Rheumatology 2015;54:1056–1064
doi:10.1093/rheumatology/keu453
Advance Access publication 27 November 2014
RHEUMATOLOGY
Original article
Reliability of histopathological salivary gland biopsy
assessment in Sjögren’s syndrome: a multicentre
cohort study
Sebastian Costa1, Isabelle Quintin-Roué1, Agnès Lesourd2, Sandrine JousseJoulin3,4, Jean-Marie Berthelot5, Eric Hachulla6, Pierre-Yves Hatron6,
Vincent Goeb7, Olivier Vittecoq7, Jacques Olivier Pers4, Pascale Marcorelles1,
Emmanuel Nowak8, Alain Saraux3,4 and Valérie Devauchelle-Pensec3,4
Abstract
Objective. The aim of this study was to assess intraobserver and interobserver reliability of minor salivary
gland biopsy (MSGB) in SS.
Methods. All MSGBs available from the Tolerance and Efficacy of Rituximab in Primary Sjögren’s
Syndrome (TEARS) study were subjected to a standardized blinded assessment by a single specifically
trained pathologist twice at a 2 month interval; both the Chisholm–Mason (CM) grade and the focus score
(FS) were determined. Baseline histopathological reports by local pathologists at each study centre were
compared with the first standardized blinded assessment. Agreement was assessed for the dichotomized
FS (dFS) and dichotomized CM (dCM) grade, as well as for nine other histopathological features.
CLINICAL
SCIENCE
Results. Eighty-nine MSGBs were studied. Intraobserver k values were 1 for dFS, 0.80 for dCM, 0.67 for
germinal centre-like structures, 0.44 for fibrosis and 0.29 for confluent foci. Most of the local histopathological reports based their diagnosis on the CM grade, although the FS was often reported or the data
needed to determine it were provided. Interobserver agreement k values were 0.71 for dFS, 0.64 for dCM,
0.46 for focal lymphocytic sialadenitis, 0.42 for non-specific chronic inflammation and 0.16 for fibrosis.
Conclusion. Although FS reliability was good, disparities were noted in the assessment methods used by
local pathologists. The protocol for FS determination was not followed routinely, with the result that the FS
was often overestimated. Germinal centre-like structures, which predict lymphoma, showed good reliability but were inconsistently reported.
Key words: Sjögren’s syndrome, minor salivary gland biopsy, Chisholm, focus score, reliability.
Rheumatology key messages
.
.
.
Overall, focus score evaluation for SS diagnosis has substantial reliability.
Disparity in minor salivary gland biopsy evaluation can lead to overestimation of SS.
Germinal centre–like structures are insufficiently assessed, despite their importance for SS prognosis.
1
Pathology Department, Brest University Hospital, Brest, France,
Pathology Department, Vannes General Hospital, Vannes, 3Brest
Occidental University, Rheumatology Unit, Hôpital de la Cavale
Blanche, 4EA 2216, ESPRI 29, Brest University, Brest, 5Rheumatology
Department, Hôtel Dieu University Hospital, Nantes, 6Department of
Internal Medicine, Nord-de-France University, Claude-Huriez Hospital,
Lille, 7Rheumatology Department, Rouen University Hospital, Rouen
and 8Brest Occidental University, Clinical Investigation Centre, Institut
National de la Santé et de la Recherche Médicale, Brest, France
2
Submitted 19 May 2014; revised version accepted 29 September 2014
Correspondence to: Valerie Devauchelle-Pensec, Rheumatology Unit,
Hôpital de la Cavale Blanche, BP 824, F 29609 Brest cedex, France.
E-mail: [email protected]
Introduction
SS is a common chronic autoimmune disease
characterized by lachrymal and salivary gland dysfunction
due in part to lymphocytic infiltration and destruction of
the gland parenchyma [1–5]. Diagnosis relies on a combination of clinical and laboratory findings, among which
minor salivary gland biopsy (MSGB) abnormalities are a
major criterion [6–9]. MSGB has been added to preliminary European classification criteria to improve their
! The Author 2014. Published by Oxford University Press on behalf of the British Society for Rheumatology. All rights reserved. For Permissions, please email: [email protected]
Minor salivary gland biopsy reliability in SS
diagnostic performance and to exclude other entities
resembling SS [10, 11]. In 1968, Chisholm and Mason
introduced an MSGB grading system with five grades
[1], 0–4, based on the presence of diffuse lymphocytic
infiltration and/or lymphocytic foci, with a focus defined
as an aggregate of at least 50 lymphocytes. All aggregates were taken into account, whatever the aspect of
the adjacent glandular parenchyma. The result was expressed as foci per 4 mm2 of tissue. Greenspan and
Daniels [12, 13] subsequently added the focus score
(FS) concept, with the focus as the number of foci per
4 mm2 of normal-appearing tissue. The presence of foci
adjacent to apparently normal mucous acini defines focal
lymphocytic sialadenitis (FLS), as opposed to non-specific
chronic sialadenitis (NSCS), in which glands may have
lymphocytic aggregates but also present acinar atrophy,
fibrosis and dilated ducts. Other grading systems were
suggested subsequently [14–16], generating debate
about the best histological method for assessing MSGB.
Nevertheless, FLS is strongly associated with the serological and ocular manifestations of SS [17] and an FS
value 51 rather than <1 [18]. Although still widely used,
the Chisholm–Mason (CM) grading system fails to distinguish FLS from non-specific chronic sialadenitis, an inflammatory disorder that does not involve autoimmunity.
Furthermore, sialadenitis and the FS can be challenging to
evaluate even for experienced pathologists. In addition,
the FS is not determined routinely by all pathologists, a
fact that results in diagnostic discrepancies [19, 20]. Thus
a blinded FS-based review of 60 MSGBs from 58 individuals resulted in revision of the initial diagnosis in 53% of
cases, established previously undiagnosed SS in 12 patients and refuted a previous diagnosis of SS in 8 patients
based on the work of Tavoni et al. [19]; in 58 patients, the
initial evaluation had not involved FS determination. Other
histopathological features sometimes found in MSGBs include germinal centre-like structures, which may predict
lymphoma and deserve incorporation into severity classification systems for SS and monitoring criteria for specific
patient subgroups [21–23].
Despite these uncertainties, a recent systematic review
confirmed the good diagnostic value of MSGB, with sensitivity and specificity values ranging from 63.5% to
93.7% [24], but this article does not report the methods
used to assess the MSGBs. Specificity cannot be perfect,
since MSGB alterations are found in healthy elderly patients. Among 68 patients free of systemic disease and
12–78 years of age, 6 (8.8%) had FS values >1 [25, see
Table 4]. Similarly, among 54 apparently healthy patients
with a mean age of 31.5 years, 8 (14.8%) had FS values
>1 [26]. The significance of these findings is unknown, but
hypotheses include trauma, lip-biting habit, infections and
early SS. An obstacle to the interpretation of published
data is the use of various grading systems. Thus, of the
nine studies identified in the above-mentioned systematic
review [24], four used the CM scale, four used the Daniels
scale and one used both. Achieving a consensus about a
standardized method for obtaining, processing and assessing MSGB is crucial.
www.rheumatology.oxfordjournals.org
The primary objective of this study was to assess the
intraobserver reliability of each histopathological criterion
used to assess MSGBs. We also determined the level of
agreement between the initial pathologist reports of
MSGB findings and a centralized standardized evaluation
by a pathologist specifically trained to assess MSGBs. To
achieve this objective, we used data from the recently
published randomized multicentre Tolerance and
Efficacy of Rituximab in Primary Sjögren’s Syndrome
(TEARS) study, designed to evaluate the efficacy of rituximab in primary SS [27].
Patients and methods
Patients
The TEARS study patients were recruited at 14 university
hospitals in France if they fulfilled the new
American–European Consensus Group (AECG) criteria
for SS [6] and had active disease, defined as scores of
at least 50 mm on at least two of four visual analogue
scales (VASs) (from 0 = none to 100 = worst possible) for
global disease, pain, fatigue and dryness. Additional
requirements were either SS symptom onset within the
last 10 years and biologically active SS [defined as
autoantibodies (anti-Ro-SSA antibodies or RF), cryoglobulinaemia, hypergammaglobulinaemia, b2-microglobulin
elevation or hypocomplementaemia] or systemic SS with
at least one extraglandular manifestation or current
parotid gland enlargement.
MSGB was performed routinely at 4 of the 14 participating centres (Brest, Lille, Nantes and Rouen) at study
inclusion and after 24 weeks. The MSGB specimens were
immediately fixed in formalin, processed and embedded
in paraffin according to local standardized laboratory
methods. MSGBs in paraffin blocks, the corresponding
slides and the local pathologist reports were sent to the
coordinating centre in Brest (Fig. 1). The TEARS study was
approved by the appropriate ethics committee (CPP
Ouest VI) and all patients gave written consent before
study enrolment. The protocol is registered on
ClinicalTrials.gov (NCT00740948).
Training session
Intraobserver and interobserver reliability of each histopathological feature was assessed by two of the
authors—S.C., with 2 years of experience evaluating
MSGBs, and I.Q., with >20 years of experience evaluating
MSGBs. The training session consisted of both observers
evaluating 32 MSGBs from the Brest centre histopathology specimen bank (Brest). None of these 32 MSGBs
were from TEARS study patients.
Intraobserver reliability of MSGB review by a single
trained pathologist at the study centre
All MSGBs were read twice, 2 months apart, in random
order by S.C., who used a standardized blinded procedure to evaluate 11 histopathological features. All slides
were anonymized via the use of a random-number
sequence.
1057
Sebastian Costa et al.
FIG. 1 Flow chart
The TEARS cohort included 93 minor salivary gland
biopsies (MSGBs) from 60 patients, of which 89 were sent
to the Brest coordinating centre and entered into the
study and 77 had available baseline pathology reports.
The first and second evaluations by a single trained
pathologist at the Brest coordinating centre were done 2
months apart to assess intraobserver agreement. The first
evaluation by this pathologist was compared with the
baseline reports by the local pathologists at each centre
to assess interobserver agreement.
Interobserver reliabilities of baseline MSGB
evaluations by local pathologists
Baseline histopathological reports written by the local
pathologists at the four study centres that performed
MSGBs were sent to the coordinating centre and reviewed. The number of reports indicating the presence
(coded 1) or absence (coded 0) of each feature was recorded. We considered the following seven features,
which could be described using various synonyms:
acinar depletion, acinar atrophy or acinar metaplasia;
ductular hypertrophy, duct dilatation, ectasia or hyperplasia; NSCS, diffuse fibrosis, fibrous scar or sclerosing
lymphocytic sialadenitis; fibrosis, perilobular fibrosis or
intralobular fibrosis; and adiposis, fatty degeneration or
fatty replacement.
To evaluate interobserver reproducibility, we compared
the presence or absence of each of these seven features
according to the local baseline histopathological reports
and to a standardized review by S.C., who did not have
access to the local baseline pathologist reports.
Histopathological criteria
The 11 histopathological features evaluated during the
centralized reviews by S.C. and I.Q. are described
below and in Fig. 2. The digital slide scanner
NanoZoomer (Hamamatsu, Hamamatsu City, Japan) was
used to digitize the slides and the biopsy surface areas
1058
were measured using the freehand surface area tool in
NanoZoomer Digital Pathology Viewer software
(Hamamatsu). This method ensures highly accurate
standardized
surface
area
measurement.
The
dichotomized FS (dFS) was defined as the lymphocytic
FS as described by Greenspan and Daniels [24, 25] and
at http://sicca.ucsf.edunot (see http://sicca.ucsf.edu/
LSG_bx_Grading_SOP.pdf) [17, 18], dichotomized by
scoring values <1 as 0 and values 51 as 1. The
dichotomized CM (dCM) grade was obtained by scoring
CM grades [1] as 0 if equal to 0, 1 or 2 and as 1 if equal to
3 or 4. FLS was defined as the presence of one or more
dense aggregates of 550 lymphocytes adjacent to apparently normal mucous acini located within lobes or lobules
free of generalized duct dilation or fibrosis and containing
few or no plasma cells, either in the absence of other
evidence of inflammation or predominating markedly
over any other evidence of inflammation. NSCS and sclerosing lymphocytic sialadenitis (intense NSCS) were
included in a single category and defined as scattered
or focal infiltrates of lymphocytes and macrophages that
were not immediately adjacent to apparently normal acini
and that were located within gland lobules, gland lobes or
entire glands exhibiting some combination of mild to moderate acinar atrophy, interstitial fibrosis and dilated ducts,
with these last often filled with inspissated mucus. Withinnormal-limits glands were defined as normal architecture,
densely arranged normal acini and scattered or small aggregates of plasma cells with few or no lymphocytes.
Fibrosis was defined as collagenous fibrosis sheathing
the ducts or forming tracts that dissociated the lobules
and encased the acini. Duct dilatation was defined as a
duct lumen enlargement with epithelial flattening. Acinar
depletion (or atrophy) was defined as de-differentiation of
the acinar epithelium and/or loss of gland parenchyma,
with substitution by adipocytes or fibrosis. Adiposis was
defined as the replacement of normal parenchyma by adipocytes. Germinal centre-like structures were defined as
well-circumscribed mononuclear inflammatory cell infiltrates (550 cells) exhibiting a tightly packed dark zone
and a more loosely packed light zone within an otherwise
normal salivary gland epithelium [28]. Confluent foci were
defined as focal lymphocytic infiltrates that were too large
or numerous to be counted separately.
Statistical analysis
Statistical analysis was performed with SPSS version 22
(IBM, Armonk, NY, USA). Cohen’s k statistic was used to
assess interobserver agreement while taking chance
agreement into account [29]. The 95% CIs were computed. Values of k were interpreted as follows [28]:
0.81–1.00, almost perfect agreement; 0.61–0.80, substantial agreement; 0.41–0.60, moderate agreement;
0.21–0.40, fair agreement; 0–0.20, slight agreement;
1.0–0, no agreement. The intra- and interclass correlation coefficients were computed to evaluate the quantitative variable of focus score. Bland–Altman plots were
constructed to assess agreement between the CM
grade and the Greenspan–Daniels FS [30]. Since the CM
www.rheumatology.oxfordjournals.org
Minor salivary gland biopsy reliability in SS
FIG. 2 Main histopathological features of minor salivary gland biopsies in patients with primary syndromes
Minor salivary gland biopsy stained with haematoxylin, eosin and saffron. (A) Normal parenchyma with mucinous acini.
The cells have abundant cytoplasm with clear mucin (10). (B) Dense aggregate of >50 lymphocytes defining a focus
adjacent to mucous acini of normal appearance (10). (C) Duct dilation (10). (D) Focal lymphocytic sialadenitis (FLS)
with at least five foci and confluent foci adjacent to mucous acini of normal appearance (4). (E) Non-specific chronic
sialadenitis (NSCS) with scattered infiltrates of lymphocytes and moderate acinar atrophy and interstitial fibrosis (10).
(F) Focal lymphocytic sialadenitis with a single prominent germinal centre–like structure (10).
www.rheumatology.oxfordjournals.org
1059
Sebastian Costa et al.
TABLE 1 Intraobserver and interobserver reliability of minor salivary gland biopsy assessment
Intraobserver agreement
(centralized reading by S.C.) (n = 89)
i (95% CI)
dFS
dCM
Acinar depletion
Duct dilatation
Fibrosis
Adiposis
FLS
NSCS
WLN
Confluent foci
Germinal centre–like
structures
1.00
0.80
0.73
0.59
0.44
0.53
0.71
0.64
0.78
0.29
0.67
(1.00,
(0.73,
(0.66,
(0.50,
(0.35,
(0.43,
(0.64,
(0.56,
(0.66,
(0.18,
(0.56,
1.00)
0.87)
0.80)
0.68)
0.53)
0.63)
0.78)
0.72)
0.90)
0.40)
0.79)
Cases,
n (%)
89
89
89
89
89
89
89
89
89
89
89
Interobserver agreement
(baseline local reading vs centralized
reading by S.C.) (n = 77)
Cases with
the feature
at second
reading, n
i (95% CI)
47
66
50
32
52
21
45
54
8
8
13
0.71 (0.62, 0.79)
0.64 (0.55, 0.74)
0.41 (0.31, 0.51)
0.12 ( 0.2, 0.04)
0.16 ( 0.01, 0.34)
1 (1.00, 1.00)
0.46 (0.23, 0.70)
0.42 (0.23, 0.69)
NS (NS)
NS (NS)
NS (NS)
(100)
(100)
(100)
(100)
(100)
(100)
(100)
(100)
(100)
(100)
(100)
Cases,
n (%)
63
75
67
51
35
10
14
24
3
3
5
Cases with
the feature in
the baseline
local
pathologist’s
report, n
(81.8)
(97.4)
(87)
(66.2)
(45.4)
(12.9)
(18.1)
(31.1)
(3.8)
(3.8)
(6.4)
50
50
18
3
26
2
12
2
3
2
5
dCM: dichotomized Chisholm–Mason grade; dFS: dichotomized focus score; FLS: focal lymphocytic sialadenitis; NS: not
significant; NSCS: non-specific chronic sialadenitis; WLN: within limits of normal.
grade takes all the foci into account, we expect overestimation of the number of foci.
Results
Study population
Of the 122 TEARS study patients with primary SS, 60
underwent MSGB and were included in the present
study. Of these 60 patients, 33 had two MSGBs at baseline and week 24, 23 had a single MSGB at baseline and 4
had a single MSGB at week 24. A total of 93 MSGBs,
89 biopsy specimens and their corresponding slides
were sent to the Brest study coordinating centre and
included in the present study. The specifically trained
pathologist (S.C.), to avoid variability of evaluation between different levels of tissue, evaluated the same
slides as the local pathologists. If multiple MSGB sections
were available, the level with the higher FS was retained,
as recommended [18, 31]. The mean FS in our 60
patients was 2.8/4 mm2 and 31 of 60 patients (51%) had
a FS 51.
Centralized histopathological assessment
Training session: interobserver agreement between the
specifically trained pathologist (S.C.) and senior pathologist (I.Q.)
S.C. and I.Q. read the 32 MSGBs. Interobserver k between S.C. and I.Q. was 0.80 for dFS and 0.93 for dCM.
The same slides were read twice by S.C. and the intraobserver k was 1 for dFS and 0.72 for dCM.
1060
Intraobserver agreement
S.C. read the 89 MSGBs twice at an interval of 2 months.
Table 1 reports the k values, which indicated almost perfect agreement (k = 1.0) for dFS and substantial agreement (k = 0.80) for dCM. The intraclass correlation
coefficient (ICC) for the FS was 0.76 (95% CI 0.66, 0.84)
(not shown in Table 1). Agreement was substantial for
FLS, NSCS and germinal centre–like structures (k = 0.71,
k = 0.64 and k = 0.67, respectively); moderate for fibrosis
and duct dilatation (k = 0.44 and k = 0.59, respectively)
and fair for confluent foci (k = 0.29).
Baseline histopathological reports by local
pathologists
Reported findings
Table 1 lists the findings described in the reports. Of the
93 baseline local pathology reports, 77 were available for
the study; among these, 13 (17%) failed to report the
number of foci instead using terms such as many foci or
significant number of foci. The CM grade was given in the
conclusion in 75 of 77 reports, but 63 of 77 reports gave
the FS value. The presence or absence of FLS and NSCS
was indicated in only 18.1% and 31.1% of reports, respectively. Germinal centre structures were found in 13
of 89 (14.6%) MSGBs from 10 of 60 (16.6%) patients.
The presence or absence of fibrosis was indicated in 35
(45.5%) reports, including 26 describing the presence of
fibrosis. The frequent lack of completeness of the local
reports may be ascribable to failure of the pathologists
to report the absence of features. Acinar depletion was
www.rheumatology.oxfordjournals.org
Minor salivary gland biopsy reliability in SS
FIG. 3 Bland–Altman assessment of agreement between focus counts by Chisholm–Mason grading and focus score
determination
The difference in focus count values obtained using the two methods (y-axis) is plotted against the mean of the values
obtained using the two methods (x-axis). Bias is defined as the mean difference and shown by the blue line. Bias reflects
the mean of the systematic deviation between the two sets of values. CM: Chisholm–Mason; FS: focus score.
the most commonly reported feature [67 of 77 (87%)] and
was present in 18 of 67 (27%) cases.
Interobserver agreement
Interobserver agreement between the local pathologists
and S.C. (Table 1) was substantial for dFS and dCM
(k = 0.71 and k = 0.64, respectively). The ICC for FS was
0.66 (95% CI 0.49, 0.78) (not shown in Table 1). Of 63
MSGBs, 55 were concordant for dFS between the local
pathologist and S.C. Importantly, the re-evaluation by
S.C. led to a change in 8 of 63 (12.6%) diagnoses of the
MSGBs. Six (9.5%) MSGBs had FS changed to <1 and
two (3.1%) to 51. Agreement was absent for duct dilatation (k = 0.12), slight for fibrosis (k = 0.16) and moderate
for acinar depletion (k = 0.41), FLS (k = 0.46) and NSCS
(k = 0.42). Sample sizes were small, however, for FLS
and NSCS and too small for a valid assessment of agreement regarding the other features.
Difference between CM grade and FS
The mean number of foci was 3.63 by CM grading and 3
by FS determination, yielding a difference of 0.63. The
Bland–Altman plot (Fig. 3) showed systematic bias with
overestimation of the number of foci by CM grading.
This finding is not surprising, as CM grading takes foci
within sites of non-specific chronic sialadenitis into account, thereby inducing a risk of SS overdiagnosis.
Discussion
Using data from our previous TEARS study of rituximab in
primary SS, we showed here that the reliability for a
standardized evaluation of MSGBs from patients with
www.rheumatology.oxfordjournals.org
primary SS was substantial. Review by the trained pathologist resulted in a different diagnosis for 12.6% of the
patients. MSGB is of prime importance for the diagnosis,
classification and prognosis of SS. Validated classification
criteria sets for SS incorporate histopathological findings,
reflecting the definition of SS as an autoimmune disease
involving the exocrine epithelia [30]. However, the methods used to read MSGBs and to write the reports are not
standardized. Nevertheless, standardization can be
achieved using the detailed protocol described by
Daniels et al. [18]. Furthermore, little information is available on the reliability of these new criteria. In keeping with
our findings, a 2012 report from Italy indicated that CM
grading was still used in most of the seven experienced
rheumatology centres [19]. AECG and ACR criteria refer to
the method developed by Daniels and Greenspan [6, 7]
and specify that the lymphocytic focus count should be
performed throughout the normal-appearing mucosa. The
CM method does not meet this requirement and consequently cannot be substituted for FS determination. Thus
CM grading fails to distinguish between FLS and NSCS.
Moreover, CM grading is a categorical scale that does not
evaluate severity in patients with FS values >1. CM grading overestimates the focus count, as all foci are taken
into account.
Reliability may be slightly better for the FS than for CM
grading. In the study carried out at seven Italian centres,
interobserver agreement for CM grade ranged from 0.44
to 0.92 [19]. A blinded review of the FS in 60 MSGBs from
58 individuals resulted in revision of the initial diagnosis in
53% of cases, although aggregate glandular surface area,
gross foci and FS showed high reliability as assessed by
regression equations [20]. In another study involving the
1061
Sebastian Costa et al.
independent review of 56 MSGBs by two pathologists,
agreement was high for the number of foci [ICC 0.97
(95% CI 0.96, 0.99)] and FS [ICC 0.96 (95% CI 0.94,
0.99)] [18]. Discrepancies in focus count can be ascribed
to two main factors: a large focus may be counted as
several nearly confluent foci or as a single focus and
determining whether a lymphocyte aggregate is sufficiently dense to constitute a focus may be challenging.
Surface area measurement may be difficult but benefits
considerably from the use of an eyepiece graticule or
computer-based tools.
Germinal centre–like structures predict severe SS and
the development of lymphoma [28]. Their prevalence
varied between 18% and 59% in Risselada et al.’s [33]
review in which many studies had a FS 51 in all
MSGBs. Germinal centre–like structures were common
in studies in populations with a mean FS >3 or with
MALT lymphoma [32, 34, 35]. In this systematic review
of 13 cohorts of SS patients, germinal centre–like structures were seen in 25.1% (S.D. 5) of patients [33]. Germinal
centre–like structures should be sought routinely, as
they may indicate greater disease severity with higher
frequencies of lymphadenopathy, leucopenia, hypergammaglobulinaemia and extraglandular complications,
as well as higher antibody titres [21, 23, 33, 36].
Furthermore, germinal centre–like structures are associated with non-Hodgkin lymphoma [23]. Thus, of seven
patients who developed non-Hodgkin lymphoma during
1855 patient-years of follow-up, six had germinal
centre–like structures at the diagnosis of primary SS,
and the negative predictive value of this feature for lymphoma was high [22].
FLS is the distinctive feature of SS, whereas NSCS is
not usually associated with SS [26]. Lymphocytic infiltration of the ductal structures could also be a specific feature of SS and might be a useful criteria [37]. In our study,
many local pathologists failed to specify whether FLS and
NSCS were present or absent. Challenges may arise when
both features are present in the same gland. If FLS predominates, then the FS should be determined. However, it
is unclear whether the focus count should include all foci
within the gland or only those in FLS areas.
The natural history of minor salivary gland alterations in
SS patients remains unclear. Until now, the FS and fibrosis were believed to increase over time and with disease
progression [38, 39]. However, in a recent study of serial
MSGBs from 28 patients with SS and a median follow-up
of 55 months, neither the infiltration grade nor the FS
values changed significantly, although significant disease
progression occurred [5]. Neither did significant changes
develop for germinal centre–like structures, fibrosis or adiposis. This would indicate that SS lesions are slowly progressing alterations. Another study established that FLS
was the key histopathological feature in SS and did not
usually progress to fibrosis, although patients with chronic
sialadenitis were excluded from this work [17].
Nevertheless, fibrosis of NSCS features has been reported in patients meeting AECG or ACR criteria for SS.
Significant associations linking stimulated salivary flow to
1062
both the FS and fibrosis have been reported, and neither
feature varied with the degree of atrophy [40]. The correlations between oral manifestations and histological features remain unclear. Circular reasoning limits the validity
of many studies. The identification of sensitive, specific
and reliable criteria has become a key objective since
the introduction of immunosuppressant therapy for SS.
Among responders to the immunosuppressant mizoribine,
those with marked lymphocytic infiltration and little atrophy and fibrosis significantly increased their salivary secretion volume in contrast to those with limited
lymphocytic infiltration and major atrophy and fibrosis
[41]. No other published data are available on the relationships between MSGB features and the response to immunosuppressants in patients with SS [42, 43].
MSGB is a key tool for the diagnosis of SS. In everyday
practice, many of the useful histopathological features are
underevaluated. Given the major role for MSGB findings in
the classification and outcome prediction of patients with
SS, standardization is required. Reliability was good in our
study for most of the main MSGB features after a training
session. Further effort is needed to standardize the evaluation of lymphocytic infiltrates and the FS, particularly for
monitoring the response of SS to immunosuppressant
therapy.
Acknowledgements
We thank the French rheumatologists and pathologists
who referred their patients and biopsies to the TEARS
subcohort. We are grateful to Marie Jezequel (Clinical
Investigation Centre) for centralizing the biopsies. We
thank the staff and particularly the pathology department
technicians (Brest Teaching Hospital) for the time and
dedication they gave to this study. Rituximab for the
TEARS study was donated by Roche (Boulogne
Billancourt, France).
Funding: No specific funding was received from any funding bodies in the public, commercial or not-for-profit sectors to carry out the work described in this article.
Disclosure statement: The authors have declared no
conflicts of interest.
References
1 Chisholm DM, Mason DK. Labial salivary gland biopsy in
Sjögren’s disease. J Clin Pathol 1968;21:656–60.
2 Chisholm DM, Waterhouse JP, Mason DK. Lymphocytic
sialadenitis in the major and minor glands: a correlation in
postmortem subjects. J Clin Pathol 1970;23:690–4.
3 Tzioufas AG, Vlachoyiannopoulos PG. Sjogren’s syndrome: an update on clinical, basic and diagnostic therapeutic aspects. J Autoimmun 2012;39:1–3.
4 Christodoulou MI, Kapsogeorgou EK, Moutsopoulos HM.
Characteristics of the minor salivary gland infiltrates in
Sjögren’s syndrome. J Autoimmun 2010;34:400–7.
5 Kapsogeorgou EK, Christodoulou MI, Panagiotakos DB
et al. Minor salivary gland inflammatory lesions in Sjogren
www.rheumatology.oxfordjournals.org
Minor salivary gland biopsy reliability in SS
syndrome: do they evolve? J Rheumatol 2013;40:
1566–71.
6 Vitali C. Classification criteria for Sjögren’s syndrome: a
revised version of the European criteria proposed by the
American-European Consensus Group. Ann Rheumat Dis
2002;61:554–8.
7 Shiboski S. American College of Rheumatology
Classification Criteria for Sjögren’s syndrome: a datadriven, expert consensus approach in the Sjögren’s
International Collaborative Clinical Alliance Cohort.
Arthritis Care Res 2012;64:475–87.
8 Gálvez J, Sáiz E, López P et al. Diagnostic evaluation and
classification criteria in Sjögren’s syndrome. Joint Bone
Spine 2009;76:44–9.
9 Alegria GC, Guellec D, Devauchelle-Pensec V, Saraux A.
Is there specific neurological disorders of primary
Sjögren’s syndrome? Joint Bone Spine 2014, Jun 20. pii:
S1297-319X(14)00104-3. doi: 10.1016/
j.jbspin.2014.04.002 [Epub ahead of print].
10 Daniels TE, Silverman S Jr, Michalski JP et al. The oral
component of Sjögren’s syndrome. Oral Surg Oral Med
Oral Pathol 1975;39:875–85.
11 Goules AV, Tzioufas AG, Moutsopoulos HM. Classification
criteria of Sjögren’s syndrome. J Autoimmun 2014;48–49:
42–5.
12 Daniels TE. Labial salivary gland biopsy in Sjögren’s syndrome. Assessment as a diagnostic criterion in 362 suspected cases. Arthritis Rheum 1984;27:147–56.
13 Greenspan JS, Daniels TE, Talal N, Sylvester RA. The
histopathology of Sjögren’s syndrome in labial salivary
gland biopsies. Oral Surg Oral Med Oral Pathol 1974;37:
217–29.
14 Tarpley TM Jr, Anderson LG, White CL. Minor salivary
gland involvement in Sjögren’s syndrome. Oral Surg Oral
Med Oral Pathol 1974;37:64–74.
15 Le Charpentier Y, Auriol M, Boutin TH et al.
[Histopathologic lesions of the accessory salivary glands
in Gougerot-Sjögren syndrome: re-evaluation of the
diagnostic criteria of Chisholm and Mason and of
Chometter et al]. Rev Stomatol Chir Maxillofac 1994;95:
207–12.
16 Jonsson R, Kroneld U, Tarkowski A. Histological and
functional features of salivary glands in rheumatic patients
with oral sicca symptoms. Scand J Rheumatol 1988;17:
387–91.
17 Daniels TE, Whitcher JP. Association of patterns of labial
salivary gland inflammation with keratoconjunctivitis
SICCA. Arthritis Rheum 1994;37:869–77.
18 Daniels TE, Cox D, Shiboski CH et al. Associations
between salivary gland histopathologic diagnoses and
phenotypic features of Sjögren’s syndrome among
1,726 registry participants. Arthritis Rheum 2011;63:
2021–30.
19 Tavoni AG, Baldini C, Bencivelli W et al. Minor salivary
gland biopsy and Sjögren’s syndrome: comparative analysis of biopsies among different Italian rheumatologic
centers. Clin Exp Rheumatol 2012;30:929–33.
20 Vivino FB, Gala I, Hermann GA. Change in final diagnosis
on second evaluation of labial minor salivary gland biopsies. J Rheumatol 2002;29:938–44.
www.rheumatology.oxfordjournals.org
21 Le Pottier L, Devauchelle V, Fautrel A et al. Ectopic germinal centers are rare in Sjogren’s syndrome salivary
glands and do not exclude autoreactive B cells. J Immunol
2009;182:3540–7.
22 Bombardieri M, Barone F, Humby F et al. Activationinduced cytidine deaminase expression in follicular dendritic cell networks and interfollicular large B cells supports functionality of ectopic lymphoid neogenesis in
autoimmune sialoadenitis and MALT lymphoma in
Sjögren’s syndrome. J Immunol 2007;179:4929–38.
23 Theander E, Vasaitis L, Baecklund E et al. Lymphoid organisation in labial salivary gland biopsies is a possible
predictor for the development of malignant lymphoma in
primary Sjögren’s syndrome. Ann Rheum Dis 2011;70:
1363–8.
24 Guellec D, Cornec D, Jousse-Joulin S et al. Diagnostic
value of labial minor salivary gland biopsy for Sjögren’s
syndrome: a systematic review. Autoimmun Rev 2013;12:
416–20.
25 De Wilde PC, Baak JP, van Houwelingen JC, Kater L,
Slootweg PJ. Morphometric study of histological changes
in sublabial salivary glands due to aging process. J Clin
Pathol 1986;39:406–17.
26 Radfar L, Kleiner DE, Fox PC, Pillemer SR. Prevalence and
clinical significance of lymphocytic foci in minor salivary
glands of healthy volunteers. Arthritis Care Res 2002;47:
520–4.
27 Devauchelle-Pensec V, Mariette X, Jousse-Joulin S et al.
Treatment of primary Sjögren syndrome with rituximab: a
randomized trial. Ann Intern Med 2014;160:233–42.
28 Jonsson MV, Skarstein K, Jonsson R, Brun JG.
Serological implications of germinal center-like structures
in primary Sjögren’s syndrome. J Rheumatol 2007;34:
2044–9.
29 Fleiss JL, Levin B, Paik MC. Statistical Methods for Rates
and Proportions. Hoboken, NJ, USA: John Wiley & Sons,
2013.
30 Altman DG, Bland JM. Measurement in medicine: the
analysis of method comparison studies. Statistician 1983;
32:307.
31 Al-Hashimi I, Wright JM, Cooley CA, Nunn ME.
Reproducibility of biopsy grade in Sjögren’s syndrome.
J Oral Pathol Med 2001;30:408–12.
32 Vitali C, Carotti M, Salaffi F. Is it the time to adopt salivary
gland ultrasonography as an alternative diagnostic tool for
the classification of patients with Sjögren’s syndrome?
Comment on the article by Cornec et al. Arthritis Rheum
2013;65:1950.
33 Risselada AP, Looije MF, Kruize AA, Bijlsma JWJ, van
Roon JAG. The role of ectopic germinal centers in the
immunopathology of primary Sjögren’s syndrome: a systematic review. Semin Arthritis Rheum 2013;42:368–76.
34 Daridon C, Pers J-O, Devauchelle V et al. Identification of
transitional type II B cells in the salivary glands of patients
with Sjögren’s syndrome. Arthritis Rheum 2006;54:
2280–8.
35 Szodoray P, Alex P, Jonsson MV et al. Distinct profiles of
Sjögren’s syndrome patients with ectopic salivary gland
germinal centers revealed by serum cytokines and BAFF.
Clin Immunol 2005;117:168–76.
1063
Sebastian Costa et al.
36 Zuckerman NS, Hazanov H, Barak M et al. Somatic
hypermutation and antigen-driven selection of B cells are
altered in autoimmune diseases. J Autoimmun 2010;35:
325–35.
inflammation and damage in minor salivary gland biopsy
specimens from patients with primary Sjögren’s syndrome
but not patients with sicca. Arthritis Rheum 2011;63:
2014–20.
37 Leroy JP, Pennec YL, Letoux G, Youinou P. Lymphocytic
infiltration of salivary ducts: a histopathologic lesion specific for primary Sjögren’s syndrome? Arthritis Rheum
1992;35:481–2.
41 Nakayamada S, Fujimoto T, Nonomura A et al. Usefulness
of initial histological features for stratifying Sjögren’s syndrome responders to mizoribine therapy. Rheumatology
2009;48:1279–82.
38 Leroy JP, Pennec YL, Soulier C et al. Follow up study of
labial salivary gland lesions in primary Sjögren’s syndrome. Ann Rheum Dis 1992;51:777–80.
42 Hamza N, Bootsma H, Yuvaraj S et al. Persistence
of immunoglobulin-producing cells in parotid salivary
glands of patients with primary Sjögren’s syndrome
after B cell depletion therapy. Ann Rheum Dis 2012;71:
1881–7.
39 Jonsson R, Kroneld U, Bäckman K, Magnusson B,
Tarkowski A. Progression of sialadenitis in Sjögren’s syndrome. Br J Rheumatol 1993;32:578–81.
40 Bookman AAM, Shen H, Cook RJ et al. Whole stimulated
salivary flow: correlation with the pathology of
1064
43 Cornec D, Jamin C, Pers J-O. Sjögren’s syndrome: Where
do we stand, and where shall we go? J Autoimmun 2014;
51:109–14.
www.rheumatology.oxfordjournals.org