Chronic Sclerosing Sialadenitis of the Submandibular Gland and its Histopathological Spectrum in the IgG4-Related Disease: a Series of 17 Cases

Abstract

Purpose

This study aimed to characterize the histopathological immunohistochemical features of chronic sclerosing sialadenitis, emphasizing the IgG4-related disease.

Methods

Seventeen cases of chronic sclerosing sialoadenitis were examined for histopathological aspects, (inflammation, fibrosis, glandular parenchyma, and lymphoid follicles) and immunohistochemistry (BCL2, CD3, CD20, CD34, CD163, p63, cyclin D1, mast cell, SMA, S100A4, IgG, and IgG4) which were scored. IgG4-related disease features were investigated. Demographic and clinical data were also collected.

Results

Males predominated (10:7), with an average lesion size of 3.9 cm. Common histopathological findings included reduced acinar parenchyma, lymphoid follicle formation, and ductular proliferation. CD3-positive T lymphocytes and CD34- and SMA-positive stromal fibroblasts were abundant. Nine cases (53%) showed sialoliths and three cases met the criteria for IgG4-related disease.

Conclusion

CSS of the submandibular gland represents a reactive pattern rather than IgG4-RD as only 3 cases seemed to be related to IgG4-RD. The immunohistochemical profile revealed an abundant population of CD3-positive T lymphocytes, as opposed to regulatory proteins such as cyclin D1, demonstrating that populations of CD34- and SMA-positive stromal fibroblasts contribute to the fibrosis characteristic of CSS. In addition, our results provide a comprehensive insight into the study of CSS and its relationship with IgG4-RD.

Supplementary Information

The online version contains supplementary material available at 10.1007/s12105-024-01651-4.

Introduction

Chronic sclerosing sialadenitis (CSS), or Küttner’s tumor, is a fibroinflammatory disease that occurs almost exclusively in the submandibular gland. It is the most common condition affecting this anatomical site [1–3]. CSS manifests as a uni- or bilateral hard swelling [2, 4] that cannot be clinically differentiated from a neoplasm [4, 5]. The condition is more common among middle-aged adults, and there is a discrete male predilection [6].

CSS is routinely associated with sialoliths at a frequency that ranges from 29–83% [6–8]. Sialoliths have been indicated as the etiological factor of CSS and are a pathological characteristic of the disease [8]. CSS was described for the first time by Küttner in 1896 [7]. However, since it is a fibroinflammatory disease, “chronic sclerosing sialadenitis” instead of “Küttner’s tumor” became the most appropriate nomenclature [9].

Histopathologically, CSS (Küttner’s tumor) was classified by Seifert in 1992 [10] into four stages according to the degree of severity of the inflammatory components, fibrosis, and lymphoid follicle formation. However, CSS is generally characterized by a preserved lobular architecture, interlobular fibrosis, marked lymphoplasmacytic inflammation, lymphoid follicle formation, acinar atrophy, obliterative phlebitis, and the absence of lymphoepithelial lesions [5, 6, 11, 12]. By the end of the 1990s, CSS started to be associated with idiopathic retroperitoneal fibrosis [13], sclerosing cholangitis [14], and sclerosing pancreatitis [15, 16].

In addition to the histopathological similarity of these diseases with CSS, sclerosing pancreatitis is also characterized by high levels of IgG4 [15, 16]. Thus, CSS was considered a probable manifestation of IgG4-related disease (IgG4-RD) [3, 6, 17, 18], which also involves different tissues and organs and causes swelling [15, 19–21], especially in the pancreas [15, 21]. In the head and neck region, orbital structures and lacrimal and submandibular glands are the most affected areas, respectively [21]; lymphadenopathy and swellings may also be observed [19, 21].

Therefore, this study provides a detailed histopathological description of 17 cases of CSS highlighting these characteristics in the context of IgG4-RD. Although histopathological findings alone are not sufficient to define IgG4-RD, they may draw pathologists’ and clinicians’ attention to its differential diagnosis, as these alterations are still unknown in Brazil. In addition, an immunohistochemical profile of lymphoid cells and stromal tissue are shown.

Materials and Methods

Sample Characterization

After approval of the study by the Ethics Committee of the School of Dentistry, Federal University of Bahia (FO-UFBA) (CAAE: 45701121.6.0000.5024, Opinion number 4.834.624), 17 cases of CSS were diagnosed according to the criteria proposed by Seifert [10]. The cases were selected among cases with previous histopathological diagnoses of calculous sialadenitis, chronic sialadenitis, and CSS of the submandibular gland, stored in the archives of the surgical pathology laboratories of FO-UFBA and the collaborating institutions State University of Feira de Santana (UEFS), Federal University of Minas Gerais (UFMG), and State University of Campinas (School of Medicine and School of Dentistry), as well as other affiliated laboratories. All hematoxylin/eosin-stained histological slides were evaluated by two examiners under a light microscope. Clinical data such as sex, age, affected submandibular gland, size, and clinical suspicions were also collected. This is a cross-sectional cohort study that performed a descriptive and exploratory analysis of a series of cases. Written informed consent was obtained from each subject following the Declaration of Helsinki.

Histopathological Study

This study sought to identify the four stages of CSS based on the following histopathological characteristics described by Seifert [10]: focal chronic inflammation, foci of lymphocytic infiltration around the salivary ducts, a diffuse lymphocytic infiltrate, periductal fibrosis, epimyoepithelial islands, reduced secretory parenchyma, secondary lymphoid follicles with reactive germinal centers, ductular proliferation, and glandular architecture with cirrhotic appearance. Other pertinent histopathological features were also analyzed: lobular remnants of serous acini with eosinophilic granules, remnants of mucous acini, vascular congestion, fat deposits, ductal metaplasia, reactive atypia in duct remnants, primary lymphoid follicles, neutrophilic infiltrate causing acinar and ductal injury, ductal ectasia, periductal concentric fibrocellular arrangement, and perineural lymphoplasmacytic infiltration (LPI). The presence of the histopathological features was classified as follows: absent (-), mild (+), moderate (++), and abundant (+++).

Immunohistochemical Technique

Immunohistochemistry was performed for BCL2, CD3, CD20, CD34, CD163, p63, cyclin D1, mast cell, SMA 1A4, S100A4, IgG, and IgG4. Supplementary Table 1 shows the specifications after the standardization of the steps and reagents. First, the sections were deparaffinized in xylene and rehydrated in a decreasing alcohol series. Endogenous peroxidase was blocked by incubation in 3% hydrogen peroxide for 10 min (2x). Antigen retrieval for exposure of the epitopes was performed in moist heat by incubating the sections in a water bath for 20 min with the substances described in Supplementary Table 1.

The sections were incubated with the primary antibodies overnight (18 h) at 4ºC and then subjected to the HRP polymer amplification system (Cell Marque, Rocklin, CA, USA), as described by the manufacturer. The reactions were developed with 3,3-diaminobenzidine (Dako, Carpinteria, CA, USA) as chromogen for 2 min in a dark chamber. After dehydration in alcohol and clearing in xylene, the slides were counterstained with Harris hematoxylin for 1 min and mounted. The positive controls are listed in Supplementary Table 1.

Immunohistochemical Analysis

The slides were evaluated by an experienced examiner under an Axio Lab.A1 virtual digital microscope (Zeiss, Germany, 2008) at 400x magnification. The images were captured with a digital camera (AxioCam ICc3, Zeiss, Germany, 2008) and transferred to a video monitor by a computer system. For the evaluation of IgG4 and IgG antibodies, three high-power fields (HPF) of IgG4-positive plasma cells were selected at a final magnification of 400x to obtain the mean number of positive cells in the selected fields using the Image J program (National Institutes of Health), according to the criteria of Umehara [22]. For the other antibodies, all cases were classified by assigning the following scores according to the proportion of staining, based on the criteria described by Maccagno [23]: absent (-), mild (+), moderate (++), and abundant (+++).

Diagnostic Criteria for IgG4-RD

The criteria proposed by Umehara [22] were used: (1) diffuse or localized edema in one or more organs or a mass or nodule characteristic of IgG4-RD; (2) serum IgG4 levels > 135 mg/dL, and (3) positive for two of the following three criteria: (a) dense LPI with fibrosis, (b) mean number of IgG4-positive plasma cells per HPF > 10 and an IgG4+/IgG + ratio > 40% and (c) typical tissue fibrosis, particularly storiform fibrosis or obliterative phlebitis. According to the cited authors, the presence of items 1 and 2 defines the possible diagnosis of IgG4-RD, the presence of items 1 and 3 the probable diagnosis of IgG4-RD, and the presence of items 1, 2 and 3 the definitive diagnosis of IgG4-RD.

Statistical Analysis

Statistical analysis was performed using the Minitab 14® software (Minitab Inc., Pennsylvania, USA). Data were submitted to descriptive analysis (mean, median, and standard deviation). Parametric and non-parametric tests were applied according to the distribution found and sample size. The Wilcoxon test was used to compare continuous variables between groups. A level of significance of 5% was adopted.

Results

Clinical and Histopathological Features

Seventeen cases of CSS of the submandibular gland that were submitted to excisional biopsy were selected. The demographic data, affected gland, clinical findings, dimensions, type of biopsy, and diagnostic suspicion are compiled in Table 1. The histopathological features were subdivided into findings necessary for the definition of CSS stages (Table 2) and other additional findings (Table 3).

Table 1.

Clinical features of the cases with chronic sclerosing sialadenitis

Case	Age (years)	Gender	Affected gland	Clinical finding	Dimensions (cm)	Type of biopsy	Diagnostic suspicion
I	38	M	Left submandibular	Swelling	3.5 × 3.0 × 2.5	Excisional	Sialolith
II	62	F	Right submandibular	Swelling	4.5 × 2.3 × 2.0	Excisional	Sialolithiasis
III	41	M	Submandibular (side NS)	Swelling	4.0 × 3.0 × 2.7	Excisional	Sialolith
IV	41	M	Right submandibular	Swelling	3.2 × 2.5 × 2.0	Excisional	Pleomorphic adenoma
V	58	M	Submandibular (side NS)	Swelling	3.0 × 1.8 × 1.8	Excisional	Tumor (NS)
VI	57	M	Submandibular (side NS)	Swelling	4.0 × 2.5 × 2.0	Excisional	Tumor (NS)
VII	58	M	Left submandibular	Swelling	NS	Excisional	NS
VIII	61	F	Right submandibular	Swelling	4.0 × 4.0 × 3.5	Excisional	CSS
IX	69	M	Left submandibular	Swelling	3.8 × 2.2 × 1.7	Excisional	NS
X	61	M	Left submandibular	Swelling	3.0 × 2.5 × 1.5	Excisional	Sialolithiasis
XI	36	M	Left submandibular	Swelling	4.0 × 3.5 × 2.5	Excisional	Sialolithiasis
XII	37	F	Left submandibular	Swelling	5.5 × 3.0 × 2.5	Excisional	Recurrent sialadenitis
XIII	78	F	Right submandibular	Swelling	2.8 × 2.3 × 1.0	Excisional	Chronic sialadenitis and sialolithiasis
XIV	59	F	Right submandibular	Swelling	4.5 × 3.5 × 1.2	Excisional	Chronic sialadenitis and sialolithiasis
XV	72	F	Left submandibular	Swelling	4.0 × 3.5 × 1.8	Excisional	Sialadenitis with lithiasis
XVI	58	F	Right submandibular	Swelling	4.5 × 3.5 × 0.7	Excisional	Malignant neoplasm
XVII	62	M	Right submandibular	Swelling	3.6 × 2.5 × 2.0	Excisional	Sialadenitis

Legend: (F) female; (M) male; (NS) not specified; (CSS) chronic sclerosing sialadenitis

Table 2.

Distribution of the histopathological features used for the classification of the stages of chronic sclerosing sialadenitis (CSS)

Histopathological feature	Absent		Mild		Moderate		Abundant
	n	%	n	%	n	%	n	%
Focal chronic inflammation	3	18%	7	41%	2	12%	5	29%
Foci of lymphocytic infiltration around salivary ducts	1	6%	6	35%	4	24%	6	35%
Diffuse lymphocytic infiltrate	4	24%	6	35%	5	29%	2	12%
Periductal fibrosis	0	0%	8	47%	3	18%	6	35%
Reduced secretory parenchyma	2	12%	1	6%	3	18%	11	64%
Secondary lymphoid follicles with reactive germinal centers	2	12%	7	41%	3	18%	5	29%
Ductular proliferation	1	6%	4	24%	3	18%	9	52%
Glandular architecture with cirrhotic appearance	13	76%	3	18%	1	23%	0	0%
Stages of CSS	Stage 1		Stage 2		Stage 3		Stage 4
	1	6%	2	12%	10	58%	4	24%

Table 3.

Distribution of other additional histopathological features

Histopathological feature	Absent		Mild		Moderate		Abundant
	n	%	n	%	n	%	n	%
Lobular remnants of serous acini with eosinophilic granules	6	35%	7	41%	1	6%	3	18%
Remnants of mucous acini	9	53%	6	35%	2	12%	0	0%
Vascular congestion	0	0%	7	41%	3	18%	7	41%
Fat deposits	1	6%	7	41%	5	29%	4	24%
Ductal metaplasia	12	71%	5	29%	0	0%	0	0%
Reactive atypia in duct remnants	16	94%	1	6%	0	0%	0	0%
Primary lymphoid follicles	1	6%	9	53%	5	29%	2	12%
Neutrophil infiltrate causing acinar and ductal injury	10	59%	6	35%	1	6%	0	0%
Ductal ectasia	2	12%	8	47%	4	23%	3	18%
Periductal concentric fibrocellular arrangement	7	41%	6	35%	2	12%	2	12%
Perineural lymphoplasmacytic infiltration	5	29%	9	53%	3	18%	0	0%

All cases were encapsulated and fibrous septa emerged from the capsule that divided the lesions into lobules. The resemblance with features of the developing salivary gland was observed in most cases (n = 13). Other less frequently observed features were oncocytic duct metaplasia (n = 1), the presence of a sebaceous gland (n = 1); the presence of tortuous secondary lymphoid follicles (n = 1), clustered eosinophilic infiltrate or sometimes surrounding ducts (n = 1), and inflammation around a periductal concentric fibrocellular arrangement (n = 2).

The relevant histopathological and immunohistochemical findings regarding IgG4-RD are described in (Table 4). Furthermore, non-dense LPI was present in some cases. Likewise, phlebitis was not always found in the obliterative form.

Table 4.

Histopathological and immunohistochemical features in the spectrum of IgG4-related disease

Histopathological features Case	OP	SF	Dense LPI	Dense fibrosis	Eosinophilic infiltrate	Sialolith	IgG4 + cells per HPF (n)	IgG + cells per HPF (n)	IgG4+/IgG + ratio (%)	Probable IgG4-RD?
I	-	-	++	+++	-	Y	-	73	-	N
II	+	-	+	+++	-	Y	-	94	-	N
III	+	-	-	+++	-	Y	-	140	-	N
IV	-	-	-	++	-	Y	-	45	-	N
V	+	-	-	+	-	N	-	-	-	N
VI	++	-	-	+++	-	N	-	52	-	N
VII	++	+	++	++	-	N	-	77	-	Y
VIII	-	-	++	+	-	Y	15	59	25	N
IX	+	++	++	+++	-	N	96	222	43	Y
X	+	-	-	++	-	N	-	-	-	N
XI	++	-	+++	+++	-	N	26	214	12	N
XII	-	-	-	+	++	N	-	185	-	N
XIII	+	-	-	+++	-	N	-	86	-	N
XIV	++	+	+++	++	-	Y	-	103	-	Y
XV	-	-	++	++	-	Y	-	55	-	N
XVI	+	-	+++	++	-	Y	13	120	11	N
XVII	-	-	+	++	-	Y	31	226	14	N

The table is based on the study by Peuraharju et al. (2019) and the criteria of Umehara et al. (2020). Legend: (OP) obliterative phlebitis; (SF) storiform fibrosis; (LPI) lymphoplasmacytic infiltration; (HPF) high-power field; (IgG4-RD) IgG4-related disease; (-) absent; (+) mild presence; (++) moderate presence; (+++) abundant presence; (Y) yes; (N) no

Immunohistochemical Features

Regarding IgG and IgG4 immunostaining, plasma cells were the predominantly stained cells. Three cases met the criteria for a probable diagnosis of IgG4-RD (Table 4) described by Umehara [22]. Analysis of correlations between the antibodies used (Wilcoxon test) showed positive correlations between CD34 and SMA1A4 (p = 0.00) and between CD34 and SMA1A4 (p = 0.02), as well as a negative correlation between CD34 and mast cell antibodies (p = 0.02). The other results are described in Table 5.

Table 5.

Distribution of immunostaining according to antibody

Antibody distribution		BCL2	CD3	CD20	CD34	CD163	Cyclin D1	Mast cell	p63	SMA1A4	S100A4	IgG4	IgG
Absent	n	0	0	0	0	0	9	0	0			12	2
	%	0%	0%	0%	0%	0%	53%	0%	0%	0%	0%	71%	12%
Mild	n	6	3	3	0	8	7	10	14			Mean ± SD (range)
	%	35%	18%	18%	0%	47%	41%	59%	82%	0%	0%
Moderate	n	4	3	6	2	5	1	5	3
	%	24%	18%	35%	12%	29%	6%	29%	18%	0%	0%	37 ± 34 (13 to 102)	116 ± 65 (45 to 226)
Abundant	n	7	11	8	15	4	0	2	0
	%	41%	64%	47%	88%	24%	0%	12%	0%	6%	6%

Immunostaining for CD20 was detected in the following structures: lymphoid follicles, B lymphocytes around ducts, focal clusters, and primary follicles. CD3 staining was observed in lymphocytes of primary and secondary lymphoid follicles, around blood vessels and glandular ducts, as well as in focal lymphocytes and lymphocyte clusters. Regarding BCL2, staining was found around germinal centers and focal areas of a lymphocytic inflammatory infiltrate, as well as in plasma cells, inflammatory cells around periductal fibrosis, and primary lymphoid follicles. Cyclin D1 was detected only in some ductal cells, while p63 was mainly expressed in basal cells of ductal and acinar remnants, as well as excretory ducts.

CD163 antibody was detected in focal clusters, including septal fibrosis, around nerves, and in follicular centers, highlighting the only case classified as potentially related to IgG4-RD. Structures exhibiting CD34 immunostaining included vessels inside germinal centers; staining was also noted around follicles and in septal fibrosis, as well as in fibroblasts and collagen fibers around ducts, lymphoid follicles, and around periductal hyalinization. The main sites of mast cell immunostaining were around vessels and ducts, near inflammatory processes, in the mantle zone and inside germinal centers, and sometimes in areas of septal fibrosis, inside the ductal epithelium, and around the periductal concentric fibrocellular arrangement.

Immunostaining for SMA was observed in some fibroblasts of septal fibrosis, in myoepithelial cells of glandular ducts and acinar remnants, in blood vessels, and around periductal fibrosis/hyalinization and surrounding periductal concentric fibrocellular arrangement. S100A4 was detected in stromal areas, intralobular fibroblasts, periductal fibrosis, and septal fibrosis, around vessels and ducts, and in the periductal concentric fibrocellular arrangement.

The mean number of IgG4-positive cells was 37 ± 34 (range 13 to 102), while the mean number of IgG-positive cells was 116 ± 65 (range 45 to 226). The mean IgG4/IgG ratio was 21 ± 13% (range 11 to 46%). Finally, Supplementary Table 2 shows a survey of studies on CSS that reported at least 10 cases in the submandibular gland, highlighting the main histopathological features and those of the IgG4-RD spectrum. Figures 1 and 2, and 3 provide images of the main histopathological and immunohistochemical findings.

Fig. 1.

General histopathological features of chronic sclerosing sialadenitis. (A) Lobules of salivary gland parenchyma intersected by thick fibrous septa displaying a cirrhotic feature. Note the branching glandular ducts, absence of acini, and accumulation of fat at the periphery of the lobules. (B) Lobules of salivary gland parenchyma intersected by fibrous septa of varying thickness. Note the slightly dilated central excretory duct and mild periductal fibrosis. (C) Salivary gland parenchyma consisting of interconnecting and branching glandular ducts, lymphoid follicles, and chronic inflammation. Note the absence of acini and the fibrous septum on the right erased by inflammation. (D) Fibrous septa of marked thickness. (E) Marked replacement of glandular parenchyma showing lymphoplasmacytic inflammation and lymphoid follicle formation. (F) Intercalated duct-like small glandular ducts, mild to moderate inflammation rich in plasma cells, and absence of acini. (G) Nodular structure without acini showing a glandular duct surrounded by a concentric fibrocellular arrangement that is delimited by a dense lymphocyte infiltrate simulating a primary lymphoid follicle. (H) and (I) Glandular ducts formed by a double layer of luminal and extraluminal cells and surrounded by a hyaline matrix conferring a concentric outline. Note the mild to moderate inflammation rich in lymphocytes and absence of acini. (J) Thick fibrous septum limiting a lobule composed of ductal parenchyma and marked lymphoplasmacytic inflammation. K) Note the marked chronic inflammation composed of lymphocytes and plasma cells, as well as remnants of serous acini and mucous semi-moons. L) Marked lymphoplasmacytic inflammation and glandular ducts surrounded by hyaline lines. M) Marked perineural inflammation composed of plasma cells. N) Cystically dilated excretory duct exhibiting pseudostratified columnar luminal epithelium and progressive squamous metaplasia. Note the mild chronic inflammation and absence of acini. O) Slightly dilated excretory duct that is surrounded by lymphoid follicle formations with prominent germinal centers. Note the absence of acini

Fig. 2.

Histopathological features of chronic sclerosing sialadenitis within the spectrum of IgG4-related disease. (A) Salivary gland lobules are separated by fibrous and hypercellular stroma composed of lymphoid follicle formations with prominent, sometimes tortuous, germinal centers and glandular atrophy. (B) Detail of the previous image showing a lymphoid follicle with a well-developed germinal center surrounded by chronic inflammation, acinar atrophy, and glandular duct remnants. Note the striated duct. (C) Tortuous lymphoid follicles with well-developed germinal centers conferring a pseudolymphomatous pattern. Note the parallel glandular duct remnants and mild fibrosis. (D) Lobules are separated by fibrocellular stroma composed of lymphoid follicles, sometimes with irregular contours, glandular atrophy, and chronic inflammation. (E) A band of fibrocellular stroma that limits the chronic inflammation replacing the glandular parenchyma on the left and storiform fibrosis on the right. (F) Storiform fibrocellular pattern with a pseudoneoplastic appearance. (G) Stroma composed of fibroblasts and parallel collagen fibers. Note the pseudonodular arrangement surrounding an obliterated blood vessel. (H) Detail of the previous image showing an obliterated blood vessel resulting from fibrosis and periductal lymphoplasmacytic inflammation

Fig. 3.

Immunohistochemical features of chronic sclerosing sialadenitis. (A) BCL2 negativity in germinal centers and positivity in mantle cells. (B) Numerous CD3-positive cells, especially around germinal centers and surrounding glandular ducts. (C) CD20-positive cells especially in germinal centers. (D) Fibrocellular stromal tissue that separates nodules of well-developed tortuous lymphoid follicles with glandular atrophy was negative for CD34 but positive around glandular duct remnants (case with characteristics of IgG4-related disease). (E) Fibrocellular stromal tissue that separates nodules of tortuous lymphoid follicles with CD163-positive macrophages. Note the presence of some of these cells adjacent to glandular duct remnants (case with characteristics of IgG4-related disease). (F) p63-positive basal cell nuclei of glandular ducts. Note the periductal hyaline fibrosis on the right. (G) Cyclin D1-positive ductal cells. (H) Mast cells in focal areas and surrounding periductal fibrohyaline tissue. (I) Nodules of lymphoid follicles surrounded by fibrocellular stroma positive for SMA (case with characteristics of IgG4-related disease). (J) Protein S100A4 positivity in stromal fibroblasts and around glandular ducts. L) IgG-positive cells (case with characteristics of IgG4-related disease). M) IgG-positive cells (case with characteristics of IgG4-related disease)

Discussion

The results of the present study showed 17 cases of CSS of the submandibular gland. Three cases (VII, IX, and XIV) met the histopathological criteria proposed by Umehara [22] for the diagnosis of probable IgG4-RD. It is possible that positive-IgG4 cells do not always represent an association with IgG4-RD. However, IgG4-RD should be investigated regarding the histopathological findings that can be found in the CSS. Even regarding clinical and laboratory, histopathological, and immunohistochemical data, its diagnosis continues to be difficult. Despite these aspects, although serum IgG4 levels are elevated in most patients diagnosed with IgG4-RD, about 40% of these patients have normal serum levels [24]. Taken together, all these aspects show that CSS in the context of IgG4-RD is complex.

The present study showed a male predilection (59% of cases) and the mean age was 55 years, as also reported by other authors [2, 6, 17, 25–27]. There was unilateral involvement of the submandibular gland [28]. The mean size was 3.9±5.5 cm, in agreement with the findings of Kitagawa [17].

Regarding the rarity of CSS associated with IgG4-RD demonstrated the features of the condition reported in the present study contribute to attracting the attention of clinicians and pathologists. Despite the criteria for this condition, some authors have used the Boston consensus statement criteria [19] which considers the same ratio but with a number of IgG4-positive cells per HPF greater than 100. It is important to highlight that the results vary markedly. Maccagno et al [23] used this criterion, obtaining a density of at least 50 IgG4-positive plasma cells per HPF, although others have adopted different criteria [17, 29]. Kitagawa [17] observed 100.2 to 323.6 IgG-positive cells per HPF, while the number of IgG4-positive plasma cells ranged from 62.2 to 232.4 per HPF in five HPF. Li et al [29] found 167 ± 42 IgG-positive plasma cells (range 79 to 246) and 124 ± 42 IgG4-positive plasma cells (range 56 to 215) per HPF; the IgG4/IgG ratio was 73.7 ± 12.8% (range 50.4 to 93.9%). In our study, the IgG4/IgG ratio could be calculated in five cases, and the mean ratio was 21 ± 13% (range 11 to 46%).

It is important to state that many clinical, imaging, and laboratory data were not available, which might be a limitation of this study. However, other authors have focused exclusively on histopathological features [23]. Recently, the European Alliance of Associations for Rheumatology (EULAR) and the American College of Rheumatology adopted an extensive list of criteria for a case to be potentially considered IgG4-RD, including a compiled score ≥ 20 to confirm IgG4-RD [28]. On the other hand, IgG4-RD may be diagnosed in the absence of any classic histopathological or immunohistochemical characteristics of the disease. Thus, the characterization of IgG4-RD is complex since it requires a large amount of patient information, which may only be achievable in large hospital centers.

Application of the criteria of Wallace et al. [28] identified five cases (cases VII, IX, XI, XVI, and XVII) in our study that could be diagnosed as potential lesions associated with IgG4-RD. The diagnosis of IgG4-RD could not be ruled out in the remaining cases, and further clinical, imaging, and serological data are necessary. Furthermore, the global variation in the results found, including those of the present study, suggests that genetic and regional factors may somehow influence the manifestation of IgG4-RD, at least regarding the involvement of the submandibular glands.

The cases described here had lesions characterized by preserved salivary gland lobules intersected by fibrous and sclerosing septa of varying thickness, although a cirrhotic appearance was also observed. Reduced acinar parenchyma, foci of periductal lymphocyte and plasma cell infiltration, and lymphoid follicle formation were the predominant findings, as also described by previous authors [12, 17, 25, 30–33]. However, the individual histological features often varied from case to case and also within the same case.

Although it is clear that CSS indeed exhibits varying levels of severity, it is difficult to affirm the need for categorizing the condition into these stages, even when considering its main differential diagnoses, since they do not influence surgical resection. We observed that stage 3 corresponded to 58% of cases. Among the 17 cases of CSS, 53% were associated with the presence of sialoliths. However, its frequency ranges from 29–83% [7, 8]. Some authors attribute sialoliths to the etiopathogenesis of CSS [8, 27], while others suggest an immunoinflammatory origin or attribute them to disturbances of salivary secretion [7, 10]. Although morphological features do not appear to differ between patients with and without sialoliths, only one case that met the criteria for IgG4-RD had sialoliths.

The main histopathological findings of IgG4-RD are dense LPI rich in IgG4-positive plasma cells, obliterative phlebitis, storiform fibrosis, and, frequently, elevated serum IgG4 levels [15, 19, 22, 23, 34]. However, these levels can be normal in a considerable proportion of cases [24] and should, therefore, not be used alone as a determinant for the diagnosis of IgG4-RD [9]. We detected obliterative phlebitis, storiform fibrosis, and dense LPI in 65%, 18%, and 59% of cases, respectively. The simultaneous presence of these histological features was observed in only three cases, all of them classified as probable manifestations of IgG4-RD. The frequency of storiform fibrosis observed in our study was lower than that reported by Peuraharju et al. [27] and Maccagno et al. [23] who found this condition in 58% and 100% of cases, respectively. On the other hand, the frequency of obliterative phlebitis was similar to the 50% reported by Maccagno et al. [23] In contrast, Peuraharju et al. [27] identified obliterative phlebitis in 13% of cases. In addition, the frequency of dense LPI was lower in our study when compared to that reported by Peuraharju et al. [27] who observed this feature in 85% of cases. Therefore, all histopathological features observed in this study, in addition to the IgG/IGg4 ratio, allow us to suggest that CSS is a fibroinflammatory disease that may share characteristics originally described for Küttner’s tumor and that is also associated with IgG4-RD. Features such as obliterative phlebitis, storiform fibrosis, and dense LPI are evident in the latter but are not as common in the former.

The mechanisms that could explain interlobular fibrosis and sclerosis and concentric fibro cellular arrangement and cirrhotic features in CSS have not been explored. However, some authors report that mast cells produce fibrogenic proteins that influence the synthesis of extracellular matrix and myofibroblast differentiation [35]. In addition, TFGfbeta, CD163, and CD34-positive cells could also influence this process of fibrosis [36]. Thus, we studied stromal cells of CSS regarding these proteins and those related to lymphoid components.

In addition to CD34 indicating the marked presence of blood vessels of different calibers, interestingly, intralobular stromal tissue, periductal fibrosis and hyalinization, and septal and intralobular fibroblasts were also positive for this protein. In contrast, other authors did not observe CD34-positive fibroblasts in chronic, although not sclerosing, sialadenitis of the submandibular gland [37]. In parallel, SMA-positive fibroblasts were present in these regions but were absent in the intralobular stromal tissue and were less common in fibrous septa. These findings show that, despite the presence of CD34 fibroblasts in fibrous septa, some of them are also SMA and S100A4 positive in this region. These features were exuberant in one of the cases potentially related to IgG4-RD in which simultaneous immunolocalization of CD34 S100A4 and CD163-positive macrophages was observed. There are reports that CD163-positive macrophages are more frequent in IgG4-RD than in Sjögren’s syndrome and chronic sialadenitis caused by sialoliths [38] and are the main factor responsible for fibrosis in IgG4-RD [39]. We detected a moderate to abundant population of macrophages in 53% of the CSS studied, including those potentially related to IgG4-RD. It is important to note that macrophages are a major source of angiogenic factors [40].

One case potentially related to IgG4-RD showed histological similarity of the case with Kimura’s disease. The case reported here is different considering the absence of eosinophils in germinal centers, the absence of regular germinal centers, and the non-remarkable presence of mast cells, as described by Zhu et al. [41]. Staining for mast cells was predominantly mild (59%). However, other authors found an increased number of mast cells in CSS [14, 37]. Mast cells may also contribute to fibrosis [35], as they have also been detected around periductal fibrosis.

The immunohistochemical results showed the absence of BCL2 staining in the follicular germinal centers and positive staining in some periductal lymphocytes, while CD20 B lymphocytes were positive in germinal centers and around ducts, as previously reported [42], and negative for cyclin D1. These findings are suggestive of reactive phenomena and rule out follicular lymphoma [3, 43], MALT lymphoma, and other types of lymphomas. On the other hand, CD3-positive T lymphocytes were more abundant and were located in lymphoid follicles and around blood vessels. Although we did not investigate cytotoxic T lymphocytes, they are numerous in acini and ducts in CSS of the submandibular gland [44].

Cyclin D1 was negative in almost half of the cases, similar to the reports of other authors [45], and the staining was mild in the positive cases. This finding can be attributed to the reactive capacity of ductal cells to fibroinflammatory lesions since cell proliferation is greater in sialadenitis of the parotid gland than in the normal salivary gland [46]. In addition, normal salivary gland basal cells are also positive for p63 [47].

Conclusion

This study showed the histopathological and immunohistochemical spectrum of 17 cases of CSS of the submandibular gland as a result of a reactive pattern rather than IgG4-RD. Their association with IgG4-RD seems to be rare as 3 cases seemed to be related to IgG4-RD. The immunohistochemical profile revealed an abundant population of CD3-positive T lymphocytes, as opposed to regulatory proteins such as cyclin D1, demonstrating that populations of CD34- and SMA-positive stromal fibroblasts contribute to the fibrosis characteristic of CSS. In addition, our results provide a comprehensive insight into the study of CSS and its relationship with IgG4-RD.

Electronic Supplementary Material

Below is the link to the electronic supplementary material.

Author Contributions

Vinícius Rio Verde Melo Muniz: Methodology; validation; investigation; formal analysis; data curation; writing (original draft); visualization; Albina Altemani: Methodology; validation; investigation; formal analysis; data curation; writing (original draft); visualization; Valéria Souza Freitas: Conceptualization; methodology; writing (review & editing); visualization; supervision. Bruno Cunha Pires: Validation; data curation; writing (review & editing). Dandara Andrade de Santana: Validation; methodology; investigation; data curation; writing (review & editing); funding acquisition. Larissa Abbehusen Couto: methodology; investigation; data curation; writing (review & editing); . Maria Cristina Teixeira Cangussu: Visualization; validation; formal analysis; writing (review & editing). Ricardo Santiago Gomez: Visualization; validation; investigation; data curation; writing (review & editing). Suzana Catanhede Orsine Machado de Souza: Visualization; validation; formal analysis; writing (review & editing). Pablo Augustin Vargas: Methodology; investigation; formal analysis; writing (review & editing); Patrícia Ramos Cury: Methodology; investigation; formal analysis; writing (review & editing); Iguaracyra Barreto de Araújo: Validation; investigation; data curation; writing (review & editing). Roberta Rayra Martins Chaves: Methodology; investigation; formal analysis; writing (review & editing); Felipe Paiva Fonseca: Visualization; validation Methodology; investigation; formal analysis; writing (review & editing); Jean Nunes dos Santos: Conceptualization; methodology; investigation; formal analysis; writing (original draft); writing (review & editing); supervision; project administration; funding acquisition.

Funding

This work was supported by the “Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)”; “Coordenação de Aperfeiçoamento de Ensino Superior (CAPES),” and “Fundação de Amparo à Pesquisa do Estado da Bahia (FAPESB).

Data Availability

No datasets were generated or analysed during the current study.

Declarations

Ethical Approval

All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. The study protocol was reviewed and approved by the Ethics Committee of the School of Dentistry, Federal University of Bahia (FO-UFBA) (CAAE: 45701121.6.0000.5024, Opinion number 4.834.624).

Informed Consent

Informed consent was obtained from all individual participants included in this study following the Declaration of Helsinki. The purpose, nature, and potential risks and benefits of participation were explained to each participant, and they were provided with an opportunity to ask questions before providing consent.

Consent for Publication

Consent for publication was obtained for every individual person’s data included in the study.

Conflict of Interest

The authors declare that they have no conflict of interest.