Effects of potent neutralizing antibodies from convalescent plasma in patients hospitalized for severe SARS-CoV-2 infection

Abstract

In a randomized clinical trial of 86 hospitalized COVID-19 patients comparing standard care to treatment with 300mL convalescent plasma containing high titers of neutralizing SARS-CoV-2 antibodies, no overall clinical benefit was observed. Using a comprehensive translational approach, we unravel the virological and immunological responses following treatment to disentangle which COVID-19 patients may benefit and should be the focus of future studies. Convalescent plasma is safe, does not improve survival, has no effect on the disease course, nor does plasma enhance viral clearance in the respiratory tract, influence SARS-CoV-2 antibody development or serum proinflammatory cytokines levels. Here, we show that the vast majority of patients already had potent neutralizing SARS-CoV-2 antibodies at hospital admission and with comparable titers to carefully selected plasma donors. This resulted in the decision to terminate the trial prematurely. Treatment with convalescent plasma should be studied early in the disease course or at least preceding autologous humoral response development.

Fig. 1. Predicted probabilities of WHO COVID-19 disease severity score.

A Disease severity score after 15 days. B Disease severity score after 30 days. Predicted probabilities of belonging to each outcome category of the WHO COVID-19 8-point Disease Severity Scale over different values of age and per treatment group. The predicted probabilities are based on the proportional ordinal logistic regression model adjusted for age, sex, CRP, and whether a patient was admitted to the intensive care unit at enrollment. The values of these factors are set to the median value (if continuous) or to the most frequent value (if categorical). WHO COVID-19 8-point Disease Severity Scale: 0 = no clinical or virological evidence of infection; 1 = no limitation of activities; 2 = limitation of activities; 3 = hospitalized, no oxygen therapy; 4 = oxygen by mask or nasal prongs; 5 = non-invasive ventilation or high-flow oxygen; 6 = intubation and mechanical ventilation; 7 = ventilation + additional organ support = vasopressors, renal replacing therapy, extracorporeal membrane oxygenation (ECMO); 8 = death.

Fig. 2. Kaplan–Meier curves of probability of discharge.

Shaded area indicates 95% CI around the Kaplan–Meier estimate of discharge for the two treatment groups (standard of care: red dashed line; convalescent plasma: blue solid line) after enrollment (D = 1) and table with number of subjects at risk of discharge. Death is not accounted for as competing risk.

Fig. 3. Inflammatory markers in patients.

A CRP. B Ferritin. C Lymphocytes. CRP, ferritin and lymphocytes were measured* in the serum of COVID-19 patients (standard of care: red; convalescent plasma: blue) on day 1 of enrollment, between days 2–7 and days 8–14 after enrollment. Reported data are the highest value for CRP and ferritin, and the lowest value for lymphocytes. Box plots indicate median (middle line), 25th, 75th percentile (box), and 5th and 95th percentile (whiskers), as well as outliers (single points). *Only measured if it was part of routine care.

Fig. 4. Antibodies against receptor-binding domain and viral neutralization capacity in patients and donors.

A SARS-CoV-2 total Ig against SARS-CoV-2 receptor-binding domain (RBD) measured by Wantai ELISA. B SARS-CoV-2 IgM against SARS-CoV-2 RBD measured by Wantai ELISA. C Viral neutralization capacity measured as PRNT50 titer. SARS-CoV-2 total Ig and IgM against SARS-CoV-2 RBD and viral neutralization capacity were evaluated in the serum of COVID-19 patients (gray) at enrollment (day 1) and serum of donors at day of plasma donation (all 115 donors tested: green; donors of whom plasma was selected for use in the study: blue). Box plots indicate median (middle line), 25th, 75th percentile (box), and 5th and 95th percentile (whiskers), as well as outliers (single points). Dashed line indicates positive cutoff at 1.0 optical density (OD) ratio for both total Ig and IgM.

Fig. 5. SARS-CoV-2 antibodies in patients, donors, and healthy controls.

A SARS-CoV-2 total Ig against SARS-CoV-2 receptor-binding domain (RBD) measured by Wantai ELISA. B SARS-CoV-2 IgM against SARS-CoV-2 RBD measured by Wantai ELISA. C Nucleocapsid IgG antibodies. D Nucleocapsid IgM antibodies. An independent Mann–Whitney U-test was used to evaluate SARS-CoV-2 total Ig and IgM against SARS-CoV-2 RBD measured by Wantai ELISA in the serum of COVID-19 patients (standard of care: red; convalescent plasma: blue) at enrollment (day 1). Nucleocapsid IgM and IgG antibodies were measured in the serum of COVID-19 patients (gray) at enrollment (day 1) in serum of donors (green) at week 6 post infection and in serum of healthy uninfected controls (brown). Box plots indicate median (middle line), 25th, 75th percentile (box), and 5th and 95th percentile (whiskers), as well as outliers (single points). Dashed line indicates positive cutoff at 1.0 OD ratio for both total Ig and IgM Wantai ELISA, and 11 units for both IgM and IgG nucleocapsid antibodies.

Fig. 6. Relative frequencies of PRNT50 titers in patients and selected donors.

PRNT50 titers were measured in the serum of COVID-19 patients (gray) on day 1 of enrollment and in selected donors (blue). Height of the bars indicates the relative frequency of PRNT50 titer in that group.

Fig. 7. IFNγ, TNFα, and IL-6 in patients and healthy controls.

A Cytokine concentrations in serum. B Fold change. Cytokines IL-6, IFNγ, and TNFα were measured in the serum of nine convalescent plasma and ten standard of care patients (standard of care: red; convalescent plasma: blue) at enrollment (day 1) and on day 7 and day 14 after enrollment, and in the serum of healthy controls (brown). Dots represent median and vertical lines represent IQR.

Fig. 8. SARS-CoV-2 (predicted) viral load in patients.

A SARS-CoV-2 viral load in patients. B Predicted SARS-CoV-2 viral load in patients. C Predicted SARS-CoV-2 viral load in selected patients. SARS-CoV-2 viral load (copies/mL) measured by RT-PCR from nasopharyngeal swabs of COVID-19 patients (standard of care: red; convalescent plasma: blue) at enrollment (day 1) and day 3, day 7, day 10, and day 14 after enrollment. Box plots indicate median (middle line), 25th, 75th percentile (box), and 5th and 95th percentile (whiskers), as well as outliers (single points). Predicted evolution mean (solid line) and 95% CI (shaded area) of SARS-CoV-2 log(Viral Load) in log(copies/ml) per day since enrollment (D = 1) for COVID-19 patients (standard of care: red; convalescent plasma: blue). Predicted evolution mean (solid line) and 95% CI (shaded area) of absolute log(Viral Load) in log(copies/ml) per day since enrollment (D = 1) for COVID-19 patients (standard of care: red; convalescent plasma: blue) excluding subjects who had SARS-CoV-2 viral load equal to zero at day 1 of enrollment.