Splenic marginal zone lymphoma

INTRODUCTION — Splenic marginal zone lymphoma (SMZL) is an uncommon subtype of non-Hodgkin lymphoma that usually presents with splenomegaly and lymphocytosis and is characterized by a marginal zone growth pattern in the spleen. SMZL was previously termed splenic lymphoma with circulating villous lymphocytes.

The epidemiology, pathogenesis, diagnosis, and management of SMZL will be discussed here. Extranodal marginal zone lymphoma and nodal marginal zone lymphoma are distinct neoplasms that are discussed separately.

●(See "Clinical manifestations, pathologic features, and diagnosis of extranodal marginal zone lymphoma of mucosa associated lymphoid tissue (MALT)".)

●(See "Nodal marginal zone lymphoma".)

●(See "Primary cutaneous marginal zone lymphoma".)

●(See "Treatment of extranodal marginal zone lymphoma of mucosa associated lymphoid tissue (MALT lymphoma)".)

PATHOGENESIS — SMZL is postulated to arise from a post-germinal center, memory B cell of splenic type [1]. The pathogenesis of SMZL is incompletely understood, but acquired mutations in oncogenes and tumor suppressor genes and chronic immunologic stimulation appear to be involved.

Molecular studies suggest alterations to pathways that result in the proliferation and commitment of mature B cells to marginal zone differentiation and homing to the spleen [2-5]. The following mutations have been identified in SMZL and are thought to play a role in this process:

●Mutations in genes encoding Notch pathway components in up to 40 percent

●Mutations in genes encoding NF-KB pathway components in 35 to 40 percent

●KLF2 mutations in 20 to 40 percent

●MYD88 mutations in 3 to 15 percent

●CARD11 mutations in 5 to 10 percent

●KMT2D (previously known as MLL2) mutations in 7 percent

Involvement of Notch signaling in SMZL echoes the requirement for Notch2 in the development of marginal zone B cells [6]. MYD88 and CARD11 mutations likely enhance B cell receptor (BCR) signaling. SMZL also has a microRNA signature that reproduces that of the normal marginal zone and memory B cell program [7]. This microRNA signature likely influences the expression of genes involved in the NF-kB pathway, which is affected by recurrent gain-of-function mutations in SMZL. Efforts to subclassify SMZL based on its molecular characteristics are underway, in the hope of better defining its prognosis and identifying therapeutic targets in individual patients [8].

Epidemiologic studies have identified an association between SMZL and infection with viruses, such as hepatitis C virus (HCV) [9] and Kaposi sarcoma-associated herpes virus (human herpesvirus type 8, HHV-8) [10]. Antigen receptors for HCV have been identified on some tumors [11]. Treatment of coexisting HCV infection can induce regression of SMZL in some patients, and some studies have demonstrated a correlation between lymphoma disease activity and HCV load [9]. Thus, like other subtypes of marginal zone lymphoma, SMZL may be driven in some cases by chronic immunologic stimulation, at least early in the disease course. (See 'Antiviral therapy for hepatitis C' below.)

The association between HCV and lymphomagenesis is discussed in more detail separately. (See "Extrahepatic manifestations of hepatitis C virus infection", section on 'Lymphoma'.)

EPIDEMIOLOGY — SMZL constitutes <1 percent of all non-Hodgkin lymphomas [12,13], 1 to 2 percent of indolent lymphoid leukemias found on bone marrow examination, and up to 25 percent of low grade B cell neoplasms in splenectomy specimens [14]. The estimated incidence in the United States is 1.3 cases per million persons per year and appears to be increasing [13,15,16].

SMZL occurs at a median age of 65 to 70; the disease is uncommon before age 50. SMZL occurs in all geographic locations, but the incidence in White individuals is approximately twice that of other races [13]. There is no sex predominance.

There is an association with autoimmune conditions that lead to B cell activation (eg, Sjögren’s disease, systemic lupus erythematosus, hemolytic anemia, pernicious anemia), asthma (with or without other atopy), and chronic hepatitis C virus infection [17,18]. Studies have also suggested associations with the use of permanent hair dye and occupation as a metal worker [18]. (See "Extrahepatic manifestations of hepatitis C virus infection", section on 'Lymphoma'.)

CLINICAL FEATURES — Patients typically present with splenomegaly, lymphocytosis, and cytopenias (eg, anemia, thrombocytopenia) [19-22]. Many patients are asymptomatic at the time of diagnosis; a minority has symptoms related to splenomegaly (eg, early satiety, abdominal discomfort) or cytopenias (eg, bleeding). Rare patients present with lymphocytosis as the sole marker of disease [22].

Unlike most other non-Hodgkin lymphomas, lymphadenopathy and significant involvement of extralymphatic organs is uncommon [23]. Systemic B symptoms (eg, fatigue, night sweats, weight loss) and elevations in lactate dehydrogenase (LDH) are rare and should lead to a more detailed investigation for large cell transformation. (See 'Large cell transformation' below.)

The cytopenias are most commonly due to hypersplenism but may be immune mediated. Approximately 20 percent of patients with SMZL have an autoimmune manifestation, including autoimmune hemolytic anemia, immune thrombocytopenia, cold agglutinin disease, antiphospholipid antibodies, acquired von Willebrand disease, and acquired C1-esterase inhibitor deficiency [24-26]. (See 'Managing autoimmune complications' below.)

Approximately one-third of patients have a small amount of circulating monoclonal protein, often immunoglobulin M (IgM). In those with hepatitis C infection, mixed cryoglobulinemia and an associated small vessel vasculitis may be present [27]. (See "Overview of cryoglobulins and cryoglobulinemia".)

PATHOLOGIC FEATURES

Morphology — SMZL is a neoplasm of small B lymphocytes that surround and replace splenic white pulp germinal centers, merging with an outer marginal zone of cells with more abundant pale cytoplasm. Variable numbers of cells with distinct nucleoli resembling immunoblasts may be admixed. Both small and large cells infiltrate red pulp, and "villous" lymphocytes often are seen circulating in the peripheral blood (picture 1) [28].

●Peripheral blood – Involvement of the peripheral blood is common in SMZL. While there is morphologic heterogeneity, the typical tumor cell has a round nucleus, condensed chromatin, and abundant agranular basophilic cytoplasm with small surface "villous" projections (picture 1) [19,22,28]. These projections are often unevenly distributed around the circumference of the cell. Extended exposure of these cells to anticoagulants used for specimen collection can result in loss of these projections, and examination of a smear made from freshly drawn blood may be required to identify projections in suspected cases [22]. Less commonly, tumor cells have a plasmacytoid appearance.

●Spleen – Pathologic examination of the spleen demonstrates involvement of both the white and red pulp in practically all cases of SMZL. In the white pulp, the neoplastic cells occupy both the mantle and marginal zones, sometimes with a central residual germinal center, which may be either atrophic or hyperplastic (picture 2) [14]. Both the mantle and marginal zones are expanded. The cells in the mantle zone are small, with slight nuclear irregularity and scant cytoplasm, while those in the marginal zone have more dispersed chromatin and abundant, pale cytoplasm, resembling normal marginal zone cells, and are admixed with variable numbers of larger activated cells resembling immunoblasts. In some tumors there is also a degree of plasmacytic differentiation.

The red pulp demonstrates both a diffuse and micronodular pattern of involvement with sinus infiltration. Activated "epithelioid" macrophages may be present singly or in clusters, and, particularly in the bone marrow, may be so conspicuous as to suggest an infectious process associated with granulomatous inflammation.

●Lymph nodes – Splenic hilar lymph nodes are often involved; the neoplastic cells form vague nodules, often without a central germinal center; a marginal zone pattern may or may not be present [29]. Sinuses are often preserved and dilated.

●Bone marrow – Bone marrow involvement, if present, is often mild and can be subtle. The marrow usually contains discrete lymphoid aggregates, which may also have a marginal zone pattern, with or without diffuse intrasinusoidal infiltration by tumor cells [19,20]. The nodular pattern of involvement can be helpful in distinguishing SMZL from hairy cell leukemia, which always involves the marrow in a diffuse interstitial pattern. (See "Clinical features and diagnosis of hairy cell leukemia", section on 'Bone marrow'.)

Immunophenotype — The following immunophenotype is typical for SMZL (table 1) [1,14]:

●Express surface immunoglobulin (eg, IgM, IgD), B cell antigens (CD19, CD20, CD22), and BCL2. CD5 is usually negative but can be positive in a minor subset of cases.

●Do not express CD10, CD25, CD103, or annexin A1.

●Variable expression of CD21, CD23, CD43, and CD35.

In addition, the majority of SMZL expresses the adhesion molecules LFA-1, CD44, CD49d, CD29, and ICAM-1, but not L-selectin. The cells are cyclin D1 negative by immunoperoxidase staining [30].

As described in more detail below, immunophenotyping and the pattern of tissue involvement can help to distinguish SMZL from other small B cell malignancies:

●Lack of CD5 usually helps to distinguish SMZL from chronic lymphocytic leukemia/small lymphocytic lymphoma and mantle cell lymphoma, but a significant minority of cases of SMZL are CD5 positive and must be distinguished based on other genetic, immunophenotypic, and morphologic findings [31]. The absence of cyclin D1 staining also helps to distinguish SMZL from mantle cell lymphoma, which is strongly cyclin D1 positive in a large majority of cases, as well as from hairy cell leukemia, which is weakly cyclin D1 positive in most cases. (See 'Differential diagnosis' below.)

●Lack of CD103, CD25, and annexin A1 are particularly helpful in distinguishing SMZL from hairy cell leukemia, which is typically positive for these markers.

Genetic features — Up to 80 percent of SMZL have an abnormal karyotype, often with complex chromosomal changes [22,32,33]. No specific cytogenetic abnormalities have been identified.

Analysis of the immunoglobulin variable region genes shows somatic mutations in approximately half of cases, consistent with a post-germinal center stage of B cell development [2,34,35]. Ongoing mutation of V region genes, similar to germinal center cells, has been reported [36].

Trisomy 3, found in nodal and extranodal marginal zone lymphoma, is detected in up to 39 percent of cases of SMZL [37,38], although, unlike in these other disorders, the t(11;18) is lacking in SMZL [39,40]. BCL2 rearrangements and the t(11;14)(q13;q32) (involving the cyclin D1 gene) are not present.

Other common structural chromosomal aberrations include gains of 12q, deletions of 7q and 6q, and various translocations involving the IgH locus (14q32) and regions on chromosome 1q and 8q [31]. By fluorescence in situ hybridization (FISH) analysis, del(7q) is the most common aberration in some series [41]; although not pathognomonic, among indolent B cell tumors del(7q) is most common in SMZL.

Molecular studies have identified recurrent mutations in several genes, including gain-of-function mutations in NOTCH2 and in genes encoding components of the NF-KB signaling pathway. This is described in more detail separately. (See 'Pathogenesis' above.)

DIAGNOSIS — SMZL is usually suspected in an adult presenting with unexplained splenomegaly and lymphocytosis. The diagnostic evaluation should include a complete blood count with differential and review of the peripheral smear, histologic review of the bone marrow aspirate and biopsy, immunophenotypic analysis of the tumor cells, and measurement of serum immunoglobulins. Other studies that may be of use in determining the diagnosis include a histologic review of splenic tissue and genetic mutation analysis.

The diagnosis of SMZL is made based on evaluation of tumor cell morphology and immunophenotype, cytogenetic analysis, bone marrow histology, and spleen histology, when available [22]. The diagnosis of SMZL can most readily be made according to one of the following scenarios:

●The splenic histology and immunophenotypic pattern are consistent with SMZL, as described above.

●If splenic tissue is not available for histologic review, the diagnosis of SMZL can be made in a patient with splenomegaly and typical morphologic and immunophenotypic findings on blood smear and bone marrow biopsy.

DIFFERENTIAL DIAGNOSIS — Most patients with SMZL present with splenomegaly and lymphocytosis. There are numerous infectious (eg, infectious mononucleosis, pertussis, toxoplasmosis) and neoplastic conditions that can present with these findings. (See "Approach to the child with lymphocytosis or lymphocytopenia" and "Splenomegaly and other splenic disorders in adults", section on 'Splenomegaly'.)

Reactive follicular hyperplasia is much more common than SMZL in children, adolescents, and young adults, but is generally mild in degree. The enlarged spleen may have expanded white pulp on cut section, producing a superficial resemblance to lymphoma, but on microscopy the white pulp is expanded by normal-appearing reactive follicles and there is little or no lymphoid expansion in red pulp zones. The polyclonal nature of suspected reactive hyperplasia can be confirmed by flow cytometry and/or molecular studies of antigen receptor gene rearrangements.

The most common neoplastic conditions that mimic SMZL are other low grade B cell proliferations involving the spleen (table 2). The most reliable means of distinguishing these tumors is immunophenotyping, but these distinctions are not absolute, and in some instances the diagnosis must be based on the combination of immunophenotypic, genetic, and morphologic findings.

Extranodal and nodal marginal zone lymphoma — Some cases of marginal zone lymphoma (MZL) originating at mucosal sites (extranodal MZL) can disseminate to nodal sites, spleen, and bone marrow [1]. In such cases it may be unclear exactly where the disease originated. One definitive clue is detection of t(11;18), t(14;18) (in this case involving the MALT1 gene on chromosome 18), t(1;14), or t(3;14), all of which may be seen in extranodal MZL, but not SMZL. By contrast, NOTCH2 mutations have not been reported in extranodal MZL, but testing for this abnormality is not routinely performed. (See "Clinical manifestations, pathologic features, and diagnosis of extranodal marginal zone lymphoma of mucosa associated lymphoid tissue (MALT)".)

Nodal MZL is a primary nodal lymphoma with features identical to lymph nodes involved by extranodal MZL, but without evidence of extranodal disease [1]. Nodal MZL does not have splenic involvement, but closely resembles SMZL genetically, including the presence of frequent mutations in NOTCH2, MLL2, and KLF2 [42]. Nodal MZL has also been reported to frequently harbor mutations in the gene PTPRD, which encodes a receptor-type tyrosine phosphatase that regulates cell growth [42]; it has been proposed that these mutations are specific for nodal MZL, but additional study is needed to confirm this genetic distinction.

Other splenic B cell neoplasms — SMZL must be distinguished from other low grade B cell proliferations involving the spleen (table 1 and table 2). These include:

●Hairy cell leukemia (HCL) – Both HCL and SMZL can present with splenomegaly and circulating lymphocytes with cytoplasmic projections but are easily differentiated from each other by immunophenotype. Unlike HCL, SMZL tumors do not express CD103, CD25, or annexin A1. In addition, splenic white pulp involvement or nodular bone marrow involvement excludes the diagnosis of HCL. Classic HCL is virtually always associated with a V600E mutation involving the serine/threonine kinase BRAF, while up to 50 percent of a related neoplasm called HCL variant has activating mutations in MAP2K1 (MEK1). These mutations are by contrast rare or nonexistent in SMZL. (See "Clinical features and diagnosis of hairy cell leukemia".)

●Lymphoplasmacytic lymphoma (LPL) – Both LPL and SMZL can have plasma cell differentiation and similar patterns of spleen and bone marrow involvement. The presence of prominent "monocytoid" B cells in the splenic white pulp marginal zones is typical of SMZL and can be helpful in making this distinction, as can the presence of "villous" lymphocytes in the peripheral blood. Although SMZL may be associated with a small monoclonal immunoglobulin spike, large spikes (>0.5 gm/dL) are not seen and, if present, strongly support the diagnosis of LPL. In addition, virtually all LPLs have mutations in MYD88, while MYD88 mutations are relatively uncommon in SMZL. (See "Clinical manifestations, pathologic features, and diagnosis of lymphoplasmacytic lymphoma".)

●Mantle cell lymphoma (MCL) – Both MCL and SMZL can present with splenomegaly and peripheral blood involvement and can infiltrate both the red and white pulp of the spleen. In general, these two can be distinguished based on histology and immunophenotype. On histology, MCL tumor cells are typically monomorphous small to medium-sized B lymphocytes while SMZL often contains a population of large immunoblast-like cells, cells with abundant pale cytoplasm (monocytoid B cell-like cells), and/or cells exhibiting plasmacytic differentiation. MCL typically expresses CD5, while only a minority of SMZL cases express CD5. In cases of SMZL that express CD5, MCL can be excluded based on the absence of cyclin D1 and SOX11 expression and the absence of the t(11;14)(q13;q32). (See "Mantle cell lymphoma: Epidemiology, pathobiology, clinical manifestations, diagnosis, and prognosis".)

●Chronic lymphocytic leukemia (CLL) – While most cases of splenic involvement in CLL demonstrate diffuse infiltration of both the red and white pulp with effacement of the follicles, there are some cases that appear micronodular and involve the marginal zones of the white pulp. In addition, both CLL and SMZL can express CD5, CD23, CD43, and IgD, although expression of these is much more typical of CLL. In difficult cases, evaluation of the hilar lymph nodes, peripheral blood, and bone marrow must be used in concert to determine the most likely diagnosis. In particular, the presence of proliferation centers is pathognomonic for CLL and excludes the diagnosis of SMZL. (See "Clinical features and diagnosis of chronic lymphocytic leukemia/small lymphocytic lymphoma".)

●Follicular lymphoma (FL) – While both SMZL and FL have a nodular growth pattern, the nodules of SMZL tend to be more uniform in size. FL cells are most often found in the white pulp and consist of a mixture of centrocytes (small cleaved cells) and centroblasts. Unlike SMZL, FL cells express CD10 and BCL6, and frequently harbor BCL2 gene rearrangements. (See "Clinical manifestations, pathologic features, diagnosis, and prognosis of follicular lymphoma".)

●Splenic B cell neoplasms – There are a small number of splenic B cell neoplasms comprised predominantly of small cells that do not fit any of the typical diagnostic categories. One entity described in the World Health Organization classification and the International Consensus Classification carries the descriptive name "splenic diffuse red pulp small B cell lymphoma" [43,44]. As the name implies, this tumor shows diffuse involvement of red pulp cords and sinusoids and does not involve the white pulp follicles or marginal zones. Marrow involvement most commonly takes the form of sinusoidal infiltrates; while nodular interstitial infiltrates can also be seen in the marrow, marginal zone growth around reactive-appearing follicles, often seen in SMZL, is absent.

Monoclonal B cell lymphocytosis — The term monoclonal B cell lymphocytosis (MBL) is used to categorize individuals who have an absolute increase in the number of clonal B lymphocytes in the peripheral blood that does not exceed 5000/microL (5 x 10⁹/L) and who have no lymphadenopathy, organomegaly, cytopenias, or disease-related symptoms. It has been recognized that MBL with a CLL-like phenotype (CD5+, CD23+) is present years prior to the diagnosis in virtually all patients with CLL. (See "Monoclonal B cell lymphocytosis".)

MBL with a SMZL-like phenotype (CD20+, CD49d+, CD10–, CD5– or weak) can be detected using high sensitivity flow cytometry in some otherwise asymptomatic patients with persistent lymphocytosis. In a retrospective analysis of 102 patients with MZL-like MBL and a median follow-up of five years, the vast majority (83 percent) remained stable and a minority (15 percent) developed splenomegaly [45]. All were seronegative for hepatitis C. Only three patients required treatment for progression to symptomatic disease.

The clinical significance of SMZL-like MBL is unclear. As only a small minority of these persons develop a lymphoproliferative disorder, it is important not to cause undue anxiety. We follow persons identified with MBL at yearly intervals with repeat flow cytometry of blood lymphocytes performed according to the physician's clinical judgment. Perhaps more important is the judicial use of flow cytometry in asymptomatic patients in order to avoid over-diagnosis of clinically insignificant lymphoproliferative conditions.

MANAGEMENT

Pretreatment testing — The pretreatment evaluation of a patient with SMZL is focused on determining the extent of disease and comorbidities that may impact on the decision to initiate therapy and the choice of therapy. In general, we perform the following examinations:

●Physical examination with determination of the patient's performance status by the Eastern Cooperative Oncology Group (ECOG) or Karnofsky performance scales (table 3 and table 4).

●Laboratory testing includes a complete blood count (CBC) with differential and platelet count, chemistries with liver and renal function and electrolytes, lactate dehydrogenase (LDH), serum protein electrophoresis, quantitative immunoglobulin testing, and serology for hepatitis C virus (HCV). Patients should also be tested for hepatitis B virus prior to the use of rituximab therapy. Women of childbearing age should undergo a pregnancy test if chemotherapy is planned.

●Unilateral bone marrow biopsy and aspirate.

●Imaging should include a contrast-enhanced computed tomography (CT) scan of the chest, abdomen, and pelvis. We obtain positron emission tomography (PET) scan if large cell transformation is suspected clinically. Unlike in some lymphoma subtypes, FDG PET/CT has a low sensitivity for identifying bone marrow involvement in SMZL. In a series of 42 patients with SMZL of whom 83 percent had pathologically confirmed bone marrow involvement, FDG PET/CT had the following test characteristics for identifying bone marrow involvement: sensitivity 31 percent, specificity 100 percent, positive predictive value 100 percent, negative predictive value 23 percent [46]. (See 'Large cell transformation' below.)

●Stage is determined with the Lugano modifications of the Ann Arbor staging system (table 5).

Therapeutic strategy

Indications for therapy — Not all patients with SMZL require immediate treatment since SMZL is not curable and generally associated with reasonably long survival, measured in years, even if initially untreated [47,48]. Asymptomatic patients without splenomegaly, anemia, thrombocytopenia, or leukopenia are observed initially (algorithm 1).

We agree with consensus guidelines that offer therapy to patients with splenomegaly and one or more of the following signs or symptoms [22,49,50]:

●Local symptoms related to splenomegaly (eg, a sense of fullness or discomfort in the left upper quadrant, pain referred to the left shoulder, or early satiety)

●Cytopenias due to extensive bone marrow infiltration or hypersplenism

In addition, the following comorbidities or complications require specific therapy:

●Autoimmune hemolytic anemia or thrombocytopenia should be treated with specific therapies for these complications. (See 'Managing autoimmune complications' below.)

●Patients with coexisting hepatitis C infection should be evaluated for antiviral therapy. (See 'Antiviral therapy for hepatitis C' below.)

Asymptomatic patients — Asymptomatic patients without splenomegaly, anemia, thrombocytopenia, or leukopenia are observed initially (algorithm 1). A retrospective analysis evaluated a watch and wait strategy in 63 asymptomatic patients with SMZL [51]. The median treatment-free interval was 58 months. At 10 years, 30 percent of patients still did not require therapy.

We follow asymptomatic patients in clinic every three months for the first year and then every three to six months thereafter until progressive disease is noted. At these appointments we perform a history, physical examination, and laboratory studies including a CBC with differential and LDH. Imaging studies are repeated only if clinically indicated. Progressive disease is defined by an enlarging spleen or worsening cytopenias due to bone marrow infiltration or hypersplenism.

Up to 10 percent of SMZL will undergo large cell transformation. Histologic transformation should be suspected if there is a rapid progression of splenomegaly, lymphadenopathy, infiltration of extranodal sites, the development of systemic symptoms, or an elevated serum LDH in the absence of autoimmune hemolytic anemia. Biopsy of a suspicious area is indicated in such circumstances. (See 'Large cell transformation' below.)

Managing autoimmune complications — Approximately 20 percent of patients with SMZL have an autoimmune manifestation that requires specific management described separately [24-26]. If these complications persist despite usual management, we proceed with treatment directed at the SMZL.

●Autoimmune hemolytic anemia – (See "Warm autoimmune hemolytic anemia (AIHA) in adults" and "Warm autoimmune hemolytic anemia (AIHA) in adults", section on 'Initial management'.)

●Immune thrombocytopenia – (See "Immune thrombocytopenia (ITP) in adults: Clinical manifestations and diagnosis" and "Initial treatment of immune thrombocytopenia (ITP) in adults".)

●Cold agglutinin disease – (See "Cold agglutinin disease".)

●Antiphospholipid antibodies – (See "Diagnosis of antiphospholipid syndrome", section on 'Other conditions associated with antiphospholipid antibodies'.)

●Acquired von Willebrand disease – (See "Acquired von Willebrand syndrome".)

●Angioedema as a result of acquired C1-esterase inhibitor deficiency presents as episodes of angioedema lasting two to five days and unresponsive to antihistamines or glucocorticoids – (See "Acquired C1 inhibitor deficiency: Clinical manifestations, epidemiology, pathogenesis, and diagnosis".)

Antiviral therapy for hepatitis C — Antiviral treatment is a standard part of the management of patients with virologic evidence of chronic HCV infection (ie, detectable HCV viral level over a six-month period). Therapy directed at the HCV infection may result in regression of SMZL. An initial trial of treatment directed at the HCV is indicated for patients who are asymptomatic or mildly symptomatic from their lymphoma (algorithm 1). (See "Overview of the management of chronic hepatitis C virus infection".)

Data supporting the use of HCV-directed therapy in SMZL comes from observational studies that have noted an association between the two diseases and case series demonstrating clinical remissions in some patients [52-54].

In one study, 46 patients with indolent B cell lymphoma and chronic HCV infection were treated with interferon-free HCV treatment regimens (direct-acting antivirals) [55]. Responses were seen in 33 of 37 patients with marginal zone lymphoma. Of the 17 patients with SMZL, 4 achieved a complete remission, 7 a partial response, and 5 had stable disease. After a median follow-up of eight months, estimated progression-free survival (PFS) and overall survival (OS) rates at one year were 75 and 98 percent, respectively.

Similar treatment results have been noted in other reviews [56,57], and in two series of patients with HCV infection and various forms of lymphoma, including follicular, plasmacytoid, as well as marginal zone lymphomas of the splenic, nodal, and extranodal types [58,59].

Treatment

Choice of therapy — As described above, not all patients with SMZL require immediate therapy directed at the malignant cells. The management of asymptomatic patients is focused on the control of autoimmune complications (eg, autoimmune hemolytic anemia, immune thrombocytopenia, cold agglutinin disease). Those with HCV infection are evaluated for antiviral therapy.

The management of symptomatic patients is principally determined by the extent of involvement and whether there is HCV infection (algorithm 1):

●For minimally or moderately symptomatic patients with HCV infection, we suggest a trial of antiviral therapy rather than rituximab or splenectomy. Antiviral therapy is not appropriate for patients without HCV infection. (See 'Antiviral therapy for hepatitis C' above.)

●For minimally or moderately symptomatic patients without HCV infection, we suggest single agent rituximab rather than splenectomy, rituximab plus chemotherapy, or the use of novel agents. (See 'Single agent rituximab' below.)

●For patients with severe local symptoms due to splenomegaly (eg, abdominal pain, early satiety with weight loss), we suggest splenectomy rather than other options. Splenectomy is not expected to benefit patients with disseminated nodal involvement outside of the splenic hilum or those with cytopenias due to extensive bone marrow involvement. (See 'Splenectomy' below.)

●Splenectomy is also indicated in patients with suspected large cell transformation within the spleen (eg, nodular splenic lesion with increased activity on PET). (See 'Large cell transformation' below.)

●The combination of rituximab plus chemotherapy and novel agents are usually reserved for patients with SMZL that has relapsed after or is refractory to treatment with single agent rituximab. (See 'Rituximab plus chemotherapy' below and 'Targeted therapies' below.)

There have been no randomized trials comparing these approaches and there are few prospective trials in patients with SMZL. Data regarding treatment come largely from retrospective series and from extrapolation of experience in other indolent B cell lymphomas. A choice for an individual patient is largely made based upon the anticipated side effects and risks.

Often there is no better therapy to offer a patient than enrollment onto a well-designed, scientifically valid, peer-reviewed clinical trial. Additional information and instructions for referring a patient to an appropriate research center can be obtained from the United States National Institutes of Health.

Systemic therapy — The following sections describe the data supporting the use of various systemic therapies in SMZL.

Single agent rituximab — Single agent rituximab is our preferred systemic therapy for most patients with symptomatic SMZL (algorithm 1). This preference is based on high response rates seen in retrospective series and the expected decrease in short- and long-term complications when compared with splenectomy. While not curative, >90 percent of patients will have improvements in splenomegaly and normalization of absolute lymphocyte counts [60-66]. Responses are usually seen within three weeks and are sustained for more than five years.

While different schedules have been used, a reasonable initial schedule is rituximab 375 mg/m² for four weekly doses [67-69]. If well tolerated, maintenance therapy may be offered with four additional doses of rituximab administered at two-month intervals. These doses and schedules are extrapolated from those used in follicular lymphoma. Maintenance therapy is likely to prolong PFS but has not been shown to improve OS. (See "Initial treatment of stage II to IV follicular lymphoma", section on 'Immunotherapy alone'.)

The major toxicities of rituximab include infusion reactions (ie, fevers, rigors, and hypotension) and infections related to immunosuppression. In our experience, the incidence of infusion reactions is higher in SMZL than in other indolent non-Hodgkin lymphomas. In an effort to decrease infusion reactions, we use split-dose rituximab, which is administered over two days. On the first day of cycle one, rituximab 50 mg/m² is infused over four hours at a continuous rate [70]. The remainder of the first dose (325 mg/m²) is administered on the second or third day of cycle one.

Initial studies suggest a risk of HCV reactivation with rituximab, although the magnitude of this risk is poorly understood. In one small study of HCV-infected patients with various malignancies, HCV reactivation occurred in 18 of 50 patients with hematologic malignancies [71]. Hepatitis flare was seen in a little less than half of reactivations and no patients had liver failure or liver-related death within 36 weeks after initiation of cancer treatment. Independent variables associated with HCV reactivation included the use of rituximab (odds ratio 9), the use of high dose steroids (odds ratio 5), and higher baseline viral load (odds ratio 0.12). In another small, retrospective study rituximab-containing chemotherapy was associated with an increase in HCV viral load [72].

Rituximab also imposes a risk of hepatitis B reactivation among patients positive for hepatitis B surface antigen (HBsAg) or antibodies against hepatitis B core antigen (anti-HBc). (See "Infusion-related reactions to therapeutic monoclonal antibodies used for cancer therapy", section on 'Rituximab' and "Secondary immunodeficiency induced by biologic therapies", section on 'Rituximab' and "Hepatitis B virus reactivation associated with immunosuppressive therapy".)

Data regarding the use of rituximab in SMZL come from retrospective series. As always, caution must be used in comparing treatment options based on data generated in a nonrandomized fashion. As examples:

●A retrospective study reported the outcomes of 108 patients with symptomatic SMZL treated with six weekly infusions of rituximab [66]. Of the 98 patients responding to initial rituximab, 76 received maintenance rituximab administered every two months for one to two years. Overall, responses were seen in 92 percent, with 65 percent complete responses or unconfirmed complete responses. After a median follow-up of 57 months, estimated 5- and 10-year PFS rates were 71 and 64 percent. Corresponding OS rates were 93 and 85 percent. Maintenance was associated with improved PFS, but not OS.

●In a single institution retrospective analysis, patients with SMZL were treated with rituximab alone (26 patients), chemotherapy alone (11 patients), or rituximab plus chemotherapy (6 patients) [60]. When compared with those who received chemotherapy alone, patients who received single agent rituximab had superior rates of overall response (88 versus 55 percent), three-year OS (86 versus 45 percent), and three-year failure-free survival (86 versus 45 percent). Outcomes in patients who received both rituximab and chemotherapy were similar to those seen for patients who received rituximab alone.

●In another retrospective analysis, patients with symptomatic SMZL were treated with single agent rituximab (58 patients) or splenectomy (27 patients) [61]. Rituximab was associated with an overall response rate of 95 percent (71 percent complete) with a median time to clinical response of three weeks. At a median follow-up of three years, estimated rates of OS and PFS at five years were 92 and 73 percent, respectively. Splenectomy resulted in an overall response rate of 85 percent with estimated PFS and OS at five years of 58 and 77 percent, respectively.

Rituximab plus chemotherapy — The combination of rituximab plus chemotherapy is usually reserved for patients with SMZL that has relapsed after or is refractory to treatment with single agent rituximab (algorithm 1). This is primarily because chemotherapy adds toxicity, and it is not clear that the addition of chemotherapy improves outcomes in previously untreated patients. In retrospective analyses, outcomes in patients who received both rituximab and chemotherapy as initial therapy were similar to those seen for patients who received rituximab alone [60].

Data regarding the efficacy of various regimens comes from clinical trials in which patients with SMZL accounted for a minority of cases and from retrospective series. Most reports describe the efficacy for all subtypes of marginal zone lymphoma as a group, rather than for SMZL in particular. Data regarding the efficacy of these therapies and the choice of regimen is presented in detail separately. (See "Nodal marginal zone lymphoma", section on 'Treatment'.)

An international phase II trial (BRISMA/IELSG36) evaluated the efficacy of six cycles of bendamustine plus rituximab as initial systemic therapy in 56 patients with symptomatic SMZL [73]. Responses were documented in 91 percent of patients and 73 percent achieved a complete response. Estimated PFS and OS at three years were 90 and 96 percent, respectively. Toxicity was similar to other trials of bendamustine plus rituximab in indolent lymphoma. Phase III trials evaluating this regimen in marginal zone lymphoma is discussed in more detail separately. (See "Nodal marginal zone lymphoma", section on 'Incorporation of chemotherapy'.)

Targeted therapies — Prospective studies of targeted therapies (eg, Bruton tyrosine kinase inhibitors, PI3K inhibitors, lenalidomide) have included small numbers of patients with SMZL. Several targeted agents are approved by the US Food and Drug Administration for the treatment of relapsed marginal zone lymphoma. Most reports describe the efficacy for all subtypes of marginal zone lymphoma as a group, rather than for SMZL in particular. The use of novel agents in marginal zone lymphoma is discussed in more detail separately. (See "Nodal marginal zone lymphoma", section on 'Treatment'.)

Splenectomy — While systemic therapies (eg, rituximab) are preferred for most patients with SMZL, splenectomy is an acceptable alternative for a subset of patients with SMZL and may be preferred in certain scenarios. As examples:

●Splenectomy is indicated in patients with suspected large cell transformation within the spleen (eg, nodular splenic lesion with increased activity on PET). (See 'Large cell transformation' below.)

●Splenectomy is an acceptable alternative to other therapies in patients with isolated SMZL plus splenomegaly with severe localized symptoms such as abdominal pain, early satiety, or weight loss. Patients who otherwise meet these criteria but also demonstrate minimal bone marrow involvement are sometimes considered for splenectomy as well.

●Splenectomy is not expected to benefit patients with disseminated nodal involvement outside of the splenic hilum or those with cytopenias due to extensive bone marrow involvement.

Splenectomy can be performed as an open or laparoscopic procedure and allows for a definitive pathologic diagnosis, relieves abdominal symptoms due to splenomegaly, and reverses cytopenias due to splenic sequestration and/or immune destruction. Perioperative complications are seen in approximately one-quarter of patients and may include major bleeding, pulmonary complications, and thromboses [74]. Long-term complications include immune suppression with an increased risk for infectious complications, and thromboembolism. (See "Elective (diagnostic or therapeutic) splenectomy".)

Whenever elective splenectomy is considered, patients should undergo appropriately timed preoperative immunization. (See "Prevention of infection in patients with impaired splenic function", section on 'Vaccinations'.)

Splenectomy is not curative. Large retrospective studies have reported improvement of cytopenias and relief of localized symptoms sustained for more than five years with splenectomy alone [20,21,38,75-81]. It is not clear whether this translates into a survival benefit. Some but not all of these retrospective studies suggested improvements in OS for those treated with splenectomy when compared with those treated with rituximab-containing and non-rituximab-containing chemotherapy. In contrast, splenectomy did not appear to offer an OS or lymphoma-specific survival advantage in an analysis of the Surveillance Epidemiology and End Results (SEER) registry that included 1671 patients with SMZL, approximately 40 percent of whom had undergone splenectomy [82].

A few case series have reported improvements in patient symptoms with splenic irradiation in those who are not candidates for splenectomy [83-85].

POST-TREATMENT MANAGEMENT — After completion of initially planned treatment of SMZL, patients should be evaluated to determine the disease response to treatment and should be followed longitudinally for relapse.

Evaluating response to therapy — The main goal of treatment is to reverse the symptoms and signs that lead to the initiation of therapy. We evaluate patients three months following the completion of therapy to assess response and determine a new baseline of disease with which to compare at the time of suspected progression. At this evaluation, we usually perform a complete blood count with differential and a contrast-enhanced computed tomography (CT) scan of the chest, abdomen, and pelvis. Responses are designated using the Lugano response criteria (table 6) [86,87]. Specific response criteria for SMZL have also been proposed for clinical trials [22]. Using these latter criteria, a complete response is achieved when splenomegaly has been resolved, blood cell counts are normalized, flow cytometry on blood is negative, and bone marrow histology is negative by immunohistochemistry.

Monitoring for relapse — Patients are seen at periodic intervals to monitor for treatment complications and assess for possible relapse. The frequency and extent of these visits depends on the comfort of both the patient and physician. There have been no prospective, randomized trials comparing various schedules of follow-up. When planning the post-treatment surveillance strategy, care should be taken to limit the number of CT scans, particularly in younger individuals, given concerns about radiation exposure and the risk for second malignancies. We generally follow patients every three to four months for the first year and then every six months, if stable. (See "Radiation-related risks of imaging".)

Patients with disease progression after initial therapy are candidates for systemic therapy with rituximab plus chemotherapy or targeted agents. (See 'Rituximab plus chemotherapy' above and 'Targeted therapies' above.)

Often there is no better therapy to offer a patient than enrollment onto a well-designed, scientifically valid, peer-reviewed clinical trial. Additional information and instructions for referring a patient to an appropriate research center can be obtained from the United States National Institutes of Health.

Large cell transformation — Approximately 5 to 10 percent of patients undergo transformation to diffuse large B cell lymphoma [79,88-90]. Large cell transformation should be suspected in patients with systemic B symptoms (eg, fatigue, night sweats, weight loss), elevations in lactate dehydrogenase, and/or rapidly enlarging lymph nodes or splenic lesions. If large cell transformation is suspected clinically, a positron emission tomography (PET) scan can help guide biopsy site selection.

There is little data to guide the selection of therapy in the setting of large cell transformation. If the PET scan is consistent with large cell transformation in the spleen, we suggest splenectomy for definitive diagnosis and local control. We manage large cell transformation occurring outside of the spleen in a similar fashion to histologic transformation of follicular lymphoma. This is discussed in more detail separately. (See "Histologic transformation of follicular lymphoma".)

In a retrospective analysis of 36 patients with SMZL with transformation, median survival from transformation was approximately five years [90]. Management was not standardized and chemoimmunotherapy (21 patients), chemotherapy (6 patients), splenectomy (3 patients), and radiotherapy (1 patient).

PROGNOSIS — The course of SMZL is generally indolent, with a median overall survival in excess of 10 years, and estimated survival rates at 5, 10, and 15 years of 74, 53, and 38 percent, respectively [82]. The survival rates have improved with the incorporation of rituximab [51,91].

However, some patients appear to have a more aggressive course with a median survival of 18 months [92]. Two main prognostic indices have been developed for patients with SMZL [93,94]:

●The SMZL Study Group (SMZLSG) prognostic index (calculator 1)

●The Intergruppo Italiano Linfomi (IIL) prognostic index

Of these, the SMZLSG prognostic index appears to be more cumbersome, but better able to discriminate between low- and intermediate-risk groups. However, these indices are neither 100 percent sensitive nor 100 percent specific for identifying high-risk patients.

The IIL prognostic index was developed based on data from 309 patients with SMZL diagnosed between 1989 and 2004 (table 7) [93]. The overall cause-specific five-year survival rate was 76 percent, with the following three factors being predictive of shorter survival on multivariate analysis:

●Hemoglobin <12 g/dL

●Serum lactate dehydrogenase (LDH) level greater than normal

●Serum albumin level <3.5 g/dL

Patients who had zero, one, or two to three of these adverse risk factors had five-year cause-specific survival rates of 88, 73, and 50 percent, respectively. When applied to an expanded cohort of 450 patients with SMZL, corresponding five-year survival rates were 95, 90, and 79 percent, respectively [94].

The SMZLSG prognostic index was developed using data from 366 patients with SMZL and validated in an independent cohort of 227 patients (calculator 1) [94]. This score is determined by entering the hemoglobin (g/L), platelet count (10⁹/L), LDH (LDH score one point if high, zero points if normal), and the presence of extrahilar lymphadenopathy (LAD score, one point if present, zero points if absent) into the following equation:

Using this information, patients were separated into three prognostic groups with the following estimated lymphoma-specific survival at five years:

●Low risk (PI ≥2.6) – 94 and 96 percent in derivation and validation cohorts, respectively

●Intermediate risk (PI ≥0.9 and <2.6) – 78 and 88 percent in derivation and validation cohorts, respectively

●High risk (PI <0.9) – 69 and 44 percent in derivation and validation cohorts, respectively

The SMZLSG prognostic index was compared with the IIL prognostic index using a cohort of 450 patients with SMZL [94]. While both scores were able to distinguish the high-risk group from the other two, the SMZLSG was better able to separate the low- and intermediate-risk groups.

Other prognostic factors have been investigated, including abnormalities in p53 (deletion and/or expression) and expression of NF-KB signature genes, but such tests are not currently clinically applicable [2,31,92,95].

Similar to other indolent lymphomas, marginal zone lymphoma has the potential to transform into a high-grade lymphoma. Such patients have a poor prognosis. The diagnosis and treatment of histologic transformation to an aggressive lymphoma is discussed in more detail separately. (See 'Large cell transformation' above and "Histologic transformation of follicular lymphoma".)

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Lymphoma diagnosis and staging" and "Society guideline links: Marginal zone lymphoma".)

SUMMARY AND RECOMMENDATIONS

●Epidemiology and presentation – Splenic marginal zone lymphoma (SMZL) is an uncommon B cell non-Hodgkin lymphoma of adults that usually presents with splenomegaly and lymphocytosis. The median age at diagnosis is 65 years. There is no sex predominance. (See 'Epidemiology' above.)

Patients typically present with splenomegaly, lymphocytosis, and cytopenias (often due to hypersplenism). Unlike most other non-Hodgkin lymphomas, lymphadenopathy, involvement of extralymphatic organs, systemic B symptoms, and elevated lactate dehydrogenase levels are uncommon. (See 'Clinical features' above.)

●Diagnosis – The diagnosis of SMZL is made based on evaluation of the lymphocyte morphology (picture 1), immunophenotype, cytogenetic analysis, bone marrow histology, and spleen histology (picture 2), when available. (See 'Diagnosis' above and 'Morphology' above.)

The tumor cells express surface immunoglobulin (eg, IgM, IgD), B cell antigens (CD19, CD20, CD22), and Bcl-2. They typically do not express CD10, CD25, or CD103, and also usually (but not always) are negative for CD5, CD23, and CD43. (See 'Immunophenotype' above.)

The differential diagnosis includes infectious (eg, infectious mononucleosis, pertussis, toxoplasmosis) and neoplastic conditions other than SMZL that can present with splenomegaly and lymphocytosis (table 2). (See 'Differential diagnosis' above.)

●Management – Not all patients require immediate treatment (algorithm 1). For asymptomatic patients with SMZL without splenomegaly, anemia, thrombocytopenia, or leukopenia, we suggest observation rather than treatment directed at the underlying tumor (Grade 2C). Autoimmune complications (eg, autoimmune hemolytic anemia, immune thrombocytopenia) may need specific management. (See 'Indications for therapy' above and 'Managing autoimmune complications' above.)

Those with hepatitis C virus (HCV) infection are offered antiviral therapy (as otherwise indicated). Therapy directed at the HCV infection may result in regression of SMZL. For patients with HCV who are asymptomatic or mildly symptomatic from their lymphoma, we suggest a trial of antivirals with deferral of other therapies (Grade 2C). (See 'Antiviral therapy for hepatitis C' above.)

For most symptomatic patients who do not have concomitant HCV, we suggest treatment with rituximab rather than splenectomy or observation (Grade 2C). Splenectomy is an acceptable alternative and may be preferred in patients with localized symptoms due to splenomegaly who do not have involvement beyond the hilar lymph nodes. (See 'Single agent rituximab' above and 'Splenectomy' above.)

The combination of rituximab plus chemotherapy is usually reserved for patients with SMZL that has relapsed after or is refractory to treatment with single agent rituximab. For SMZL progressing after rituximab, we usually prefer to add bendamustine (ie, bendamustine plus rituximab) before using targeted therapies (eg, Bruton tyrosine kinase inhibitors, PI3K inhibitors, lenalidomide). (See 'Rituximab plus chemotherapy' above and 'Targeted therapies' above.)

●Prognosis – The course of SMZL is usually indolent, with a median overall survival of greater than 10 years. However, some patients have a more aggressive course. Levels of hemoglobin, lactate dehydrogenase, and albumin have been proposed as potential predictors of survival. (See 'Prognosis' above.)

Similar to other indolent lymphomas, SMZL has the potential to transform into a high-grade lymphoma. If imaging suggests large cell transformation in the spleen, we suggest splenectomy for definitive diagnosis and local control (Grade 2C). This is discussed in more detail separately. (See 'Large cell transformation' above.)