Author:
Wick Katherine D.,Leligdowicz Aleksandra,Willmore Andrew,Carrillo Sidney A.,Ghale Rajani,Jauregui Alejandra,Chak Suzanna S.,Nguyen Viet,Lee Deanna,Jones Chayse,Dewar Robin,Lane H. Clifford,Kangelaris Kirsten N.,Hendrickson Carolyn M.,Liu Kathleen D.,Sinha Pratik,Erle David J.,Langelier Charles R.,Krummell Matthew F.,Woodruff Prescott G.,Calfee Carolyn S.,Matthay Michael A.,Abe-Jones Yumiko,Beagle Alexander,Bhide Sharvari,Fragiadakis Gabriela K.,Gonzalez Ana,Jamdar Omid,Jones Norman,Lea Tasha,Leroux Carolyn,Milush Jeff,Pierce Logan,Prasad Priya,Rashid Sadeed,Rodriguez Nicklaus,Sigman Austin,Ward Alyssa,Wilson Michael,
Abstract
Abstract
Background
Studies quantifying SARS-CoV-2 have focused on upper respiratory tract or plasma viral RNA with inconsistent association with clinical outcomes. The association between plasma viral antigen levels and clinical outcomes has not been previously studied. Our aim was to investigate the relationship between plasma SARS-CoV-2 nucleocapsid antigen (N-antigen) concentration and both markers of host response and clinical outcomes.
Methods
SARS-CoV-2 N-antigen concentrations were measured in the first study plasma sample (D0), collected within 72 h of hospital admission, from 256 subjects admitted between March 2020 and August 2021 in a prospective observational cohort of hospitalized patients with COVID-19. The rank correlations between plasma N-antigen and plasma biomarkers of tissue damage, coagulation, and inflammation were assessed. Multiple ordinal regression was used to test the association between enrollment N-antigen plasma concentration and the primary outcome of clinical deterioration at one week as measured by a modified World Health Organization (WHO) ordinal scale. Multiple logistic regression was used to test the association between enrollment plasma N-antigen concentration and the secondary outcomes of ICU admission, mechanical ventilation at 28 days, and death at 28 days. The prognostic discrimination of an externally derived “high antigen” cutoff of N-antigen ≥ 1000 pg/mL was also tested.
Results
N-antigen on D0 was detectable in 84% of study participants. Plasma N-antigen levels significantly correlated with RAGE (r = 0.61), IL-10 (r = 0.59), and IP-10 (r = 0.59, adjusted p = 0.01 for all correlations). For the primary outcome of clinical status at one week, each 500 pg/mL increase in plasma N-antigen level was associated with an adjusted OR of 1.05 (95% CI 1.03–1.08) for worse WHO ordinal status. D0 plasma N-antigen ≥ 1000 pg/mL was 77% sensitive and 59% specific (AUROC 0.68) with a positive predictive value of 23% and a negative predictive value of 93% for a worse WHO ordinal scale at day 7 compared to baseline. D0 N-antigen concentration was independently associated with ICU admission and 28-day mechanical ventilation, but not with death at 28 days.
Conclusions
Plasma N-antigen levels are readily measured and provide important insight into the pathogenesis and prognosis of COVID-19. The measurement of N-antigen levels early in-hospital course may improve risk stratification, especially for identifying patients who are unlikely to progress to severe disease.
Funder
National Institutes of Health
Genentech
Publisher
Springer Science and Business Media LLC
Subject
Critical Care and Intensive Care Medicine