Rheumatology International

https://doi.org/10.1007/s00296-018-4006-4
Rheumatology
INTERNATIONAL

BIOMARKERS

CXCL13 levels in serum but not in saliva are elevated in Asian Indian

patients with primary Sjögren’s syndrome
Santosh Kumar Mandal1,2   · Pulukool Sandhya1,3   · Jayakanthan Kabeerdoss1   · Janardana Ramya1,4 ·
Gowri Mahasampath5 · Debashish Danda1

Received: 8 January 2018 / Accepted: 7 March 2018

© Springer-Verlag GmbH Germany, part of Springer Nature 2018

Abstract
Human and animal model studies suggest CXCL13 is a potential biomarker in primary Sjögren’s syndrome (pSS). CXCL13
has not been studied in Indian patients with pSS. pSS cases classified by American European Consensus Group (AECG) or
American college of Rheumatology(ACR) 2012 criteria, attending rheumatology clinic between July 2014 and July 2015
were included. Hospital staff and healthy, non-blood related family members of patients constituted the control group. pSS
cases underwent clinical evaluation, laboratory investigations, ESSDAI and ESSPRI scoring. Unstimulated saliva was col-
lected by the spitting method. Salivary and serum CXCL13 were quantified by indirect ELISA. CXCL13 positivity was
determined using Receiver Operator Characteristic (ROC) curve. STATA13.1 (StataCorpLP,Texas,USA) software was used
for statistical analysis. In this study, 45 pSS cases and 42 healthy controls were recruited. In pSS, median levels of serum
CXCL13, but not salivary CXCL13 was significantly higher as compared to the corresponding levels in healthy controls
(p < 0.001). Using cutoff of 43.03 pg/ml obtained by ROC, serum CXCL13 positivity was seen in 31/43(72.1%) cases and
10/34 (29.4%) controls, respectively. Serum CXCL13 levels among pSS patients on treatment, treatment naïve patients and
healthy controls were statistically different. Serum CXCL13 positivity was associated with oral symptoms (p = 0.02), ocular
signs (p = 0.03) and hyperglobulinemia (p = 0.01). There was no association of salivary CXCL13 level with any of the clinical
variables. While serum CXCL13 was elevated in pSS, salivary CXCL13 was not. In conclusion, serum CXCL13 positivity
was found to be associated with oral symptoms, ocular signs and hyperglobulinemia in pSS.

Keywords  Saliva · Serum · CXCL13 · Sjogren’s syndrome · India

Pulukool Sandhya and Jayakanthan Kabeerdoss would like to be
known as joint second authors.
* Debashish Danda
debashisdandacmc@hotmail.com;
debashish.danda@cmcvellore.ac.in
1
Department of Clinical Immunology and Rheumatology,
Christian Medical College, Vellore, Tamil Nadu, India
2
Present Address: Department of Rheumatology,
Rabindranath Tagore international institute of cardiac
sciences, Kolkata, West Bengal, India
3
Present Address: St. Stephens Hospital, Tis Hazari,
New Delhi 110054, India
4
Present Address: St. John’s Medical College, Sarjapur Road,
John Nagar, Koramangala, Bengaluru, Karnataka 560034,
India
5
Department of Biostatistics, Christian Medical College,
Vellore, Tamil Nadu, India
Introduction
Primary Sjogren’s syndrome (pSS) is a chronic inflamma-
tory autoimmune disease, with prevalence rates ranging
from 0.1 to 4.6% [1]. Diagnosis of pSS is based on presence
of sicca symptoms, objective findings suggestive of salivary
and lacrimal gland dysfunction, positive serology for anti-
Ro/SSA and anti-La/SSB autoantibodies and /or histopatho-
logical finding of periductal lymphocytic infiltrate in minor
salivary gland tissue [2, 3].
Salivary gland B-cell infiltration increases with disease
progression [4]. Chemokines direct the process of sequestra-
tion of B and T cells at the site of inflammation [5, 6]. The
chemokine CXC ligand 13 protein (CXCL13), also known
as B-cell-attracting chemokine-1 or B-lymphocyte chemoat-
tractant (BLC), is the only chemokine which is specifically
chemoattractant for B cells via interaction with its receptor
CXCR5. Thus, CXCL13-CXCR5 interaction controls the

13
Vol.:(0123456789)

organization of B cells within follicles of lymphoid tissues
[7]. Moreover, overexpression of CXCL13 lead to forma-
tion of ectopic lymphoid tissue composed primarily of B
cells [8].
Several studies have demonstrated upregulation of
CXCL13 expression in the salivary glands of patients
afflicted with pSS [6, 9–12]. In these studies, CXCL13 was
localized within B-cell rich areas of salivary glands [6, 9], in
close association with glandular acini and ductal epithelial
cells [11, 12]. In mouse models of SS, the serum CXCL13
level increased with the disease progression. However, neu-
tralizing anti-CXCL13 antibody led to diminished disease
severity and progression [13]. Saliva and serum CXCL13
were found to be elevated in 74% of Caucasian patients
with pSS [13]. CXCL13 has not been studied till date in
Asian–Indian patients. We estimated salivary and serum
levels of CXCL13 in Asian–Indian patients with SS to fill
in the gap in this area and also to evaluate its correlation/
association with clinical presentations.
Patients and methods
Recruitment of cases and controls
This study was performed after approval by the Institutional
review board. (IRB No. 8945 dated 07.7.12014 & IRB No.
8986 dated 04.8.2014). pSS patients with age of 18 years
and above satisfying the American European Consensus
Group (AECG) 2002 or American College of Rheumatol-
ogy (ACR) 2012 criteria were included in the study [2, 3].
Healthy controls were recruited from non-genetically related
family members, friends of patients and consenting hospital
staff matched for age and sex.
Cases and controls were recruited from inpatient or out-
patient services of Rheumatology department between July
2014 and July 2015. Written informed consent was obtained
from all individuals. Individuals with an established diag-
nosis of another connective tissue disease were excluded.
Patients with the following conditions were also excluded:
acute or chronic infection, diabetes mellitus, malignancy,
poor dental hygiene, candidiasis, smokers, tobacco/“pan”
chewers, and those on cytotoxic therapy.
Clinical features, disease activity scoring and lab param-
eters of pSS patients were noted. Important clinical param-
eters that were recorded include age, duration of disease
prior to presentation, presence of sicca symptoms and
extra-glandular features. For ocular signs, Schirmer’s test
was performed. Assessment of salivary flow rate was not
done in any of our patients. EULAR Sjogren’s Syndrome
Disease Activity Index (ESSDAI) and EULAR Sjogren’s
Syndrome Patient Reported Index (ESSPRI) were used to
assess the disease activity status [14]. Results of anti-nuclear
13
Rheumatology International
antibodies (ANA), Rheumatoid Factor (RF), anti-SSA &
-SSB antibody, complement (C3 and C4) levels and total
globulin levels were noted. Chisholm and Mason histopatho-
logical grading score were noted [15].
Collection of samples and estimation of CXCL13
levels
Samples were collected as described previously [16, 17].
All participants were asked to refrain from eating, drink-
ing, chewing gums or oral hygiene procedures for at least
2 h prior to sample collection. At least 3 ml of saliva was
collected by spitting method. Saliva was processed by cen-
trifugation at 17,849×g for 15 min at room temperature and
clear supernatant was immediately frozen at − 80 °C. 5 ml of
blood was also collected from all participants, in vacutainer
tube. Blood was allowed to clot for 2 h and centrifuged at
5829×g for 15 min. Then serum was removed and stored at
− 80 °C.
CXCL13 in serum and saliva were quantitated by com-
mercially available sandwich ELISA kits following manu-
facturer’s recommended protocol (DY801, R&D Systems,
MN, USA) [13].
Statistical analysis
The categorical variables were expressed as frequencies with
percentage, while continuous variables were expressed as
median with interquartile range (IQR). To compare CXCL13
levels between the groups, Mann–Whitney U test was used.
The Receiver Operator Characteristic (ROC) curve was
constructed based on the values of CXCL13 and a cutoff
value was selected to determine CXCL13 positivity. Chi-
square test was used to analyze differences in clinical and
serologic variables of pSS patients between CXCL13 posi-
tive and CXCL13 negative patients. p < 0.05 was considered
as statistically significant. The statistical analysis was done
using STATA 13.1 (StataCorp LP, Texas, USA) statistical
package. Graphpad Prism was used for graphical representa-
tion (GraphPad Software, Inc. CA, USA).
Results
In this study, 45 patients with pSS and 42 healthy controls
were recruited. Paired salivary and serum samples were
available for 42 patients; while unpaired samples, i.e.,
only saliva was available for one and only serum for two
patients, respectively. Paired salivary and serum sample
were available in 30 healthy subjects; while unpaired sam-
ples, only saliva was available for six and only serum for
four healthy subjects. Median age of healthy controls was
35.5 (22–60) years and male to female ratio was 1:12.4.
Rheumatology International

The baseline characteristics of patients are shown in cutaneous vasculitis was seen in 12 and 6 patients, respec-
Table 1. All patients also fulfilled the 2016 American Col- tively. Interstitial lung disease, neuropathy, and cytopenia
lege of Rheumatology (ACR)/European League Against were seen in 1, 3 and 2 patients, respectively. None of the
Rheumatism (EULAR) classification criteria [18]. The patients had lymphoma.
most common extra-glandular feature was arthritis and
was seen in 22 (48.9%) patients. Renal involvement and
Comparison of CXCL13 levels between cases
and controls:
Table 1  Baseline characteristics of pSS cases
Parameters N = 45 The median levels of serum CXCL13 in pSS cases were
significantly higher as compared to healthy controls,
Age (IQR) 40 (32–45)
but same was not true for salivary levels as depicted in
Duration of disease (IQR) in months 60 (27–80)
Fig. 1a, b. Moreover, difference in serum levels between
Extra-glandular features (%) 37/45 (80.22)
pSS patients on treatment (immunomodulation and immu-
Median ESR in mm/1st Hour 46 (36–55)
nosuppression) and treatment naïve patients as well as
Median CRP in mg/dl (IQR) 3 (1–8.4)
healthy controls were also significant as depicted in Fig. 2.
RF positivity (%) 28/36 (77.78)
The cutoff values for serum and salivary CXCL13
ANA positivity (%) 43/45 (95.56)
were obtained by ROC as depicted in Fig. 3. A cutoff of
Anti-SSA antibody positivity (%) 35/45 (77.78)
43.03 pg/ml for serum CXCL13 had a sensitivity and spec-
Anti-SSB antibody positivity (%) 24/44 (54.55)
ificity of 72.09 and 70.59%, respectively. Using this cut-
High Globulin levels (%)a 27/41 (65.81)
off, serum CXCL13 positivity was noted in 31/43 (72.1%)
Positive MSG histopathology (%)b 41/44 (93.18)
cases and 10/34 (29.4%) controls, respectively.
Median ESSDAI (IQR) 1.00 (0–7)
In case of salivary CXCL13, cutoff of 14 pg/ml had sen-
Median ESSPRI (IQR) 2.3 (0–3.37)
sitivity and specificity of 50.00 and 59.46%, respectively.
Treatment
Salivary CXCL13 positivity was seen in 21/43 (48.83%)
 Treatment naïve 15
cases and 15/36 (41.67%) controls, respectively.
 Immunomodulationc 25
Cases with serum CXCL13 positivity as defined above
 Immunosuppressiond 5
was found to be associated with oral symptoms, ocular
a
 Globulin levels ≥ 3.5 gm/dl signs and hyperglobulinemia (Table 2). No association was
b
 Positive histopathology defined as presence of focus score ≥ 1 on found between any of the clinical features and salivary
minor salivary gland histopathology CXCL13 levels (Table 2).
c
 Immunomodulation includes those on low dose of steroids (< 10 mg
prednisolone), hydroxychloroquine/chloroquine, methotrexate, dap-
sone
d
 Immunosuppresion includes those on moderate (11–30 mg predniso-
lone/ day), high doses of steroids(≥ 30  mg prednisolone), mycophe-
nolate

Fig. 1  Median serum and salivary CXCL13 levels in healthy controls and pSS

13

Fig. 2  Median serum and salivary CXCL13 levels in patients with treatment naïve pSS, pSS patients on treatment and healthy controls
Fig. 3  The Receiver operating characteristic (ROC) curve plotted for
serum and salivary CXCL13
Discussion
Studies on humans and experimental mice models empha-
size the role of CXCL13 in formation of ectopic germinal
center, a highly relevant lesion with role in pathogenesis of
pSS [6, 9, 11–13]. Studies on serum and salivary CXCL13
in patients with pSS have provided conflicting results.
The utility of CXCL13 as a biomarker has not yet been
explored in Asian Indian patients.
In an earlier study from United States of America on
27 pSS patients and 21 healthy controls, serum CXCL13
positivity was reported in 56% of patients with pSS [13].
Using the same ELISA kit, CXCL13 positivity was 72%
in our cohort [13]. In a French study using multiplex
13
Rheumatology International
immunoassay, mean levels of serum CXCL13 were sig-
nificantly elevated in pSS as compared to controls [19].
On the other hand, a Chinese study could not find any
difference in serum CXCL13 levels between SS and con-
trols [20]. These differences could be attributed to ethnic
differences as well as methodology adopted to determine
cutoff in different populations. The previous study used
mean ± standard deviation, whereas the present study
relied on ROC method [13]. Further, a difference in the
median levels among patients on treatment, those treat-
ment naive as well as healthy controls was statistically sig-
nificant. While this novel finding raises the possibility that
serum CXCL13 can be used to assess treatment response,
larger longitudinal studies are needed for validation of this
observation.
Higher frequency of oral sicca symptoms and ocu-
lar signs of dryness was detected in those with elevated
serum CXCL13 as compared to those with negative serum
CXCL13 levels. This association has not been described
previously. We could not find any association of serum
CXCL13 with disease activity or rheumatoid factor posi-
tivity as reported previously in the French population [19].
However, association of CXCL13 with hyperglobulinemia
was reconfirmed in our study. Hyperglobulinemia is an
important predictor of extra-glandular disease and disease
activity in SS [21]. CXCL13 has been found to be an inde-
pendent predictor of lymphoma [19]. This aspect could
not be studied in our cohort, as there was no case with
lymphoma. Likewise, we could not find any association of
CXCL13 with pulmonary, glandular or lymphadenopathy
domains of ESSDAI as reported earlier [22].
Rheumatology International

Table 2  Association of clinical and laboratory characteristics with CXCL13 in pSS

Variables Serum CXCL13 Serum CXCL13 p value Saliva CXCL13 Saliva CXCL13 p value
positivity (n = 31) negativity (n = 12) positivity (n = 21) negativity (n = 23)
(%) (%) (%) (%)

Age (> 35) 19 (61.29) 7 (58.33) 0.86 13 (61.90) 13 (56.52) 0.72

Disease duration in months (> 60) 17 (54.84) 6 (50.00) 0.78 12 (57.14) 12 (52.17) 0.74
Oral symptoms 29 (93.55) 8 (66.67) 0.02 18 (85.71) 20 (86.96) 0.91
ocular signs 24 (88.9) 3 (50) 0.03 9 (75) 18 (85.7) 0.44
RF positivity 20 (83.33) 7 (63.64) 0.2 12 (80.00) 15 (75.00) 0.73
SSA positivity 23 (74.19) 9 (75.00) 0.96 17 (80.95) 16 (69.57) 0.38
SSB positivity 18 (60.00) 5 (41.67) 0.28 13 (61.90) 11 (50.00) 0.43
Globulins positivity 22 (78.57) 4 (36.36) 0.01 12 (60.00) 17 (70.00) 0.51
Positive histopathology 30 (96.77) 10 (83.33) 0.33 19 (90.5) 21 (91.3) 0.35
ESSDAI positivity 13 (41.94) 5 (41.67) 0.99 11 (52.38) 7 (30.43) 0.14
Treatment 18 (58.06) 10 (83.33) 0.12 13 (61.9) 17 (73.91) 0.39

Significant values are in bold (p < 0.05)

High-expression of CXCL13 has been found in minor
salivary gland of pSS and it was also found to correlate
with disease activity [23]. Kramer et  al. [13] reported
increased salivary CXCL13 levels in pSS as compared to
controls. We have used the same ELISA kit as Kramer
et al. [13], however, salivary CXCL13 was not found to be
elevated in our pSS patients. Low salivary CXCL13 lev-
els were reported by Hernendez et al. in a Mexican study
using Luminex bead-based technology, as well as Delaleu
et al. in salivary proteome study in the Norwegian popu-
lation [24, 25]. Causes for low CXCL13 in saliva despite
the role played by CXCL13 in ectopic germinal centre
formation can only be speculative at this stage [24]. This
could possibly be due to rapid degradation of chemokines
in saliva. We had not added protease inhibitors while
processing of salivary sample and this remains to be one
limitation of our study [26]. The other limitation is lack of
disease control in our study such as rheumatoid arthritis
or lupus patients. As CXCL13 is known to have role in
formation of germinal center, the salivary gland biopsies
with higher focus score and germinal center are expected
to have higher CXCL13 levels. In this study we could not
look at such a correlation which is another limitation of
the study.
In conclusion, level of serum CXCL13 was higher in
Indian patients with pSS as compared to healthy controls.
However, there was no difference in salivary CXCL13 levels
between patients with pSS and healthy controls. In addition,
our pSS patients with oral symptoms, ocular signs, hyper-
globulinemia as well as the treatment naive patients had
higher frequency of serum CXCL13 positivity.
Acknowledgements  We acknowledge the funding received from Chris-
tian Medical College (CMC) fluid research grant for the study.
Compliance with ethical standards 
Conflict of interest  The authors report that they have no conflicts of
interests.
Ethical approval  This study was approved by the Institutional review
board (IRB) & Ethics committee and was conducted in accordance with
the Helsinki declaration of 1975, as revised in 2008.
Informed consent  Informed consent has been obtained from all par-
ticipants.
References
1. Gabriel SE, Michaud K (2009) Epidemiological studies in inci-
dence, prevalence, mortality, and comorbidity of the rheumatic
diseases. Arthritis Res Ther 11:229. https:​ //doi.org/10.1186/ar266​
9
2. Vitali C, Bombardieri S, Jonsson R et al (2002) Classification
criteria for Sjögren’s syndrome: a revised version of the European
criteria proposed by the American-European Consensus Group.
Ann Rheum Dis 61:554–558. https:​ //doi.org/10.1136/ard.61.6.554
3. Shiboski SC, Shiboski CH, Criswell LA et al (2012) American
College of Rheumatology classification criteria for Sjögren’s syn-
drome: a data-driven, expert consensus approach in the Sjögren’s
International Collaborative Clinical Alliance cohort. Arthritis
Care Res 64:475–487
4. Christodoulou MI, Kapsogeorgou EK, Moutsopoulos HM (2010)
Characteristics of the minor salivary gland infiltrates in Sjögren’s
syndrome. J Autoimmun 34:400–407. https​://doi.org/10.1016/j.
jaut.2009.10.004
5. Xanthou G, Polihronis M, Tzioufas AG, et  al (2001) “Lym-
phoid” chemokine messenger RNA expression by epithelial
cells in the chronic inflammatory lesion of the salivary glands
of Sjögren’s syndrome patients: possible participation in lym-
phoid structure formation. Arthritis Rheum 44:408–418.
https://doi.org/10.1002/1529-0131(200102)44:2<408::AID-
ANR60>3.0.CO;2-0

13

6. Barone F, Bombardieri M, Manzo A et al (2005) Association of
CXCL13 and CCL21 expression with the progressive organiza-
tion of lymphoid-like structures in Sjögren’s syndrome. Arthritis
Rheum 52:1773–1784. https​://doi.org/10.1002/art.21062​
7. Ansel KM, Ngo VN, Hyman PL et al (2000) A chemokine-driven
positive feedback loop organizes lymphoid follicles. Nature
406:309–314. https​://doi.org/10.1038/35018​581
8. Luther SA, Lopez T, Bai W et al (2000) BLC expression in pan-
creatic islets causes B cell recruitment and lymphotoxin-depend-
ent lymphoid neogenesis. Immunity 12:471–481
9. Amft N, Curnow SJ, Scheel-Toellner D et  al (2001) Ectopic
expression of the B cell-attracting chemokine BCA-1 (CXCL13)
on endothelial cells and within lymphoid follicles contributes to
the establishment of germinal center-like structures in Sjögren’s
syndrome. Arthritis Rheum 44:2633–2641
10. Hjelmervik TOR, Petersen K, Jonassen I et al (2005) Gene expres-
sion profiling of minor salivary glands clearly distinguishes pri-
mary Sjögren’s syndrome patients from healthy control subjects.
Arthritis Rheum 52:1534–1544. https:​ //doi.org/10.1002/art.21006​
11. Salomonsson S, Jonsson MV, Skarstein K et al (2003) Cellular
basis of ectopic germinal center formation and autoantibody pro-
duction in the target organ of patients with Sjögren’s syndrome.
Arthritis Rheum 48:3187–3201. https:​ //doi.org/10.1002/art.11311​
12. Salomonsson S, Larsson P, Tengnér P et al (2002) Expression
of the B cell-attracting chemokine CXCL13 in the target organ
and autoantibody production in ectopic lymphoid tissue in the
chronic inflammatory disease Sjögren’s syndrome. Scand J Immu-
nol 55:336–342
13. Kramer JM, Klimatcheva E, Rothstein TL (2013) CXCL13 is ele-
vated in Sjögren’s syndrome in mice and humans and is implicated
in disease pathogenesis. J Leukoc Biol 94:1079–1089. https:​ //doi.
org/10.1189/jlb.01130​36
14. Seror R, Gottenberg JE, Devauchelle-Pensec V et al (2013) Euro-
pean league against rheumatism Sjögren’s syndrome disease activ-
ity index and European League Against Rheumatism Sjögren’s
syndrome patient-reported index: a complete picture of primary
Sjögren’s syndrome patients. Arthritis Care Res 65:1358–1364.
https​://doi.org/10.1002/acr.21991​
15. Chisholm DM, Mason DK (1968) Labial salivary gland biopsy
in Sjögren’s disease. J Clin Pathol 21:656–660. https​://doi.
org/10.1136/jcp.21.5.656
16. Jayakanthan K, Ramya J, Mandal SK et al (2016) Younger patients
with primary Sjögren’s syndrome are more likely to have sali-
vary IgG anti-muscarinic acetylcholine receptor type 3 antibodies.
13
Rheumatology International
Clin Rheumatol 35:657–662. https​: //doi.org/10.1007/s1006​
7-016-3186-0
17. Kabeerdoss J, Sandhya P, Mandal SK et al (2016) High salivary
soluble L-selectin and interleukin-7 levels in Asian Indian patients
with primary Sjögren’s syndrome. Clin Rheumatol 35:3063–3067.
https​://doi.org/10.1007/s1006​7-016-3406-7
18. Shiboski CH, Shiboski SC, Seror R et al (2016) 2016 American
College of Rheumatology/European League Against Rheumatism
classification criteria for primary Sjögren’s syndrome A consensus
and data-driven methodology involving three international patient
cohorts. Ann Rheum. https​://doi.org/10.1136/annrh​eumdi​s-2016-
21057​1 (Dis annrheumdis–2016–210571)
19. Nocturne G, Seror R, Fogel O et al (2015) CXCL13 and CCL11
serum levels and lymphoma and disease activity in primary
Sjögren’s syndrome. Arthritis Rheumatol Hoboken NJ 67:3226–
3233. https​://doi.org/10.1002/art.39315​
20. Jin L, Yu D, Li X et al (2014) CD4 + CXCR5+ follicular helper T
cells in salivary gland promote B cells maturation in patients with
primary Sjogren’s syndrome. Int J Clin Exp Pathol 7:1988–1996
21. García-Carrasco M, Mendoza-Pinto C, Jiménez-Hernández C et al
(2012) Serologic features of primary Sjögren’s syndrome: clinical
and prognostic correlation. Int J Clin Rheumatol 7:651–659. https​
://doi.org/10.2217/ijr.12.64
22. Nishikawa A, Suzuki K, Kassai Y et al (2016) Identification of
definitive serum biomarkers associated with disease activity in
primary Sjögren’s syndrome. Arthritis Res Ther 18:106. https​://
doi.org/10.1186/s1307​5-016-1006-1
23. Lee K-E, Kang J-H, Yim Y-R et  al (2017) Predictive sig-
nificance of CCL21 and CXCL13 levels in the minor salivary
glands of patients with Sjögren’s syndrome. Clin Exp Rheumatol
35(2):234–240
24. Hernández-Molina G, Michel-Peregrina M, Hernández-Ramírez
DF et al (2011) Chemokine saliva levels in patients with primary
Sjögren’s syndrome, associated Sjögren’s syndrome, pre-clinical
Sjögren’s syndrome and systemic autoimmune diseases. Rheuma-
tol Oxf Engl 50:1288–1292. https​://doi.org/10.1093/rheum​atolo​
gy/ker01​9
25. Delaleu N, Mydel P, Kwee I et al (2015) High fidelity between
saliva proteomics and the biologic state of salivary glands defines
biomarker signatures for primary Sjögren’s syndrome. Arthritis
Rheumatol 67:1084–1095. https​://doi.org/10.1002/art.39015​
26. Henson BS, Wong DT (2010) Collection, storage, and processing
of saliva samples for downstream molecular applications. In: Oral
Biology. Humana Press, Totowa, pp 21–30