The most well-described conditions of chronic inflammation that have been associated with a risk of developing MALT lymphoma include systemic and organ-specific autoimmune diseases. In some series, up to 40 % of patients with MALT lymphoma have a history of autoimmune disease (Wohrer et al. 2007). Of these, Sjögren’s syndrome and Hashimoto’s thyroiditis have the strongest correlation with the development of MALT lymphomas of the salivary and thyroid glands, respectively (Ansell et al. 1999; Diss et al. 1995; Kassan et al. 1978; Pertovaara et al. 2001). Patients with Sjögren’s syndrome, a condition associated with inflammation of the lacrimal and salivary glands leading to dry eyes and dry mouth, are 44 times more likely to develop non-Hodgkin lymphoma than the general population, and this risk appears to be increased further in patients with elevated β₂-microglobulin levels and evidence of reactive lymphadenopathy and/or splenomegaly (Kassan et al. 1978; Pertovaara et al. 2001). Biopsies of salivary and lacrimal glands from patients with Sjögren’s syndrome are notable for CD4 and CD8 T cell and CD20 B-cell infiltrates with increased expression of T- and B-cell chemokines CXCL12, CXCL13, and CCL21; the degree of elevation of these chemokines directly correlates with the extent of lymphocytic infiltration and organization (Barone et al. 2005; Salomonsson et al. 2003). The result is a local inflammatory environment marked by high interferon-γ, interleukin-2, interleukin-10, and B-cell activating factor levels, all of which promote the transcription of genes involved in cell survival and proliferation and the ultimate development of a malignant B-cell clone (Fox et al. 1994; Groom et al. 2002). A similarly organized and prolific lymphocyte infiltrate with elaboration of a T-helper-1 cytokine profile can be seen in biopsy specimens from patients with Hashimoto’s thyroiditis (Armengol et al. 2001; Del Prete et al. 1989). As a result, patients with Hashimoto’s thyroiditis are 67 times more likely to develop lymphoma of the thyroid than the general population, though thyroid MALT is a very rare disease (Holm et al. 1985). Clonal B-cell populations have been found in patients with autoimmune thyroid disease without evidence of lymphoma, and this did not progress to overt lymphoma in three such patients with over a decade follow-up, however (Saxena et al. 2004). Other autoimmune diseases have been shown to be associated with an increased risk of lymphoma and specifically marginal zone lymphoma, but to a lesser degree, including systemic lupus erythematosus which carries an eightfold increased risk of developing marginal zone lymphoma. Data regarding the risk of rheumatoid arthritis and lymphoma is conflicting, and a recent meta-analysis suggests that there is no significant association between the two diseases (Baecklund et al. 2006; Ekstrom Smedby et al. 2008; Smedby et al. 2006). Patients who develop MALT lymphomas in the context of autoimmune disease are more commonly female and of younger age, and the lymphomas are more often non-gastric and negative for the t(11;18) translocation or for trisomy 3; their prognosis appears to be similar to other MALT lymphoma patients (Wohrer et al. 2007).

As discussed previously, epidemiologic studies support a causal role of H. pylori infection in the development of MALT lymphomas, specifically of the stomach. In Northeastern Italy, where the incidence of H. pylori infections is high, there is an associated increased incidence of gastric lymphomas (Doglioni et al. 1992). Similarly, a case-controlled study documented an increased risk of gastric lymphoma in patients with H. pylori infection (Parsonnet et al. 1994). In addition, in some populations, the incidence of gastric MALT lymphoma and proportion of H. pylori-associated gastric MALT lymphoma is decreasing with increasing recognition and treatment of H. pylori infections (Luminari et al. 2010). As will be outlined in Sect. 14.5.1.1 of this chapter, the most compelling evidence for a causal association between H. pylori infection and gastric MALT lymphoma comes from the efficacy of H. pylori treatment in the treatment of these lymphomas (Bayerdorffer et al. 1995; Chen et al. 2005; Fischbach et al. 2004; Neubauer et al. 1997; Roggero et al. 1995; Stathis et al. 2009; Steinbach et al. 1999; Wotherspoon et al. 1993; Wundisch et al. 2005).

Infections with C. psittaci, B. burgdorferi, C. jejuni, and HCV have also been associated with an increased risk of MALT lymphoma in some populations. In a cohort of Italian patients with MALT lymphoma of the ocular adnexa, there was an increased frequency of C. psittaci in the tumor tissue and peripheral blood mononuclear cells than in healthy individuals (Ferreri et al. 2004). Seven of these patients were treated with doxycycline with eradication of the organism, and two of four evaluable patients had a documented tumor response. After this first report, many studies have been conducted showing high prevalence variations among different geographic regions (Chanudet et al. 2006; Mulder et al. 2006; Rosado et al. 2006). There are similar reports of an association between B. burgdorferi infection and cutaneous MALT lymphoma, C. jejuni infection and small bowel MALT lymphoma, and HCV infection and MALT lymphoma (Ascoli et al. 1998; Cerroni et al. 1997b; Lecuit et al. 2004; Luppi et al. 1996; Roggero et al. 2000; Zucca et al. 2000a, b). While an association between B. burgdorferi and primary cutaneous MALT lymphoma has been described in Europe, this has not been replicated in North American and Asian studies (Goodlad et al. 2000; Jelic and Filipovic-Ljeskovic 1999; Li et al. 2003; Wood et al. 2001). Immunoproliferative small intestinal disease is a lymphoma arising from small bowel MALT that is most commonly seen in the Middle and Far East, Mediterranean basin, and Africa (Lecuit et al. 2004). This geographic pattern of disease and the fact that reports of early stage disease response to antibiotics prompted a search for an infectious etiologic agent in its genesis. In seven patients who had responded to antimicrobial treatment, there was evidence of C. jejuni infection in four patients and H. pylori infection in no patients, thus establishing a potential link between C. jejuni infection and lymphomagenesis. The relative risk for patients with HCV infection to develop marginal zone lymphoma (particularly splenic and extranodal marginal zone lymphoma) is 2.5 times higher than the general population in a large intercontinental study (de Sanjose et al. 2008). Although this association is perhaps best described in patients with splenic marginal zone lymphoma, there is a high frequency of HCV infection in MALT lymphomas, and there have been reports of advanced MALT lymphomas of the salivary gland and intestines in patients with HCV infection that responded to treatment of the viral infection (Arcaini et al. 2004, 2009; Kelaidi et al. 2004; Svoboda et al. 2005). There is an association between patients presenting with a subcutaneous subtype of MALT lymphoma and HCV infection as well, with some patients responding to antiviral therapy (Paulli et al. 2010). In patients with MALT lymphoma, the incidence of HCV infection has been reported to be 35–43 %, most frequently in lymphomas of the skin, salivary glands, and orbit (Arcaini et al. 2007, 2009). Interestingly, these lymphomas frequently harbor the classical t(14;18) translocation joining the genes for BCL2 and the immunoglobulin heavy chain (Libra et al. 2004). The presence of HCV infection does not appear to influence the prognosis of the MALT lymphoma (Arcaini et al. 2007).

The clinical presentation of MALT lymphoma depends in large part on the site of disease. Gastric and intestinal MALT lymphomas may present with symptoms of dyspepsia and abdominal pain, sometimes with signs and symptoms of bowel obstruction but rarely with bleeding. These lymphomas are diagnosed on endoscopy with biopsies from multiple areas of endoscopically abnormal tissue as well as random sampling of macroscopically uninvolved mucosa. Lymphomas involving the small bowel may require a capsule video endoscopy for visualization. Involvement of the salivary and lacrimal glands, on the other hand, can result in Sjögren-like syndromes of dry eyes and mouth. MALT lymphomas involving the ocular adnexa typically present with painless conjunctival injection and photophobia, resembling allergic conjunctivitis. In the latter case, lesions are often bilateral and multifocal. Bronchus-associated lymphoid tissue (BALT) lymphoma involving the lungs and bronchi is a disease that most often affects older men (>60 years) (Fiche et al. 1995; Li et al. 1990). Unlike other MALT lymphomas, just over half of patients are symptomatic at diagnosis, with symptoms including cough, fever, and/or weight loss. This disease is often multifocal, spreading to other areas of the lungs and to other mucosal sites (Cordier et al. 1993). Other sites of disease often present with an obstructing mass. Some patients are diagnosed incidentally, either because of imaging studies or an exam of the eye or gastrointestinal track done for another reason or as part of an evaluation for a monoclonal gammopathy, which is present in approximately 25–35 % of MALT lymphoma patients; this feature is generally associated with plasmacytoid differentiation (Wohrer et al. 2004). B symptoms are rare in this disease (Armitage and Weisenburger 1998). Bone marrow involvement is present in a minority of patients, so cytopenias are rare, as is disease in the peripheral blood (Armitage and Weisenburger 1998).

The initial evaluation of a patient newly diagnosed with a MALT lymphoma should include a complete physical exam with attention to Waldeyer’s ring, including an assessment of performance status with either the Karnofsky performance scale or Eastern Cooperative Oncology Group performance status assessment. Laboratory evaluation should include a complete blood count with differential, comprehensive metabolic panel and liver function tests, lactate dehydrogenase, and an assessment of complete hepatitis B virus markers in patients who may be treated with rituximab. HCV testing may also be performed given its association with MALT lymphoma; a human immunodeficiency virus test is advised. Additional laboratory studies to consider include a β₂-microglobulin, serum protein electrophoresis and immunofixation, and serum light chains. Staging is done with computed tomography (CT) scans of the chest, abdomen, and pelvis, as well as imaging of the neck, including the parotids and salivary glands, and orbits with CT or MRI. Although reports on the utility of positron emission tomography (PET) scans have been conflicting in this disease, with some demonstrating a lack of PET-avid signal in involved tissues and nodes and others suggesting that PET scans result in the upstaging of the disease at diagnosis, current recommendations of an International Harmonization Project in 2007 recommend against the use of PET scanning in the initial evaluation of lymphomas that are not routinely PET avid, including marginal zone lymphoma, based on a lack of sufficient evidence supporting its use (Beal et al. 2005; Fueger et al. 2009; Hoffmann et al. 1999; Juweid et al. 2007). A bone marrow biopsy should be considered for patients with multifocal disease, and an evaluation of the gastric mucosa is reasonable for all patients with non-gastric MALT lymphoma given the documented high rate of gastric involvement in these patients (Raderer et al. 2006).

Gastric and intestinal MALT lymphomas are staged by the Lugano staging system for gastrointestinal lymphomas (Table 14.1) (Rohatiner et al. 1994). By this system, stage I disease is limited to the gastrointestinal track, whereas stage II disease involves extraintestinal lymph nodes within the abdomen, and stage IV disease involves supradiaphragmatic nodal or disseminated extranodal tissue; there is no stage III disease. Endoscopic ultrasound has allowed for an estimate of the depth of infiltration into the gastric wall, which correlates with the extent of lymph node involvement and prognosis (Steinbach et al. 1999). Although the majority are localized at diagnosis, these lymphomas can spread to other parts of the gastrointestinal tract and the splenic marginal zone, perhaps due to H. pylori-activated T-cell-induced overexpression of the mucosal-homing integrin α₄β_7, whose ligand is expressed by these two tissues (Briskin et al. 1997; Du et al. 1997; Kraal et al. 1995). Staging for non-gastric MALT lymphoma is by the Ann Arbor staging system for lymphoma, which takes the extent of lymph node stations involvement as well as extranodal involvement into account when assigning a stage (Table 14.1) (Carbone et al. 1971; Lister et al. 1989).

Unlike for other non-Hodgkin lymphomas, there exists no specific prognostic scoring system for marginal zone or MALT lymphomas. The Follicular Lymphoma International Prognostic Index (FLIPI) was developed as a prognostic tool in follicular lymphoma and is often applied to other indolent lymphomas (Buske et al. 2006; Solal-Celigny et al. 2004). This index incorporates age (>60 years), Ann Arbor stage (III–IV), hemoglobin level (<12 g/dL), lactate dehydrogenase serum level (> upper limit of normal), and number of involved nodal areas (>4) into a 5-point scoring scale that corresponds to low (0–1 points), intermediate (2 points), and high (3 or more points) risk. A revised FLIPI has been developed, the FLIPI2, based on the results of a prospective study of almost 1,000 patients with newly diagnosed follicular lymphoma who underwent therapy that incorporates age (>60 years), bone marrow involvement, hemoglobin level (<12 g/dL), longest diameter of the largest involved lymph node (>6 cm), and a serum β2-microglobulin level (> upper limit of normal) into a 5-point scoring scale corresponding to low- (0 points), intermediate- (1–2 points), and high- (3 or more points) risk disease (Federico et al. 2009).

MALT lymphomas overall, however, have a good prognosis with a 5-year overall and failure-free survival of 81 % and 65 %, respectively (Nathwani et al. 1999). Patients with stage III or IV disease assessed before the era of rituximab and treated with cyclophosphamide, adriamycin, vincristine, and prednisone (CHOP) chemotherapy had a median overall survival of 5 years with median failure-free survival of 3 years (Fisher et al. 1995). In one retrospective series of a heterogeneous cohort of patients treated for MALT lymphoma, relapse rates approached 40 % at 4-year follow-up (Raderer et al. 2005). Non-gastric lymphomas had a higher rate of relapse than their gastric counterparts (48 vs. 22 %). In these patients, a high IPI and lymph node involvement has been associated with worse outcomes, whereas multiple mucosal sites were not (Thieblemont et al. 1997; Zucca et al. 2000a, b). Other series have shown that poor prognostic features in non-gastric MALT lymphomas include multiple MALT lymphoma sites, advanced stage disease, bone marrow and nodal involvement, and MALT lymphomas outside the skin and ocular adnexa (Arcaini et al. 2006). Primary cutaneous marginal zone lymphomas are particularly indolent and unlikely to disseminate with 5-year survival rates of 98–100 % (Hoefnagel et al. 2005). The presence of systemic symptoms in BALT lymphomas has been associated with a poorer prognosis (Cordier et al. 1993).

In gastric MALT lymphomas, the presence of a t(11;18) translocation is associated with an increased risk of disseminated disease, whereas trisomy 18 is associated with a risk of advanced stage disease in extragastric MALT lymphoma (Raderer et al. 2006). However, whereas t(11;18)-positive MALT lymphomas rarely are associated with additional clonal aberrations, a majority of t(11;18)-negative MALT lymphomas have allelic imbalances that are also present in gastric DLBCL (Starostik et al. 2000; Starostik et al. 2002). Thus, t(11;18)-positive lymphomas appear to be genetically stable and therefore at lower risk for transformation into a more aggressive DLBCL. Instead, certain genetic alterations appear to increase the risk of histologic transformation into an aggressive lymphoma including loss or deletion of TP53, hypermethylation or deletion of CDKN2A, and other chromosomal gains or losses (Du et al. 1995; Martinez-Delgado et al. 1997; Neumeister et al. 1997). Aberrant expression of CD5 detected by flow cytometry or immunohistochemistry is likewise associated with a more aggressive phenotype, as is high-grade histology (de Jong et al. 1997, 2000; Wenzel et al. 2001). A grading system for MALT lymphoma has been devised that assigns a letter grade (A to D) based on the number of blast or blast clusters seen histologically, ranging from <5 % blasts in clusters of up to ten cells to pure DLBCL without a low-grade component (de Jong et al. 1997). Gastric lymphomas that do not respond to H. pylori eradication are enriched for higher-grade histology, although many patients with high-grade histology do respond to antibiotic therapy (Bayerdorffer et al. 1995; Chen et al. 2001). In one series of 16 patients with stage I–II H. pylori-positive DLBCL of the stomach, half achieved a complete remission with antibiotic therapy alone, and another 13 % had a partial remission convert to a complete remission after the addition of rituximab monotherapy (Govi et al. 2011b). After 53-month follow-up, nine of the ten complete responders were still in remission. Grade similarly correlated with prognosis in a small number of patients with MALT lymphoma of the thyroid gland treated in the pre-rituximab era; all patients with low-grade lesions were alive at a median follow-up of 26 months, whereas patients with high-grade lesions appeared to do worse than patients with pure DLBCL of the thyroid gland with a low 5-year overall survival rate of 25 % owing to a higher rate of more advanced stage disease at diagnosis (Skacel et al. 2000).

Early stage or localized MALT lymphoma of the stomach is an indolent disease that is often associated with H. pylori infection. Eradication of H. pylori is effective treatment for many patients with good long-term disease control and overall survival and is recommended for most patients with H. pylori-positive gastric MALT lymphomas that do not harbor a t(11;18) translocation (Bayerdorffer et al. 1995; Chen et al. 2005; Fischbach et al. 2004; Neubauer et al. 1997; Roggero et al. 1995; Stathis et al. 2009; Steinbach et al. 1999; Wotherspoon et al. 1993; Wundisch et al. 2005). In patients with H. pylori-negative lymphomas, MALT lymphomas with a t(11;18) translocation, or lymphomas that fail to respond to H. pylori therapy, radiation therapy is the preferred treatment modality (Hitchcock et al. 2002; Schechter et al. 1998; Tomita et al. 2009; Tsang et al. 2001, 2003). Chemotherapy, immunotherapy, or chemoimmunotherapy is active in this disease but is generally reserved for patients with relapsed or refractory disease to antibiotic or radiation therapy or patients with more advanced stage or aggressive disease (Conconi et al. 2003, 2011; Hammel et al. 1995; Jager et al. 2002; Martinelli et al. 2005; Nakamura et al. 2005; Raderer et al. 2003).

Treatment of early stage non-gastric MALT lymphomas follows many of the same paradigms as for early stage gastric MALT lymphomas in Sect. 14.5.1, except that there is no association between a particular infectious organism and these lymphomas that is as strong or penetrant as the association between H. pylori and gastric MALT lymphoma. MALT lymphomas outside the stomach are also positive for H. pylori, in up to 45 % of cases in one series, but H. pylori treatment only led to regression of one colonic tumor in 16 such patients (Grunberger et al. 2006). Although antibiotic approaches against B. burgdorferi for cutaneous lymphomas and C. jejuni for intestinal lymphomas have been investigated and successful in some cases, the treatment of choice for most localized, early stage disease is radiation therapy (Hitchcock et al. 2002; Isobe et al. 2007; Tomita et al. 2009; Tsang et al. 2001, 2003). The use of doxycycline for treatment of C. psittaci-associated MALT lymphoma of the ocular adnexa has been resulted in high response rates in certain patient populations and is a reasonable upfront treatment option, in addition to radiation therapy (Govi et al. 2011a). For many of these lymphomas, doses of 24–30 Gy are sufficient and associated with minimal toxicity. One prospective series of 37 patients with stage IE non-gastric MALT lymphoma treated with a median of 30.6 Gy of radiation reported a 92 % complete response rate, with a 3-year progression-free and overall survival of 92 % and 100 %, respectively (Isobe et al. 2007). Many of the relapses involved the contralateral paired organ with nearly 100 % local control (Tsang et al. 2001, 2003). Surgical resection is appropriate for tumors that are not amenable to radiation and has been reported in patients with lymphomas involving the salivary glands, thyroid, skin, breasts, lung, genitourinary tract, and dura (Ambrosetti et al. 2004; Cerroni et al. 1997a; Ferraro et al. 2000; Gogas et al. 2002; Kees et al. 2005; Kempton et al. 1997; Zinzani et al. 2003). Surgery may be performed before the diagnosis of MALT lymphoma is known, and if complete excision is achieved, these patients should be observed without further treatment; if margins remain positive though, adjuvant radiation should be administered when feasible. Improving on outcomes following radiation therapy with adjuvant chemotherapy has not been beneficial: the addition of adjuvant anthracycline-based chemotherapy following radiation for stage IE orbital MALT lymphomas did not improve outcomes, whereas such adjuvant chemotherapy in stage III–IV non-gastric MALT lymphomas improved only the complete response rate but not progression-free survival (Aviles et al. 2006; Oh et al. 2007). Chemotherapy and/or immunotherapy approaches as were outlined for early stage gastric MALT lymphomas have been studied in non-gastric MALT lymphomas as well and are appropriate for cases of relapsed or refractory disease to local therapies like surgery or radiation, multifocal disease that is not amenable to local therapies, or more aggressive or advanced stage disease (Conconi et al. 2003, 2011; Hammel et al. 1995; Jager et al. 2002; Raderer et al. 2003). Although perhaps less effective for non-gastric compared to gastric MALT lymphomas, rituximab is a reasonable first option for some of these patients. Single-agent alkylating agents like oral cyclophosphamide or chlorambucil or purine analogues like cladribine have activity in these diseases as well (Conconi et al. 2003, 2011; Hammel et al. 1995; Jager et al. 2002). The response rates and duration of responses seen with these single agents can be low, making combination chemoimmunotherapy with regimens like rituximab and bendamustine or rituximab and cyclophosphamide, vincristine, and prednisone more appropriate for some patients. For patients who have no signs or symptoms related to their disease and who have either relapsed following local therapies or are not candidates for local therapies, observation with close follow-up is the preferred option (Ardeshna et al. 2003). The remainder of this section will outline the approaches that have been investigated for MALT lymphomas of specific disease sites: ocular adnexa, thyroid, breast, skin, lung, dura, and genitourinary tract.