1
Reduced neutralization of SARS-CoV-2 variants by convalescent plasma and hyperimmune 1
intravenous immunoglobulins for treatment of COVID-19 2
3
Juanjie Tang
#
, Youri Lee
#
, Supriya Ravichandran
#
, Gabrielle Grubbs
#
, Chang Huang, 4
Charles Stauft, Tony Wang, Basil Golding, Hana Golding, and Surender Khurana
*
5
6
Division of Viral Products, Center for Biologics Evaluation and Research (CBER), FDA, Silver 7
Spring, Maryland, 20993, USA. 8
# These authors contributed equally to this manuscript. 9
10
*
Corresponding author: 11
*
Surender Khurana, Ph.D. 12
Division of Viral Products, Center for Biologics Evaluation and Research (CBER) 13
Food and Drug Administrationa (FDA) 14
10903 New Hampshire Avenue 15
Silver Spring, MD, 20993, USA 16
Phone- 240-402-9632, Fax- (301) 595-1125 17
E. mail- Surender.Khurana@fda.hhs.gov 18
19
20
105 and is also made available for use under a CC0 license.
(which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC
The copyright holder for this preprintthis version posted March 19, 2021. ; https://doi.org/10.1101/2021.03.19.436183doi: bioRxiv preprint
2
ABSTRACT 21
Hyperimmune immunoglobulin (hCoV-2IG) preparations generated from SARS-CoV-2 22
convalescent plasma (CP) are under evaluation in several clinical trials of hospitalized COVID-19 23
patients. Here we explored the antibody epitope repertoire, antibody binding and virus neutralizing 24
capacity of six hCoV-2IG batches as well as nine convalescent plasma (CP) lots against SARS-25
CoV-2 and emerging variants of concern (VOC). The Gene-Fragment Phage display library 26
spanning the SARS-CoV-2 spike demonstrated broad recognition of multiple antigenic sites 27
spanning the entire spike including NTD, RBD, S1/S2 cleavage site, S2-fusion peptide and S2-28
heptad repeat regions. Antibody binding to the immunodominant epitopes was higher for hCoV-29
2IG than CP, with predominant binding to the fusion peptide. In the pseudovirus neutralization 30
assay (PsVNA) and in the wild-type SARS-CoV-2 PRNT assay, hCoV-2IG lots showed higher 31
titers against the WA-1 strain compared with CP. Neutralization of SARS-CoV-2 VOCs from 32
around the globe were reduced to different levels by hCoV-2IG lots. The most significant loss of 33
neutralizing activity was seen against the B.1.351 (9-fold) followed by P.1 (3.5-fold), with minimal 34
loss of activity against the B.1.17 and B.1.429 (<2-fold). Again, the CP showed more pronounced 35
loss of cross-neutralization against the VOCs compared with hCoV-2IG. Significant reduction of 36
hCoV-2IG binding was observed to the RBD-E484K followed by RBD-N501Y and minimal loss 37
of binding to RBD-K417N compared with unmutated RBD. This study suggests that post-exposure 38
treatment with hCoV-2IG is preferable to CP. In countries with co-circulating SARS-CoV-2 39
variants, identifying the infecting virus strain could inform optimal treatments, but would likely 40
require administration of higher volumes or repeated infusions of hCOV-2IG or CP, in patients 41
infected with the emerging SARS-CoV-2 variants. 42
43
105 and is also made available for use under a CC0 license.
(which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC
The copyright holder for this preprintthis version posted March 19, 2021. ; https://doi.org/10.1101/2021.03.19.436183doi: bioRxiv preprint
3
INTRODUCTION 44
An expedited access to treatment of COVID-19 patients with convalescent plasma was 45
issued by FDA via Emergency Use Authorization on August 23, 2020. Additional studies, 46
including randomized, controlled trials, have provided data to further inform the safety and 47
efficacy of COVID-19 convalescent plasma. Based on assessment of these data, potential clinical 48
benefit of transfusion of COVID-19 convalescent plasma in hospitalized patients with COVID-19 49
is associated with high neutralizing titer units administered early in the course of disease(1, 2). 50
Intravenous immunoglobulins (IVIG) are a more concentrated form of IgG preparations 51
fractionated from large number of plasma units that are prescreened for the presence of high titer 52
anti-spike antibodies and predetermined SARS-CoV-2 neutralization titers. Several hCoV-2IG lots 53
are currently being evaluated in clinical trials. The effectiveness of hCoV-2 IG products may be 54
hampered by evolving SARS-CoV-2 and the emergence of new variants with high transmissibility 55
rates and mutations in the Receptor Binding Domain (RBD) which are less susceptible to 56
antibodies from recovered COVID-19 patients. The main variants of concern (VOC) are the 57
B.1.1.7 spreading from the UK, the B1.351 spreading in South Africa (SA), and the P.1 that 58
appeared in northeast Brazil and found in Japan (JP). In the US, several variants were identified 59
recently including California (CA) variant B.1.429 (3-6). 60
The phage display technique is suitable to display properly folded and conformationally 61
active proteins, as it has been widely used for display of large functionally-active antibodies, 62
enzymes, hormones, and viral and mammalian proteins. We have adapted this Genome Fragment 63
Phage Display Library (GFPDL) technology for unbiased, comprehensive approach for multiple 64
viral pathogens including SARS-CoV-2, Ebola virus, highly pathogenic avian influenza virus, 65
105 and is also made available for use under a CC0 license.
(which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC
The copyright holder for this preprintthis version posted March 19, 2021. ; https://doi.org/10.1101/2021.03.19.436183doi: bioRxiv preprint
4
respiratory syncytial virus and Zika virus, to define both the linear and conformational antibody 66
epitope repertoire of post-vaccination/infection samples (7-11) 67
In the current study we probed the antibody epitope repertoires of 6 hCoV-2IG products 68
using SARS-CoV-2-spike GFPDL. Surface Plasmon Resonance (SPR) was used to measure 69
antibody binding to SARS-CoV-2 antigenic site peptides identified in the GFPDL analyses and to 70
spike protein receptor binding domain (RBD) representing WA-1 as well RBD mutants engineered 71
to express key amino acid mutations of the VOCs. Neutralization capacity of the hCoV-2IG lots 72
against the SARS-CoV-2 WA-1 strain and several VOC (CA, UK, JP, SA) was measured in 73
pseudovirion neutralization assay (PsVNA) as well as classical PRNT assay. For comparison with 74
hCoV-2IG, we evaluated nine convalescent plasma from recovered COVID-19 patients and 16 75
IVIG preparations that were manufactured with pre-pandemic plasma units prior to August 2019. 76
77
RESULTS 78
SARS-CoV-2 spike antibody epitope repertoires of six hCoV-2IG batches 79
The spike protein is the antigen of choice for development of vaccines and therapeutics 80
against SARS-CoV-2. To decipher the epitope-specificity of the SARS-CoV-2 spike-specific 81
antibodies in an unbiased manner, we subjected the six hCoV-2IG lots to antibody epitope 82
profiling with a highly diverse SARS-CoV-2 spike GFPDL with >10
7.1
unique phage clones 83
displaying epitopes of 18-500 amino acid residues across the SARS-CoV-2 spike. During GFPDL 84
characterization, GFPDL based epitope mapping of monoclonal antibodies (MAbs) targeting 85
SARS-CoV-2 spike or RBD identified the expected linear or conformation-dependent epitopes 86
recognized by these MAbs. Recently, we showed that SARS-CoV-2 spike GFPDL can recognize 87
105 and is also made available for use under a CC0 license.
(which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC
The copyright holder for this preprintthis version posted March 19, 2021. ; https://doi.org/10.1101/2021.03.19.436183doi: bioRxiv preprint
5
both linear, conformational and neutralizing epitopes in the post-vaccination sera of rabbits (11) 88
and post-SARS-CoV-2 infection sera in the adults and elderly (10, 12). 89
Six hCoV-2IG lots were used for SARS-CoV-2 GFPDL based epitope mapping as 90
previously described (7-11, 13). Similar numbers of phages were bound by IgG of these hCoV-91
2IG batches (3.4 x 10
4
– 2.1 x 10
5
) (Fig. 1A). The bound phages demonstrated a diverse epitope 92
repertoire spanning the entire SARS-CoV-2 spike protein including N-terminal domain (NTD) and 93
RBD in S1, and the fusion peptide (FP), β-rich connector domain (CD), heptad repeat 1 (HR1) and 94
2 (HR2) in S2 (Fig. 1B and Fig. S1). Peptides representing the key immunodominant antigenic 95
sites identified by GFPDL analysis were chemically synthesized and used to evaluate binding of 96
each of the six hCoV-2IG batches, 16 pre-pandemic 2019-IVIG lots and 9 COVID-19 97
convalescent plasma (Fig. 1C). As expected 2019-IVIG demonstrated minimal to no binding to 98
the SARS-CoV-2 spike peptides. In aggregate, the convalescent plasma showed lower binding to 99
epitopes spanning the entire spike in comparison with the hCoV-2IG (Fig. 1C). In agreement with 100
GFPDL analysis, the hCoV-2IG demonstrated highest antibody binding to the spike peptide 790-101
834 that contains the fusion peptide sequence (residues 788-806), which is unchanged among the 102
major VOCs ((Fig. S2 and Table S1). Most of the GFPDL-identified antigenic site sequences 103
recognized by hCoV-2IGs are conserved in the spike protein of various SARS-CoV-2 VOC (Table 104
S1). 105
106
Neutralization capacity of CP and hCoV-2IG against the SARS-CoV-2 WA-1 and B.1.429, 107
B1.1.7, P.1, B.1.351 VOCs 108
105 and is also made available for use under a CC0 license.
(which was not certified by peer review) is the author/funder. This article is a US Government work. It is not subject to copyright under 17 USC
The copyright holder for this preprintthis version posted March 19, 2021. ; https://doi.org/10.1101/2021.03.19.436183doi: bioRxiv preprint