Clinicopathologic and Genetic Profile of Intracranial Marginal Zone Lymphoma: A Primary Low-Grade CNS Lymphoma That Mimics Meningioma
http://www.100md.com
《临床肿瘤学》
the Division of Neuropathology and Section of Hematopathology, Washington University School of Medicine, St Louis, MO
Division of Neuropathology and Department of Neurology, Mayo Clinic, Rochester, MN
Department of Pathology, Children's Hospital of Philadelphia, University of Pennsylvania School of Medicine, Philadelphia, PA
Division of Neurosurgery, University of Toronto, Toronto, Ontario, Canada
Division of Neuropathology, University of Florida College of Medicine, Gainesville, FL
Department of Pathology, The Johns Hopkins University School of Medicine, Baltimore, MD
ABSTRACT
PURPOSE: Although rare overall, marginal zone B-cell lymphoma (MZBCL) is the most common primary low-grade CNS lymphoma reported in the literature. The aim of this study is to elucidate the biology and genetic features of this unusual tumor.
PATIENTS AND METHODS: Fifteen CNS MZBCLs were studied clinically, pathologically, and genetically, including fluorescent in situ hybridization analyses with commercially available MALT1 and IgH break-apart and centromere 3, 7, 12, and 18 probes.
RESULTS: CNS MZBCLs preferentially affect middle-aged women (female-to-male ratio, 4:1), with 93% presenting as dural-based masses mimicking meningioma. Ten patients with 1 to 7.6 years of follow-up after diagnosis showed no evidence of disease after radiation and/or chemotherapy. Like MZBCLs outside of the CNS, they consisted of CD20+, CD3– small B lymphocytes with varying degrees of plasmacytic differentiation and predominantly light-chain restriction (78%). Lymphoid follicles with follicular colonization were seen in three patients and deposition of amyloid was noted in samples from two patients, one of which was tumefactive. Neither Bcl-6 protein nor Epstein-Barr virus–encoded RNA was expressed. Trisomy 3 was detected in six of 12 patients, with no rearrangements of MALT1 or IgH and no trisomies of 7, 12, or 18 detected.
CONCLUSION: Our data suggest that intracranial MZBCL is an indolent primary CNS lymphoma that typically presents as a meningioma-like dural-based mass. Trisomy 3, but not MALT1 or IgH translocation, is a common genetic abnormality that may contribute to the pathogenesis of this CNS lymphoma.
INTRODUCTION
Marginal zone B-cell lymphoma (MZBCL) was first described in the mucosa-associated lymphoid tissue (MALT) of the gastrointestinal tract by Isaacson and Wright1 as an indolent low-grade lymphoma. It was subsequently recognized in the mucosa of other organs including lung, bladder, salivary gland, conjunctiva, and lacrimal gland, as well as tissue sites without a mucosa, such as thyroid, breast, thymus, orbit, skin, liver, and even the CNS.2-15 MZBCLs were accepted as a distinct non-Hodgkin’s lymphoma (NHL) in the Revised European and American Lymphoma classification in 1994. The current WHO classification16 further subdivides this lymphoma into three types: extranodal or MALT, nodal, and splenic MZBCLs. The three types of MZBCLs have been estimated to constitute approximately 7.6%, 1.8%, and less than 1% of NHL, respectively. Although these three types share similar histopathologic and immunophenotypic features, they differ in terms of clinical behavior.17 Briefly, MALT-type MZBCL typically presents as a long-standing local mass in tissues that acquire MALT during an immune reaction; bone marrow and lymph node involvement is rare. In contrast, patients with nodal MZBCLs usually present at an advanced clinical stage, characterized by lymphadenopathy and bone marrow disease. Splenic MZBCL is characterized by splenomegaly in addition to peripheral-blood and marrow involvement,18 and may be associated with infection with hepatitis C virus.19 As with other low-grade NHLs, advanced clinical stage does not necessarily portend a poor outcome; newly diagnosed stage 3 patients have an 80% 10-year survival rate. Recently, a leukemic variant of MZBCL without any involvement of MALT-containing tissue, lymph node, or spleen has been reported, also with a surprisingly indolent course.20
The MZBCL is believed to derive from postgerminal center marginal zone B-cells, best visualized surrounding secondary follicles of mesenteric lymph node; spleen and acquired MALT typically occur in organs where exposure to antigens is considerable.16,21 Both the normal cells and their neoplastic counterparts are characterized by abundant clear cytoplasm and an irregular, centrally located nucleus (monocytoid B cells). They express pan–B-lymphocyte markers (CD19, CD20, and CD79a), some complement receptors (eg, CD21, CD35), and surface immunoglobulin, but do not express CD3, CD5, CD10, CD23, or cyclin D1.17
Recently, a number of cytogenetic abnormalities have been identified in MZBCLs. The most common karyotypic abnormalities are trisomies of chromosomes 3, 7, 12, and 18, as well as translocations such as t(11;18) (q21;q21), t(3;14)(q27;q32), t(1;14)(p22;q32), t(14;18)(q32;q21), and t(3;14)(p14.1;q32).22-24 The frequency of these abnormalities varies among the organs involved and the tumor types. For example, the t(11;18)(q21;q21) chromosomal abnormality is seen in approximately 20% to 30% of MALT-type MZBCL, but is not found in the nodal and splenic types.25,26 In fact, the t(11;18)(q21;q21) translocation is considered specific to MZBCLs, and thus far, has not been encountered in any other types of lymphoma. The translocation results in an API2-MALT1 fusion gene encoding a chimeric protein consisting of the N-terminal portion of API2 linked to the C-terminus of the paracaspase, MALT1.27,28 The normal MALT1 protein functions downstream of Bcl-10 to activate nuclear factor kappa B, and is required for the generation or maintenance of marginal zone B cells.29 The API2-MALT1 fusion protein is suspected to promote tumorigenesis through auto-oligomerization, which results in constitutive activation of nuclear factor kappa B signaling in marginal zone B cells. In contrast, trisomy 3 has been detected in all three types of MZBCLs with similar, albeit widely ranging frequencies of 15% to 60%.17,22 Most recently, immunoglobulin heavy chain (IgH) rearrangements have also been described, most commonly involving MALT-type MZBCL of the thyroid and orbital regions.24
Within the CNS, the vast majority of primary lymphomas are diffuse large B-cell lymphomas presenting in periventricular or deep brain parenchymal regions.30,31 Primary low-grade lymphomas occur at a much lower incidence, appearing in the literature as small series or single case reports. Among the low-grade group, MZBCL is by far the most common primary CNS lymphoma. To our knowledge, 19 well-documented examples have been reported previously.2-15 However, no genetic data have been published to date. In this study, we report the clinicopathologic features of 15 additional patients, including genetic analyses using fluorescent in situ hybridization (FISH). Our data further support the female predilection, frequent dural localization radiographically mimicking meningioma, favorable clinical behavior, and occasional tumefactive amyloid deposition. Genetically, our patients lacked MALT1- or IgH-containing translocations, but half harbored FISH-detectable trisomy 3.
PATIENTS AND METHODS
Patient and Tumor Cohort
A search for the diagnosis of MZBCL of the CNS yielded 16 patients from the pathology files at Washington University (St Louis, MO), University of Pennsylvania (Philadelphia, PA), Johns Hopkins University (Baltimore, MD), Mayo Clinic (Rochester, MN), and the University of Florida (Gainesville, FL). On subsequent review, one patient was classified as having a low-grade lymphoplasmacytic lymphoma and was excluded from further study. All available routine and immunohistochemistry slides were reviewed and a representative block was chosen from each available patient. Unstained slides were cut at 5-μm thickness and mounted onto precoated superfrost/plus glass slides for FISH studies (Fisher Scientific, Pittsburgh, PA). This study was performed after approval by the Human Studies Committee of each institute and in accord with the regulations approved by the Department of Health and Human Services.
FISH
Dual-color FISH experiments were performed using commercially available MALT1 and IgH break-apart, as well as centromere-enumerating DNA probes (CEP), CEP3, CEP7, CEP8, CEP12 (Vysis Inc, Downers Grove, IL) as described previously.32 Briefly, after deparaffinization, the sections were subjected to target retrieval by steam cooking in citrate buffer for 20 minutes followed by a 20-minute cool-down period and a 5-minute wash (distilled water). This was followed by pepsin (4 mg/mL) digestion at 37°C for 30 minutes. The slides were then washed in 2x saline sodium citrate (SSC) and allowed to air dry. SpectrumOrange (Vysis Inc) -labeled CEP3 probe was paired with SpectrumGreen (Vysis Inc) -labeled CEP12; SpectrumOrange-labeled CEP7 was similarly paired with SpectrumGreen-labeled CEP8. The probes were diluted in DenHyb hybridization buffer (Insitus Laboratories, Albuquerque, NM) at either 1:25 or 1:50 dilution. The hybridization mix was applied to the sections, followed by simultaneous denaturation of the probe and target DNA at 90°C for 13 minutes. Overnight hybridization at 37°C took place in a humidified chamber. Posthybridization washes in 50% formamide/1x SSC (5 minutes) and 2x SSC (5 minutes) were performed at room temperature, and the slides were allowed to air dry.
4',6-Diamidino-2-phenylindole (Insitus Laboratories) was used as a nuclear counterstain, and the sections were viewed under an Olympus BX60 fluorescent microscope with appropriate filters (Olympus, Melville, NY). Two of the authors (P-h.T. and A.P.) each independently scored 100 intact nonoverlapping nuclei for the number of fluorescent red and green signals derived from each hybridization. Nuclei with more than two signals were rarely seen in non-neoplastic controls; therefore, trisomies and polysomies (gains) were arbitrarily defined as more than 5% nuclei containing three or more signals. For the MALT1 and IgH break-apart probe cocktails, nuclei were scored for the presence or absence of split red and green signals, with two yellow or red-green fusion signals indicating a lack of chromosomal rearrangement. Samples with more than 5% nuclei containing split signals were considered positive for a MALT1- or IgH-containing chromosomal translocation.
Hybridizations were digitally photographed using a high-resolution black and white COHU charge-coupled device camera with a Z-stack motor programmed to capture images at sequential 4',6-diamidino-2-phenylindole (one level), fluorescein isothiocyanate (10 levels), and rhodamine (10 levels) filter settings. Reconstruction into a single superimposed image with blue, green, and red pseudocolors was accomplished using a CytoVision basic workstation (Applied Imaging, Santa Clara, CA).
Immunohistochemistry Staining
Immunohistochemistry was performed on 5-μm-thick sections of paraffin-embedded, formalin-fixed tissue with a DAKO autostainer (DAKO, Carpinteria, CA) with antigen retrieval using 1 mmol/L EDTA, pH 8.0. Reagents were used as supplied in the Envision Plus Detection Kit (DAKO). The antibodies used in this study were CD3 (clone UCHT1; DAKO), CD5 (clone SP19; NeoMarkers, Fremont, CA), CD20 (clone L26; DAKO), CD23 (clone MHM6; DAKO), CD43 (clone DF-T1; DAKO), immunoglobin light chains (IgLs) and rabbit polyclonal antisera (DAKO), cyclin D1 (clone SP4; NeoMarkers), and Bcl-6 (clone P1F6; Neomarkers). The concentrations of individual antibodies were optimized by the immunohistochemistry laboratory of each medical center.
Epstein-Barr Virus–Encoded Nontranslated RNA In Situ Hybridization
In situ hybridization was performed on 5-μm-thick sections of paraffin-embedded formalin-fixed with the INFORM Epstein-Barr virus (EBV) –encoded nontranslated RNA (EBER) probe (Ventana Medical Systems Inc, Tucson, AZ), which contained a cocktail of oligonucleotide probes, and the signals were visualized using the ISH iVIEW Blue Detection Kit (Ventana) with Nitro Blue Tetrazolium and an ISH Red Counterstain. All of the procedures were performed with the BenchMark XT autostainer (Ventana) according to the manufacturer’s instructions.
RESULTS
Clinical Characteristics
Clinical data from the 15 patients are summarized in Table 1. Twelve patients (80%) were women, yielding a female-to-male ratio of 4:1. The mean age at diagnosis was 55 years (range, 29 to 70 years). Presenting symptoms were either nonspecific (headache, nausea, vomiting, visual disturbance) or localized (focal seizures, motor/sensory deficits, ataxia, gait disturbance, otalgia). Patient 2 also had a long history of relapsing/remitting multiple sclerosis (MS). All but one patient (patient 3) presented with a contrast-enhancing, extra-axial, dural-based or leptomeningeal mass, often with a dural tail sign, which is radiologically consistent with meningioma (Fig 1). An en plaque pattern of meningeal thickening was reported in two patients. The tumors arose at various intracranial sites, although the cerebral convexity was most common. The falx was involved in three patients, the tentorium was involved in two patients, and the sellar/parasellar region was involved in one patient. No definitive evidence of tumor outside of the CNS was found in any of the patients.
All of the patients underwent craniotomy with tumor resections. Of the nine patients with available information about adjuvant therapy, six received whole-brain radiation alone; two had combined chemotherapy and radiation, and one received chemotherapy alone (Table 1). Various regimens including procarbazine, vincristine, intravenous cytarabine, leucovorin, and/or intravenous or intrathecal methotrexate were used. Patient 2 was initially treated with methotrexate, but suffered a recurrence 4 months later. The recurrent tumor was subsequently treated with fludarabine with an excellent response, resulting in complete remission. These data indicate that surgery plus radiation and/or chemotherapy is sufficient to achieve complete remission. The longest followup in our series was 7.6 years, which suggests that long-term remission is probable.
Pathologic Features
Sections revealed a mixed cellular infiltrate primarily involving dura and leptomeninges, with extension along Virchow-Robin spaces in some patients (Fig 2A). Focal infiltration of the neoplastic cells into the underlying brain parenchyma was noted in two patients. The architectural and cytologic features were otherwise identical to those of MZBCLs in other sites. They were composed of sheets of small- to medium-sized lymphocytes with moderate amounts of cytoplasm and irregular nuclei, bearing a close resemblance to the small-cleaved cells or centrocytes of the lymphoid follicles (Fig 2B). Many of these cells exhibited the so-called monocytoid appearance characterized by clear cytoplasm and well-defined cell borders. Scattered among these cells were occasional immunoblast-like cells with large vesicular nuclei and prominent nucleoli. Plasma cells were identified in every patient, ranging from rare to plentiful. Reactive lymphoid follicles were identified in three patients (patients 4, 6, and 7; Fig 2C). In patient 4, some follicles were replaced partially or completely by the neoplastic cells, mimicking follicular lymphoma focally.
Mitotic figures were generally hard to find, but were focally increased in patients 1 and 15. Scattered islands with an increase in the number of large immunoblast-like cells were also identified in these two patients. However, no definitive features of transformation to a more aggressive histology were seen (Fig 2D). Numerous plasma cells with eosinophilic intracytoplasmic inclusions (Russell bodies) were noted in patient 2 (Fig 3A). Amyloid deposition with foreign body giant cell reaction was identified in two patients (patients 2 and 9; Fig 3B and inset). The amount of the deposited amyloid material could be scant (patient 2) or massive (patient 9), the latter forming tumefactive thioflavin-S–positive amyloid pools (Figs 3C and 3D) that were large enough to detect radiographically (the hypodense regions within the tumor in Figs 1C and 1D).
Immunohistochemical and genetic findings are summarized in Table 2. The neoplastic cells were CD20+ and CD3–, consistent with a B-cell origin (Figs 4A and 4B) Monoclonality was demonstrated by immunohistochemical stains for IgL. In our patients, light-chain restriction was most common (10 of 14 patients; Figs 4C and 4D; Table 2). The amyloid pools in patient 9 were also highlighted by the immunostain for light chain (Fig 4C), but not by that for light chain (Fig 4D), suggesting that the deposited amyloid was derived from secreted IgL. Ten patient samples available for the cyclin D1 immunostain showed no immunoreactivity and ruled out the possibility of mantle-cell lymphoma.
Characterization of the Genetic Alterations by FISH
As an initial attempt to elucidate the associated genetic changes, FISH analyses were performed using probes targeting previously reported alterations. The results of these studies are summarized in Table 2. Hybridizations with the MALT1 and IgH break-apart probes each produced two fused yellow or juxtaposed red and green signals in all 12 and 10 patients tested, respectively (Fig 5A). Therefore, there was no evidence of translocations involving the MALT1 or IgH genes (Table 2). Hybridizations using the CEP 7, CEP12, and CEP18 typically produced no more than two signals, consistent with a normal disomic state (except for patient 12, who had polysomy or gains of chromosome 7). However, a subset of neoplastic cells in six of the 12 patients (50%) yielded three CEP3 signals, consistent with trisomy 3 (Fig 5B). Cells with three CEP3 signals ranged from 9% to 53%, with an average of 29%. The lack of other associated chromosomal abnormalities suggests that the trisomy 3 is a specific genetic alteration, rather than a nonspecific chromosomal gain associated with an overall state of polyploidy.
Bcl-6 Is Not Involved in the Pathogenesis of CNS MZBCL
The gene encoding Bcl-6 protein is located on the chromosome 3q27 and has been suggested to be a potential oncogene involved in MZBCL.33 To test if it has a role in the pathogenesis of the CNS MZBCL, immunohistochemistry was performed in 10 of 15 patients, five of whom had trisomy 3 and the rest of whom had normal chromosome 3 copy numbers. The neoplastic cells in all 10 patients were immunonegative for Bcl-6. However, small clusters of positive cells that had vesicular nuclei larger than those of the neoplastic cells were noted in two patients (data not shown). These cells were interpreted as representing remnant germinal center elements, whereas the small- to medium-sized neoplastic marginal zone lymphocytes appeared negative. Thus, our data do not support a role for Bcl-6 in the pathogenesis of the CNS MZBCLs.
EBV Expression Is Not Involved in the Pathogenesis of CNS MZBCL
EBV has been demonstrated as an important pathogen in primary CNS NHLs, in both immunocompetent and immunocompromised patients, particularly in the latter.34 To examine its potential role in the CNS MZBCLs, in situ hybridization for the EBER was performed in 10 patients with available tissue. No EBER signal was detected in any of these cases (data not shown). These findings strongly argue against the involvement of EBV in the pathogenesis of the CNS MZBCLs.
DISCUSSION
Primary dura-based lymphomas are distinct from the great majority of primary intraparenchymal CNS lymphomas in their predominance of low-grade forms, including small lymphocytic lymphoma,30,35 lymphoplasmacytic lymphoma,36,37 and follicular lymphoma.38,39 Although still a rare entity, CNS MZBCL is the most common primary CNS low-grade lymphoma, with 19 well-documented occurrences in the English literature, most of which were dural-based. To our knowledge, the current clinicopathologic study represents the largest series to date and the first to incorporate genetic data. Both our data and those previously published indicate that CNS MZBCL predominantly afflicts middle-aged women and presents as a solitary dural mass, radiographically indistinguishable from meningioma. In fact, 13 of our 15 patients carried a preoperative diagnosis of meningioma. Radiosurgery (RS) is increasingly being used to treat meningiomas based on a presumptive radiologic diagnosis alone. RS is widely available, cost-effective, and highly reproducible, and exhibits a wide safety margin. However, the use of RS for meningioma is critically dependent on an accurate imaging diagnosis. The strikingly different histology and its management implications in our patients should remind the clinician to proceed with caution in younger patents, patients whose history is short, or those whose presentation includes systemic symptoms.
Another common clinical scenario is that of atypical plasmacytoid lymphocytes encountered in CSF cytology studies. Serial CSF studies are often pursued to rule out low-grade lymphoma; sometimes attempts are made to pool CSF specimens for sufficient cells to perform flow cytometry. Because the vast majority of primary low-grade CNS lymphomas, including MZBCL, present as mass lesions, we propose that similar patients should be investigated further radiologically, rather than cytologically. If no mass is detected, then additional work-up for low-grade lymphoma is likely not warranted.
Most CNS MZBCLs have had an insidious onset and relatively long history of symptoms before diagnosis. A notable exception was the patient reported by Goetz et al,5 who had an acute onset of presentation and was thought to represent an acute subdural hematoma preoperatively. It is interesting to note that this tumor disappeared on computed tomography scan within 48 hours after low-dose oral dexamethasone treatment, but a separate lesion of similar size appeared 4 days later. Thus, this report suggests that MZBCL may follow a more rapid and dramatic clinical progression in rare cases. CNS MZBCLs generally share histologic features with those encountered in other anatomic sites. Deposition of amyloid has similarly been reported in MZBCLs outside of the CNS.40 In our series, two of the 15 patients had amyloid accumulation, with one patient forming radiologically visible tumefactive accumulations. Interestingly, another CNS MZBCL with massive amyloid deposition has been reported previously.10
All three types of MZBCLs outside of CNS are considered low-grade lymphomas, which follow a relatively indolent course. For example, the MALT-type MZBCLs have an overall 5-year survival rate of 81%. However, the long-term survival of the CNS MZBCLs remains to be defined. In this series, the mean follow-up period was 2.7 years, with the 7.6-year follow-up of patient 2 representing the longest reported to date. Nonetheless, all of our patients with clinical follow-up achieved complete remission after surgery and adjuvant therapy; all are currently alive with no evidence of disease. These data suggest that long-term survival is probably the rule and therapeutic cure may be possible, although longer follow-up will be necessary to confirm the latter. A single tumor in our series recurred (patient 2), although a complete remission was nevertheless accomplished after a change in chemotherapy. A variety of therapeutic treatments have also been applied in previously published examples. Thus far, CNS MZBCLs have responded to all of the therapeutic strategies applied. Therefore, although a conclusion on the optimal therapy cannot be determined based on the current data, it appears that surgery combined with radiation and/or chemotherapy is likely to achieve complete remission.
Outside the CNS, MZBCLs are believed to result from chronic inflammation induced by infectious microorganisms such as Helicobacter pylori, Borrelia burgdorferi, Campylobacter jejuni, and Chlamydia psittaci, or autoimmune diseases such as Hashimoto’s disease and Sj?gren’s syndrome. Hepatitis C virus has also been implicated as an important infectious etiology in the pathogenesis of nodal, splenic,19,41 and even CNS MZBCLs.4 Of interest, therapeutic eradication of Helicobacter pylori or hepatitis C virus leads to regression of gastric and splenic MZBCLs, respectively,42,43 and argues strongly for a causal role of these infections in the pathogenesis of MZBCLs. Whether or not a chronic inflammatory stimulus is germane to CNS MZBCLs remains to be defined. It is interesting to note, however, that patient 2 had a history of MS with several relapses and a previously reported case with Sj?gren’s syndrome.6
Recent epidemiologic evidence suggests a link between EBV and the risk of MS, as well as an increased incidence of both Hodgkin’s lymphomas and NHLs in MS patients.44,45 Therefore, it is interesting to determine if EBV or other infectious agents might play a role in the pathogenesis of CNS MZBCLs. Several EBV-encoded genes, such as EBV nuclear antigens, latent membrane proteins, and EBER, expressed during latency are useful makers for EBV infection and are known to play an important pathogenic role in other lymphomas. The latter is thought to act by stimulating the growth of infected B cells through enhanced secretion of interleukin-10 and the suppression of cytotoxic T cells.46 When we used in situ hybridization with an EBER probe in the current study, we did not find any evidence of EBV infection in the 10 patients tested. This argues against a tumorigenic role for EBV in CNS MZBCLs. In terms of its site of origin, clearly there is no CNS-associated mucosa. However, some have suggested that dura-associated lymphoid tissue may arise from chronic inflammatory stimuli, analogous to the MALT encountered in other organ systems.6,9
Several genetic abnormalities have been found in the MZBCLs. Although the t(11, 18)(q21;q21) or API2-MALT1 translocation has been shown to be the most specific genetic change and plays an important role in the pathogenesis of MZBCLs of MALT-type, this alteration is essentially not identified in either nodal- or splenic-type MZBCLs.47 These findings clearly indicate that the pathogenesis of MZBCLs is heterogeneous. Our data also show that MALT1-containing translocations are not a feature of the CNS MZBCLs, consistent with the notion that MALT1 translocations are primarily limited to certain MALT-type MZBCLs, especially the gastric variant. Trisomy 3, either partial or complete, represents the most common numeric chromosomal abnormality in MZBCLs outside of the CNS, reportedly detected in 15% to 60% by classical techniques and 60% to 80% by interphase cytogenetic techniques, such as FISH.17 In our series, CNS MZBCLs similarly had evidence of trisomy 3 in six (50%) of 12 patients analyzed by FISH. Of interest, trisomy 3 and API2-MALT1 fusions have been shown to be mutually exclusive in MZBCLs outside of the CNS.48,49 Our data are consistent with that interpretation and suggest that the pathogenesis of CNS MZBCLs is similar to that of MZBCLs encountered in sites other than the GI tract and lungs. Two chromosomal translocations t(14;18)(q32;q21) (ie, IgH-MALT1 translocation) and t(3;14)(p14.1;q32) (ie, IgH-FOXP1 translocation), were reported in a subset of the non-GI MZBCLs, with frequencies much higher in patients also harboring trisomy 3.23,24 However, we did not find any IgH-containing translocations in our patients, suggesting molecular genetic differences between MALT-type MZBCL in the CNS versus those of other sites, such as thyroid and orbital regions where IgH rearrangements have been described in 20% to 50% of patients.24
The Bcl-6 proto-oncogene is mapped to 3q27 and frequently is involved in the diffuse large B-cell lymphomas.50 It encodes a 95-kd nuclear phosphoprotein that functions as a potent transcriptional repressor, the function of which is required for the proliferation of mature B cells in germinal centers. It is suspected to have a causal role in the pathogenesis of a subset of MZBCLs with t(3;14)(q27;q32).33 Therefore, it is logical to hypothesize that trisomy 3 could potentially lead to an overexpression of Bcl-6, with subsequent tumorigenesis associated with an extra copy of this gene. However, our study failed to find any evidence for overexpression of the Bcl-6 protein, arguing against this possibility in CNS MZBCLs with and without trisomy 3. Therefore, trisomy 3 may either represent an epigenetic phenomenon that is associated with MZBCLs or may exert its effect on tumorigenesis through gene(s) other than the Bcl-6. Additional studies are needed to clarify the molecular pathogenesis of CNS MZBCLs.
Authors' Disclosures of Potential Conflicts of Interest
The authors indicated no potential conflicts of interest.
Acknowledgment
We thank Keith Rich, MD, of the Neurosurgery Department at Washington University for his help with clinical follow-up, as well as Ruma Banerjee, Kevin Selle, Kevin Keith, and Rodney Brown for their excellent technical assistance with in situ hybridization and immunostains.
NOTES
Authors' disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
Isaacson P, Wright DH: Malignant lymphoma of mucosa-associated lymphoid tissue. A distinctive type of B-cell lymphoma. Cancer 52:1410-1416, 1983
Altundag KM, Ozisik Y, Yalcin S, et al: Primary low-grade B-cell lymphoma of the dura in an immunocompromised patient. J Exp Clin Cancer Res 19:249-251, 2000
Bodi I, Hussain A, Gullan RW, et al: 56-year-old female with right frontal tumor of the dura. Brain Pathol 13:417-418, 2003
Estevez M, Chu C, Pless M: Small B-cell lymphoma presenting as diffuse dural thickening with cranial neuropathies. J Neurooncol 59:243-247, 2002
Goetz P, Lafuente J, Revesz T, et al: Primary low grade B-cell lymphoma of mucosa-associated lymphoid tissue of the dura mimicking the presentation of an acute subdural hematoma. J Neurosurg 96:611-614, 2002
Itoh T, Shimizu M, Kitami K, et al: Primary extranodal marginal zone B-cell lymphoma of the mucosa-associated lymphoid tissue type in the CNS. Neuropathology 21:174-180, 2001
Kambham N, Chang Y, Matsushima AY: Primary low-grade B-cell lymphoma of mucosa-associated lymphoid tissue (MALT) arising in dura. Clin Neuropathol 17:311-317, 1998
Garcia-Serra A, Mendenhall NP, Hinerman RW, et al: Management of neurotropic low-grade B-cell lymphoma: Report of two cases. Head Neck 25:972-976, 2003
Kumar S, Kumar D, Kaldjian EP, et al: Primary low-grade B-cell lymphoma of the dura: A mucosa associated lymphoid tissue-type lymphoma. Am J Surg Pathol 21:81-87, 1997
Lehman NL, Horoupian DS, Warnke R, et al: Dural marginal zone lymphoma with massive amyloid deposition: Rare low-grade primary central nervous system B-cell lymphoma. J Neurosurg 96:368-372, 2002
Lima VS, Leite EB, Fonseca RP, et al: Primary low grade mucosa-associated B-cell lymphoma of the dura. Diagn Oncol 21:4058-4063, 2003
Menniti A, Moschettoni L, Liccardo G, et al: Low-grade primary meningeal lymphoma: Case report and review of the literature. Neurosurg Rev 28:229-233, 2005
Roonprapunt C, Retech H, LaSala P et al: Primary diffuse B-cell lymphoma arising in the dura in an immunocompetent individual: Case report. J Neuropathol Exp Neurol 62:574, 2003 (abstr 146)
Rottnek M, Strauchen J, Moore F, et al: Primary dural mucosa-associated lymphoid tissue-type lymphoma: Case report and review of the literature. J Neurooncol 68:19-23, 2004
Sanjeevi A, Krishnan J, Bailey PR, et al: Extranodal marginal zone B-cell lymphoma of malt type involving the cavernous sinus. Leuk Lymphoma 42:1133-1137, 2001
Jaffe ES, Harris NL, Stein H, et al: World Health Organization Classification of Tumours. Pathology & Genetics. Tumours of Haematopoietic and Lymphoid Tissues. Lyon, France, IARC Press, 2001
Maes B, de Wolf-Peeters C: Marginal zone cell lymphoma: An update on recent advances. Histopathology 40:117-126, 2002
Franco V, Florena AM, Lannitto E: Splenic marginal zone lymphoma. Blood 101:2464-2472, 2003
Arcaini L, Paulli M, Boveri E, et al: Splenic and nodal marginal zone lymphomas are indolent disorders at high hepatitis C virus seroprevalence with distinct presenting features but similar morphologic and phenotypic profiles. Cancer 100:107-115, 2004
Tam CS, Prince HM, Westerman D, et al: Leukemic subtype of marginal zone lymphoma: A presentation of three cases and literature review. Leuk Lymphoma 45:705-710, 2004
Martin F, Kearney JF: Marginal-zone B cells. Nat Rev Immunol 2:323-335, 2002
Dierlamm J, Wlodarska I, Michaux L, et al: Genetic abnormalities in marginal zone B-cell lymphoma. Hematol Oncol 18:1-13, 2000
Streubel B, Lamprecht A, Dierlamm J, et al: T(14;18)(q32;q21) involving IgH and MALT1 is a frequent chromosomal aberration in MALT lymphoma. Blood 101:2335-2339, 2003
Streubel B, Vinatzer U, Lamprecht A, et al: T(3;14)(p14.1;q32) involving IgH and FOXP1 is a novel recurrent chromosomal aberration in MALT lymphoma. Leukemia 19:652-658, 2005
Remstein ED, James CD, Kurtin PJ: Incidence and subtype specificity of API2-MALT1 fusion translocations in extranodal, nodal and splenic marginal zone lymphomas. Am J Pathol 156:1183-1188, 2000
Rosenwald A, Ott G, Stilgenbauer S, et al: Exclusive detection of the t(11;18)(q21;q21) in extranodal marginal zone B cell lymphomas (MZBL) of MALT type in contrast to other MZBL and extranodal large B cell lymphomas. Am J Pathol 155:1817-1821, 1999
Akagi T, Motegi M, Tamura A, et al:, A novel gene, MALT1 at 18q21, is involved in t(11;18)(q21;q21) found in low grade B-cell lymphoma of mucosa-associated lymphoid tissue. Oncogene 18:5785-5794, 1999
Dierlamm J, Baens M, Wiodarska I, et al: The apoptosis inhibitor gene API2 and a novel 18q gene, MLT, are recurrently rearranged in the t(11;18)(q21;q21) associated with mucosa-associated lymphoid tissue lymphoma. Blood 93:3601-3609, 1999
Ruefli-Brasse AA, French DM, Dixit VM: Regulation of NF-B-dependent lymphocyte activation and development by paracaspase. Science 302:1581-1584, 2003
Lachance DH, O'Neill BP, Macdonald DR, et al: Primary leptomeningeal lymphoma: Report of 9 cases, diagnosis with immunocytochemical analysis and review of the literature. Neurology 41:95-100, 1991
Schlegel U, Schmidt-Wolf IgH, Deckert M: Primary CNS lymphoma: Clinical presentation, pathological classification, molecular pathogenesis and treatment. J Neurol Sci 181:1-12, 2000
Rajaram V, Leuthardt EC, Singh PK, et al: 9p21 and 13q14 dosages in ependymomas: A clinicopathologic study of 101 cases. Mod Pathol 17:9-14, 2004
Dierlamm J, Pittaluga S, Stul M, et al: BCL6 gene rearrangements also occur in marginal zone B-cell lymphoma. Br J Haematol 98:719-725, 1997
Nakhleh RE, Manivel JC, Copenhaver CM, et al: In situ hybridization for the detection of Epstein-Barr virus in central nervous system lymphomas. Cancer 67:444-448, 1991
Kai Y, Kuratsu J, Ushio Y: Primary malignant lymphoma of the brain in childhood. Neurol Med Chir 38:232-237, 1998
Murphy JK, O'Brien CJ, Ironside JW: Morphologic and immunophenotypic characterization of primary brain lymphomas using paraffin-embedded tissue. Histopathology 15:449-460, 1989
Ellie E, Vergier B, Duche B, et al: Local amyloid deposits in a primary central nervous system lymphoma: Study of a stereotactic brain biopsy. Clin Neuropathol 9:231-233, 1990
Beriwal S, Hou JS, Miyamoto C, et al: Primary dural low grade BCL-2 negative follicular lymphoma: A case report. J Neurooncol 61:23-25, 2003
Subramanian S, Bahceci E, Kanji A: Primary follicular lymphoma of the meninges: A case report. Proc Am Soc Clin Oncol 14:153, 1995 (abstr 308)
Gotteri G, Ranaldi R, Pileri SA, et al: Localized amyloidosis and gastrointestinal lymphoma: A rare association. Histopathology 32:348-355, 1998
Weng W, Levy S: Hepatitis C virus (HCV) and lymphomagenesis. Leuk Lymphoma 44:1113-1120, 2003
Hermine O, Lefrere F, Bronowicki JP, et al: Regression of splenic lymphoma with villous lymphocytes after treatment of hepatitis C virus infection. N Engl J Med 347:89-94, 2002
Wotherspoon AC, Doglioni C, Diss TC, et al: Regression of primary low-grade B-cell gastric lymphoma of mucosa-associated lymphoid tissue type after eradication of Helicobacter pylori. Lancet 342:575-577, 1993
Hjalgrim H, Rasmussen S, Rostgaard K, et al: Familial clustering of Hodgkin lymphoma and multiple sclerosis. J Natl Cancer Inst 96:780-784, 2004
Vineis P, Crosignani P, Vigano C, et al: Lymphoma and multiple sclerosis in a multicenter case-control study. Epidemiology 12:134-135, 2001
Küpper R: B cells under influence: Transformation of B cells by Epstein-Barr virus. Nat Rev Immunol 3:801-812, 2003
Muller-Hermelink HK: Genetic and molecular genetic studies in the diagnosis of B-cell lymphomas: Marginal zone lymphomas. Hum Pathol 34:336-340, 2003
Ott G, Kalla J, Steinhoff A, et al: Trisomy 3 is not a common feature in malignant lymphomas of mucosa-associated lymphoid tissue type. Am J Pathol 153:689-694, 1998
Remstein ED, Kurtin PJ, James CD, et al: Mucosa-associated lymphoid tissue lymphomas with t(11;18)(q21;q21) and mucosa-associated lymphoid tissue lymphomas with aneuploidy develop along different pathogenetic pathways. Am J Pathol 161:63-71, 2002
Pasqualucci L, Bereschenko O, Niu H, et al: Molecular pathogenesis of non-Hodgkin's lymphoma: The role of Bcl-6. Leuk Lymphoma 44:S5-S12, 2003 (suppl 3)(Pang-hsien Tu, Caterina G)
Division of Neuropathology and Department of Neurology, Mayo Clinic, Rochester, MN
Department of Pathology, Children's Hospital of Philadelphia, University of Pennsylvania School of Medicine, Philadelphia, PA
Division of Neurosurgery, University of Toronto, Toronto, Ontario, Canada
Division of Neuropathology, University of Florida College of Medicine, Gainesville, FL
Department of Pathology, The Johns Hopkins University School of Medicine, Baltimore, MD
ABSTRACT
PURPOSE: Although rare overall, marginal zone B-cell lymphoma (MZBCL) is the most common primary low-grade CNS lymphoma reported in the literature. The aim of this study is to elucidate the biology and genetic features of this unusual tumor.
PATIENTS AND METHODS: Fifteen CNS MZBCLs were studied clinically, pathologically, and genetically, including fluorescent in situ hybridization analyses with commercially available MALT1 and IgH break-apart and centromere 3, 7, 12, and 18 probes.
RESULTS: CNS MZBCLs preferentially affect middle-aged women (female-to-male ratio, 4:1), with 93% presenting as dural-based masses mimicking meningioma. Ten patients with 1 to 7.6 years of follow-up after diagnosis showed no evidence of disease after radiation and/or chemotherapy. Like MZBCLs outside of the CNS, they consisted of CD20+, CD3– small B lymphocytes with varying degrees of plasmacytic differentiation and predominantly light-chain restriction (78%). Lymphoid follicles with follicular colonization were seen in three patients and deposition of amyloid was noted in samples from two patients, one of which was tumefactive. Neither Bcl-6 protein nor Epstein-Barr virus–encoded RNA was expressed. Trisomy 3 was detected in six of 12 patients, with no rearrangements of MALT1 or IgH and no trisomies of 7, 12, or 18 detected.
CONCLUSION: Our data suggest that intracranial MZBCL is an indolent primary CNS lymphoma that typically presents as a meningioma-like dural-based mass. Trisomy 3, but not MALT1 or IgH translocation, is a common genetic abnormality that may contribute to the pathogenesis of this CNS lymphoma.
INTRODUCTION
Marginal zone B-cell lymphoma (MZBCL) was first described in the mucosa-associated lymphoid tissue (MALT) of the gastrointestinal tract by Isaacson and Wright1 as an indolent low-grade lymphoma. It was subsequently recognized in the mucosa of other organs including lung, bladder, salivary gland, conjunctiva, and lacrimal gland, as well as tissue sites without a mucosa, such as thyroid, breast, thymus, orbit, skin, liver, and even the CNS.2-15 MZBCLs were accepted as a distinct non-Hodgkin’s lymphoma (NHL) in the Revised European and American Lymphoma classification in 1994. The current WHO classification16 further subdivides this lymphoma into three types: extranodal or MALT, nodal, and splenic MZBCLs. The three types of MZBCLs have been estimated to constitute approximately 7.6%, 1.8%, and less than 1% of NHL, respectively. Although these three types share similar histopathologic and immunophenotypic features, they differ in terms of clinical behavior.17 Briefly, MALT-type MZBCL typically presents as a long-standing local mass in tissues that acquire MALT during an immune reaction; bone marrow and lymph node involvement is rare. In contrast, patients with nodal MZBCLs usually present at an advanced clinical stage, characterized by lymphadenopathy and bone marrow disease. Splenic MZBCL is characterized by splenomegaly in addition to peripheral-blood and marrow involvement,18 and may be associated with infection with hepatitis C virus.19 As with other low-grade NHLs, advanced clinical stage does not necessarily portend a poor outcome; newly diagnosed stage 3 patients have an 80% 10-year survival rate. Recently, a leukemic variant of MZBCL without any involvement of MALT-containing tissue, lymph node, or spleen has been reported, also with a surprisingly indolent course.20
The MZBCL is believed to derive from postgerminal center marginal zone B-cells, best visualized surrounding secondary follicles of mesenteric lymph node; spleen and acquired MALT typically occur in organs where exposure to antigens is considerable.16,21 Both the normal cells and their neoplastic counterparts are characterized by abundant clear cytoplasm and an irregular, centrally located nucleus (monocytoid B cells). They express pan–B-lymphocyte markers (CD19, CD20, and CD79a), some complement receptors (eg, CD21, CD35), and surface immunoglobulin, but do not express CD3, CD5, CD10, CD23, or cyclin D1.17
Recently, a number of cytogenetic abnormalities have been identified in MZBCLs. The most common karyotypic abnormalities are trisomies of chromosomes 3, 7, 12, and 18, as well as translocations such as t(11;18) (q21;q21), t(3;14)(q27;q32), t(1;14)(p22;q32), t(14;18)(q32;q21), and t(3;14)(p14.1;q32).22-24 The frequency of these abnormalities varies among the organs involved and the tumor types. For example, the t(11;18)(q21;q21) chromosomal abnormality is seen in approximately 20% to 30% of MALT-type MZBCL, but is not found in the nodal and splenic types.25,26 In fact, the t(11;18)(q21;q21) translocation is considered specific to MZBCLs, and thus far, has not been encountered in any other types of lymphoma. The translocation results in an API2-MALT1 fusion gene encoding a chimeric protein consisting of the N-terminal portion of API2 linked to the C-terminus of the paracaspase, MALT1.27,28 The normal MALT1 protein functions downstream of Bcl-10 to activate nuclear factor kappa B, and is required for the generation or maintenance of marginal zone B cells.29 The API2-MALT1 fusion protein is suspected to promote tumorigenesis through auto-oligomerization, which results in constitutive activation of nuclear factor kappa B signaling in marginal zone B cells. In contrast, trisomy 3 has been detected in all three types of MZBCLs with similar, albeit widely ranging frequencies of 15% to 60%.17,22 Most recently, immunoglobulin heavy chain (IgH) rearrangements have also been described, most commonly involving MALT-type MZBCL of the thyroid and orbital regions.24
Within the CNS, the vast majority of primary lymphomas are diffuse large B-cell lymphomas presenting in periventricular or deep brain parenchymal regions.30,31 Primary low-grade lymphomas occur at a much lower incidence, appearing in the literature as small series or single case reports. Among the low-grade group, MZBCL is by far the most common primary CNS lymphoma. To our knowledge, 19 well-documented examples have been reported previously.2-15 However, no genetic data have been published to date. In this study, we report the clinicopathologic features of 15 additional patients, including genetic analyses using fluorescent in situ hybridization (FISH). Our data further support the female predilection, frequent dural localization radiographically mimicking meningioma, favorable clinical behavior, and occasional tumefactive amyloid deposition. Genetically, our patients lacked MALT1- or IgH-containing translocations, but half harbored FISH-detectable trisomy 3.
PATIENTS AND METHODS
Patient and Tumor Cohort
A search for the diagnosis of MZBCL of the CNS yielded 16 patients from the pathology files at Washington University (St Louis, MO), University of Pennsylvania (Philadelphia, PA), Johns Hopkins University (Baltimore, MD), Mayo Clinic (Rochester, MN), and the University of Florida (Gainesville, FL). On subsequent review, one patient was classified as having a low-grade lymphoplasmacytic lymphoma and was excluded from further study. All available routine and immunohistochemistry slides were reviewed and a representative block was chosen from each available patient. Unstained slides were cut at 5-μm thickness and mounted onto precoated superfrost/plus glass slides for FISH studies (Fisher Scientific, Pittsburgh, PA). This study was performed after approval by the Human Studies Committee of each institute and in accord with the regulations approved by the Department of Health and Human Services.
FISH
Dual-color FISH experiments were performed using commercially available MALT1 and IgH break-apart, as well as centromere-enumerating DNA probes (CEP), CEP3, CEP7, CEP8, CEP12 (Vysis Inc, Downers Grove, IL) as described previously.32 Briefly, after deparaffinization, the sections were subjected to target retrieval by steam cooking in citrate buffer for 20 minutes followed by a 20-minute cool-down period and a 5-minute wash (distilled water). This was followed by pepsin (4 mg/mL) digestion at 37°C for 30 minutes. The slides were then washed in 2x saline sodium citrate (SSC) and allowed to air dry. SpectrumOrange (Vysis Inc) -labeled CEP3 probe was paired with SpectrumGreen (Vysis Inc) -labeled CEP12; SpectrumOrange-labeled CEP7 was similarly paired with SpectrumGreen-labeled CEP8. The probes were diluted in DenHyb hybridization buffer (Insitus Laboratories, Albuquerque, NM) at either 1:25 or 1:50 dilution. The hybridization mix was applied to the sections, followed by simultaneous denaturation of the probe and target DNA at 90°C for 13 minutes. Overnight hybridization at 37°C took place in a humidified chamber. Posthybridization washes in 50% formamide/1x SSC (5 minutes) and 2x SSC (5 minutes) were performed at room temperature, and the slides were allowed to air dry.
4',6-Diamidino-2-phenylindole (Insitus Laboratories) was used as a nuclear counterstain, and the sections were viewed under an Olympus BX60 fluorescent microscope with appropriate filters (Olympus, Melville, NY). Two of the authors (P-h.T. and A.P.) each independently scored 100 intact nonoverlapping nuclei for the number of fluorescent red and green signals derived from each hybridization. Nuclei with more than two signals were rarely seen in non-neoplastic controls; therefore, trisomies and polysomies (gains) were arbitrarily defined as more than 5% nuclei containing three or more signals. For the MALT1 and IgH break-apart probe cocktails, nuclei were scored for the presence or absence of split red and green signals, with two yellow or red-green fusion signals indicating a lack of chromosomal rearrangement. Samples with more than 5% nuclei containing split signals were considered positive for a MALT1- or IgH-containing chromosomal translocation.
Hybridizations were digitally photographed using a high-resolution black and white COHU charge-coupled device camera with a Z-stack motor programmed to capture images at sequential 4',6-diamidino-2-phenylindole (one level), fluorescein isothiocyanate (10 levels), and rhodamine (10 levels) filter settings. Reconstruction into a single superimposed image with blue, green, and red pseudocolors was accomplished using a CytoVision basic workstation (Applied Imaging, Santa Clara, CA).
Immunohistochemistry Staining
Immunohistochemistry was performed on 5-μm-thick sections of paraffin-embedded, formalin-fixed tissue with a DAKO autostainer (DAKO, Carpinteria, CA) with antigen retrieval using 1 mmol/L EDTA, pH 8.0. Reagents were used as supplied in the Envision Plus Detection Kit (DAKO). The antibodies used in this study were CD3 (clone UCHT1; DAKO), CD5 (clone SP19; NeoMarkers, Fremont, CA), CD20 (clone L26; DAKO), CD23 (clone MHM6; DAKO), CD43 (clone DF-T1; DAKO), immunoglobin light chains (IgLs) and rabbit polyclonal antisera (DAKO), cyclin D1 (clone SP4; NeoMarkers), and Bcl-6 (clone P1F6; Neomarkers). The concentrations of individual antibodies were optimized by the immunohistochemistry laboratory of each medical center.
Epstein-Barr Virus–Encoded Nontranslated RNA In Situ Hybridization
In situ hybridization was performed on 5-μm-thick sections of paraffin-embedded formalin-fixed with the INFORM Epstein-Barr virus (EBV) –encoded nontranslated RNA (EBER) probe (Ventana Medical Systems Inc, Tucson, AZ), which contained a cocktail of oligonucleotide probes, and the signals were visualized using the ISH iVIEW Blue Detection Kit (Ventana) with Nitro Blue Tetrazolium and an ISH Red Counterstain. All of the procedures were performed with the BenchMark XT autostainer (Ventana) according to the manufacturer’s instructions.
RESULTS
Clinical Characteristics
Clinical data from the 15 patients are summarized in Table 1. Twelve patients (80%) were women, yielding a female-to-male ratio of 4:1. The mean age at diagnosis was 55 years (range, 29 to 70 years). Presenting symptoms were either nonspecific (headache, nausea, vomiting, visual disturbance) or localized (focal seizures, motor/sensory deficits, ataxia, gait disturbance, otalgia). Patient 2 also had a long history of relapsing/remitting multiple sclerosis (MS). All but one patient (patient 3) presented with a contrast-enhancing, extra-axial, dural-based or leptomeningeal mass, often with a dural tail sign, which is radiologically consistent with meningioma (Fig 1). An en plaque pattern of meningeal thickening was reported in two patients. The tumors arose at various intracranial sites, although the cerebral convexity was most common. The falx was involved in three patients, the tentorium was involved in two patients, and the sellar/parasellar region was involved in one patient. No definitive evidence of tumor outside of the CNS was found in any of the patients.
All of the patients underwent craniotomy with tumor resections. Of the nine patients with available information about adjuvant therapy, six received whole-brain radiation alone; two had combined chemotherapy and radiation, and one received chemotherapy alone (Table 1). Various regimens including procarbazine, vincristine, intravenous cytarabine, leucovorin, and/or intravenous or intrathecal methotrexate were used. Patient 2 was initially treated with methotrexate, but suffered a recurrence 4 months later. The recurrent tumor was subsequently treated with fludarabine with an excellent response, resulting in complete remission. These data indicate that surgery plus radiation and/or chemotherapy is sufficient to achieve complete remission. The longest followup in our series was 7.6 years, which suggests that long-term remission is probable.
Pathologic Features
Sections revealed a mixed cellular infiltrate primarily involving dura and leptomeninges, with extension along Virchow-Robin spaces in some patients (Fig 2A). Focal infiltration of the neoplastic cells into the underlying brain parenchyma was noted in two patients. The architectural and cytologic features were otherwise identical to those of MZBCLs in other sites. They were composed of sheets of small- to medium-sized lymphocytes with moderate amounts of cytoplasm and irregular nuclei, bearing a close resemblance to the small-cleaved cells or centrocytes of the lymphoid follicles (Fig 2B). Many of these cells exhibited the so-called monocytoid appearance characterized by clear cytoplasm and well-defined cell borders. Scattered among these cells were occasional immunoblast-like cells with large vesicular nuclei and prominent nucleoli. Plasma cells were identified in every patient, ranging from rare to plentiful. Reactive lymphoid follicles were identified in three patients (patients 4, 6, and 7; Fig 2C). In patient 4, some follicles were replaced partially or completely by the neoplastic cells, mimicking follicular lymphoma focally.
Mitotic figures were generally hard to find, but were focally increased in patients 1 and 15. Scattered islands with an increase in the number of large immunoblast-like cells were also identified in these two patients. However, no definitive features of transformation to a more aggressive histology were seen (Fig 2D). Numerous plasma cells with eosinophilic intracytoplasmic inclusions (Russell bodies) were noted in patient 2 (Fig 3A). Amyloid deposition with foreign body giant cell reaction was identified in two patients (patients 2 and 9; Fig 3B and inset). The amount of the deposited amyloid material could be scant (patient 2) or massive (patient 9), the latter forming tumefactive thioflavin-S–positive amyloid pools (Figs 3C and 3D) that were large enough to detect radiographically (the hypodense regions within the tumor in Figs 1C and 1D).
Immunohistochemical and genetic findings are summarized in Table 2. The neoplastic cells were CD20+ and CD3–, consistent with a B-cell origin (Figs 4A and 4B) Monoclonality was demonstrated by immunohistochemical stains for IgL. In our patients, light-chain restriction was most common (10 of 14 patients; Figs 4C and 4D; Table 2). The amyloid pools in patient 9 were also highlighted by the immunostain for light chain (Fig 4C), but not by that for light chain (Fig 4D), suggesting that the deposited amyloid was derived from secreted IgL. Ten patient samples available for the cyclin D1 immunostain showed no immunoreactivity and ruled out the possibility of mantle-cell lymphoma.
Characterization of the Genetic Alterations by FISH
As an initial attempt to elucidate the associated genetic changes, FISH analyses were performed using probes targeting previously reported alterations. The results of these studies are summarized in Table 2. Hybridizations with the MALT1 and IgH break-apart probes each produced two fused yellow or juxtaposed red and green signals in all 12 and 10 patients tested, respectively (Fig 5A). Therefore, there was no evidence of translocations involving the MALT1 or IgH genes (Table 2). Hybridizations using the CEP 7, CEP12, and CEP18 typically produced no more than two signals, consistent with a normal disomic state (except for patient 12, who had polysomy or gains of chromosome 7). However, a subset of neoplastic cells in six of the 12 patients (50%) yielded three CEP3 signals, consistent with trisomy 3 (Fig 5B). Cells with three CEP3 signals ranged from 9% to 53%, with an average of 29%. The lack of other associated chromosomal abnormalities suggests that the trisomy 3 is a specific genetic alteration, rather than a nonspecific chromosomal gain associated with an overall state of polyploidy.
Bcl-6 Is Not Involved in the Pathogenesis of CNS MZBCL
The gene encoding Bcl-6 protein is located on the chromosome 3q27 and has been suggested to be a potential oncogene involved in MZBCL.33 To test if it has a role in the pathogenesis of the CNS MZBCL, immunohistochemistry was performed in 10 of 15 patients, five of whom had trisomy 3 and the rest of whom had normal chromosome 3 copy numbers. The neoplastic cells in all 10 patients were immunonegative for Bcl-6. However, small clusters of positive cells that had vesicular nuclei larger than those of the neoplastic cells were noted in two patients (data not shown). These cells were interpreted as representing remnant germinal center elements, whereas the small- to medium-sized neoplastic marginal zone lymphocytes appeared negative. Thus, our data do not support a role for Bcl-6 in the pathogenesis of the CNS MZBCLs.
EBV Expression Is Not Involved in the Pathogenesis of CNS MZBCL
EBV has been demonstrated as an important pathogen in primary CNS NHLs, in both immunocompetent and immunocompromised patients, particularly in the latter.34 To examine its potential role in the CNS MZBCLs, in situ hybridization for the EBER was performed in 10 patients with available tissue. No EBER signal was detected in any of these cases (data not shown). These findings strongly argue against the involvement of EBV in the pathogenesis of the CNS MZBCLs.
DISCUSSION
Primary dura-based lymphomas are distinct from the great majority of primary intraparenchymal CNS lymphomas in their predominance of low-grade forms, including small lymphocytic lymphoma,30,35 lymphoplasmacytic lymphoma,36,37 and follicular lymphoma.38,39 Although still a rare entity, CNS MZBCL is the most common primary CNS low-grade lymphoma, with 19 well-documented occurrences in the English literature, most of which were dural-based. To our knowledge, the current clinicopathologic study represents the largest series to date and the first to incorporate genetic data. Both our data and those previously published indicate that CNS MZBCL predominantly afflicts middle-aged women and presents as a solitary dural mass, radiographically indistinguishable from meningioma. In fact, 13 of our 15 patients carried a preoperative diagnosis of meningioma. Radiosurgery (RS) is increasingly being used to treat meningiomas based on a presumptive radiologic diagnosis alone. RS is widely available, cost-effective, and highly reproducible, and exhibits a wide safety margin. However, the use of RS for meningioma is critically dependent on an accurate imaging diagnosis. The strikingly different histology and its management implications in our patients should remind the clinician to proceed with caution in younger patents, patients whose history is short, or those whose presentation includes systemic symptoms.
Another common clinical scenario is that of atypical plasmacytoid lymphocytes encountered in CSF cytology studies. Serial CSF studies are often pursued to rule out low-grade lymphoma; sometimes attempts are made to pool CSF specimens for sufficient cells to perform flow cytometry. Because the vast majority of primary low-grade CNS lymphomas, including MZBCL, present as mass lesions, we propose that similar patients should be investigated further radiologically, rather than cytologically. If no mass is detected, then additional work-up for low-grade lymphoma is likely not warranted.
Most CNS MZBCLs have had an insidious onset and relatively long history of symptoms before diagnosis. A notable exception was the patient reported by Goetz et al,5 who had an acute onset of presentation and was thought to represent an acute subdural hematoma preoperatively. It is interesting to note that this tumor disappeared on computed tomography scan within 48 hours after low-dose oral dexamethasone treatment, but a separate lesion of similar size appeared 4 days later. Thus, this report suggests that MZBCL may follow a more rapid and dramatic clinical progression in rare cases. CNS MZBCLs generally share histologic features with those encountered in other anatomic sites. Deposition of amyloid has similarly been reported in MZBCLs outside of the CNS.40 In our series, two of the 15 patients had amyloid accumulation, with one patient forming radiologically visible tumefactive accumulations. Interestingly, another CNS MZBCL with massive amyloid deposition has been reported previously.10
All three types of MZBCLs outside of CNS are considered low-grade lymphomas, which follow a relatively indolent course. For example, the MALT-type MZBCLs have an overall 5-year survival rate of 81%. However, the long-term survival of the CNS MZBCLs remains to be defined. In this series, the mean follow-up period was 2.7 years, with the 7.6-year follow-up of patient 2 representing the longest reported to date. Nonetheless, all of our patients with clinical follow-up achieved complete remission after surgery and adjuvant therapy; all are currently alive with no evidence of disease. These data suggest that long-term survival is probably the rule and therapeutic cure may be possible, although longer follow-up will be necessary to confirm the latter. A single tumor in our series recurred (patient 2), although a complete remission was nevertheless accomplished after a change in chemotherapy. A variety of therapeutic treatments have also been applied in previously published examples. Thus far, CNS MZBCLs have responded to all of the therapeutic strategies applied. Therefore, although a conclusion on the optimal therapy cannot be determined based on the current data, it appears that surgery combined with radiation and/or chemotherapy is likely to achieve complete remission.
Outside the CNS, MZBCLs are believed to result from chronic inflammation induced by infectious microorganisms such as Helicobacter pylori, Borrelia burgdorferi, Campylobacter jejuni, and Chlamydia psittaci, or autoimmune diseases such as Hashimoto’s disease and Sj?gren’s syndrome. Hepatitis C virus has also been implicated as an important infectious etiology in the pathogenesis of nodal, splenic,19,41 and even CNS MZBCLs.4 Of interest, therapeutic eradication of Helicobacter pylori or hepatitis C virus leads to regression of gastric and splenic MZBCLs, respectively,42,43 and argues strongly for a causal role of these infections in the pathogenesis of MZBCLs. Whether or not a chronic inflammatory stimulus is germane to CNS MZBCLs remains to be defined. It is interesting to note, however, that patient 2 had a history of MS with several relapses and a previously reported case with Sj?gren’s syndrome.6
Recent epidemiologic evidence suggests a link between EBV and the risk of MS, as well as an increased incidence of both Hodgkin’s lymphomas and NHLs in MS patients.44,45 Therefore, it is interesting to determine if EBV or other infectious agents might play a role in the pathogenesis of CNS MZBCLs. Several EBV-encoded genes, such as EBV nuclear antigens, latent membrane proteins, and EBER, expressed during latency are useful makers for EBV infection and are known to play an important pathogenic role in other lymphomas. The latter is thought to act by stimulating the growth of infected B cells through enhanced secretion of interleukin-10 and the suppression of cytotoxic T cells.46 When we used in situ hybridization with an EBER probe in the current study, we did not find any evidence of EBV infection in the 10 patients tested. This argues against a tumorigenic role for EBV in CNS MZBCLs. In terms of its site of origin, clearly there is no CNS-associated mucosa. However, some have suggested that dura-associated lymphoid tissue may arise from chronic inflammatory stimuli, analogous to the MALT encountered in other organ systems.6,9
Several genetic abnormalities have been found in the MZBCLs. Although the t(11, 18)(q21;q21) or API2-MALT1 translocation has been shown to be the most specific genetic change and plays an important role in the pathogenesis of MZBCLs of MALT-type, this alteration is essentially not identified in either nodal- or splenic-type MZBCLs.47 These findings clearly indicate that the pathogenesis of MZBCLs is heterogeneous. Our data also show that MALT1-containing translocations are not a feature of the CNS MZBCLs, consistent with the notion that MALT1 translocations are primarily limited to certain MALT-type MZBCLs, especially the gastric variant. Trisomy 3, either partial or complete, represents the most common numeric chromosomal abnormality in MZBCLs outside of the CNS, reportedly detected in 15% to 60% by classical techniques and 60% to 80% by interphase cytogenetic techniques, such as FISH.17 In our series, CNS MZBCLs similarly had evidence of trisomy 3 in six (50%) of 12 patients analyzed by FISH. Of interest, trisomy 3 and API2-MALT1 fusions have been shown to be mutually exclusive in MZBCLs outside of the CNS.48,49 Our data are consistent with that interpretation and suggest that the pathogenesis of CNS MZBCLs is similar to that of MZBCLs encountered in sites other than the GI tract and lungs. Two chromosomal translocations t(14;18)(q32;q21) (ie, IgH-MALT1 translocation) and t(3;14)(p14.1;q32) (ie, IgH-FOXP1 translocation), were reported in a subset of the non-GI MZBCLs, with frequencies much higher in patients also harboring trisomy 3.23,24 However, we did not find any IgH-containing translocations in our patients, suggesting molecular genetic differences between MALT-type MZBCL in the CNS versus those of other sites, such as thyroid and orbital regions where IgH rearrangements have been described in 20% to 50% of patients.24
The Bcl-6 proto-oncogene is mapped to 3q27 and frequently is involved in the diffuse large B-cell lymphomas.50 It encodes a 95-kd nuclear phosphoprotein that functions as a potent transcriptional repressor, the function of which is required for the proliferation of mature B cells in germinal centers. It is suspected to have a causal role in the pathogenesis of a subset of MZBCLs with t(3;14)(q27;q32).33 Therefore, it is logical to hypothesize that trisomy 3 could potentially lead to an overexpression of Bcl-6, with subsequent tumorigenesis associated with an extra copy of this gene. However, our study failed to find any evidence for overexpression of the Bcl-6 protein, arguing against this possibility in CNS MZBCLs with and without trisomy 3. Therefore, trisomy 3 may either represent an epigenetic phenomenon that is associated with MZBCLs or may exert its effect on tumorigenesis through gene(s) other than the Bcl-6. Additional studies are needed to clarify the molecular pathogenesis of CNS MZBCLs.
Authors' Disclosures of Potential Conflicts of Interest
The authors indicated no potential conflicts of interest.
Acknowledgment
We thank Keith Rich, MD, of the Neurosurgery Department at Washington University for his help with clinical follow-up, as well as Ruma Banerjee, Kevin Selle, Kevin Keith, and Rodney Brown for their excellent technical assistance with in situ hybridization and immunostains.
NOTES
Authors' disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
Isaacson P, Wright DH: Malignant lymphoma of mucosa-associated lymphoid tissue. A distinctive type of B-cell lymphoma. Cancer 52:1410-1416, 1983
Altundag KM, Ozisik Y, Yalcin S, et al: Primary low-grade B-cell lymphoma of the dura in an immunocompromised patient. J Exp Clin Cancer Res 19:249-251, 2000
Bodi I, Hussain A, Gullan RW, et al: 56-year-old female with right frontal tumor of the dura. Brain Pathol 13:417-418, 2003
Estevez M, Chu C, Pless M: Small B-cell lymphoma presenting as diffuse dural thickening with cranial neuropathies. J Neurooncol 59:243-247, 2002
Goetz P, Lafuente J, Revesz T, et al: Primary low grade B-cell lymphoma of mucosa-associated lymphoid tissue of the dura mimicking the presentation of an acute subdural hematoma. J Neurosurg 96:611-614, 2002
Itoh T, Shimizu M, Kitami K, et al: Primary extranodal marginal zone B-cell lymphoma of the mucosa-associated lymphoid tissue type in the CNS. Neuropathology 21:174-180, 2001
Kambham N, Chang Y, Matsushima AY: Primary low-grade B-cell lymphoma of mucosa-associated lymphoid tissue (MALT) arising in dura. Clin Neuropathol 17:311-317, 1998
Garcia-Serra A, Mendenhall NP, Hinerman RW, et al: Management of neurotropic low-grade B-cell lymphoma: Report of two cases. Head Neck 25:972-976, 2003
Kumar S, Kumar D, Kaldjian EP, et al: Primary low-grade B-cell lymphoma of the dura: A mucosa associated lymphoid tissue-type lymphoma. Am J Surg Pathol 21:81-87, 1997
Lehman NL, Horoupian DS, Warnke R, et al: Dural marginal zone lymphoma with massive amyloid deposition: Rare low-grade primary central nervous system B-cell lymphoma. J Neurosurg 96:368-372, 2002
Lima VS, Leite EB, Fonseca RP, et al: Primary low grade mucosa-associated B-cell lymphoma of the dura. Diagn Oncol 21:4058-4063, 2003
Menniti A, Moschettoni L, Liccardo G, et al: Low-grade primary meningeal lymphoma: Case report and review of the literature. Neurosurg Rev 28:229-233, 2005
Roonprapunt C, Retech H, LaSala P et al: Primary diffuse B-cell lymphoma arising in the dura in an immunocompetent individual: Case report. J Neuropathol Exp Neurol 62:574, 2003 (abstr 146)
Rottnek M, Strauchen J, Moore F, et al: Primary dural mucosa-associated lymphoid tissue-type lymphoma: Case report and review of the literature. J Neurooncol 68:19-23, 2004
Sanjeevi A, Krishnan J, Bailey PR, et al: Extranodal marginal zone B-cell lymphoma of malt type involving the cavernous sinus. Leuk Lymphoma 42:1133-1137, 2001
Jaffe ES, Harris NL, Stein H, et al: World Health Organization Classification of Tumours. Pathology & Genetics. Tumours of Haematopoietic and Lymphoid Tissues. Lyon, France, IARC Press, 2001
Maes B, de Wolf-Peeters C: Marginal zone cell lymphoma: An update on recent advances. Histopathology 40:117-126, 2002
Franco V, Florena AM, Lannitto E: Splenic marginal zone lymphoma. Blood 101:2464-2472, 2003
Arcaini L, Paulli M, Boveri E, et al: Splenic and nodal marginal zone lymphomas are indolent disorders at high hepatitis C virus seroprevalence with distinct presenting features but similar morphologic and phenotypic profiles. Cancer 100:107-115, 2004
Tam CS, Prince HM, Westerman D, et al: Leukemic subtype of marginal zone lymphoma: A presentation of three cases and literature review. Leuk Lymphoma 45:705-710, 2004
Martin F, Kearney JF: Marginal-zone B cells. Nat Rev Immunol 2:323-335, 2002
Dierlamm J, Wlodarska I, Michaux L, et al: Genetic abnormalities in marginal zone B-cell lymphoma. Hematol Oncol 18:1-13, 2000
Streubel B, Lamprecht A, Dierlamm J, et al: T(14;18)(q32;q21) involving IgH and MALT1 is a frequent chromosomal aberration in MALT lymphoma. Blood 101:2335-2339, 2003
Streubel B, Vinatzer U, Lamprecht A, et al: T(3;14)(p14.1;q32) involving IgH and FOXP1 is a novel recurrent chromosomal aberration in MALT lymphoma. Leukemia 19:652-658, 2005
Remstein ED, James CD, Kurtin PJ: Incidence and subtype specificity of API2-MALT1 fusion translocations in extranodal, nodal and splenic marginal zone lymphomas. Am J Pathol 156:1183-1188, 2000
Rosenwald A, Ott G, Stilgenbauer S, et al: Exclusive detection of the t(11;18)(q21;q21) in extranodal marginal zone B cell lymphomas (MZBL) of MALT type in contrast to other MZBL and extranodal large B cell lymphomas. Am J Pathol 155:1817-1821, 1999
Akagi T, Motegi M, Tamura A, et al:, A novel gene, MALT1 at 18q21, is involved in t(11;18)(q21;q21) found in low grade B-cell lymphoma of mucosa-associated lymphoid tissue. Oncogene 18:5785-5794, 1999
Dierlamm J, Baens M, Wiodarska I, et al: The apoptosis inhibitor gene API2 and a novel 18q gene, MLT, are recurrently rearranged in the t(11;18)(q21;q21) associated with mucosa-associated lymphoid tissue lymphoma. Blood 93:3601-3609, 1999
Ruefli-Brasse AA, French DM, Dixit VM: Regulation of NF-B-dependent lymphocyte activation and development by paracaspase. Science 302:1581-1584, 2003
Lachance DH, O'Neill BP, Macdonald DR, et al: Primary leptomeningeal lymphoma: Report of 9 cases, diagnosis with immunocytochemical analysis and review of the literature. Neurology 41:95-100, 1991
Schlegel U, Schmidt-Wolf IgH, Deckert M: Primary CNS lymphoma: Clinical presentation, pathological classification, molecular pathogenesis and treatment. J Neurol Sci 181:1-12, 2000
Rajaram V, Leuthardt EC, Singh PK, et al: 9p21 and 13q14 dosages in ependymomas: A clinicopathologic study of 101 cases. Mod Pathol 17:9-14, 2004
Dierlamm J, Pittaluga S, Stul M, et al: BCL6 gene rearrangements also occur in marginal zone B-cell lymphoma. Br J Haematol 98:719-725, 1997
Nakhleh RE, Manivel JC, Copenhaver CM, et al: In situ hybridization for the detection of Epstein-Barr virus in central nervous system lymphomas. Cancer 67:444-448, 1991
Kai Y, Kuratsu J, Ushio Y: Primary malignant lymphoma of the brain in childhood. Neurol Med Chir 38:232-237, 1998
Murphy JK, O'Brien CJ, Ironside JW: Morphologic and immunophenotypic characterization of primary brain lymphomas using paraffin-embedded tissue. Histopathology 15:449-460, 1989
Ellie E, Vergier B, Duche B, et al: Local amyloid deposits in a primary central nervous system lymphoma: Study of a stereotactic brain biopsy. Clin Neuropathol 9:231-233, 1990
Beriwal S, Hou JS, Miyamoto C, et al: Primary dural low grade BCL-2 negative follicular lymphoma: A case report. J Neurooncol 61:23-25, 2003
Subramanian S, Bahceci E, Kanji A: Primary follicular lymphoma of the meninges: A case report. Proc Am Soc Clin Oncol 14:153, 1995 (abstr 308)
Gotteri G, Ranaldi R, Pileri SA, et al: Localized amyloidosis and gastrointestinal lymphoma: A rare association. Histopathology 32:348-355, 1998
Weng W, Levy S: Hepatitis C virus (HCV) and lymphomagenesis. Leuk Lymphoma 44:1113-1120, 2003
Hermine O, Lefrere F, Bronowicki JP, et al: Regression of splenic lymphoma with villous lymphocytes after treatment of hepatitis C virus infection. N Engl J Med 347:89-94, 2002
Wotherspoon AC, Doglioni C, Diss TC, et al: Regression of primary low-grade B-cell gastric lymphoma of mucosa-associated lymphoid tissue type after eradication of Helicobacter pylori. Lancet 342:575-577, 1993
Hjalgrim H, Rasmussen S, Rostgaard K, et al: Familial clustering of Hodgkin lymphoma and multiple sclerosis. J Natl Cancer Inst 96:780-784, 2004
Vineis P, Crosignani P, Vigano C, et al: Lymphoma and multiple sclerosis in a multicenter case-control study. Epidemiology 12:134-135, 2001
Küpper R: B cells under influence: Transformation of B cells by Epstein-Barr virus. Nat Rev Immunol 3:801-812, 2003
Muller-Hermelink HK: Genetic and molecular genetic studies in the diagnosis of B-cell lymphomas: Marginal zone lymphomas. Hum Pathol 34:336-340, 2003
Ott G, Kalla J, Steinhoff A, et al: Trisomy 3 is not a common feature in malignant lymphomas of mucosa-associated lymphoid tissue type. Am J Pathol 153:689-694, 1998
Remstein ED, Kurtin PJ, James CD, et al: Mucosa-associated lymphoid tissue lymphomas with t(11;18)(q21;q21) and mucosa-associated lymphoid tissue lymphomas with aneuploidy develop along different pathogenetic pathways. Am J Pathol 161:63-71, 2002
Pasqualucci L, Bereschenko O, Niu H, et al: Molecular pathogenesis of non-Hodgkin's lymphoma: The role of Bcl-6. Leuk Lymphoma 44:S5-S12, 2003 (suppl 3)(Pang-hsien Tu, Caterina G)