Abstract
Background/aim: To investigate the frequency and clinical
relevance of an extended autoantibody (ab) profile in patients with SSc.
Materials and Methods: In this cross-sectional study, serum
from 100 consecutive patients was subjected to indirect
immunofluorescence (HEp-20-10/primate liver mosaic) and Systemic
Sclerosis Profile by EUROIMMUN (Lübeck, Germany) to evaluate ANA and
autoantibodies against 13 different autoantibodies in patients with SSc
less than three years.
Results: 93 of 100 patients were positive for ANA by indirect
immunofluorescence (IIF). Fifty-three patients showed single positivity
(26 anti-Scl70, 16 ACA, six anti-RNAPIII, one anti-Ku ab, one
anti-PM/Scl100, two anti-PM/Scl75, one anti-Ro52), whereas 32 patients
had multiple auto-antibody positivities. Among common SSc-specific
autoantibodies, Scl70 and RNAPIII showed the highest co-occurrence
(n=4). One patient was simultaneously positive for anti-RNAPIII ab and
ACA, and one was positive for ACA and Scl70. The clinical features were
not statistically different between single and multiple
autoantibody-positivity for common SSc-specific autoantibodies (ACA,
Scl70, and RNAP III), except for digital ulcer in the multi-antibody
positive ACA group (p=0.019).
Conclusion: Based on our results, co-expression of
auto-antibodies is not uncommon in SSc patients. Although autoantibodies
specific to SSc in early disease show generally known clinical features,
it remains to be investigated how co-expression of specific and
nonspecific autoantibody positivity will affect clinical presentation.
Keywords: Systemic sclerosis, Autoantibody,
Scleroderma-specific antibodies, Immunoblot assay, Indirect
immunofluorescence assay
Introduction
Systemic sclerosis (SSc) is a rare autoimmune disease characterized by
progressive skin and internal organs fibrosis, vasculopathy, and
autoantibody production [1]. Anti-nuclear antibodies (ANA) can be
found in 90-95% of patients with SSc and SSc-specific autoantibodies in
>80% [2]. Today, at least 10 SSc-specific
autoantibodies, anticentromere, anti-Scl-70 (anti-topoisomerase I),
anti-RNA polymerase III (anti-RNAPIII), anti-U3 ribonucleoprotein
(anti-RNP), antiTh/To, anti-U11/U12 RNP, anti-PM/Scl, anti-Ku,
antiRuvBL1/2, anti-U1 RNP antibodies (ab) have been reported in SSc
patients [3, 4, 5].
The clinical course of SSc is not easily predictable because of the
clinical and serologic heterogeneity. Nevertheless, from early to
established SSc, autoantibodies are used as an indicator in diagnosis,
predicting organ involvement, determining prognosis, and making
treatment decisions [6, 7, 8]. Although recent classification
criteria have confirmed their diagnostic utility and have long been used
for prognostic stratification of patients, there is still a need to
recognize the potential interaction between autoantibodies and their
representation on clinical phenotypes of disease [9-11].
The most known and widely used autoantibodies targeting Scl-70,
centromere proteins and RNAP-III have been reported to be mutually
exclusive and strongly associated with certain clinical phenotypes
[12]. However, there is also evidence of overlap between these
auto-antibodies [3, 13]. Although the relationship between single
positivity of some specific auto-antibodies and involvement of certain
organs is well known today, the relationship between the positivity of
compound auto-antibodies and their clinical significance in SSc patients
with short disease duration has not been investigated in detail
[14].
Therefore, we aimed to investigate the frequency and clinical relevance
of autoantibodies in the early stage of SSc using an expanded panel of
autoantibodies. We also aimed to follow these patients to investigate
how the relationship between these autoantibodies and organ involvement
progressed over time. Here we report our initial baseline data.
Methods
Patient Selection
One hundred patients with SSc were recruited consecutively from six
tertiary centers in Turkey specializing in the care of patients with
SSc. We included limited or diffuse cutaneous SSc patients [10] in
the early stage of the disease with a disease duration of <3
years from the first non-Raynaud symptom [15], meeting the 2013
American College of Rheumatology/European League Against Rheumatism
(ACR/ EULAR) SSc classification criteria [16]. We included the SSc
patients with overlap syndromes and the overlap syndromes were defined
as cases meeting the classification criteria for one or more connective
tissue diseases (CTDs) concurrent with SSc. Patients with other
comorbidities that could lead to auto-antibody positivity were excluded.
Demographic characteristics and clinical and laboratory findings of the
patients were recorded. Disease duration was calculated as the time
between the onset of the first non-Raynaud symptom and the enrollment
date. Variables included Raynaud’s phenomenon, skin and musculoskeletal
involvement, pulmonary arterial hypertension (PAH), interstitial lung
disease (ILD), renal crisis (ever), gastrointestinal symptoms
(dysphagia, reflux, early satiety, constipation, diarrhea) and if
recorded any malignancy. The extent of the skin involvement was assessed
using the modified Rodnan skin score (mRSS) [17]. Patients with a
pulmonary artery pressure (PAP) above 45 mmHg on echocardiography
underwent right heart catheterization because of the strong correlation
between this estimated cut-off level and right heart catheterization
[18]. Pulmonary hypertension was defined as a mean pulmonary
arterial pressure of ≥20 mm Hg, and precapillary pulmonary hypertension
was defined as pulmonary vascular resistance of ≥2 Wood units and a
pulmonary capillary wedge pressure of ≤15 mm Hg on right-sided heart
catheterization [19]. ILD was defined as the presence of any
evidence of pulmonary fibrosis on lung imaging by high-resolution
computed tomography scan. The renal crisis was described as an abrupt
onset of severe hypertension (systolic blood pressure (BP) ≥180 mmHg
and/or diastolic BP ≥100 mmHg) without an alternate etiology, with or
without microangiopathic anemia or decline in renal function [20].
This study complied with the Declaration of Helsinki, and the study was
approved by the Kocaeli University School of Medicine Ethics Committee,
Kocaeli, Turkey, with study number KOU/GOKAEK 2017/347.
Autoantibody analysis
Sera from all patients were tested using a commercially available
indirect immunofluorescence (Anti-nuclear antibody [ANA], Mosaic
Hep-20-10/Liver; Euroimmun) assay and line immunoblot assay (Systemic
Sclerosis [Nucleoli] Profile EuroLine [IgG]; Euroimmun)
simultaneously in a single central laboratory. Serum aliquots were
stored at -80 C until the time of testing. The assays were performed
according to the manufacturer’s instructions. For ANA, results above the
dilution of 1:100 were considered positive. The Systemic Sclerosis
[Nucleoli] Profile kit contained 13 recombinant antigens: those
expressed in Escherichia coli (RNA polymerase III [RNAP III; subunits
RP11 and RP155], fibrillarin, the 90-kd nucleolar protein NOR-90, and
Th/To) or insect cells using the baculovirus system (CENP-A, CENP-B,
PM/Scl-100, PM/ Scl-75, Ku, and tripartite motif-containing protein 21
[TRIM- 21]/Ro 52) plus PDGFR expressed in mammalian cells and native
topo I (Scl-70) isolated from calf and rabbit thymus. Sera were analyzed
at a dilution of 1:101, and autoantibodies were detected using alkaline
phosphatase–labeled anti-human IgG. The EuroLine flatbed scanner was
used to provide semi-quantitative results. Readings obtained with a
signal intensity of 0-5, 6-10, 11-25, 26-50, and >50 were
defined as negative, borderline, medium, strong, and very strong bands
and were given equivalent scores of negative, (1+), 1+, 2+, and 3+,
respectively.
Statistical analysis
When determining the sample size, the following criteria were applied to
see how clinical characteristics differed between patients with and
without positive autoantibodies (two independent groups): effect
size=0.5, α=0.05, and power (1-β)=0.80. The total sample size required
to meet these criteria was 128, and 100 patients were recruited.
Descriptive statistics for clinical and demographic characteristics of
the patients are presented as frequency and percentage (%) for
categorical variables and mean with standard deviation (mean ± SD) or
median with interquartile range (median [Q3–Q1]) according to the
distribution of the continuous variables. The distribution normality was
assessed visually and through the Shapiro–Wilk test. An independent
sample T-test was used to analyze how specific autoantibodies affected
clinical outcomes in positive and negative groups and in cases of single
and multiple positivity. A chi-squared test was also performed for
categorical variables.
In evaluating semi-quantitative results, we considered a score of ≥ +1
for each autoantibody to be positive. In addition, in statistical
analyses, we assumed CENPA and/or CENPB positivity as ACA positive and,
similarly RNAP11 and/or RNA 155 positivity as RNAP III positive.
To reduce the dimensionality and increase the interpretability of the
data while minimizing the information loss, we examined the data by
principal components analysis (PCA) as performed in the R Library
FactoMineR [21]. PCA is a method that successively maximizes
variance by creating new unrelated variables. We used the intensity
scores in PCA.
Statistical analyses of further demographic and phenotypic data were
performed using SPSS, version 20.0 (IBM Inc., Chicago, IL, USA).
Two-sided p values less than 0.05 were considered statistically
significant (p<0.05).
Results
Characteristics of the study population and frequency
of autoantibodies
Demographic, clinical, and serologic characteristics of the 100 SSc
patients are presented in Table 1. The majority of the patients had
lcSSc (63%), whereas 36 had diffuse involvement, and only one patient
had sine scleroderma. Ninety-three out of 100 patients were positive for
ANA by IIF. Anti-Scl-70 ab was the most frequent (41%) auto-antibody,
followed by ACA with a rate of 27%. The anti - RNAPIII ab positivity
was 15%. Except for common autoantibody (ACA, Scl-70, or RNAP III),
anti - Ro52 was the most common SSc-specific autoantibody (22%). None
of the patients were positive for either anti-Fibrillarin or Anti-PDGF
abs.
The majority of the patients exhibited single or multiple positivities
for analyzed autoantibodies. The distribution of the autoantibody
positivities of the patients is shown in Figure 1. Of 100 patients, 53
were single positivity for any autoantibodies and 48 of which were
common SSc-specific autoantibodies (26 with anti-Scl-70 ab, 16 with ACA,
6 with anti-RNAPIII ab), and 5 were uncommon autoantibodies (1 with
anti-Ku ab, 1 with ant -PM/Scl100 ab, 2 with anti-PM/Scl75 ab and 1 with
anti-Ro52 ab). There were 32 patients with multiple antibody positivity.
Amongst the common SSc-specific autoantibodies, anti-Scl70 and
anti-RNAPIII antibodies showed the highest co-occurrence and were
simultaneously positive in 4 patients. One patient was positive for
anti-RNAPIII ab and ACA, and another for ACA and anti-Scl70 ab
simultaneously. All patients with anti-Ro52 ab except one were also
positive for the other antibodies. Eight patients (8%) were
positive for uncommon
scleroderma-specific autoantibodies with single or multiple staining. No
specific autoantibodies were detected in 15 patients (15%), and ANA was
positive in 11 of them.
Associations of autoantibodies with clinical features
The association between the common SSc-specific autoantibodies (ACA,
Scl70, and RNAP III) and clinical features was evaluated by comparing
the antibody-positive patients (regardless of being single or multiple
positive) with the rest of the study population (Table 2). Overlap
syndrome was more common in patients with ACA, and only 1 (3.7%)
patient with ACA had ILD. Anti-RNAPIII ab was associated with a common
disease subtype, ILD, and the highest mRSS among the three groups. Among
all SSc patients, two out of 3 patients with malignancy were anti-RNAP
antibody positive.
When we compared the clinical features of the patients in terms of
single and multiple antibody positivity for each of the common
autoantibodies specific to SSc (ACA, Scl70, and RNAP III) the clinical
features were not different between the subgroups (Table 3). However,
the digital ulcers were more frequent in the multiple antibody-positive
ACA group compared to single positives.
Clinical features of the patients who were positive only for uncommon
SSc-specific autoantibodies are shown in Table 4. All the patients
except one with anti-Ku (dcSSc) and another with PM/Scl100 (sine
scleroderma) had limited lcSSc. Only one patient with Ro52 had overlap
syndrome (Sjogren’s syndrome). None of these patients had either ILD,
PAH, or malignancy.
When we compared the Ro52 positive and negative patients, we found that
DUs and ACA positivity were more common in Ro52 positive patients
compared to negative ones (27.3% vs. 9%, p=0.035, and 45.5% vs.
21.8%, p=0.027, respectively). Ro52-positive patients also showed more
NOR90 positivity simultaneously (9.1% vs. 0, p=0.047). There was no
difference between Ro52 positive and negative patients regarding ILD,
PAH, GIS involvement, disease duration, or other auto-antibody
positivities.
Principle component analysis
Based on the principal component analysis (PCA), 63% of the variability
in the data was explained by the first three principal components. The
staining intensity scores of the autoantibodies were used to select the
main determinants of each component. Th/To, PM/Scl75 and PM/Scl100 were
determinants of the first dimension, ACA (CENP A and B) second, and
RNAP-III (RP11 and RP155) third dimension (Figure 2).
Discussion
In this study, we investigated an extended autoantibody profile and its
association with the clinical manifestations in a group of patients with
early-stage SSc. Consistent with the general knowledge, well-known
common SSc-specific autoantibodies (Scl-70, ACA, RNAPIII) were more
frequent among all tested auto-antibodies and exhibited the expected
clinical features. However, our results revealed that a substantial
proportion of patients were positive for more than one auto-antibody,
including common SSc-specific auto-antibodies known as mutually
exclusive. Another result that should be considered is that in patients
negative for common SSc-specific autoantibodies (8%), an extended test
profile showed the presence of another autoantibody.
The prevalence of anti-Scl70, ACA, and anti-RNAPIII abs in our patients
was 41%, 27%, and 15%, respectively. These results were close and
consistent with those previously reported in the literature
[22–25]. However, the frequency of autoantibodies may vary by
ethnicity, and data on SSc-specific autoantibodies from Turkey are
limited to the frequencies of anti-Scl70 and anti-centromere antibodies
[26]. In addition, in only one study, the frequency of RNAPIII
antibodies was reported as 2.2%, which was lower than our results
[27]. The difference in results may be because the previous study
was conducted in patients with extensive SSc and long disease duration
or because the two studies’ antibody analysis methods were different.
Our results revealed that a substantial proportion of patients were
positive for more than one auto-antibody, including common SSc-specific
auto-antibodies. Although common SSc-specific autoantibodies (ACA, topo
I, and RNAP III) are thought to be mutually exclusive and do not change
from one to another during the disease, there is evidence that they may
occur together [3, 23, 28]. With the recent advent of multiplexed
immunoassays, the notion that these autoantibodies are mutually
exclusive is slowly disappearing [29]. Consistent with these views,
we detected the co-expression of common SSc-specific autoantibodies in
some patients: anti-Scl70 ab and anti-RNAPIII Ab positivity in 4, ACA
and anti-RNAP III ab in 1 and anti-Scl70 ab and ACA in 1[30-31].
Regardless of being single or multiple positive, comparison of common
SSc-specific antibody-positive patients with the rest of the study
population showed expected clinical associations with these abs (ACA
with IcSSc, anti-Scl ab with ILD, and anti-RNAPIII ab with dcSSc and
malignancy, etc.). When we compared the single and multiple positivities
for each of common SSc-specific abs in terms of clinical involvements,
there were no significant differences between subgroups, except for the
higher occurrence of DUs in patients who were also positive to ACA.
There are uncertainties about the clinical features of multiple
antibody-positive SSc patients in the studies reported to date.
Unexpectedly, in 7 dcSSc patients with anti-RNAPIII ab and ACA, reported
by Satoh et al., none had significant organ involvement, such as renal
crisis during the disease course [32]. Similarly, in our study, none
of the patients with multiple ab positivity (1 with anti-Scl70 ab and
ACA, and 2 with anti-RNAPIII ab and ACA positivity) had severe internal
organ involvement; interestingly, all had lcSSc. It was difficult to
comment on the clinical significance of these multiple autoantibody
positivities because the number of patients with uncommon SSc-specific
antibodies needed to be higher to make detailed comparisons.
Ro52 was positive in 22% of the patients in our study. It is mainly
expressed in Sjögren’s syndrome and reported in varying frequencies in
SSc. Results from two large cohorts of the German network for systemic
Scleroderma and the Canadian Scleroderma Research Group demonstrated
that anti-Ro52 was the second most common autoantibody in patients with
SSc [33, 12]. It was our study’s third most common auto-antibody
following anti-Scl 70 and ACA antibodies. Anti-Ro52-positive SSc
patients were more likely to be older, to have ILD, and to have overlap
syndrome compared with anti-Ro52-negative patients [34]. Unlike
these results, we found anti-Ro52 was more prevalent in ACA-ab-positive
patients and more associated with DUs. Regarding the aforementioned
clinical conditions, it seems essential to re-evaluate the patients
during the follow-up period of the disease.
Although anti-RNAPIII (18%) positivity was not uncommon in our study
patients, none of them had SRC. In a large SSc cohort of 1325 patients,
more than 90% of the patients developed SRC within five years of SSc
onset [11]. The occurrence estimates of SRC were 6.5%, 7.1%, and
7.6% at 5, 10, and 15 years, respectively. SRC prevalence is lower in
North America than in Europe [35]. SRC is considered less frequent
in the Turkish SSc population, with a reported frequency of %3 in a
cross-sectional study from Turkey [22]. Therefore, the absence of
SRC in our study can be partly explained by genetic and geographical
differences and the short disease duration of the patients in our study
population.
Anti-PM/Scl antibodies were positive in 10% of the patients in our
study. They have been reported to be associated with inflammatory
myositis and calcinosis [36]. In a multinational cohort of SSc,
myositis was seen only in subjects positive for both anti-PM/Scl75 and
anti-PM/Scl100 antibodies [37]. From this perspective, the three
patients with both anti-PM/Scl75 and anti-PM/Scl100 antibodies appeared
at higher risk for myositis in our study. However, none had myositis or
calcinosis at baseline clinical evaluation at enrollment. Despite all
these results, continuous monitoring of serum creatine kinase levels in
anti-PM/Scl ab positive patients may be helpful in the further diagnosis
of myositis.
In order to look at multiple auto-antibody positivity from a different
perspective, we tried to reduce the dimensionality and increase the
interpretability of the data by the principal component analysis using
the staining intensities of auto-antibodies. Three significant
dimensions (Th/To - PM/Scl 100 - PM/Scl 75, ACA, and anti-RNAPIII
positive groups) were produced by PCA, and anti-Scl70 ab was ignored.
However, it was numerically higher in the number of patients since its
intensity was low. Therefore, we refrained from interpreting the
clinical relevance according to the results of PCA. However, we included
our results to compare with the PCA analyses of other studies in the
literature (38).
One of the limitations of our study was the small sample size. The most
important reason for this was to include patients who met the 2013
ACR/EULAR classification criteria and had as short a disease duration as
possible. This situation prevented reaching statistically significant
results in some subgroup analyses. Another limitation of our study was
that the study kits did not contain all SSc-specific auto-antibodies,
such as anti-U1RNP and anti-U11/U12 RNP. Another issue that could be
responsible for the lack of clinical involvement at study enrollment was
the short disease duration.
In conclusion, our results revealed that among SSc-specific
autoantibodies, anti-Scl70, ACA, and anti-RNAPIII are more common in
patients with early SSc, and co-expression of autoantibodies is not
infrequent. Testing a broad panel of autoantibodies yielded diagnostic
support in 8% of the patients who were negative for common SSc-specific
antibodies. Although autoantibodies specific to SSc in early disease
show generally known clinical features, it remains to be investigated
how co-expression of specific and nonspecific autoantibody positivity
will affect clinical presentation.
Acknowledgment: Actelion company contributed financially to the
study to be used during the purchase of the study kits. No other
disclosures relevant to this article were reported.
Author Contributions : DTK acquired the clinical data,
contributed to the experimental plan design, performed all the
statistical analyses, and drafted the manuscript. MH contributed to the
statistical analyses. AA contributed to the experimental plan design and
critical revision of the manuscript. AK, BF, AS, AK, YE, EK, and GK
contributed to data collection. AY, MB, DA, SSK, and AC contributed to
the critical revision of the manuscript. NK performed the laboratory
analysis and also contributed to the critical revision of the
manuscript.
Funding None.
Data availability All data have been included in the
manuscript.