Marginal zone lymphoma: old, new, targeted, and epigenetic therapies

Abstract

Marginal zone lymphoma (MZL) is an indolent B-cell lymphoma arising from marginal zone B-cells present in lymph nodes and extranodal tissues. MZL comprises 5–17% of all non-Hodgkin’s lymphomas in adults. The World Health Organization categorizes MZL into three distinct types based on their site of impact: (1) splenic marginal zone lymphoma (SMZL); (2) nodal marginal zone lymphoma (NMZL); (3) extranodal mucosa-associated lymphoid tissue (MALT) lymphoma, which can be subdivided into gastric and nongastric. The subgroups of MZL share some common features but are different in their biology and behavior. Owing to the rarity of MZL there are few randomized trials available comparing various treatment options and therefore treatment is controversial, lacking standard guidelines. Treatment should be patient tailored and can range from a ‘watchful waiting’ approach for asymptomatic patients without cytopenias to surgery or localized radiation therapy. Rituximab in combination with chemotherapy has resulted in longer failure-free survival than chemotherapy alone in patients with SMZL. Helicobacter pylori positive gastric MALT shows a good response rate to triple antibiotic therapy. Newer therapies such as bendamustine, everolimus, lenalidomide, vorinostat and phosphoinositide 3-kinase inhibitors are in clinical trials for patients with relapsed or refractory MZL and have shown promising results. We are presently conducting clinical trials testing the efficacy of the epigenetic activity of cladribine as a hypomethylating agent in combination with the histone deacetylase inhibitor (HDACi) vorinostat and rituximab in patients with MZL. Further studies with the newer agents should be done both in newly diagnosed or relapsed/refractory MZL to streamline the care and to avoid the use of toxic chemotherapies as initial treatment.

Introduction

Marginal zone lymphoma (MZL) is a group of indolent B-cell lymphomas that arises from marginal zone B cells present in lymph nodes and extranodal tissues, such as spleen and mucosal lymphoid tissues. According to the World Health Organization (WHO) classification MZL consists of the following subtypes: splenic marginal zone lymphoma (SMZL) with or without villous lymphocytes, nodal marginal zone lymphoma (NMZL) and extranodal marginal zone B-cell lymphoma, also called mucosa-associated lymphoid tissue (MALT) [Harris et al. 1994; Chan et al. 1995]. They comprise about 5–17% of all non-Hodgkin’s lymphoma (NHL) in adults [Cervetti et al. 2010]. The three distinct groups have been classified together because they all seem to originate from post-germinal center marginal zone B cells. They share common immunophenotype markers, including CD19, CD20 and CD22. They are negative for CD5, CD10 and (usually) CD23 [Harris et al. 1994; Swerdlow, 2008]. See Table 1 for further details of the immunophenotypic markers in different types of MZL.

Table 1.

Immunophenotypic markers in different types of marginal zone lymphoma (MZL).

	Splenic marginal zone lymphoma (SMZL)	Nodal marginal zone lymphoma (NMZL)	Mucosa-associated lymphoid tissue (MALT) lymphoma
Most common Markers
IgG	+	+	+
IgD	+	- (MALT type NMZL)	-
		+(Splenic type NMZL)
BCL-2	+	+	+
CD19	+	+	+
CD20	+	+	+
CD22	+	+	+
CD43	-	+ in 50% cases	variable
CD79a	+	+	+
FMC7	+	+	+
CD103	-	-	-
CD5	- Helps differentiate it from chronic lymphocytic leukemia (CLL)/mantle cell (can be positive in rare cases)	-	- (can be positive in rare cases)
CD10	- Helps differentiation from follicular	-	-
CD23	- Helps differentiation from CLL (can be positive in rare cases)	-	- (+ in rare cases)
Cyclin D1	- Helps differentiation from Mantle Cell		-

MZL can be linked to a variety of chronic infections. Helicobacter pylori has been associated with gastric MALT; the hepatitis C virus (HCV) is associated in some cases of SMZL; other microbes implicated in the pathogenesis are Campylobacter jejuni, Borrelia burgdorferi, and Chlamydia psittaci [Kahl and Yang, 2008]. A variety of autoimmune diseases have been associated with their development, such as rheumatoid arthritis, Sjögren’s syndrome, systemic lupus erythematosus and Wegner’s granulomatosis [Zucca et al. 2010a]. Initial lymphoid hyperplasia is caused by antigenic stimulation followed by additional oncogenic events, including translocation, which then activate the pathway causing the development of the disease [Kahl and Yang, 2008]. To date there are no large randomized prospective studies comparing different treatment options and therefore physicians face a challenge in their decisions about evidence-based treatment for an individual patient. In this paper we review the treatment of MZL, in particular, paying more attention to the newer therapies. The review is split into the following sections:

1. SMZL;

2. NMZL;

3. MALT lymphoma: gastric and nongastric;

4. Newer agents and future directions;

5. Conclusions.

SMZL

This disease usually affects elderly or middle-age patients with median survival of >10 years [Troussard et al. 1996; Berger et al. 2000; Kahl and Yang, 2008; Cervetti et al. 2010]. SMZL has a relatively indolent course, but some patients can suffer a more rapid course. The Integruppo Italiano Linfomi developed the prognostic categorization based on three risk factors: (1) hemoglobin (Hb) <12 g/dl; (2) abnormal lactate dehydrogenase (LDH); and (3) albumin <3.5 g/dl. The patients were grouped into three categories of increasing severity: (1) low risk when none of the risk factor are present (5-year lymphoma specific survival of 88%); (2) intermediate risk when one risk factor is present (5-year lymphoma specific survival of 73%); and (3) high risk when two or more risk factors are present (5-year lymphoma specific survival of 50%) [Arcaini et al. 2006; Arcaini and Paulli, 2010]. In their multicenter study, Salido and colleagues suggested that complexity of the karyotype, 14q aberrations, and TP53 deletions are poor prognostic indicators and should be considered together with other clinicobiological parameters to ascertain the prognosis of SMZL [Salido et al. 2010].

In addition to the cytogenetic abnormalities detected above, more recent studies have identified molecular lesions involving nuclear factor kappa B (NF-κB). Detection of active NF-κB signaling in 14/24 (58%) SMZL prompted the investigation of NF-κB molecular alterations in 101 patients with SMZL. Mutations and copy number abnormalities of NF-κB genes occurred in 36/101 (36%) SMZL, and targeted both canonical (TNFAIP3 and IKBKB) and noncanonical (BIRC3, TRAF3, MAP3K14) NF-κB pathways. Most alterations were mutually exclusive, documenting the existence of multiple independent mechanisms affecting NF-κB in SMZL. BIRC3 inactivation in SMZL was due to somatic mutations disrupting the same RING domain that in extranodal MZL is removed by the t(11;18), pointing to BIRC3 disruption as a common mechanism in marginal zone B-cell lymphomagenesis. Genetic lesions of NF-κB provide a molecular basis for the pathogenesis of over 30% SMZL, and provide a logical target for NF-κB therapeutic targeting in this lymphoma [Rossi et al. 2011].

Splenectomy is considered a gold standard for diagnosis but a combination of bone marrow histology, clinical findings and the immunophenotype can assist with diagnosis in patients who do not require splenectomy for therapeutic reasons [Arcaini and Paulli, 2010]. Treatment of SMZL is controversial. There is no clear advantage to starting therapy early when patients are asymptomatic. Roughly two thirds of patients are asymptomatic at diagnosis and only one third require treatment [Franco et al. 2003]. A watchful waiting approach should be considered in patients with favorable prognosis. If the patients are symptomatic and require treatment, the available options include splenectomy, chemotherapy or splenic irradiation. See Table 2 for a description of the available treatment options.

Table 2.

Available treatment options for marginal zone lymphoma.

SMZL, splenic marginal zone lymphoma; NMZL, nodal marginal zone lymphoma; MALT, mucosa-associated lymphoid tissue lymphoma; HCV, hepatitis C virus; IFN, interferon; LDH, lactate dehydrogenase; CHOP, cyclophosphamide, vincristine, doxorubicin, prednisone; CVP, cyclophosphamide, vincristine and prednisone; FC, fludarabine, cyclophosphamide.

Splenectomy

Splenectomy is considered the traditional treatment of choice in patients who are symptomatic and are fit for surgery. Improvement in symptoms is seen in approximately 90% of splenectomized patients [Bennett and Schechter, 2010]. Although not curative, it generally provides improvement in symptoms related to cytopenias and also provides improved disease control.

Immunotherapy and chemotherapy

Introduction of anti-CD20 antibody rituximab significantly improved the treatment outcomes. In a Greek study of 52 patients, Kalpadakis and colleagues showed that the use of rituximab as a single agent yielded an overall response rate (ORR) of 100%, with 69% complete response (CR), 19% unconfirmed response and 12% partial response (PR) [Kalpadakis et al. 2007]. The nucleoside analogs fludarabine, cladribine and pentostatin have also demonstrated effectiveness in the treatment of SMZL both as first-line and as a salvage therapy [Bolam et al. 1997; Lefrere et al. 2004; Iannitto et al. 2005].

Clinical trials comparing rituximab alone versus in combination with chemotherapy in SMZL patients have been conducted. In a prospective study of 70 SMZL patients by Tsimberidou and colleagues both the treatment arm containing rituximab alone and rituximab in combination with chemotherapy showed a major advantage when compared with chemotherapy alone [Tsimberidou et al. 2006]. The overall response rate was 88% when administering rituximab alone, 83% in combination with chemotherapy and 55% for chemotherapy alone arm. The 3-year survival rates were 95%, 100% and 55% while the 3-year failure-free rates (FFRs) were 86%, 100% and 45%, respectively. Rituximab treatment resulted in longer survival and FFR when compared with chemotherapy with fludarabine, cyclophosphamide and rituximab (FCR) or with rituximab, fludarabine, mitoxantrone and dexamethasone (R-FMD) [Tsimberidou et al. 2006]. Low-dose chlorambucil can be considered as a treatment option for stage I–II. For later stages III–IV, more aggressive therapies such as polychemotherapy should be considered. Cladribine (2CdA), a lymphocyte-specific cytotoxic agent and purine nucleoside analog was identified to be active in indolent lymphoproliferative disease and was initially used to treat multiple sclerosis and hairy cell leukemia [Delannoy, 1996; Else et al. 2009; Orciuolo et al. 2010]. A phase II study demonstrated that the use of 2CdA showed a 60% partial response and 20% complete response in patients with SMZL and was one of the initial few studies to show that this treatment is a useful alternative to splenectomy [Riccioni et al. 2003]. There were two critical studies published in the past few years about 2CdA and rituximab combination in treating MZL. The rationale behind the combination was to have synergistic activity [Orciuolo et al. 2010]. In their retrospective study of 89 patients with MZL, Orciuolo and colleagues showed that ORR and CR were better in the combination group with rituximab and 2CdA (96.5% and 60.3%, respectively) when compared to 2CdA alone (73.1% and 38.5%, respectively) [Orciuolo et al. 2010]. This study showed significant benefit in the SMZL subgroup of 26 patients. The ORR was 100%, with 75% CR in the combination arm while it was 66.7% ORR with 22.2% CR in 2CdA alone arm [Orciuolo et al. 2010]. Another study also published in 2010 evaluated the effectiveness of 2CdA with or without rituximab in 50 SMZL patients and showed that the ORR was 87% (52% achieved a complete hematological response, 16% achieved PR and 13% did not respond). A total of 62% of patients with complete hematological response also achieved molecular remission [Cervetti et al. 2010]. In patients who achieved complete hematological response, the rate of polymerase chain reaction (PCR) negativity in the cladribine alone group was 33%, compared with 81% in post-rituximab and cladribine combination arm. Furthermore, the combination was relatively well tolerated [Cervetti et al. 2010]. Based on their study, the authors postulated that addition of rituximab to cladribine significantly improved the outcome of treatment. We have also observed the synergy of cladribine and rituximab in the treatment of B-cell malignancies and our hypothesis and preliminary data support an epigenetic synergistic mechanism [Spurgeon et al. 2009, 2010a, 2010b, 2011].

In HCV-positive cases, antiviral treatment was able to induce remission and could be used in patients who are unresponsive to anti-CD20 [Kelaidi et al. 2004; Cervetti et al. 2010]. In a different study, Cervetti and colleagues also showed that amongst HCV-related NHLs, anti-CD20 antibody caused a high clinical response of 100% with 50% of CR without any significant hepatic or extrahepatic toxicity [Cervetti et al. 2005, 2010].

NMZL

Like other nodal indolent lymphomas NMZL tends to present in advanced stages with widespread but nonbulky lymphadenopathy. Bone marrow involvement is seen in approximately one third of new cases [Kahl and Yang, 2008]. B symptoms are reported in less than 15% and the majority of patients are asymptomatic at diagnosis. A small-to-moderate IgM paraprotein can be detected in approximately 10% of cases. NMZL reports have been limited to small series of patients, and specific prognostic factors are lacking. In the series of Camacho and colleagues Kaplan–Meier analysis demonstrated that a shorter FFS was associated with loss of survivin and active caspase 3, while a shorter OS was associated with increased age (>60) and overexpression of cyclin E [Camacho et al. 2003]. Transformation to diffuse large B-cell lymphoma can occur, although the incidence is not well studied. In one series, 20 of 124 (16%) patients with NMZL transformed to diffuse large B-cell lymphoma at a median time of 4.5 years from diagnosis [Berger et al. 2000]. See Table 2 for the available treatment options.

There are no reported prospective studies for treatment of NMZL. The ‘watchful waiting’ approach is used in patients with low tumor burden or when the risk associated with therapy outweighs benefits [Kahl and Yang, 2008]. As per the National Comprehensive Cancer Network (NCCN) guidelines, treatment for NMZL is currently consistent with follicular lymphoma. NMZL is only curable during its early stages and is more aggressive than SMZL.

Immunotherapy and chemotherapy

Single-agent rituximab can be considered in low tumor burden patients unwilling to consider observation or in elderly patients with symptoms who cannot tolerate chemotherapy [Kahl and Yang, 2008]. In the series reported by Nathwani and colleagues the overall 5-year survival rate was 56% and the 5-year FFR was 28% [Nathwani et al. 1999]. Fludarabine plus rituximab was evaluated in a phase II study of patients with MZL (including MALT, nodal and splenic types) and the ORR was 83% while the CR rate was 52% [Brown et al. 2007; Kahl and Yang, 2008]. While response rates were high, significant toxicity, especially hematologic, was noted. 2CdA has also shown potential when treating NMZL. A phase III clinical trial with 28% marginal cell lymphoma patients showed that progression-free survival (PFS) was 61%, 48% and 22% for combination group with cladribine and cyclophosphamide, cladribine alone group and cyclophosphamide, vincristine and prednisone (CVP) regimen group, respectively [Kalinka-Warzocha et al. 2008]. The proteosome inhibitor, bortezomib, has demonstrated activity in NMZL, achieving a partial remission in some patients who have failed previous chemotherapies [O’Connor et al. 2005]. For patients with NMZL and HCV-related chronic hepatitis who do not immediately need chemotherapy for the lymphoma, an antiviral treatment with pegylated interferon and ribavirin is recommended. In fact, as in other MZL subtypes, activity of antiviral treatment with interferon and ribavirin is reported in HCV-positive NMZL [Hermine et al. 2002; Kelaidi et al. 2004; Saadoun et al. 2005; Vallisa et al. 2005].

MALT lymphoma

MALT lymphoma is another type of MZL which arises from sites devoid of lymphoid tissues and are often preceded by sites of chronic inflammation [Kahl and Yang, 2008]. Based on its location MALT lymphomas are divided into gastric MALT and nongastric MALT lymphomas. See Table 2 for the available treatment options.

Gastric MALT lymphoma

Gastric MALT accounts for approximately 30% of all MALT lymphomas. There is compelling evidence of a pathogenic role of H. pylori infection in gastric MALT lymphoma and it has been present in 90% cases [Kahl and Yang, 2008]. t(11; 18) (q21; q21) translocation can be present in 30–40% of patients with gastric MALT lymphomas and bears the significance that these group of patients are generally H. pylori negative and thus are less likely to respond to H. pylori eradication treatment [Kahl and Yang, 2008]. A very small percentage of patients can have t(1; 14) translocation and this subgroup also does not respond to H. pylori eradication treatment.

H. pylori-positive gastric MALT treatment

H. pylori positivity and stage of the disease plays a crucial role in the management of gastric MALT lymphoma. Patients who are at stage IE of the disease in the absence of t(11; 18) translocation respond best to the H. pylori eradication therapy whereas the response rate declines drastically with advanced stage or positivity for this translocation [Liu et al. 2002]. Therefore, therapy aimed at the eradication of H. pylori is the favored initial treatment for patients with early stage H. pylori-positive gastric MALT lymphoma. With this approach, 50–80% of patients with localized gastric MALT achieve a complete histological remission, which tends to be maintained in the long term in the majority of patients [Wotherspoon et al. 1993; Bayerdorffer et al. 1995; Roggero et al. 1995; Steinbach et al. 1999; Chen et al. 2005; Wundisch et al. 2005]. After successful eradication of H. pylori, patients should undergo periodic upper endoscopy with multiple biopsies to evaluate for tumor response and monitor for relapse. The interpretation of residual lymphoid infiltrate in post-treatment gastric biopsies can be challenging and there are no uniform criteria for the definition of histological remission [Bertoni and Zucca, 2005]. Several studies of post-antibiotic treatment for H. pylori positive MALT have shown persistence of monoclonal B cells after histologic regression of lymphoma in about half of the studied cases [Bertoni et al. 2002]. Repeat biopsies remain an essential follow up and it is recommended to get endoscopy and biopsies 2–3 months post-treatment to document H. pylori eradication and nonprogression of lymphoma. Subsequently biopsies are recommended every 6 months for 2 years. In the case of persistent but stable minimal residual disease, a wait and watch policy may be safe. Tumors demonstrating the t(11; 18) translocation are unlikely to respond to H. pylori eradication and hence are candidates for alternative therapies [Liu et al. 2001, 2002]. The long-term risk of transformation of MALT lymphoma into an aggressive form is lower when compared with other indolent subtypes of NHL [Thieblemont et al. 2000; De Boer et al. 2008].

H. pylori-negative gastric MALT

There is no clear guideline for treating H. pylori-negative gastric MALT lymphoma or patients who have failed eradication treatment. However, local radiotherapy (RT) has been used to treat gastric MALT due to the single site of involvement and is beneficial in this group of patients with early stage disease (I–II). Studies demonstrate that local RT results in high ORR of 100% and CR of 96% with 87% long-term disease-free survival as a sole treatment modality [Schechter et al. 1998; Vrieling et al. 2008]. Toxicities are limited in duration and include anorexia, malaise, nausea and/or dyspepsia. Most patients with early stage H. pylori-negative gastric MALT can achieve a CR with initial RT. Surgery has not been shown to provide any added advantage over organ-preserving treatment approaches [Koch et al. 2005].

Those patients who fail to achieve a CR or have advanced diseases should be considered for treatment with single-agent chemotherapy or immunotherapy or combination. Single-agent rituximab results in ORR and CR rates of approximately 77% and 46%, respectively [Martinelli et al. 2005]. On combining rituximab with fludarabine, the ORR increases (from 85% to 100%), as do the associated toxicities (hematologic, infectious and allergic toxicities) [Brown et al. 2007; Salar et al. 2009]. In study conducted by Salar and colleagues, 12 out of 22 patients had gastric MALT. The authors concluded that the ORR was 100%, and 90% of patients achieved a CR. The PFS rate at 2 years in patients with gastric and extragastric MALT lymphoma was 100% and 89%, respectively [Brown et al. 2007; Salar et al. 2009]. In one study daily oral cyclophosphamide or oral chlorambucil for a period of 12–24 months, resulted in CR rates of approximately 75% after a median follow up for 12 months [Hammel et al. 1995]. These agents do not appear to have activity in patients with t(11; 18) translocation [Levy et al. 2005]. In an another study, 2CdA (cladribine), administered as an infusion in 5-day cycles repeated every 4 weeks, resulted in a CR rate of 84% [Jager et al. 2002]. Toxicities were mainly hematological (leukopenia and anemia) with associated infectious complications, including herpes zoster. Of note, in the same study, authors also found that CR was higher in cohort of patients with gastric MALT as compared with extragastric origin (100% versus 43%, respectively) [Jager et al. 2002]. Long-term follow up is needed to translate this into overall survival advantage. Cladribine also appears to be effective in patients with t(11; 18) translocation [Streubel et al. 2004]. In their study 8 out of 17 patients had t(11;18)(q21;q21) and the authors concluded that the presence of the API2-MALT1 fusion transcript resulting from t(11;18)(q21;q21) did not adversely affect the response of gastric MALT lymphoma to chemotherapy with 2CdA [Streubel et al. 2004]. Another clinical trial [ClinicalTrials.gov identifier: NCT00656812] is using a combination of rituximab plus 2CdA in patients with advanced or relapsed MALT lymphoma and is currently recruiting patients. The results of this study will help determine whether this combination regimen will yield a better outcome in this group of patients. Zucca and colleagues presented the results of the largest randomized trial in MZL in American Society of Hematology meeting in 2010. The study was launched in 2003 by the International Extranodal Lymphoma Study Group to compare chlorambucil alone versus the combination of chlorambucil and rituximab in the treatment of MALT lymphomas. MALT lymphoma patients with localized disease at any extranodal site who did not respond or were not suitable for local therapy (including H. pylori-negative gastric lymphomas or those who failed antibiotic therapy), as well as those with disseminated or multifocal MALT lymphoma, were eligible [Zucca et al. 2010b]. There were two arms initially and the third arm was added in 2006. In arm A, chlorambucil was given 6 mg/m² daily p.o. for 42 consecutive days (days 1–42). After restaging, responding patients and those with stable disease were kept on chlorambucil 6 mg/m² daily p.o. for 14 consecutive days every 28 days for 4 cycles. In arm B, chlorambucil was given as in arm A, together with rituximab, 375 mg/m² IV on day 1, 8, 15, 22, 56, 84, 112 and 140. A third arm with rituximab alone (with a randomization rate of 1:1:6) was introduced in 2006 because of good recruitment. A preplanned analysis was performed for 227 patients in arms A and B, prior to randomization of the third arm. Out of 227, 86% of the patients completed treatment as per protocol and outcome analysis was done on all randomized patients, on an intent-to-treat basis. Median follow up was 60 months. The primary MALT lymphoma site was the stomach in 94 patients (41%) and 133 patients (59%) had a nongastric presentation.

The 5-year event-free survival was significantly better for the patients treated with chlorambucil plus rituximab (68%; 95% CI 58–76%) in comparison with the chlorambucil alone arm (50%; 95% CI 40–59%). Overall survival at 5 years was identical in the two arms (88%). There was no outcome difference between patients with gastric and nongastric localization. The combination regimen was well tolerated and severe hematological toxicity was uncommon. Grade 3–4 neutropenia was more frequent in the chlorambucil plus rituximab arm (19 versus 2 episodes, p < 0.001). The benefit in event-free survival failed to translate into overall survival advantage.

Nongastric MALT lymphoma

Nongastric MALT lymphoma constitutes about two thirds of all cases of extranodal MZL [Zucca et al. 2000]. The most common sites include the salivary glands, the thyroid, the upper airways, the lung, the ocular adnexa, the breast, the liver, the urothelial system, the skin, the dura and other soft tissues.

Clinical symptoms are generally related to the primary location although whether the anatomic localization is of prognostic significance remains unclear. The presence of stage IV disease or nodal involvement had a worse prognosis [Zucca et al. 2003]. However, stage IV disease confined to multiple mucosal sites without involving bone marrow or lymph nodes did not adversely affect the outcome and most patients have a favorable outcome with OS at 5 years of 90% [Thieblemont et al. 2000; Zucca et al. 2003; Raderer et al. 2006a; De Boer et al. 2008]. Histologic transformation to large-cell lymphoma is reported in about 10% of cases as a late event independent of stage IV disease [Thieblemont et al. 2000; De Boer et al. 2008].

The optimal management of nongastric disease is not clearly defined and retrospective series show that patients have been treated with surgery, radiation and chemotherapy, alone or in combination. All treatment modalities resulted in excellent disease control and overall survival and no one modality was shown to be superior to another [Thieblemont et al. 2000; Tsang et al. 2003; Zucca et al. 2003; Raderer et al. 2006a; De Boer et al. 2008]. Current management guidelines recommend a ‘patient-tailored’ approach taking into account the stage, site and clinical characteristics of the individual patient.

Radiotherapy

Patients with limited disease present a long survival and therefore need a conservative approach. In general, radiation therapy is considered the treatment of choice for localized disease [Tsang and Gospodarowicz, 2005]. Localized MALT lymphoma has been successfully treated with involved-field radiation with doses of 25–36 Gy [Yahalom, 2001; Tsang and Gospodarowicz, 2005; Zucca et al. 2010c]. Physicians may initially follow patients with asymptomatic disseminated disease with observation alone. Symptomatic patients with systemic disease or those not amenable to radiation should be considered for systemic chemotherapy.

Immunotherapy and chemotherapy

Few single agents or combination chemotherapy regimens have been tested specifically. Oral alkylating agents (either cyclophosphamide or chlorambucil) can result in a higher rate of disease control and can be used as a single agent or in combination [Levy et al. 2002; Zinzani et al. 2004; Simon et al. 2006]. Phase II studies have demonstrated the antitumor activity of the purine analogs, fludarabine and 2CdA, which however may be associated with secondary myelodysplastic syndrome [Jager et al. 2002; Zinzani et al. 2004]. In phase II studies of the anti-CD-20 monoclonal antibody, rituximab has shown efficacy with a response rate of about 70% representing additional options for the treatment of systemic disease. The median response duration was 10.5 months. At a median follow up of 15 months, 9 patients (26%) relapsed. The median time to treatment failure (TTF) was 14.2 months [Conconi et al. 2003; Martinelli et al. 2005]. In a retrospective analysis of 89 patients with advanced stage MZL (18% with MALT), concomitant use of 2CdA and rituximab resulted in ORR of 91.7% and CR of 83.3%. Median TTF was not reached yet in the combination arm. The addition of rituximab to 2CdA did not affect the ORR, however a trend directed towards longer TTF during follow up [Orciuolo et al. 2010]. Bortezomib has also shown promising results. In a small phase II study involving 16 patients, it showed an ORR of 80% with 7 patients achieving a complete remission (43%), 6 PRs (37%) and 3 stable diseases. After a median follow up of 23 months (range 8–26), all patients were alive with 4 having relapses. A total of 15 patients required dose reductions due to either neuropathy or diarrhea. Bortezomib appeared to be active in patients with MALT lymphoma. However, due to an unexpectedly high rate of toxicities, it warrants an assessment of combination schedules with bortezomib at a lower dose and different schedules including weekly administration and subcutaneous dosing [Troch et al. 2009; Bringhen et al. 2010; Moreau et al. 2011]. Aggressive anthracycline-containing regimens such those used in other indolent lymphoma have also demonstrated high activity. In a retrospective analysis, 26 patients with relapsed MALT lymphoma received rituximab plus cyclophosphamide, doxorubicin/mitoxantrone, vincristine and prednisone (R-CHOP/R-CNOP) chemotherapy. A total of 20 patients (77%) achieved a complete remission and 6 patients (23%) a partial remission. Toxicities were mainly hematological, with WHO grade III/IV leukocytopenia occurring in five patients. After a median follow up of 19 months (range 10–45), all patients were alive: 22 were in ongoing remission, while 4 had relapsed between 12 and 19 months after treatment [Raderer et al. 2006b].

Newer agents and future directions

Several drugs are currently being tested in ongoing trials for the treatment of MZL. However, since MZL constitutes a very small subgroup, the majority of studies are not MZL specific and hence MZL patients are included in the indolent lymphoma groups. These agents do appear to possess activity and need to be studied further in MZL. Most of the studies are examining the newer agents in a combination regimen with immunotherapy or chemotherapy. The promising newer therapies are discussed below. See Table 3 for newer agents that are currently in ongoing trials.

Table 3.

New agents.

	Type of study	Enrollment criteria	Outcome	References
Bendamustine (120 mg/m2 on day 1, 2 q21 days for 6–8 cycles)	Single arm, open label	Refractory indolent B-cell NHL	ORR; 78% in MZL, 86% in extranodal MZL.	[Kahl et al. 2010a]
		100 pt
		(16% MZL)
Lenalidomide + dexamethasone + rituximab*	Phase II	Relapsed or refractory indolent NHL or MCL (1 pt with MZL)	ORR 57%	[Ahmadi et al. 2010]
Lenalidomide 20 mg days 1–21 + rituximab 375 mg/m2 day 1 q28days	Phase II, single arm	Untreated indolent B-cell lymphoma	ORR 90% CR 66%	[Samaniego et al. 2011]
Everolimus + lenalidomide Once daily days 1–21.	Phase I/II	Relapsed refractory NHL and HL	Ongoing	[ClinicalTrials.gov identifier: NCT01075321]
Azacitidine (75 mg/m2 on days 1–5) + Lenalidomide (15 mg/day on days 1–21).	Phase II	Relapsed/Refractory Follicular Lymphoma and MZL	Ongoing	[ClinicalTrials.gov identifier: NCT01121757]
Everolimus (10 mg/day on days 1–28).	Phase II	Relapsed or refractory marginal zone B-cell lymphomas	Ongoing	[ClinicalTrials.gov identifier: NCT01164267]
CAL-101 (PI3K inhibitor). 1–2 times per day continuously in a 28-day cycles for up to 12 cycles	Phase I	Relapsed or refractory NHL	Partial response 66%	[Kahl et al. 2010b]
Vorinostat (200 mg twice daily on days 1–14 of a 21-day cycle)	Phase II	Relapsed or refractory indolent NHL and mantle cell lymphoma.	ORR 29%	[Kirschbaum et al. 2011]
			2 MZL (1 CR, 1 PR)
Vorinostat (SAHA) days 1–14 (200 mg, 300 mg or 400 mg), + Cladribine (2-CdA) 5 mg/m2 IV on days 1–5, and Rituximab 375 mg/m2 IV on days 3, 10, 17 and 24.	Phase I/II	Relapsed B-cell malignancies	Ongoing	[Spurgeon et al. 2011]
Rituximab 375 mg/m2 IV on day 1 of every 21 days	Pilot study, single arm	Untreated FL or MZL	Ongoing	[ClinicalTrials.gov identifier: NCT00772668]
Bortezomib 1.6 mg/m2 IV on days 1 and 8 of every 21 days
Cyclophosphamide 750 mg/m2 IV on day 1 of every 21 days
Prednisone 100 mg p.o. daily on days 1–5 of every 21-day cycle
Yttrium Y 90 Ibritumomab Tiuxetan and Rituximab	Phase II	Untreated MZL	Ongoing	[ClinicalTrials.gov identifier: NCT00453102]

^*In Part I (lenalidomide + dexamethasone), pts receive two 28-day treatment cycles of lenalidomide 10 mg every day and dexamethasone 8 mg once weekly. After assessment of response to Part I, all pts receive a single course of rituximab 375 mg/m², consisting of four weekly doses during cycle 3 (Part II: lenalidomide + dexamethasone + rituximab).

FL, follicular lymphoma; MZL, marginal zone lymphoma; HL, Hodgkin’s lymphoma; NHL, non-Hodgkin’s lymphoma; ORR, overall response rate; CR, complete response; PR, partial response.

Bendamustine

Bendamustine hydrochloride, a novel alkylating agent has shown efficacy in treatment of indolent NHL. The mechanisms of actions include induction of apoptosis through activation of DNA-damage stress responses, inhibition of mitotic checkpoints and induction of mitotic catastrophe. Bendamustine contains a benzimidazole ring, which may confer purine analog-like properties in addition to the alkylating properties [Kahl et al. 2010a]. Kahl and colleagues conducted a multicenter, open-label, single-arm, clinical trial to assess the efficacy of bendamustine in rituximab-refractory indolent B-cell NHL [Kahl et al. 2010a]. The 100 eligible patients had received a median of 2 previous regimens, and 36% were refractory to their most recent chemotherapy regimen. Primary endpoints included ORR and duration of response. Secondary endpoints were safety and PFS. Patients received bendamustine at a dose of 120 mg/m² by intravenous infusion on days 1 and 2 every 21 days for 6–8 cycles. Single-agent bendamustine produced a high rate of objective response with ORR of 75% with acceptable toxicities. The median follow up was 11.8 months. In this particular study there were 16% MZL patients and in that subgroup ORR was 78% in nodal MZL and 86% in extranodal MZL. CR was 11% and 43% and PR was 67% and 43% in nodal MZL and extranodal MZL subgroups respectively with bendamustine. In this clinical trial, the frequent reversible hematologic toxicities included neutropenia (61%), thrombocytopenia (25%) and anemia (10%). The most frequent nonhematologic adverse events of any grade included nausea (77%), infection (69%), fatigue (64%), diarrhea (42%), vomiting (40%), pyrexia (36%), constipation (31%) and anorexia (24%). Some of the infections observed included 12 episodes of herpes zoster and 5 episodes of CMV infection. It is important to point out that all patients in the study were resistant to rituximab and most developed resistance after receiving treatment with a ritiximab–chemotherapy combination. This study did demonstrate the promising activity of bendamustine in rituximab-refractory patients.

Lenalidomide

Another agent that has shown some promise and is under investigation is lenalidomide (Revlimid; Celgene), a derivative of thalidomide with antiangiogenic and antineoplastic properties. It is still at the investigational stage and two phase II studies presented by Sàrl at the American Society of Hematology in December 2010 had shown promising results in patients with relapsed or refractory indolent B-cell lymphoma. One of these phase II studies looked at the use of lenalidomide and low-dose dexamethasone plus rituximab combination in patients with relapsed or refractory indolent B-cell lymphoma or mantle cell lymphoma previously resistant to rituximab. It demonstrated an average ORR of 57%, with a median follow up of 12 months but this study only had 1 MZL patient [Ahmadi et al. 2010]. In the other study, a combination of lenlidomide and rituximab was used in patients with relapsed or refractory indolent NHL with measurable disease who had received at least one prior therapy. A total of 18 patients were enrolled in the study and 2 out of 18 patients were MZL. Patients were given 25 mg of lenalidomide on days 1–21 of a 28-day cycle, continuing until disease progression. Patients also received rituximab 375 mg/m² IV on day 15 of cycle 1, repeated weekly for a total of 4 doses. A total of 17 patients were evaluable for response, the ORR was 76% (13/17), with CR of 41% (7/17) and PR of 35% (6/17). In the study, the most common grade 3 or 4 toxicities were lymphopenia (24%), neutropenia (24%), fatigue (24%) and hyponatremia (12%). In this study two of the first four patients developed grade 3 tumor lysis syndrome, and the protocol was then amended by the investigators to reduce the lenalidomide starting dose to 20 mg, and prophylaxis with allopurinol was initiated. Lenalidomide was thought to potentially overcome the previously reported lower activity of single-agent rituximab in patients with the low-affinity-type FcγRIIIa receptor [Dutia et al. 2010]. Recently Samaniego and colleagues presented the results of a phase II study of lenalidomide plus rituximab in patients with previously untreated indolent lymphoma at the American Society of Clinical Oncology annual meeting in June 2011 [Samaniego et al. 2011]. Investigators treated 75 untreated patients of NHL (41 with follicular lymphoma, 19 with MZL and 15 with chronic lymphocytic leukemia) with 20 mg/day of lenalidomide on days 1–21 and rituximab 375 mg/m² on day 1 of each 28-day cycle, for 6 cycles. They found that 90% of patients responded, with 66% achieving a CR. A total of 17 patients (25%) had a PR and stable disease was noted in 6 patients (9%). The most common grade ≥3 nonhematologic toxicities included rash, muscle pain, thrombosis and infection. Grade ≥3 neutropenia and thrombocytopenia occurred in 27% and 5% patients, respectively. Further studies are needed to evaluate the use of lenalidomide in MZL as these studies only had a small group of MZL patients.

Everolimus

Everolimus, RAD001, an investigational mTOR inhibitor, is also being studied in various solid tumors and has been FDA approved for renal cell cancer, is also being researched in the treatment of indolent NHL including MZL. Everolimus and lenalidomide in combination is also being investigated for treating patients with relapsed or refractory NHL and the clinical trial [ClinicalTrials.gov identifier: NCT01075321] is currently recruiting patients. The rationale for this study is that everolimus may stop the growth of cancer cells by blocking some of the enzymes needed for cell growth and lenalidomide may stop the growth of cancer cells by blocking blood flow to the cancer. Thus, administering everolimus together with lenalidomide may be an effective treatment for lymphoma. Another ongoing study [ClinicalTrials.gov identifier: NCT01121757] is assessing epigenetic modulation in relapsed or refractory MZL with lenalidomide and azacitidine. Another multicenter phase II clinical trial [ClinicalTrials.gov identifier: NCT01164267] is ongoing and it has been designed to evaluate the clinical activity and the safety profile of everolimus in MZL. The results of all of these studies will help us understand the role of everolimus in MZL.

CAL-101

There is a new investigational drug, CAL-101, that is an oral isoform-selective inhibitor of PI3K (phosphoinositide 3-kinase) signaling pathway and it induces apoptosis of NHL cell lines in vitro. In a phase I study this drug has been studied evaluating the safety, pharmacokinetics and activity in relapsed or refractory hematological malignancies [Kahl et al. 2010b]. A total of 55 patients were enrolled and 28 patients had indolent NHL; 3 out of these 28 NHL patients were MZL. Sequential cohorts of patients were enrolled at progressively higher dose levels with cohort expansion based on toxicity profile and plasma exposure. CAL-101 was administered orally once or twice per day continuously in a 28-day cycles for up to 12 cycles. PRs were observed at all doses and the PR rate was 62% in indolent NHL group. Median duration of response has not been reached in this group. The most common, serious adverse event was infection. Elevation in liver enzyme was reversible but dose-limiting toxicity. It is still too early to predict how well this treatment will work but this drug has certainly generated a great deal of enthusiasm.

Vorinostat and cladribine combination

There is published data as well as our own unpublished data that cladribine may have a unique epigenetic activity by blocking the methyl donation reaction involving S-adenosyl-methionine. Inhibition of methylation through this mechanism may also serve to inhibit histone methylation as well as DNA methylation, similar to the laboratory reagent DZnep (3-deazaneplanocin A) [Tan et al. 2007; Fiskus et al. 2008; Miranda et al. 2009]. Kirschbaum and colleagues demonstrated in their phase II monotherapy study that vorinostat has activity in lymphoma [Kirschbaum et al. 2011]. A total of 37 patients were enrolled with relapsed or refractory indolent B-cell lymphoma (follicular lymphoma n = 20, MZL n = 7, mantle cell lymphoma n = 8), vorinostat was given at 200 mg p.o. twice daily for 14 consecutive days on a 21-day cycle. Two out of nine MZL patients responded with one CR and one PR [Kirschbaum et al. 2011]. Published studies and ongoing work at Penn State Milton S Hershey Medical Center in collaboration with the Oregon Health and Sciences University indicates that the combined cladribine + rituximab treatment is very effective for MZL [Spurgeon et al. 2009; Orciuolo et al. 2010]. These studies have shown several patients in complete remission, one greater than 5 years out with the use of rituximab maintenance. In three patients, we have avoided splenectomy by giving rituximab and cladribine combination. The additional epigenetic activity of cladribine as a hypomethylating agent prompted this group to initiate a phase II trial adding the histone deacetylase inhibitor (HDACi) vorinostat (SAHA) to the cladribine and rituximab backbone [Spurgeon et al. 2010a, 2010b, 2011].

Other investigational agents

Bortezomib has been studied previously in the treatment of indolent NHL. University of Sylvester comprehensive cancer center is conducting an ongoing study [ClinicalTrials.gov identifier: NCT00772668], where they are assessing the use of rituximab together with cyclophosphamide, bortezomib and prednisone as a first-line therapy in treatment of patients with stage III or IV follicular lymphoma or MZL. In another phase II study [ClinicalTrials.gov identifier: NCT00453102] by the same group of investigators, Yttrium Y 90 Ibritumomab tiuxetan and rituximab combination is being used to treat patients with previously untreated MZL. The results of these studies will help us know whether these new combination treatment regimens will be effective in patients with MZL.

Ansell and colleagues identified a novel translocation in MALT lymphoma, t(X, 14)(p11; 32) and increased expression of GPR34 in tumor cells [Ansell et al. 2010]. The authors concluded that MEK inhibitor may be useful in this subset of patients. MEK is a key protein kinase in the Ras/Raf/MEK/ERK pathway, which signals for cancer cell proliferation and survival. Genetic lesions targeting the NF-κb pathway by mutation and or deletion are frequently involved in the pathogenesis of SMZL and might provide novel therapeutic options [Rossi et al. 2010].

TRU-016, a single chain anti-CD37-directed small modular immunopharmaceutical (SMIP), is another new agent currently in clinical trials for B-cell NHL that may have activity in MZL. It has been studied in combination with other therapeutic drugs such as rituximab, rapamycin or bendamustine in models of NHL with the rationale being that it increases the cancer cell killing effect in NHL [Baum et al. 2009]. The combination of TRU-016 and bendamustine displayed greater in vivo antitumor activity than either agent alone against a follicular lymphoma tumor model.

Conclusion

MZL is a rare group of NHL which is indolent in nature with subgroups, SMZL, NMZL and MALT lymphoma. These lymphomas share some common features but are different in their biological nature and hence behave differently depending on their location. Owing to the rarity of this group of diseases, there have not been any randomized trials comparing different treatment options and therefore the current standard of care for treatment is controversial without any standard guidelines. Our literature review suggests that treatment should be tailored as per the needs of individual patients. One can opt for a watchful waiting policy for less aggressive forms of the disease. Rituximab in combination with 2CdA has shown some promising results and should be studied further in MZL. Newer therapies such as bendamustine, everolimus, lenalidomide, HDACi (vorinostat, romepdesin and other agents) and phosphoinositide 3-kinase inhibitors are some of the drugs that are currently being studied in clinical trials for patients with MZL. Further studies with the newer agents should be done both in newly diagnosed or relapsed/refractory MZL.