key: cord-0804441-v3nxmwib
authors: Bryan, Andrew; Fink, Susan L; Gattuso, Meghan A; Pepper, Gregory; Chaudhary, Anu; Wener, Mark H; Morishima, Chihiro; Jerome, Keith R; Mathias, Patrick C; Greninger, Alexander L
title: SARS-CoV-2 viral load on admission is associated with 30-day mortality
date: 2020-11-03
journal: Open Forum Infect Dis
DOI: 10.1093/ofid/ofaa535
sha: 76c23ec4852c777ffbcee5df2e7f416bd654ae56
doc_id: 804441
cord_uid: v3nxmwib

SARS-CoV-2 viral load on admission was associated with a significantly increased 30-day mortality (OR 4.20 [95% CI: 1.62-10.86]), and anti-SARS-CoV-2 nucleocapisid IgG seropositivity on admission trended toward a reduced 30-day mortality (OR 0.43 [95% CI: 0.15-1.26]). Reporting of quantitative SARS-CoV-2 viral load and serologic assays may offer prognostic clinical information.

SARS-CoV-2 is a novel coronavirus associated with high morbidity and mortality that has rapidly spread across the world. Despite increasing evidence that viral load is associated with clinical outcomes [1] [2] [3] [4] [5] [6] and that high viral loads are associated with fewer tissue-culture infective viral particles [7, 8] , no FDA Emergency Use Authorization (EUA) has been issued for a quantitative assay.

Similarly, after initial concerns surrounding test characteristics, accurate serological testing for SARS-CoV-2 is increasingly becoming available in the United States. For instance, we have previously shown the Abbott Architect anti-SARS-CoV-2 nucleocapsid IgG assay to be both highly sensitive and specific for detecting prior SARS-CoV-2 infection [9] . Multiple commercial anti-SARS-CoV-2 assays are available across different antibody isotypes and with the ability to detect different antigens. The antibody responses to the different antigens tend to track together and are correlated with neutralization, if not entirely completely [10] [11] [12] [13] . Data is beginning to emerge that the active development of neutralizing antibodies is protective against SARS-CoV-2 infection [14] . Yet, most serology assays are authorized only for the reporting of a qualitative result, even though they return semi-quantitative or quantitative results. Data correlating viral load and antibody results to meaningful virologic and clinical outcomes is continuing to emerge.

Here, we examined clinical and virologic features associated with seropositivity and seroconversion to SARS-CoV-2 in a cohort of hospitalized patients in Seattle, Washington. We specifically sought to assess whether detection of anti-SARS-CoV-2 nucleocapsid IgG was associated with a better prognosis, including lower viral load and reduced 30-day all-cause mortality.

A c c e p t e d M a n u s c r i p t

The study was approved under a consent waiver by the University of Washington IRB.

Patients with positive SARS-CoV-2 RT-PCR results from nasopharyngeal swabs were identified at UW Medicine hospitals and excess serum and plasma samples were retrieved for SARS-CoV-2 antibody testing. Samples were enriched for patients that had RT-PCR results available on the same calendar date as a remnant serum or plasma sample. A total of 245 patients were identified with at least one residual serum/plasma sample and at least one clinical note available for chart review to determine days from symptom onset ( Table 1) Anti-SARS-CoV-2 nucleocapsid IgG was determined by the Abbott Architect as previously described [9] . The manufacturer's suggested cut-off of 1.40 was used for seropositivity. SARS-CoV-2 qRT-PCR was performed using Hologic Panther Fusion, DiaSorin Simplexa, Roche Cobas 6800 platforms, or a CDC-based laboratory developed test (LDT) [15] . Cycle threshold (Ct) values were available from the Hologic Panther Fusion and LDT assays and were treated interchangeably given A c c e p t e d M a n u s c r i p t their close correlation [15] . A Ct of 22 is equivalent to approximately 2,500,000 copies/mL viral transport media in these assays [16] .

The association between SARS-CoV-2 IgG index value and Ct was assessed using a linear mixed effects model with significance determined by restricted maximum likelihood ratio using R packages lme4 and lmerTest [17, 18] . To account for singularity, the model incorporated scaling and a weak Bayesian prior via the R package blme [19] . Multivariate logistic regression to determine association of Ct value and mortality was performed using the base R function. Visualization was performed using ggplot2 [20] .

A total of 181 patients had both an Abbott Architect anti-SARS-CoV-2 nucleocapsid IgG index value and a SARS-CoV-2 PCR Ct value available from the same calendar day. Several patients had quantitative PCR and serology data available from multiple days, resulting in a total of 224 total unique patient-days. Comparison of qRT-PCR and serology data revealed only one SARS-CoV-2 seropositive individual with a simultaneous SARS-CoV-2 Ct < 22 (Figures 1A-B) . IgG levels as measured by index values were found to be inversely correlated with SARS-CoV-2 viral load (p < 0.001). To substantiate this association at the individual patient level, we identified patients with more than three measures of SARS-CoV-2 IgG and viral load. Although the kinetics of SARS-CoV-2 IgG and viral load varied between individual patients, these parameters consistently trended together in individual patients (Figures 1C-H) . Lymphocyte counts increased and inflammatory markers decreased over time in these patients ( Figure S1 ), concomitant with a decreasing viral load ( Figure   1C-H) . Figure 1J ).

Our data provide further support for quantitative viral load assessment, especially on hospital admission. Our results agree with other work that has shown viral load to be associated with disease severity [1, 3, 5, 6, 21] . Currently, the FDA has only authorized reporting of qualitative results from SARS-CoV-2 qRT-PCR tests, despite nearly all tests returning some estimate of viral load.

Indeed, our clinical laboratory has reported semi-quantitative results for respiratory viruses for more than a decade and our clinicians are well versed in interpreting semi-quantitative Ct values for medical management. Quantitative reporting of SARS-CoV-2 molecular or serologic assays would require significant modifications of existing emergency use authorizations. Our data further suggest that a cycle threshold of 22 may serve as a useful discrete cut-off for significant viral replication that is associated with mortality. We note, however, that sample and swab variability across patient populations may limit the widespread use of a discrete cutoff for quantitative RT-PCR results even if using standard curves to compute copy number/genome equivalents and improve the correlation between assays.

A c c e p t e d M a n u s c r i p t Given our data that indicated a viral load at admission is a significant independent predictor of 30-day mortality, we sought to assess the anti-nucleocapsid response as a potential biomarker.

We demonstrated that detection of anti-SARS-CoV-2 nucleocapsid IgG is associated with lower viral loads in COVID-19 patients. This antibody response also tracked closely with the amount of viral nucleic acid in individual patients over time. Due to the close relationships of both IgG and viral load with days since symptom onset, we could not conclude that the viral load dependence on IgG response was independent from the passage of time, but days since symptom onset was accounted for in our mixed effects model. Individuals who were SARS-CoV-2 antibody positive on admission were less than half as likely to die within 30 days, though this relationship was not statistically significant. The inability to reach statistical significance is likely due to the limited study population size (power = 0.56 for an OR of 0.5). Higher antibody levels have been associated with more significant clinical disease and hospitalization [22, 23] . Nonetheless, our data suggests admission antibody titers, coupled with molecular testing, may be particularly helpful to assess the disease course for patients that cannot provide a clinical history. This association might only be present at the time of admission, which is when we assessed its prognostic role. While our data do not directly assess the potential for ongoing immunity against future infections of SARS-CoV-2, they indicate that high viral loads almost never coexist with SARS-CoV-2 seropositivity and suggest that persons with anti-SARS-CoV-2 antibodies on admission have a reduced 30-day all-cause mortality.

The main limitation of our study was the retrospective nature in a population enriched for hospitalized patients with acute disease [9, 24] . The retrospective nature precluded analyses of viral clearance and length of stay due to significant confounding factors associated with RT-PCR testing frequency during admission and patient discharge placement. Although we had insufficient sample size to perform separate analyses with patients who only presented to the emergency department or outpatient clinic, results appeared similar to the full data set ( Figure S2 ). Our serological test detects IgG antibodies against the nucleocapsid protein of SARS-CoV-2. Whether these antibodies can provide protection themselves or associated with protective responses is clearly unknown.

A c c e p t e d M a n u s c r i p t Variability in neutralizing responses between patients not elucidated by our assay may explain some of the variation in our data set. However, neutralizing antibody assays are in vitro methods that may or may not be associated with clinically meaningful outcomes unless performed in concert with challenge studies. In addition, non-neutralizing antibodies may also confer protection against infection in some viral infections [25] .

Our work illustrates the importance of quantitative virologic and serological testing for SARS- A c c e p t e d M a n u s c r i p t M a n u s c r i p t A c c e p t e d M a n u s c r i p t Figure -1 

Viral load dynamics and disease severity in patients infected with SARS-CoV-2 in Zhejiang province, China

Kinetics of viral clearance and antibody production across age groups in SARS-CoV-2 infected children

Impact of SARS-CoV-2 Viral Load on Risk of Intubation and Mortality Among Hospitalized Patients with Coronavirus Disease

A Narrative Systematic Review of the Clinical Utility of Cycle Threshold Values in the Context of COVID-19

SARS-CoV-2 viral load predicts COVID-19 mortality

SARS-CoV-2 RNA in plasma is associated with ICU admission and mortality in patients hospitalized with COVID-19

Duration of infectiousness and correlation with RT-PCR cycle threshold values in cases of COVID-19

Predicting infectious SARS-CoV-2 from diagnostic samples

Performance Characteristics of the Abbott Architect SARS-CoV-2 IgG Assay and Seroprevalence in

SARS-CoV-2 seroprevalence and neutralizing activity in donor and patient blood from the San Francisco Bay Area

Comparative performance of five commercially available serologic assays to detect antibodies to SARS-CoV-2 and identify individuals with high neutralizing titers

Temporal profiles of viral load in posterior oropharyngeal saliva samples and serum antibody responses during infection by SARS-CoV-2: an observational cohort study

Severe Acute Respiratory Syndrome Coronavirus 2-Specific Antibody Responses in Coronavirus Disease

Neutralizing antibodies correlate with protection from SARS-CoV-2 in humans during a fishery vessel outbreak with high attack rate

Comparison of Commercially Available and Laboratory Developed Assays for in vitro Detection of SARS-CoV-2 in Clinical Laboratories

Validation of SARS-CoV-2 detection across multiple specimen types

Fitting Linear Mixed-Effects Models Using lme4

lmerTest Package: Tests in Linear Mixed Effects Models

A nondegenerate penalized likelihood estimator for variance parameters in multilevel models

ggplot2: Elegant Graphics for Data Analysis

Kinetics of viral load and antibody response in relation to COVID-19 severity

SARS-CoV-2 neutralizing antibody responses are more robust in patients with severe disease

Antibody responses to SARS-CoV-2 in patients with COVID

Prolonged Persistence of PCR-detectable Virus During an Outbreak of SARS-CoV-2 in an Inpatient Geriatric Psychiatry Unit

Non-neutralizing Antibodies from a Marburg Infection Survivor Mediate Protection by Fc-Effector Functions and by Enhancing Efficacy of Other Antibodies

The authors would like to thank Nathan Breit of the University of Washington Department of Laboratory Medicine for assistance in obtaining data and Thomas E. Grys of the Mayo Clinic in Arizona for critical review of the manuscript. We also thank the University of Washington Medical Center (UWMC) Northwest Campus clinical laboratory staff and the UWMC Clinical Immunology staff or reserving remnant serum and plasma samples from COVID-19 PCR positive patients.

This work was supported by the Department of Laboratory Medicine at the University of Washington Medical Center.

ALG reports personal fees from Abbott Molecular, outside of the submitted work.AB,SLF,MAG,GP,AC,MHW,CM,KRJ,PCM report no conflicts of interest.

A c c e p t e d M a n u s c r i p t A c c e p t e d M a n u s c r i p t