapsid viral antigen levels with thromboinflammatory biomarkers was examined using spearman rank correlation coefficients.

Patient characteristics
Of the 77 subjects included, 17 (22%) developed severe disease (n=15) or died (n=2) 7 days after presentation, while the remaining 60 (78%) of patients were ambulatory (n=43) or hospitalized with mild disease (n=17) ( Table 1). Those who developed severe disease or death were older (mean age 70 vs 55 years, p<0.01), had higher rates of diabetes (59% vs 29%, p=0.04), but were not significantly different with respect to rates of kidney, liver, or heart disease ( Table 1). Initial platelet counts, hemoglobin, international normalized ratio (INR), activated partial thromboplastic time, and creatinine were not different between the two groups, but fibrinogen and d-dimer were significantly elevated on presentation in those who developed severe disease or death (both p<0.05) ( Table 1). APACHE-III scores on presentation were much higher in those who developed severe disease or death (59 vs 27 points, p<0.01) ( Table 1). Eighteen percent (3/17) of patients who developed severe disease had venous thromboembolism (VTE) versus 3% (2/60) in those with mild disease (Table 1). Mortality at hospital discharge was 41% (7/17) in the patients who developed severe disease at 7 days, while one patient (2%) with mild disease at 7 days subsequently deteriorated and expired (p<0.01, Table 1).

Biomarkers compared by disease severity
T h e r e w e r e s e v e r a l n o t a b l e d i ff e r e n c e s a m o n g thromboinflammatory biomarkers measured from the plasma collected at presentation between those who did and did not develop severe disease or death at day 7 ( Table 2). The SARS-CoV-2 nucleocapsid viral antigen concentrations were notably higher among those who developed severe disease or death (median [IQR] 9.1 [8. [5][6][7][8][9].8] log pg/ml vs 7.1 [5.3-8.2] log pg/ml, p<0.01) ( Table 2). Eleven of the inflammatory and endothelial related biomarkers were also significantly elevated in those who developed severe disease or death, including RAGE, TREM-1, IL-6, IL-8, IL-10, IP-10, TNFR-1, MMP-8, and soluble P-selectin (all p<0.05) ( Table 2). With respect to coagulation and fibrinolysis related markers, PAI-1 and vWF were significantly elevated in those who developed severe disease or death, but there were no differences in tPA, plasminogen, or antiplasmin (Table 2). After adjustment for multiple comparisons, SARS-CoV-2 nucleocapsid viral antigen, RAGE, IL-6, IL-10, and TNFR-1 retained statistical significance in association with development of severe disease or death (Table 2).

Multivariable model for severe/disease death
We then used stepwise modeling to build a multivariable model predictive of development of severe disease or death at 7 days, inputting biomarkers that remained significant after adjustment for multiple comparisons (SARS-CoV-2 nucleocapsid viral antigen, RAGE, IL-6, IL-10, and TNFR-1) as well as the APACHE-III score to adjust for clinical parameters of presenting disease severity. RAGE, SARS-CoV-2 nucleocapsid antigen, and APACHE-III score were retained in the final model as each being independent significant risk factors for development of severe   disease or death at 7 days (Table 3). Cut-point analysis was performed for each of the five biomarkers that retained statistical significance after adjustment for multiple comparisons. This demonstrated that at cutoffs of >3,200pmg/ml for RAGE and 4,350pg/ml for SARS-CoV-2 viral antigen on emergency department presentation there was a sensitivity >80% and specificity >90% for development of severe disease or death at day 7 (Table 4). IL-6 and TNFR-1 each had good sensitivity (88% and 82% respectively), but lower specificity (72% and 70%) for stratifying patients by development of mild versus severe disease at day 7, and IL-10 was less discriminatory with sensitivity and specificity of 71% and 68% (Table 4).

Association of viral antigen and RAGE with thromboinflammatory biomarkers
We then examined the association of the SARS-COV-2 nucleocapsid viral antigen and RAGE with each of the measured biomarkers. This reinforced the strong relationship between SARS-CoV-2 viral antigen with RAGE, and with poor outcome. Patients with severe disease or death clustered with both high RAGE and high viral antigen concentrations ( Figure 1). IL-10, IP-10, and IL-6 were correlated significantly with both SARS-CoV-2 nucleocapsid viral antigen and RAGE (Tables 5, 6). PAI-1 was significantly correlated with SARS-CoV-2 nucleocapsid viral antigen, but the trend was not statistically significant for RAGE (Tables 5, 6). RAGE, FIGURE 1 Association of SARS-CoV-2 viral antigen levels on emergency department presentation with RAGE and with subsequent clinical outcome at 7 days.  but not SARS-CoV-2 nucleocapsid viral antigen, was also correlated with Von Willebrand Factor, TNFR-1, IL-18, and P-Selectin (Table 6).

Discussion
In this study, we examined a panel of biomarkers focusing on RAGE and SARS-CoV-2 nucleocapsid viral antigen in patients presenting to the emergency department with COVID-19. Our findings are consistent with prior literature demonstrating elevations in inflammatory cytokines and chemokines such as IL-6, IL-8, IL-10, IP-10, TNFR-1 and TREM-1 in the development of severe compared to mild COVID-19 (6,(8)(9)(10)(29)(30)(31). We also found increases in some but not all markers of thrombosis in those with severe disease including elevated PAI-1 and vWF activity. The strongest associations with development of severe disease and death were observed for RAGE, SARS-CoV-2 viral antigen, IL-6, IL-10, and TNFR-1. Our study adds to the existing literature because the measurements were performed on emergency department samples prior to any therapeutics and therefore the findings may be of prognostic value as these early measures of RAGE and SARS-CoV-2 viral antigen remained independent risk factors of development of severe disease or death in a multivariable model. RAGE is an immunoglobulin prominently expressed by type I pneumocytes and is an established prognostic marker in patients with acute respiratory distress syndrome (ARDS) due to viral, bacterial and other causes (15-17, 32, 33). RAGE is also elevated in other states of lung injury such as emphysema, and in chronic diseases including diabetes and cardiovascular disease (32,34). RAGE is thought to play a pathogenic role in ARDS based on animal studies (35,36). Recent work identified that RAGE is also an important prognostic biomarker specifically in hospitalized patients with COVID-19 (3,12,13). However, few if any have reported on RAGE and ultimate disease severity when measured in patients early in their disease course. In our study, RAGE was the biomarker most strongly associated with development of severe disease or death at 7 days, suggesting this could be used as an important early prognostic biomarker in patients presenting with COVID-19 in the emergency department setting. Using the Youden index to determine an optimal cut-off value, a plasma RAGE concentration of greater than 3,200ng/ml had high sensitivity (87%) and specificity (93%) for later development of severe disease or death. This cutoff is similar to a threshold of 3,108ng/ ml reported by Lim et al. (13) to stratify the need for mechanical ventilation, and lower than a threshold of 6,800ng/ml reported by Wick et al. for predicting mortality in hospitalized patients with COVID-19 (12).
A large cohort study examining changes in a broad range biomarkers in patients with COVID-19 found markers of alveolar epithelial injury rise early in the disease course while markers of endothelial injury occur later (18). This may explain why the observed differenced in markers of endothelial injury between those who developed mild versus severe disease were not as strong in this study, given that we only measured them at a single timepoint early in the disease course. This further underscores why markers of alveolar epithelial injury such as RAGE may be the best candidates for stratifying ultimate disease severity of COVID-19 at early timepoints. For example, no significant differences were observed for Ang-1, Ang-2 or VEGF in our study, although these are known to be increased in hospitalized patients with severe compared to mild COVID-19 (37). Further studies are needed examining how the relationship between the SARS-CoV-2 nucleocapsid viral antigen, RAGE, and additional biomarkers varies over time, which was not performed as part of this investigation.
TREM-1, a mediator of inflammation and prognostic biomarker in sepsis, was elevated in patients who developed severe disease, though did not achieve statistical significance after adjustment for multiple comparisons (38, 39). Recent data in emergency department patients demonstrated TREM-1 had excellent predictive ability for patients developing severe disease (defined as intubation or mortality within 30 days), with an AUC of 0.86 (8). This relationship was not as strong in our study, but the observed trend was the same, and a larger cohort may be needed to better examine these trends.
Our study also demonstrates a strong association of SARS-CoV-2 nucleocapsid viral antigen with development of severe disease in COVID-19, consistent with several prior studies of hospitalized patients (14,19,20,40). Recent data from a large multicenter clinical trial consortium demonstrated that in patients hospitalized with COVID-19, higher SARS-CoV-2 nucleocapsid viral antigen measured within the first 72 hours was associated with worse clinical outcomes at 7 days (19). At a threshold of 1,000 pg/ml, the viral antigen was 77% sensitive and 59% specific for a worse outcome on the WHO ordinal scale at 7 days. Similarly, in an analysis of data from the Accelerating COVID-19 Therapeutic Interventions and Vaccines (ACTIV)-3 Therapeutics for Inpatients With COVID-19 (TICO) study group, a threshold of 1000ng/ml was associated with a 5 fold increased odds of worse respiratory status at 5 days and longer length of stay, but these levels were drawn 1-2 days after hospital admission (14). In our study, we identified a threshold of 4,500pg/ml to be very specific (82%) and sensitive (90%) for development of poor outcome at 7 days on the WHO ordinal scale. Although these findings should be validated in a larger cohort of patients, it is further notable that of the 11 patients who presented with a positive COVID-19 nasopharyngeal PCR test but had an undetectable viral SARS-COV-2 nucleocapsid viral antigen level, none developed severe disease or died at 7 days. A quantitative point of care test for the SARS-CoV-2 viral antigen was recently developed and shown to be accurate (41), broadening the potential clinical utility of the data presented in this study.
This study has several important limitations. First, our sample size is modest with 77 patients of whom 17 developed the primary outcome of severe disease or death at 7 days. Therefore, our findings of the potential prognostic value of RAGE and SARS-CoV-2 nucleocapsid viral antigen plasma levels need further validation in larger cohorts. This study was not a time to event analysis-the time to development of severe disease was not factored into our models, and a small proportion of patients (n=4) developed severe disease but recovered by day 7 thus were encompassed in the mild/ambulatory group. We also had limited statistical power to evaluate secondary outcomes such as venous thromboembolism due to the low absolute number of patients developing these outcomes. This analysis was exploratory and observational in nature-we tested a range of relevant thromboinflammatory biomarkers resulting in multiple hypothesis testing. To account for this, p-values were corrected using the conservative Sidak-Holm method, which is an appropriate but conservative method and may have resulted in type II errors (25). There was a higher proportion of patients with diabetes and older age in the severe disease group, which could be a source of bias in our results given RAGE may be elevated in chronic disease and advanced age (34,42). However, we adjusted for this possible confounding in our multivariable model using the APACHE-III score which includes age as part of the input, and on additional sensitivity analyses no changes to our results were observed when controlling directly for age and diabetes in our multivariable model. Our study population included some but not all patients presenting to the emergency department with symptomatic COVID-19. Some of the sickest patients who may be rapidly triaged to the intensive care unit would not have been captured by our recruitment process. The focus of our study was to evaluate potential prognostic markers where clinical equipoise exists, and therefore omission of patients clearly critically-ill was desirable for our study's purposes. Most patients in this study were enrolled prior to the widespread Omicron variant of SARS-CoV-2. Vaccination status of patients was also not included, but the majority of patients were enrolled prior to the implementation of the COVID-19 vaccine.
In conclusion, our study demonstrates the potential prognostic value of two emerging biomarkers predictive of development of severe disease in patients presenting to the emergency department for symptomatic COVID-19: RAGE and the SARS-COV-2 nucleocapsid viral antigen. These findings are highly clinically relevant given the observed relationship was identified on emergency department presentation early in patients' disease course and remained significant even after controlling for physiologic measures of disease severity. Additional studies validating these findings and incorporating point of care assays for RAGE and SARS-COV-2 nucleocapsid viral antigen are warranted to further characterize their prognostic value and determine optimal discriminatory thresholds for patient risk stratification.

Data availability statement
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

Ethics statement
The studies involving human participants were reviewed and approved by Committee on Human Research at the University of California, San Francisco (study number 20-30895). The patients/ participants provided their written informed consent to participate in this study. clinical data. CH, AK, MM assisted in study design and provided critical revisions of the manuscript. HL performed biomarker assays. LK performed study design, analyzed data, and helped prepared the manuscript. Members of the CO-ACIT study group enrolled patients in the study, collected blood samples and performed research assays, collected clinical data, and assisted with study design and execution.

Funding
This study was funded in part by the NIH though UCSF Clinical and Translational Sciences Institute COVID-19 rapid response grant UL1 TR001872. The views and conclusions contained in this document are those of the authors and should not be interpreted as representing the official policies, either expressed or implied, of the NIH.

Conflict of interest
LK consulted for Gamma Diagnostics and Cerus Corporation at the time of this study. The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Publisher's note All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.