Extranodal MALT Lymphoma in the Oral Cavity: A Series of Three Cases with Review of Literature

Abstract

Background

Diagnosis of MALT lymphoma in the oral cavity is challenging. There is a great overlap in the histopathologic, immuno-histochemical and molecular features of MALT lymphoma with reactive lymphoid proliferations. The literature shows a very few case reports of primary MALT lymphoma of oral cavity.

Methods

We discuss the histopathologic, immuno-histochemical, cytogenetic features, treatment and behavior of 3 cases of primary MALT lymphoma oral cavity along with review of literature.

Results

The age ranged from 40 to 57 years (male to female ratio = 2:1). The sites involved were hard palate, bilateral gingivobuccal sulcus and right buccal mucosa. The most common histology was centrocyte-like (2 cases). Lymphoepithelial lesions were absent. On immunohistochemistry, all tumors showed diffuse strong CD20 and bcl2 expression with strong and diffuse MNDA staining in one case. IgH; MALT1 translocation was not seen in any of these cases. One patient received local radiotherapy, one received steroids; while the case 3 received RCHOP (Rituximab, cyclophosphamide, hydroxydaunorubicin hydrochloride, vincristine and prednisone) chemotherapy. Two patients had complete remission while one had recurrence.

Conclusion

MALT lymphoma of oral cavity shows a wide spectrum of morphology with presence of transformed cells, that may lead to misdiagnosis of DLBL. Treatment guidelines are not well established but a tendency to excise MALT lymphomas of oral cavity has been observed. Nevertheless, MALT lymphoma of oral cavity appears to be an indolent disease.

Introduction

Approximately 48% of non-Hodgkin lymphomas arise in extranodal sites. The head and neck region is the second most common site of extranodal non-Hodgkin lymphoma (NHL) after the gastrointestinal tract. The common extranodal sites of NHLs in the head and neck region include Waldeyer’s ring, palatine tonsil, salivary glands, base of tongue and oropharynx [1]. Primary involvement of the oral cavity by NHL is very rare and constitutes approximately 3–5% of all NHL described in the literature. [1, 2].

The most common non-Hodgkin lymphoma occurring in the oral cavity is diffuse large B cell lymphoma followed by mantle cell lymphoma, extranodal marginal zone lymphoma of mucosa-associated lymphoid tissue (MALT lymphoma) and Burkitt lymphoma [1–3]. The extranodal MALT lymphoma is a low-grade B cell NHL that arises from the B lymphoid cells located in the marginal zone of secondary lymphoid follicles [4]. Following variety of chronic inflammatory stimuli, different sites in the oral cavity can acquire MALT and this proliferation undergoes multiple genetic aberrations leading to the subsequent malignant transformation [5]. Extranodal MALT lymphomas have good behaviour and most respond to localized treatment, either surgery and/ or radiotherapy [6].

The diagnosis of MALT lymphoma in the oral cavity becomes challenging given the overlap in the diagnostic histopathologic, immunohistochemical and molecular features with reactive proliferations [7–10]. Few case reports are seen in the literature describing primary MALT lymphoma of oral cavity [11–13]. We wish to discuss our experience with 3 cases of extranodal MALT lymphoma in the oral cavity with a view to address histopathologic spectrum and behaviour.

Materials and Methods

These three cases were selected through our archives between 2016 to 2021. The clinical details of these were retrieved from electronic medical records. Bulky disease was defined as tumor size more than 7 cm in largest dimension.

Imaging included head and neck computed tomography (CT) scan; Magnetic resonance imaging (MRI) and Positron emission tomography (PET) scans were available in the electronic medical records.

Tissue specimens were fixed in 10% neutral buffered formalin, routinely processed and embedded in paraffin. For light microscopic examination, the four-micron thick sections were stained with haematoxylin and eosin (H&E). Immunohistochemistry was performed using a Ventana automated (Benchmark TM) Stainer and a panel of antibodies included viz. CD20, CD3, CD10, PAX5, Bcl2, Bcl6, CD5, CD3, CD7, CD4, CD8, CD10, CD23, Cyclin D1, IgG, Tdt, GCET, LMO2, EBV-LMP1 and MNDA. (Table 1 shows immunohistochemical stains including their clones and dilutions).

Table 1.

Immunohistochemistry stains

Antibody	Vendor	Clone	Dilution
CD20	Dako	L26	1:400
CD3	Dako	Polyclonal	1:400
Bcl2	Cell Marque	124	1:300
CD10	Roche	SP67	Ready to use
IgG	Cell Marque	Polyclonal	1:2000
CD21	Cell Marque	2G9	1:100
Mib1	Dako	MIB1	1:400
Cyclin D1	Thermo Fisher	SP4	1:200
CD4	Cell Marque	SP35	1:100
CD8	Roche	SP57	Pre-diluted
CD5	Dako	4C7	1:400
CD7	Cell Marque	MRQ56	1:700
Tdt	Cell Marque	Polyclonal	1:100
MNDA	Novus	4H6	Pre-diluted
LMO2	Cell Marque	SP51	1:500
GCET	Thermo Fisher	Polyclonal	1:100
CD138	Dako	M115	1:250
EBV-LMP1	Dako	CS1-4	1:800
Kappa	Cell Marque	L1C1	1:100
Lambda	Cell Marque	Lamb14	1:700

Bcl2 B-cell lymphoma 2, IgG immunoglobulin G, Tdt terminal deoxynucleotidyl transferase, MNDA myeloid cell nuclear differentiation antigen, LMO2 LIM domain only 2, GCET germinal centre expressed transcript, EBV-LMP1 Epstein-Barr virus latent membrane protein

Four-micron formalin fixed; paraffin embedded sections were used for interphase-FISH studies. Centromeric probes (CEP11, CEP18) were used to confirm/exclude aneuploidy. The following probes were used:

1. Metasystems LSI IGH (14q32) break apart probe with interpretation cut off of 10%.

2. Abbott Molecular LSI BIRC3/MALT1: t(11;18) (q21;q21) dual colour fusion translocation probe with interpretation cut off of 8%.

3. Abbott Molecular LSI IGH/BCL2: t(14;18) (q32;q21) dual colour dual fusion translocation probe CEP 11 and CEP 18 were used to confirm aneuploidy. The Interpretation cut off for a positive result was 8%.

Interpretation cut off for calling extra copies and deletions is 15%.

Results

A summary of clinical details of patients is given in Table 2. The age range varied from 40 to 57 years with male to female ratio of 2:1. The sites involved were hard palate, bilateral gingivobuccal sulcus and right buccal mucosa respectively. In one patient (Case 3), Hepatitis B core antigen was positive. No patient gave history of autoimmune disease. There was no history of oral tobacco in any patient. One patient was a chronic smoker. There were no B symptoms in any of the cases. All cases were Stage I Ann Arbor stage. Serum LDH levels were normal in 2 cases, while elevated in one case. All patients had ECOG (Eastern Cooperative Oncology group) performance score of 0/1. In all three cases, liver, spleen and bone marrow were unremarkable as confirmed by PET CT scan. Sub-centimetre regional lymph nodes were involved in two cases.

Table 2.

Clinical, pathologic and imaging features

Case no	Age/gender	Clinical features	Imaging	Histopathologic features	Immunohistochemistry	Cytogenetics	Treatment	Recurrence	Follow-up
1	57/M	Swelling hard palate for 2 years; LDH = 150U/l	MRI: 3.8 × 3.6 × 3.3 cm soft tissue sub-mucosal mass in the left maxilla involving pterygoid fossa and extending into infratemporal fossa; no regional lymph nodes Ann Arbor stage: 1AE	Monocytoid lymphoid cells admixed with scattered large cells	Positive for CD20, Bcl2, MNDA (focal), CD21 highlighted residual germinal centres; Negative for CD10, EBV-LMP1, Mib-1 labelling index ~ 10%; Few plasma cell noted with no light chain restriction	No rearrangement of IGH, IGH/BCL2 and BIRC3/MALT1 One copy of BCL2 and MALT1 was observed in 21% cells	EBRT to left infratemporal fossa to a dose of 36 Gray over 1 month	No	Disease free for 1 year
2	40/F	Swelling in both sides of GBS for 5 months; LDH = 179 U/l	PET scan showed FDG avid soft tissue thickening involving bilateral GBS (largest extent 8 cm) with regional sub-centimetric lymph nodes; Ann Arbor stage: IAEX	Nodular monomorphic infiltrate of centrocyte like lymphoid cells. Scattered large or transformed cells were seen	Positive for CD20, Bcl2, MNDA (strong and diffuse). Negative for CD5, Mib-1 labelling index ~ 10%. Few plasma cell noted, which were polyclonal on kappa & lambda immunohistochemistry	No rearrangement of IGH, IGH/BCL2 and BIRC3/MALT1. (Fig. 5B)	Chemotherapy-6 cycles of Bendamustine & Rituximab	No	Disease free for 6 months
3	54/M	Swelling in the right cheek for 2 months HBc antigen was positive Serum LDH levels = 250.8 U/l	PET-CT: 4.2 × 4 cm soft tissue lesion in the right buccal space involving right maxilla and pterygopalatine fossa with sub-centimetric regional lymph nodes; revised Ann Arbor stage: IAE	Diffuse sheets of centrocyte-like cells with interspersed transformed cells	Positive for CD20, bcl2, IgG, MNDA (focal) Negative for CD10, Tdt, Cyclin D1, LMO2, GCET, CD138, CMYC Mib-1 labelling ~ 20%. A few plasma cell noted, which showed no light chain restriction by immuno histochemistry	No rearrangement of IGH, IGH/BCL2 and BIRC3/MALT1 3–4 copies of IGH, BCL2 and MALT1 were observed in 18% cells (tetrasomy)	6 cycles of R-CHOP chemotherapy followed by IFRT to a dose of 45 Gray over one month	After 4 years of disease-free interval	Re-RT at the dose of 20 cycles each of 36 Gy over 4 weeks Disease free for 2 months

M male, F female, MRI magnetic resonance imaging, PET scan positron emission tomography scan, FGD fluorodeoxyglucose, EBRT external beam radiation therapy, GBS gingivo-buccal sulcus, HBc hepatitis B core antigen, LDH lactate dehydrogenase, PAX5 paired box protein 5, Bcl2 B-cell lymphoma 2, MNDA myeloid cell nuclear differentiation antigen, EBV-LMP1 Epstein-Barr virus latent membrane protein 1, FISH fluorescence in-situ hybridization, IgH immunoglobulin heavy chain, MALT1 mucosa associated lymphoid tissue lymphoma 1, BIRC3 baculoviral IAP repeat containing 3, API2 apoptosis inhibitor 2, Tdt terminal deoxynucleotidyl transferase, LMO2 LIM domain only 2, GCET germinal centre expressed transcript, R-CHOP rituximab, cyclophosphamide, hydroxydaunorubicin hydrochloride, Oncovin, prednisone, IFRT involved field radiation therapy, RT radiotherapy

Pathology Findings (Table 2)

Punch biopsies showed diffuse submucosal infiltrate of atypical small lymphoid cells admixed with scattered large transformed cells. The monocytoid morphology was seen in one case (case1) (Fig. 1A–C), and other two cases showed centrocyte-like cells (cases 2, 3) (Figs. 2A–C and 3B–D). Due to the lack of glandular epithelium, lympho-epithelial lesions could not be demonstrated. On immunohistochemistry, all tumours showed B cells with diffuse strong CD20 (Figs. 1D, 2D, 3E) and bcl2 expression (Fig. 3F). Strong and diffuse MNDA staining was seen in one case (Fig. 2E) and focal (Fig. 1E) in other 2 cases. IgG staining was seen in the background plasma cells. CD5 was negative in all the cases. MIB1 ranged from 10 to 20% (Fig. 2F). CD21 was negative but highlighted residual germinal centres within the infiltrate (Fig. 1F). LEF1, CyclinD1, LMO2, and GCET were negative. CD3 highlighted variable number of reactive T cells in the background. The T cell density varied from dense in case 1 to minimal in case 3. No loss of T cell antigens was seen.

Fig. 1.

(Case1): A Microphotograph showing diffuse infiltrate of atypical lymphoid cells entrapping a nerve (H&E*, × 40). B, C Microphotographs showing diffuse sheets of monocytoid lymphoid cells admixed with scattered transformed cells in a prominent vaso-formative background. (H&E*, × 200, × 400). D Immunohistochemistry microphotograph showing tumor cells to be diffusely positive for CD20. E Immunohistochemistry microphotograph showing tumor cells to be focally (~ 30%) positive for MNDA, F Immunohistochemistry microphotograph showing residual germinal centres as highlighted by CD21 *Hematoxylin and eosin

Fig. 2.

(Case 2): A Microphotograph showing vague nodular infiltrate of atypical lymphoid cells (H&E*, × 40). B, C Microphotographs showing sheets of centrocyte-like cells admixed with few large transformed cells (H&E*, × 400, × 800). D Immunohistochemistry microphotograph showing tumor cells to be diffusely positive for CD20. E Immunohistochemistry microphotograph showing tumor cells to be strongly and diffusely positive for MNDA. F Immunohistochemistry microphotograph showing Mib-1 labelling index of ~ 10%. *Hematoxylin and eosin

Fig. 3.

(Case3): A PET-CT scan image showing soft tissue lesion in right buccal mucosa. B Microphotograph showing diffuse infiltrate of atypical lymphoid cells (H&E*, × 100). C, D Microphotographs showing sheets of centrocyte-like cells admixed with numerus large transformed cells (H&E*, × 400, × 600). E Immunohistochemistry microphotograph showing tumor cells to be diffusely positive for CD20. F Immunohistochemistry microphotograph showing tumor cells to be positive for Bcl2. *Hematoxylin and eosin

Primary biopsy of case 3 showed clusters of large transformed cells giving a false appearance of high-grade tumor (Fig. 3C, D), which led to misdiagnosis as diffuse large cell lymphoma. Recurrent sample was available in this case and showed a lower grade tumor as compared to the primary.

Fluorescent in situ hybridisation (Fig. 4A–C): IgH MALT1 translocation was not seen in any case.

Fig. 4.

A Case1: Interphase-FISH with Abott Molecular IGH/BCL2 dual colour dual fusion probe microphotograph showing one copy of BCL2 (Red signals). B Case2: Interphase-FISH with Abott Molecular IGH/BCL2 dual colour dual fusion probe microphotograph showing two copies of BCL2 (Red signals) and two copies of IGH (Green signals). C Interphase-FISH with Abott Molecular BIRC3 (API2)/MALT1 dual colour dual fusion probe microphotograph showing 2 copies of BIRC3 (API2) (Green signals) and 4 copies of MALT1 (Red signals) indicating tetrasomy 18

Treatment and follow up: As documented in Table 2, there was heterogeneity in patient therapy. One patient received local radiotherapy (36 Gy over 1 month) only, one received steroids; while the case 3 was over diagnosed as diffuse large B cell lymphoma and received RCHOP (Rituximab, cyclophosphamide, hydroxydaunorubicin hydrochloride, vincristine and prednisone) chemotherapy. The first two had complete remission while the case 3 recurred in spite of chemotherapy. This patient had bilateral extensive local disease and it is likely that residual re-growth was the cause for recurrence. All the three patients are alive and disease free during the follow up period.

Discussion

The MALT lymphoma was originally described by Isaacson and Wright in the gastrointestinal tract in 1983 [14]. Mucosa associated lymphoid tissue (MALT) is a specialized lymphoid tissue, that is found in an association with the mucosal surfaces viz. gastrointestinal tract, nasopharynx, oropharynx. The sites that are normally devoid of native MALT such as the stomach, salivary glands, orbit, and thyroid can acquire MALT under various chronic inflammatory conditions like Sjogren syndrome, Helicobacter pylori infection in the stomach, etc. subsequently leading to multiple genetic aberrations [5, 6, 14–17]. Similarly chronic sinusitis, Sjogren’s syndrome, benign lymphoepithelial lesion or myoepithelial sialadenitis have been suggested as precursors of MALT lymphoma in head and neck region [3]. The acquired MALT can undergo malignant transformation at these extra nodal sites [18, 19].

To the best of our knowledge, 26 cases of oral primary MALT lymphoma have been reported so far (Table 3). A wide range of age was seen from 13 to 83 years but its commoner in middle aged adults. A distinct female predominance was seen. The age group in our series was similar but we could not document female predominance, probably due to referral bias. The most common site involved in the literature is hard palate followed by soft palate and buccal mucosa [2, 7, 18–29].

Table 3.

Review of literature

Study	No. of cases	Age/gender	Site	Treatment	Follow-up
Kolokotronis et al. [20]	1	73/F	Hard palate	Surgery	Patient died of disease after 15 months
Shah et al. [18]	1	55/F	Hard palate	Surgery	Disease free after 2 years
Jain et al. [21]	1	49/M	Sub-mucosal lesion in bilateral buccal mucosa with sub-centimetric regional lymph nodes	IFRT	Disease free for 1 year
Tauber [19]	1	71/F	Sub-mucosal mass in hard palate and soft palate	Surgery	Disease free for 4 years
Kojima et al. [7]	7	Age 23–83 years 4 females, 3 males	3 Cases-soft palate 3-Buccal mucosa,1-gingiva In 2 cases, concurrent involvement of orbit and parotid in 2 cases	Surgery alone in 1case, surgery with RT in 1 case, RT alone in 3 cases, CT in 2 cases	Recurrence in 2 cases 5 Cases disease free
Saharan et al. [22]	1	38/F	Soft palate	RT	Disease free
Frazier et al. [23]	1	15/M	Upper lip	Surgery	Disease free for 3 months
Triantafillidou et al. [2]	7	13–44 years 5 Males, 2 females	4 in minor salivary glands in oral cavity, 3 in mucosa of maxilla	Surgery	Alive
Tanaka et al. [24]	1	66/F	Right buccal mucosa	Surgery	Disease free after 1 year
Yonal Hindilerden et al. [25]	1	61/F	Hard palate and right parotid	RCHOP	Responded and disease free
Dunn et al. [26]	1	64/F Scleroderma	Hard palate, parotid, nasopharynx	CT + RT	Disease free for 57 months
Pijpe et al. [27]	1	42/F	Hard palate, parotid	CT	Disease free for 6 months
Sakuma et al. [28]	1	70/F	Hard palate	Spontaneous regression	Disease free for 38 months
Abe et al. [29]	1	64/F	Hard palate	Surgery	No follow up
Present study	3	40 to 57 years with M:F ratio of 2:1. One case was HBc antigen positive	Left palate, bilateral gingivobuccal sulcus, right buccal mucosa	RT in 1 case, CT in 1 case, Combined RT + CT in 1 case	On case showed recurrence All cases disease free

M male, F female, CT chemotherapy, R-CHOP rituximab, cyclophosphamide, hydroxydaunorubicin hydrochloride, oncovin, prednisone, IFRT involved field radiation therapy, RT radiotherapy

Histopathological features of extra nodal MALT lymphoma have been well characterized. They re-capitulate the features seen in benign, acquired MALT tissue [30]. The lymphoid cell morphology can be centrocyte-like or monocytoid admixed with scattered transformed cells. The centrocyte-like cells exhibit oval, reniform or more irregular nuclei and scant cytoplasm. The monocytoid cells exhibit round nuclei and abundant pale staining cytoplasm. Histopathology findings in our study are similar to that already reported in MALT from other sites with few differences [31, 32].

Lympho-epithelial lesions (LEL) are seen in great majority of cases of extra nodal MALT lymphoma. However, lymphoepithelial lesions are not essential for the diagnosis of MALT lymphoma, as they can be seen in reactive infiltrates as well [9]. In our study due to lack of glandular epithelium, assessment of LEL was not possible. Light-chain restricted plasma cells are reported in MALT lymphomas [33] but were polyclonal in our study.

The immunohistochemical profile of MALT lymphoma is not specific. There is expression of B-cell markers viz CD20, CD19, CD22, CD79a, PAX5. The plasma cell component is highlighted by CD138 and MUM1. Bcl2 expression is seen commonly in MALT lymphoma. The Mib-1 proliferation index is usually low in these indolent B cell lymphomas [17].

Myeloid cell nuclear differentiation (MNDA) is a helpful marker in distinguishing MALT lymphoma from follicular lymphoma. It regulates NF-kB signalling pathway [34, 35]. MNDA is used to distinguish nodal marginal zone lymphomas with follicular colonization, from follicular lymphoma with marginal zone differentiation [36]. The relevance of MNDA in MALT lymphoma of oral cavity is not discussed so far in the literature. We have included MNDA in our immunohistochemistry panel and all 3 cases showed staining with a strong and diffuse expression of MNDA in one case.

The differential diagnoses on morphology include benign lymphoid hyperplasia, and other low-grade lymphomas notably follicular lymphoma [8, 9]. Histology of benign lymphoid hyperplasia and low-grade MALT lymphoma overlaps, however, the larger the extent of lymphoid infiltrate, the greater is the likelihood of MALT lymphoma. But in a small punch biopsy this may be difficult to assess [18]. We believe that strong diffuse expression of MNDA will help distinguish MALT lymphoma from reactive lymphoid proliferations. The other differentials include mantle cell lymphoma, which can be distinguished by immunohistochemical expression of cyclin D1 and SOX11. MALT lymphoma may show a striking infiltrate of T lymphoid cells with a T follicular helper phenotype, which raises the possibility of angioimmunoblastic T cell lymphoma [9].

CD10 and bcl6 may help separate MALT lymphoma with follicular colonization from follicular lymphoma. However, there can be upregulation of these germinal centre markers by neoplastic marginal zone cells, when they colonize the germinal centres making the distinction really challenging. In these instances, a careful assessment of the extrafollicular component may be helpful, where MALT lymphoma cells do not express germinal centre markers. Further differentiation can be done by FISH for t (14;18) and t (11;18), though a negative result may not be helpful [9].

On cytogenetics evaluation, t (11; 18) (q21; q21) is the most common chromosomal translocation identified in the MALT lymphoma. [10] API2-MALT1 fusion transcript is not found to be associated with MALT lymphoma of oral cavity [7]. However, t (1; 14) (q21; q21) is found to be strongly associated with intra-oral MALT lymphoma [7].

Kojima et al. [7] have found no specific translocation associated with intra-oral MALT lymphoma studied in their series of cases. Tanaka et al. [24] have found t (11; 18) (q21; q21) in a single case report of primary MALT lymphoma of buccal mucosa. In our series of 3 cases, we studied for two translocations-t (11; 18) (q21; q21) and t (14; 18) (q32; q21), which were negative. However, we found tetrasomy 18 in one case, and loss of MALT1 and Bcl2 in other cases.

Extranodal MALT lymphoma remains localized for a long period of time. Variable data on chemotherapy, immunotherapy or a combination of these in treating patients with extra nodal marginal zone lymphoma is available. A few patients with aggressive clinical course with suspected histologic transformation can be helped with anthracycline based chemotherapy [37, 38]. One patient reported in the literature has undergone spontaneous regression [28].

Review of literature (Table 3) reveals a heterogeneity in the therapy in oral MALT lymphoma. There is a clear trend to operate MALT lymphoma and there may be difficulty in diagnosis in a punch biopsy. While IFRT is common treatment modality for MALT lymphoma at any site, this appears less popular in the oral cavity and this may be due to potential local side effects of radiotherapy. Only one case of palatal MALT lymphoma died from the disease, confirming its indolent behaviour.

In our series (Table 2), two patients received chemotherapy and all received IFRT. One case in our series relapsed in spite of anthracycline based RCHOP chemotherapy suggesting that local therapies may be more relevant than chemotherapy.

Conclusion

We discuss the histology, cytogenetics and behaviour of 3 cases of extranodal MALT lymphoma of the oral cavity along with review of literature. On histology, MALT lymphoma of oral cavity shows a wide spectrum of morphology with presence of transformed cells, that may lead to misdiagnosis of DLBL. Treatment guidelines are not well established but a tendency to excise MALT lymphomas of oral cavity has been observed. Nevertheless, MALT lymphoma of oral cavity appears to be an indolent disease.

Author Contributions

RK: Data acquisition, analysis, interpretation and drafting of manuscript; DS: Analysis of data; BPB: Critical revision of the manuscript; SG: Reporting pathologist for one of the cases; MS: Critical revision of the manuscript; SL: Critical revision of the manuscript; LN: Critical revision of the manuscript; TS: Conceptualization and design, reporting pathologist for 2 of the cases, analysis and interpretation of data, critical revision of the manuscript.

Funding

The authors received no financial support for the research, authorship and/or publication of this article.

Data Availability

All the data and material were available from the electronic medical records of Tata Memorial Hospital, Mumbai.

Code Availability

The data supporting the results reported in this article are available on request from the corresponding author. The data are not publicly available due to privacy restrictions.

Declarations

Conflict of interest

None of the authors have any conflict of interest or competing interest to declare.

Ethical Approval

Not applicable.

Consent to Participate

Verbal consent taken.

Consent for Publication

All authors have approved the manuscript.