Predicting COVID-19 Infection and Its Severity with T-Cell-Mediated Immunity in Immunosuppressed and Non-Immunosuppressed Individuals

Abstract

Background: Cell-mediated immunity to SARS-CoV-2 may define infection-risk in healthy and chronically immunosuppressed transplant patients. Methods: Blood samples from 64 unexposed healthy subjects including 21 with solid organ transplants, and 66 COVID-19 patients, including 26 with transplants were tested. Frequencies of SARS-CoV-2-reactive CD3+T-cells CD4+T-helper, CD8+T-cytotoxic and CD19+B-cells, which express CD154 were measured after stimulation with peptide mixtures representing the spike protein S antigen, its conserved C-terminal S2, and less conserved N-terminal S1 components. Findings: COVID-19 patients were sampled at median 11 days after diagnosis. S-reactive T- and B-cells were present in all samples and were significantly fewer in COVID-19 compared with healthy subjects. In logistic regression analysis using training and test sets, S-reactive CD3 and CD8 cells, age, race, and transplantation status distinguished COVID-19 patients from healthy subjects (test set AUC 0.9 [95%CI 0.83-0.97]). Among 66 COVID-19 patients S-reactive CD8 cells and age predicted respiratory failure (AUC 0.73 [95%CI 0.62-0.86]). S2-reactive T-cells also predicted COVID-19 infection (test set AUC 0.89 [95%CI 0.80-0.97]) and respiratory failure (AUC 0.73 [95%CI 0.60-0.86]). S1 antigen elicited minimal responses. COVID-19 subjects developed anti-spike-protein IgG antibody. Transplanted COVID-19 patients had significantly lower incidence of IgG to the receptor-binding-domain of the S1 antigen. Interpretations: In immunosuppressed and non-immunosuppressed individuals, SARS-CoV-2-spike-antigen-specific cellular immunity is directed toward conserved sequences. This immunity is decreased early after COVID-19 infection, more so with respiratory failure, and identifies individuals at-risk for COVID-19. Humoral immunity to less conserved viral sequences is impaired in chronically immunosuppressed patients early after infection. Funding: Intramural support from all participating institutions, NSF#2033307, NIH Grant number UL1 TR001450 (MUSC). Declaration of Interests: University of Pittsburgh Patent 9606019, author: RS, describes CMI testing for CMV, is licensed exclusively to Plexision, in which University and RS own equity. RS and CA developed Plexision’s patent-pending multi-variate CMI assay for SARS2. RS is Professor of Surgery at the University of Pittsburgh and Chief Scientific Officer of Plexision by permission of COI committee at the University. CA is a paid consultant to Plexision. All other authors declare no conflict of interest. Ethics Approval Statement: Human Subjects: COVID-19 patients were enrolled under IRB-approved protocols 2017-0365, Pro00101915, and 1551551 respectively, at the three centers in Washington, DC, Charleston, SC, and Edinburg, TX. De-identified residual cryopreserved PBL samples were tested under IRB-exempt protocol, and samples from healthy-NT subjects were tested under IRB approved protocol 6774 in the reference laboratory at Plexision.