Frequency of Chromosomal Aberrations Involving MALT1 in Mucosa-Associated Lymphoid Tissue Lymphoma in Patients with Sjögren’s Syndrome

Mucosa-associated lymphoid tissue (MALT) lymphoma is a distinct lymphoma entity with unique features (1, 2). One of the most striking clinical characteristics is the association of MALT lymphoma with chronic antigenic stimulation, as exemplified by Helicobacter pylori infection in patients with gastric MALT lymphoma (2, 3). Although the large majority of cases arise in the stomach, MALT lymphomas may be encountered in virtually every organ of the human body (2, 4), with a predilection for the salivary glands and the lung.

An association between various autoimmune conditions and the occurrence of lymphoproliferative diseases has repeatedly been reported previously (5). Especially in case of autoimmune thyroiditis and Sjögren’s syndrome (SS), a high risk for the development of MALT lymphomas in the primarily affected organs has been demonstrated. Accordingly, a 67-fold increased risk for thyroid MALT lymphoma and a 44-fold increased risk for parotid lymphoma is being attributed to autoimmune thyroiditis and SS (6, 7), and 1 in 5 deaths in patients with primary SS is due to the development of lymphoma (8).

Several genetic aberrations of MALT lymphoma have been identified, some of which appear to play an important role in the pathogenesis and also influence the prognosis of the disease. Two of these changes involve the MALT1 gene located on chromosome 18. The t(11;18)(q21;q21) is considered to be specific for MALT lymphomas because it has not been found in nodal and splenic marginal zone B-cell lymphomas or other types of lymphoma (9, 10). This translocation results in the synthesis of the API2-MALT1 fusion protein, which has recently been shown to activate nuclear factor-κB, a transcription factor for a number of survival-related genes (11). Up-regulation of these molecules promotes cellular proliferation and resistance to apoptotic signals. As a consequence, patients with gastric MALT lymphoma carrying t(11;18)(q21;q21) have been reported to be largely resistant to eradication of H. pylori and display a more aggressive clinical course (11).

Recently, we have reported the occurrence of t(14;18)(q32;q21) involving IGH and MALT1 in patients with MALT lymphomas of the liver, skin, salivary glands, and ocular adnexa (12). It was neither detected in gastric, intestinal, pulmonary, and thyroid lymphomas nor in splenic marginal zone lymphomas. Although the potential function of this particular translocation in the pathogenesis of MALT lymphomas has not yet been elucidated, these findings together with the data published on t(11;18)(q21;q21) suggest MALT1 rearrangements in up to 50% of all patients with MALT lymphoma.

A study from our institution has demonstrated an impaired response rate in patients with gastric MALT lymphoma and an additional autoimmune disease (AD) despite successful H. pylori eradication (13). Among these 6 patients initially described, three had gastric MALT lymphoma and a history of SS, suggesting AD to be an independent factor in the development of gastric MALT lymphoma. However, no analysis of genetic events involving MALT1, i.e., t(11;18)(q21;q21) and t(14;18)(q32;q21), in MALT lymphoma associated with SS has been performed thus far. This has prompted us to investigate the frequency of t(11;18)(q21;q21) and t(14;18)(q32;q21) in MALT lymphomas of various localizations arising in patients with SS.

A retrospective single-center analysis of patients diagnosed with MALT lymphoma and a history of SS was performed. The information on lymphoma diagnoses was based on entries in the Vienna Lymphoma Registry between 1997 and November 2002, which was searched for all cases identified as having MALT lymphoma. All patients enrolled in this study were staged according to our extensive protocol (14) and treated at our institution with full clinical information available. Eight patients, all of whom with MALT lymphoma of the parotid, have been part of a previous study (12). Information analyzed included the histological diagnosis, localization of the lymphoma, staging, and the presence of SS. The diagnosis of SS was based on characteristic clinical symptoms, i.e., presence of mouth and/or eye dryness, as well as additional features such as a positive lip biopsy or serological changes as required in the recently updated United States-European criteria (15). Assessment of clinical and immunological parameters to diagnose an underlying autoimmune condition is routinely performed since 1997 at our institution in all patients with a diagnosis of MALT lymphoma. Only formalin-fixed, paraffin-embedded specimens were available. The presence of tumor cells was evaluated in each tissue block on H&E-stained sections cut before and after the sections used for reverse transcriptase-PCR or fluorescence in situ hybridization (FISH). Only blocks containing >50% lymphoma cells were analyzed.

Before the molecular investigations, the histological diagnosis of MALT lymphoma was reassessed according to the criteria as outlined by Isaacson et al. (1) in the WHO classification. RNA was isolated from formalin-fixed, paraffin-embedded lymphoma tissues obtained either by endoscopic biopsies or surgery. Total RNA was extracted from 10-μm sections with a high pure RNA paraffin kit (Roche Diagnostics, Mannheim, Germany). First-strand cDNA was synthesized from 1 μg of total RNA with a superscript first-strand synthesis system (Invitrogen, Carlsbad, CA) using random hexamers as primers. For evaluation of t(11;18)(q21;q21), reverse transcriptase-PCR was performed according to Inagaki et al. (16) with one difference: first round reverse transcriptase-PCR products were amplified in a second round separately and not as multiplex nested PCRs. β-Actin was amplified as a positive internal control.

Formalin-fixed, paraffin-embedded tissues were screened by FISH for t(14;18)(q32;q21) involving IGH and MALT1. For a reliable interpretation of the hybridization signals, we preferred the analysis of single cell suspensions over thin sections. After deparaffinization in xylene, two 30-μm thick slices were incubated for 25 min in 4% pepsin (pH 1.5) at 37°C. After a rapid wash in PBS, cells were incubated for 30 min in 0.075 m KCl and fixed twice in methanol/acetic acid (3/1) and dropped on slides. FISH was performed on interphases with a probe spanning the MALT1 gene and flanking regions (PAC 152M5; Ref. 17) and an IGH probe (BAC158A2; Ref. 18) picked from a bacterial artificial chromosome library. The IGH probe was directly labeled with SpectrumGreen, the MALT1 probe with SpectrumOrange by nick translation (Vysis, Downer’s Grove, IL). A t(14;18) involving IGH and MALT1 would result in a split orange MALT1 signal with fusion to a green IGH signal. Five hundred cells were analyzed in each case. FISH procedure was performed as described elsewhere (12). The cutoff value for the diagnosis of a rearrangement involving IGH and MALT1 was 5.3%, which is above the mean percentage of cells with a false-positive signal constellation plus 3 SDs, as assessed on tissue from 20 reactive lymph nodes processed as described for the lymphomas. Moreover, in all t(14;18)(q32;q21)+ lymphomas, verification of the rearrangement was performed with dual color break apart rearrangement probes for IGH (Vysis) and MALT1 (12). As a result of this probe design, any translocation with a breakpoint at IGH or MALT1 should produce separate orange and green signals.

On the single cell suspensions of 12 patients with MALT lymphoma of the salivary gland and no MALT1 rearrangement, additional FISH was performed with probes for IGH and BCL2 (Vysis). The orange BCL2 probe covers the entire BCL2 gene and additionally 250 kb both 5′ and 3′ of the gene. The 1.5-Mb green IGH probe covers the entire IGH locus. A t(14;18) involving IGH and BCL2 would therefore result in one single green, one single orange, and two green/orange fusion signals in interphases. The cutoff value was 3.7%.

A total of 26 patients with MALT lymphoma and a clear-cut diagnosis of SS according to the criteria as outlined above (15) was identified (Table 1). As expected, the majority of patients were female (20 female and 6 male) and were aged between 31–80 years. Fifteen patients had lymphoma located in the salivary glands (parotid in 14, submandibular in 1 case), 2 had MALT lymphoma involving the lacrimal gland, whereas 9 patients suffered from gastric MALT lymphoma restricted to the stomach (stage EI disease). In all 9 cases with gastric lymphoma, evidence for H. pylori infection was present either histologically or serologically.

The reference gene β-actin was successfully amplified by reverse transcriptase-PCR in all 26 patients. In total, 7 patients (27%) were found to harbor t(11;18)(q21;q21). Interestingly, only 1 of 17 patients (6%) with extragastrointestinal lymphoma was found to be positive, whereas 6 of 9 patients (67%) with MALT lymphoma of the stomach had a positive reverse transcriptase-PCR for t(11;18)(q21;q21; Fig. 1). The breakpoints for API2 (GenBank accession no. NM_001165) were at nucleic acids 2345 in 6 cases and at 2642 in 1 case; in MALT1(AF130356), it varied from 541 (patients 1 and 2), 814 (patients 3 and 10), 1123 (patients 4 and 5), and 1151 (patient 6; Table 1).

Screening for a rearrangement of MALT1 by a t(14;18)(q32;q21) was performed using dual-color FISH (Fig. 2). In total, t(14;18)(q32;q21) was found in 4 patients (Table 1). The percentage of tumor cells evaluated in each tissue block on H&E-stained slides correlated with the number of cells with t(14;18)(q32;q21) assessed by the three independent FISH assays (IGH/MALT1 dual color fusion, IGH dual color break apart, and MALT1 dual color break apart). Three t(14;18)(q32;q21)+ tumors were of extragastrointestinal origin (two parotid and one lacrimal MALT lymphoma), whereas one originated in the stomach.

In 12 patients with MALT lymphoma of the salivary glands, rearrangements of MALT1 either by a t(11;18)(q21;q21) or t(14;18)(q32;q21) were absent. These patients were additionally tested for t(14;18)(q32;q21) involving IGH and BCL2. All 12 patients were negative for IGH/BCL2 rearrangement (Table 1).

t(11;18)(q21;q21) is found in MALT lymphomas arising at a variety of mucosal sites and is clinically important in gastric MALT lymphoma. The significance of analyzing t(11;18)(q21;q21) in gastric MALT lymphoma is underlined by its association with advanced disease and resistance to H. pylori eradication therapy (19). Studies using reverse transcriptase-PCR in comprehensive series of patients have demonstrated t(11;18)(q21;q21) in 24–40% of unselected gastric MALT lymphomas (20, 21, 22, 23, 24, 25). However, these studies did not discriminate between tumors arising from proliferating B-cell populations induced by H pylori infection or originating in the background of AD. Although salivary gland lymphomas arising in patients with AD have repeatedly been recognized, the occurrence of primary gastric MALT lymphoma in such patients has not been given due consideration. To our knowledge, no prospective systematic screening for autoimmune conditions in patients with (gastric) MALT lymphomas has been reported in the literature. As of the beginning of 1997, at our institution, all patients with a diagnosis of MALT lymphoma irrespective of localization are evaluated clinically and serologically for autoimmune conditions, including SS. Because this has not consistently been done before, one might speculate that previous series have overlooked such an association.

In the present study, 9 primary gastric MALT lymphomas in patients with SS were investigated for t(11;18)(q21;q21). According to a rigorous staging system applied at our institution, secondary spread from extragastric lymphoma can be ruled out almost with certainty. All of these patients had evidence of infection with H. pylori. Remarkably, 67% of these lymphomas were t(11;18)(q21;q21)+, which is a much higher percentage than reported in series from unselected cases of H. pylori-positive gastric MALT lymphomas and much higher than the 29.9% (20 of 69 cases) at our institution (unpublished data). At the moment, we cannot offer a definite explanation for this finding. Speculatively, MALT lymphoma cells in patients with SS in the context of H. pylori infection might be more susceptible to acquire the t(11;18), and the acquisition might occur at an earlier clinical stage of the disease. It is noteworthy in this regard that all 9 SS-associated gastric MALT lymphomas were so-called early lymphomas and thus confined to the mucosa and submucosa of the gastric wall.

As opposed to gastric and pulmonary MALT lymphomas, t(11;18)(q21;q21) is found at much lower frequencies in salivary gland and orbital tumors (9, 21, 22). Similarly, only one of the extragastrointestinal MALT lymphomas in our cohort of a patient with SS, a parotideal MALT lymphoma, revealed t(11;18)(q21;q21).

Apart from a fusion with API2, MALT1 has recently been shown to be rearranged with IGH (12, 26, 27). This novel translocation was found in MALT lymphomas of the liver (4 of 4 cases), skin (3 of 11), ocular adnexa (3 of 8), and salivary glands (2 of 11), whereas 10 MALT lymphomas of the stomach and 9 of the intestine were negative (12). Moreover, it was demonstrated that t(11;18)(q21;q21) and t(14;18)(q32;q21) were mutually exclusive. In the present series, 1 of 2 MALT lymphomas of the ocular adnexa and 2 of 15 salivary gland MALT lymphomas carried t(14;18)(q32;q21). Interestingly, a t(11;18)(q21;q21)-negative MALT lymphoma of the stomach showed a rearrangement of IGH and MALT1, representing the first observation of a t(14;18)(q32;q21) in a gastric MALT lymphoma. Taken together, MALT1 translocation either t(11;18)(q21;q21) or t(14;18)(q32;q21) occurred in 7 of 9 (78%) of gastric MALT lymphomas of SS patients.

In two previous studies of lymphomas in patients with SS, t(14;18)(q32;q21) was detected by PCR (28, 29). In these patients, however, BCL2 and not MALT1 was juxtaposed to IGH. BCL2 rearrangements are highly associated with follicular lymphoma but are absent in MALT lymphoma (30, 31). In one of these studies, 1 of 8 SS patients had a BCL2 translocation (28). This patient suffered from a pulmonary MALT lymphoma, which showed rapid transformation into a diffuse large B-cell lymphoma with consecutive nodal involvement, which also harbored t(14;18)(q32;q21). In the other study (29), BCL2 translocations were detected in 5 of 7 SS-associated B-cell lymphomas involving the salivary glands. However, the exact histological classification of these lymphomas was not provided. Our assay for the detection of the t(14;18)(q32;q21) involving IGH and MALT1 would not detect the t(14;18)(q32;q21) involving BCL2 despite localization at the same chromosomal band. We therefore performed a FISH assay with probes for BCL2 and IGH, which should disclose t(14;18) generated by BCL2 breakpoints of major breakpoint and minor cluster regions as well as by additional breakpoints. All 12 cases negative for MALT1 rearrangements were negative for t(14;18)(q32;q21) involving BCL2.

To summarize, the frequency of translocations involving MALT1 appears to be low in patients with extragastrointestinal MALT lymphoma associated with SS. By contrast, MALT1 rearrangement is frequently present in patients with gastric MALT lymphoma and SS (78%), which might explain at least in part why gastric MALT lymphomas in patients with AD are largely resistant to H. pylori eradication therapy (13).