False Positive Results in SARS-CoV-2
Serological Tests for Samples
From Patients With Chronic
Inflammatory Diseases
Nastya Kharlamova
1,2
†
, Nicky Dunn
1,2
†
, Sahl K. Bedri
1,2
, Svante Jerling
1,2
,
Malin Almgren
1,2
, Francesca Faustini
3
, Iva Gunnarsson
3
, Johan Rönnelid
4
,
Rille Pullerits
5,6
, Inger Gjertsson
5
, Karin Lundberg
2,3
, Anna Månberg
7
, Elisa Pin
7
,
Peter Nilsson
7
, Sophia Hober
8
, Katharina Fink
1,9,10
and Anna Fogdell-Hahn
1,2
*
1
Department of Clinical Neuroscience, Karolinska Institutet, Stockholm, Sweden,
2
Center for Molecular Medicine, Karolinska
Institutet, Stockholm, Sweden,
3
Department of Medicine Solna, Division of Rheumatology, Karolinska Institutet and
Rheumatology, Karolinska University Hospital, Stockholm, Sweden,
4
Department of Immunology, Genetics and Pathology,
Uppsala University, Uppsala, Sweden,
5
Department of Rheumatology and Inflammation Research, Institution of Medicine,
Sahlgrenska Academy at University of Gothenburg, Gothenburg, Sweden,
6
Department of Clinical Immunology and
Transfusion Medicine, Sahlgrenska University Hospital, Gothenburg, Sweden,
7
Division of Affinity Proteomics, Department of
Protein Science, KTH Royal Institute of Technology, SciLifeLab, Stockholm, Sweden,
8
Department of Protein Science, KTH
Royal Institute of Technology, Stockholm, Sweden,
9
Department of Neurology, Karolinska University Hospital,
Stockholm, Sweden,
10
Centrum for Neurology, Academical Specialist Centrum, Stockholm, Sweden
Patients with chronic inflammatory diseases are often treated with immunosuppressants
and therefore are of particular concern during the SARS-CoV-2 pandemic. Serological
tests will improve our understanding of the infection and immunity in this population,
unless they tests give false positive results. The aim of this study was to evaluate the
specificity of SARS-Cov-2 serological assays using samples from patients with chronic
inflammatory diseases collected prior to April 2019, thus defined as negative. Samples
from patients with multiple sclerosis (MS, n=10), rheumatoid arthritis (RA, n=47) with or
without rheumatoid factor (RF) and/or anti-cyclic citrullinated peptide antibodies (anti-
CCP2) and systemic lupus erythematosus (SLE, n=10) with or without RF, were analyzed
for SARS-CoV-2 antibodies using 17 commercially available lateral flow assays (LFA), two
ELISA kits and one in-house developed IgG multiplex bead-based assay. Six LFA and the
in-house validated IgG assay correctly produced negative results for all samples.
However, the majority of assays (n=13), gave false positive signal for samples from
patients with RA and SLE. This was most notable in samples from RF positive RA patients.
No false positive samples were detected in any assay using samples from patients with
MS. Poor specificity of commercial serological assays could possibly be, at least partly,
due to interfering antibodies in samples from patients with chronic inflammatory diseases.
For these patients, the risk of false positivity should be considered when interpreting
results of the SARS-CoV-2 serological assays.
Keywords: SARS-CoV-2, autoimmunity, autoantibod ies, diagnost ics, rheu matoid arthritis, systemic lupus
erythematosus, multiple sclerosis, rheumatoid factor
Frontiers in Immunology | www.frontiersin.org May 2021 | Volume 12 | Article 6661141
Edited by:
Dimitrios Petrou Bogdanos,
University of Thessaly, Greece
Reviewed by:
Ana Barrera-Vargas,
Instituto Nacional de Ciencias Me
dicas
y Nutricio
n Salvador Zubira
n
(INCMNSZ), Mexico
Tao Li,
National Center of Biomedical Analysis
(NCBA), China
*Correspondence:
Anna Fogdell-Hahn
Anna.Fogdell-Hahn@ki.se
†
These authors have contributed
equally to this work and
share first authorship
Specialty section:
This article was submitted to
Autoimmune and
Autoinflammatory Disorders,
a section of the journal
Frontiers in Immunology
Received: 09 February 2021
Accepted: 06 April 2021
Published: 03 May 2021
Citation:
Kharlamova N, Dunn N, Bedri SK,
Jerling S, Almgren M, Faustini F,
Gunnarsson I, Rönnelid J, Pullerits R,
Gjertsson I, Lundberg K, Månberg A,
Pin E, Nilsson P, Hober S, Fink K and
Fogdell-Hahn A (2021) False Positive
Results in SARS-CoV-2 Serological
Tests for Samples From Patients With
Chronic In flammatory Diseases.
Front. Immunol. 12:666114.
doi: 10.3389/fimmu.2021.666114
ORIGINAL RESEARCH
published: 03 May 2021
doi: 10.3389/fimmu.2021.666114
INTRODUCTION
Severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2)
is the causative agent of the coronavirus disease 2019 (COVID-
19), which emerged as a pandemic late 2019 (1). The cumulative
number of infected and fatal cases can be followed at the Johns
Hopkins University COVID-19 Dashboard (2). Patients with
chronic inflammatory disease are often treated with
immunomodulatory treatments and therefore potentially more
susceptible to infec tions (3) . As a result, there has been
substantial concern during the pandemic as to the potential
increased risk COVID-19 disease severity and mortality among
these patient groups (4). There is limited evidence about their
risk of severe COVID-19, or knowledge of how their disease or
immunomodulatory treatment may affect either their pre-
existing immunity or ability to develop protective immunity
following inf ection (5, 6). Approximately 6% of th e world’s
population are affected by chronic inflammatory diseases
which i ncludes conditions such as multiple sclerosis (MS),
rheumatoid arthritis (RA) and systemic lupus erythematosus
(SLE) (7). These are generally progressive diseases and although
for the majority there are no cures, treatment is centered around
slowing disease progression with immunomodulatory
treatments. The hallmarks of autoimmune diseases are
inflammation, loss of self-tolerance and the presenc e of
autoantibodies. MS is a chronic inflammatory disorder
restricted to the central nervous system, characterized by
demyelination, axonal loss and the formation of sclerotic
plaques. The worldwide prevalence is estimated to be 2.2
million cases, but with large geographical variation (8). RA is a
heterogeneous chronic inflammatory disease, which affected
close to 5 million people globally by 2010 and with prevalence
increasing due to the increased aging of the human population
(9). The disease is characterized by synovial inflammation and
the formation of the pannus, which causes cartilage and bone
destruction, joint dysfunction, pain and disability. Rheumatoid
factor (RF) and anti-citrullinated protein antibodies (ACPA),
often detected as a nti-cyclic citrullinated peptide (CCP)
antibodies, are the most frequent and the most studied RA-
related autoantibodies. RF is an antibody reactive with the Fc
portion of IgG, mainly consisting of IgM in Caucasian RA
populations, but also IgG and IgA RF are present. Although
RF is detected in approximately 70% of RA patients, the presence
of RF is not specific for RA. These autoantibodies are also present
in a variety of other diseases as well as in the general population
and may increase with age, smoking and chronic infection ( 10,
11). SLE is a systemic inflammatory disease of the connective
tissue, characterized by a loss of self-tolerance and leading to
production and deposition of a large panel of autoantibodies and
immune complexes formation (12). Clinical manifestation of
SLE is heterogeneous and can affect multiple organs.
Approximately 25% of SLE patients have RF (13), but these
patients can also have anti-nuclear antibodies (ANA) and anti-
double-stranded (ds) DNA antibodies.
Serological tests are useful for determining past infection and
present immunity. The presence of IgM antibodies indicates a
recent infection, whereas presence of IgG antibodies indicates
possible long-lasting immunity (14). Important information can
be achieved by having access to reliable serological methods
during a pandemic; to identify seropositive people for
convalescent plasma donations; guide policies and ease
restrictions on human mobility based on sero-epidemiological
evidence; ensure immunity to allow key workers to return to
work after exposure; and evaluate vaccine development studies
and vaccine strategies.
Due to the substantial global demand, SARS-CoV-2
serological testing has been rapidly developed and released to
the market. The assays are validated before release and also often
independently verified before being approved (15, 16). However,
the panel of samples used to d etermine specificity is often
focused on r uling out cross-reactivity with othe r viral
infections an d might not include serum from patien ts with
chronic inflammatory diseases, even though it is recommended
(
16). Based on experience from development and validation of
serology assays for measuring anti-drug antibodies (ADA) in
persons with chronic inflammatory disease, it is recommended
to show specificity against drug naïve patient serum, as
antibodies present in patients with autoimmune diseases are
known to interact with reagents in serological assays and give
unspecific signal (17, 18). Given the significant role serological
tests may have as useful wide-spread screening tools for
immunity, it is important to verify the specificity of SARS-
CoV-2serologicaltestsinasimilarwayforspecificity i n
patient groups with autoimmune diseases, using samples that
were collected before the pandemic.
The aim of this pilot study was to verify the specificity in a
number of the commercially available SARS-CoV-2 serological
tests, using a panel of samples from patients with different
chronic inflammatory diseases collected before the SARS-CoV-
2 outbreak as negative controls, to get an indication of the extent
of the issue for further developments, validatio ns and
verifications of serology assays.
MATERIALS AND METHODS
Patient Serum Samples
To evaluate specificity of SARS-C oV-2 s erolo gical assays in
patients wit h chron ic inflammatory diseases, a selection of
negative control samples was retrieved from the biobank
(n=68). To exclude individuals with risk of previous exposure
to SARS-CoV-2 infection, only samples collected before April
2019, (1-22 years since time of c ollection, Table 1), were
included in the study. Serum samples were se lected from
patients with MS (n=10), RA (n=47), of which 2 samples were
from the same patient), or SLE (n=10) (Table 1).
TheMSpatientswerediagnosedaccordingtothe2017
updated McDonald criteria (19). The RA diagnoses were
determined according to the 1987 revised American College of
Rheumatology disease classification criteria by rheumatologists,
within 12 months after the first symptoms of joint disease (20).
For the SLE patients, the diagnoses were determined according to
the American College of Rheumatism (21) and/or The Systemic
Lupus Erythematosus Collaborating Clinics (22).
Kharlamova et al. False Positive SARS-CoV-2 Serological Tests
Frontiers in Immunology | www.frontiersin.org May 2021 | Volume 12 | Article 6661142
MS patient samples were collected in a research laboratory
providing routine testing for anti-drug antibodies (ADAs) at the
Centre for Molecular Medicine, Karolinska Institutet in
Stockholm and had been treated with interferon beta (IFNb).
Three MS samples were ADA positive. Of the RA samples, 40
were from the Swedish population-based case control study
Epidemiological Investigation of RA (EIRA) and had not been
treated with any disease modifying anti-rheumatic drug
(DMARD) (23). Of these patients, 20 were RF and anti-CCP2
positive (50%); six were RF negative but anti-CCP2 positive
(15%), and 14 were both RF and anti-CCP2 negative (35%) (24).
The additional seven RA patient samples were retrieved from a
prospective study cohort (Sahlgrenska University Hospital,
Gothenburg) and were infliximab (IFX) treated. Of these seven
patient samples, three were RF and anti-CCP2 positive; two were
RF negative but anti-CCP2 positive; one was RF positive but
anti-CCP2 negative, and one sample was both RF and anti-CCP2
negative. T he SLE samples were obtained from a study
investigating the development of ADA against rituximab
(RTX), and therefore all patients were RTX treated. Five of ten
samples was anti-rituximab ADA positive (Table S1). All of the
SLE patients had ANA and four of them anti-dsDNA.
This retrospective cohort study was approved by the Ethics
Review Authority in Stockholm and Gothenburg (2020–23/04,
dnr 2020-01649, 2012/1550-31/3, dnr 96-174). Samples and data
were collected with informed consent in compliance with the
Helsinki Declaration.
Assays
Rheumatoid Factor Detection Methods
Analysis IgA, IgG and IgM isotypes of RA samles from the EIRA
cohort and SLE samples was performed using the EliA
immunoassay on the Phadia 2500 instrument and the cutoff
values as stated in the manufacturer’s instructions (Phadia GmbH,
Uppsala, Sweden) (24). Serum samples of RA patients treated with
IFX were analyzed for IgM RF using laser nephelometry technique.
Anti-Cyclic Citrullinated Peptide Assay
Anti-CCP2 IgG in EIRA was previously determined using the
Immunoscan CCPlus
®
ELISA (Euro-Diagnostica , Malmö,
Sweden), in accordance with the manufacturer´s instructions.
SARS-CoV-2 Serological Detection Methods
A total of 19 commercially available serological assays were
evaluated in this study and compared to an in-house assay.
Two Enzyme-Linked Immunosorbent Assays (ELISA) and 17
rapid diagnostic lateral flow assays (LFA) were included. These
tests were assigned a letter from A – S(Table 2) and referred to
as such in text and figures in this study. The brand name, antigen,
manufacturer determined specificity and sensitivity, are outlined
in Table 2. All tests were performed according to manufacturer
instructions and using serum.
Commercial Lateral Flow SARS-CoV-2 Assays
LFAs are designed to enable point of care analyses and can
generate immediate results with read-outs as bands in small
cartridges. These rapid lateral flow tests are developed for whole
blood, serum and plasma. At time of testing, the appropriate
volume of serum was applied to the designated well and then the
buffer was added. After the recommended incubation period, the
presence and intensity of the bands were investigated and graded
from negative to four levels of positivity by the same operator.
An In-House Validated SARS-CoV-2
Serological Assay
The results were compared to an in-house multiplex bead-based
and validated SARS-CoV-2 serological assay developed at
SciLifeLab and KTH Royal Institute of Technology as previously
described (27). In brief, IgG r eactivity was analyzed in a
high-throughput and multiplex bead-based format utilizing
384-well plates and Fle xMap3D in strumentations (Luminex
Corp) for read-out (27). Reactivity against three different
in-house produced viral protein variants was used to differentiate
between positive and negative samples: Spike trimers comprising
the prefusion-stabilized spike glycoprotein ectodomain (28)
(expressed in HEK and purified using a C-terminal Strep II tag),
Spike S1 subunit (expressed in CHO and purified with HPC4 tag),
and the Nucleocapsid protein (expressed in E. coli and purified
using an N-terminal His-tag). The antigens were immobilized on
magnetic color coded beads (MagPlex, Luminex Corp) and
plasma/serum IgG that bound to the antigens were detected by
an R-phycoerythrin conjugated goat anti-hIgG (Invitrogen,
H10104). Reactivity against at least two out of the three viral
antigens included in the panel was required for positive read out.
The cut-offfor seropositivity was defined as signals above the mean
+6 SD of the 12 negative controls included in each assay. The
method utilizing the combination of the three antigens has been
found to have 99.2% sensitivity (99.6%, 99.2%, 96.7%, respectively,
for the three antigens individually) and 99.8% specificity (98.9%,
99.1%, 98.4%, respectively, for the three antigens individually)
based on 243 positive controls (defined as >16 days after onset or
TABLE 1 | Patients’ characteristics.
Rheumatoid
Arthritis
Multiple
Sclerosis
Systemic
Lupus
Erythematosus
Patients (n) 47 10 10
Age (years median,
min - max)
53 (18-71) 46 (39-70) 35.5 (30-60)
Female (n, %) 33 (70) 7 (70) 9 (90)
IgM RF positive (n, %) 24 (51) n/a 0
Anti-CCP2 positive
(n, %)
31 (66) n/a n/a
Treated with IFX
(n, %)
7 (15) 0 0
Treated with RTX
(n, %)
0 0 10 (100)
Treated with IFN
beta-1a (n, %)
0 10 (100) 0
ADA positive (n, %) n/a 3 (30) 5 (50)
Time period of
sampling
1998 - 2006* 2003 - March
2019
2003 - 2018
n, number; RF, rheumatoid factor; anti-CCP, anti-cyclic citrullinated p eptide; IFX,
infliximab; RTX, rituximab; IFN interferon; ADA, anti-drug antibodies; n/a, not applicable;
RA, rheumatoid arthritis.
*Time period of sampling for RA patients treated with IFX: 2018 - March 2019.
Kharlamova et al. False Positive SARS-CoV-2 Serological Tests
Frontiers in Immunology | www.frontiersin.org May 2021 | Volume 12 | Article 6661143
positive PCR) and 442 negative controls (defined as collected 2019
and earlier) samples.
Commercially Available SARS-CoV-2 ELISA Kits
The two included ELISAs were pe rformed ac cording to the
manufacturers’ instructions. The first ELISA used to detect IgG
against SARS-CoV-2 (test A, Table 2) was the recomWell SARS-
CoV-2 IgG Elisa kit (Mikrogen Diagnostik GmbH, Germany). This
assay is an indirect ELISA which uses highly purified recombinant
nucleocapsid protein from SARS-CoV-2 as an antigen. The
manufacturer had determined the pote ntial interference of
antibodies against other pathogens that might induce clinical
symptoms similar to those of a SARS-CoV-2 infection (including
for example seasonal coronaviruses, influe nza A virus, RSV,
Mycoplasma pneumoniae, Chlamydia pneumoniae). In addition,
they also tested specificity using samples from people with
conditions that present with atypical immune system activity
including EBV infection, pregnancy, ANA and RF-positive
subjects. The cut-off for positivity was calculated according to the
manufacturer’s instructions.
The second ELISA test was the EDI Novel Coronavirus COVID-
19 IgG Elisa Kit (Epitope Diagnostics, Inc., San Diego, USA) to
detect IgG (test B, Table 2). This is an in vitro diagnostic and CE
marked indirect ELISA with plates coated with peptides from the
SARS-CoV-2 nucleocapsid antigen. Specificity of this assay was
determined by the manufacturer using anti-influenza A, anti-
influen za B, Hepatitis C virus (HCV), ANA and respiratory
syncytial virus (RSV). The cut-off for positivity was determined
according to the manufacturer’s instructions. The manufacturer
states that a positive result may be due to past or present infection
TABLE 2 | Description of the SARS-CoV-2 serological assays and the test codes used in this study.
Test
Code
Manufacturer Kit Name Antigen/Target Catalogue
number
Company reported
assay specifi city
A Mikrogen Diagnostik GmbH,
Germany*
recomWell SARS-CoV-2 IgG ELISA kit Nucleocapsid protein 7304 IgG: 98.7%
B Epitope Diagnostics, Inc., San
Diego, USA
EDI Novel Coronavirus COVID-19 IgG ELISA Kit Nucleocapsid protein KT-1032 IgG: 100%
C Jiangsu Medomics medical
technology Co., Ltd, China
Rapid IgM-IgG combined Antibody Test Kit for SARS-CoV-
2 (ICA)
Spike protein
(RBD MK201027)
201030 Not specified
D Salafa Oy, Salo, Finland Salacor (Biohit) SARS-CoV-2 IgG/IgM rapid test kit Nucleocapsid protein COV-01-S IgM: 99.2%
IgG: 99.9%
E Salafa Oy, Salo, Finland Sienna SARS-CoV-2 IgG/IgM rapid test kit Spike protein (RBD) 102222 IgM: 100%
IgG: 98.8%
F Liming Bio-Products Co., Ltd.
Jiangsu, China
StrongStep_SARS-CoV-2 IgM/IgG_REF502090_ Antibody
Rapid Test
Nucleocapsid and
Spike protein
502090 IgM: 100%
IgG: 98,7%
G Zhejiang Orient Gene Biotech
Co., Ltd. (China)
COVID-19 IgG/IgM Rapid Test Cassette alt.
HEALGEN_ COVID-19 IgG/IgM Rapid Test Cassette (Whole
Blood/Serum/Plasma_REF GCCOV-402a
Nucleocapsid and
Spike protein (25)^
GCCOV-
402a
IgM: 98.46%
IgG: 98.46%
H InTec Products inc., Haicang
Xiamen, China
INTEC_ Colloidal Gold (whole blood/Serum/Plasma) Rapid
SARS-CoV-2 Antibody (IgM/IgG)
Nucleocapsid protein
(25)
ITP16001-
TC25
Combined IgM+IgG:
98%
I Sugentech Inc., South Korea* SGTi-flex COVID-19 IgM/IgG Nucleocapsid and
Spike protein^
COVT025E IgM: 98.3% (90% FDA
August 2020)
IgG:100%
J Xiamen Biotime Biotechnology
Co., Ltd. China
SARS-CoV2 IgG/IgM Rapid Qualitative Test Spike protein^ BT1301 Not specified
K Zhuhai Livzon Diagnostics Inc.
(China)
COVID-19 IgG/IgM Lateral flow Rapid Test Cassette Nucleocapsid protein Not
specified
IgM: 99.7%
IgG: 99.4%
L Abbott Point of Care Inc. USA* Panbio COVID-19 lgG/lgM Rapid Test Device Nucleocapsid protein
(26)
ICO -T402 IgM: 92.8%
IgG: 92.8%
M SureScreen Diagnostics Ltd, UK SureScreen Diagnostics COVID-19 IgG/IgM Rapid Test
Cassette (Whole blood/serum/Plasma)
Spike protein/RBD COVID19C IgM: 99.2%
IgG: 99.2%
N Wuhan Easy Diagnosis
Biomedicine Co., Ltd (China)*
COVID-19 (SARS-CoV-2) IgM/IgG Antibody Test kit Not specified SA-2-D IgM: 100%
IgG: 100%
O Zhuhai Encode Medical
Engineering Co., Ltd., China*
SARS-CoV-2 IgG/IgM Rapid test S1-RBD and
nucleocapsid protein
RCD-422 IgM: 100%
IgG: 100%
P Jiangsu SuperbioBiomedical Co.,
Ltd, China
SARS-CoV-2 (COVID-19) IgM/IgG Antibody Fast Detection
Kit (Colloidal Gold)
Spike and
nucleocapsid protein^
B00502 IgG: 95.8%
IgM: 95.8%
Q Lumigenex (Suzhou) Co., Ltd.
China
Lumigenex SARS-CoV-2 IgG/IgM Antibody Rapid Test Kit Spike and
nucleocapsid protein
Not
specified
Not specified
R Wondfo, Guangzhou, China Wondfo Biotech SARS-CoV-2 Antibody Test Not specified W195 Combined: IgM
+IgG:99.57%
S Innovita (Tangshan) Biological
Technology Co Ltd China*
2019-nCoV Ab Test (Colloidal Gold) Spike and
Nucleocapsid
protein^
Not
specified
IgM: 100%
IgG: 100%
^Emergency Use Authorization (EUA) Serology Test Performance | Food and Drug Administration (FDA).
*Stated in the instructions to have tested interference with RA and/or RF or other autoantibodies.
As of January 2021 tests C, G, I, K, N, O, P are not FDA approved.
For updated status of assays visit https://www.finddx.org/covid-19/pipeline.
Kharlamova et al. False Positive SARS-CoV-2 Serological Tests
Frontiers in Immunology | www.frontiersin.org May 2021 | Volume 12 | Article 6661144
with SARS-CoV-2 but not due to other coronavirus strains, such as
coronavirus HKU1, NL63, OC43, or 229E.
Statistical Analyses
Rate of false positive signals were determined as the number of
positive samples divided by the total number of samples tested
for each assay. Statistical analyses and figures were generated
using GraphPad Prism (version 8.2.1). The statistical difference
between RF positive and RF negative RA subsets were calculated
with Fishers exact test. The other groups were too small to make
any meaningful statistical evaluations and thus these results are
only presented as descriptive analyses.
RESULTS
Commercial LFA and ELISA Assays
Serum samples from 47 RA patients (with two samples from one
of the patients), 10 SLE and 10 MS patients were evaluated using
19 SARS-CoV-2 commercial serological assays and compared to
an in-house developed multiplex bead-based assay (27).
The overall results of all 68 samples are illustrated in Figure
1. A total of six commercial LFAs (test G, H, J, K, R and S)
reached 100% specificity for both IgG and IgM including all
chronic inflammatory disease cohorts’ patients (n=67). Notably,
all samples from MS patients (n=10) were negative for both IgM
and IgG in all 20 assays.
For the 17 LFAs evaluated for specificity using 25 RA samples
(from 24 patients of which 20 were treatment naïve and 4 were
treated with infliximab) that were positive for RF, 10 assays had
unspecific signal detected for at least one immunoglobulin
isotype (Figures 2 and 3). Five assays had unspecific signal for
both IgM and IgG in a few up to a majority of the RA samples
(test C: IgM 19/20, IgG 8/20; test D: IgM 19/20, IgG 2/20; test M:
IgM 4/20, IgG 3/20; test N: IgM 6/20, IgG 1/20; and test P: IgM 1/
20, IgG 1/20). Unspecific IgM signal, without unspecific IgG
signal, was detected in four LFAs (test E: 5/20; test F: 16/20; test
O: 20/20; and test Q: 19/20). In one LFA, only the IgG test gave
unspecific signal (test L: 1/20). In contrast, only five assays
detected unspecific signal in RA samples that were RF negative
(n= 23), with five detecting IgM and one detecting IgG (test D:
IgM 1/23; test F: IgM 1/23; test M: IgM 2/23, IgG 2/23; for test N:
FIGURE 1 | Overview of false positive results in all samples for 19 different serological tests. Six LFA tests (G, H, J, K, R & S) gave no false positive results. The false
positivity rate of the remaining tests ranged between 2 - 45%. The test code keys are described in Table 2. The two ELISA assays (test A and B) were only tested
for IgG.
Kharlamova et al. False Positive SARS-CoV-2 Serological Tests
Frontiers in Immunology | www.frontiersin.org May 2021 | Volume 12 | Article 6661145