Recurrent genomic imbalances in B-cell splenic marginal-zone lymphoma revealed by comparative genomic hybridization

Cancer Genetics and Cytogenetics 156 (2005) 122–128

Recurrent genomic imbalances in B-cell splenic marginal-zone lymphoma revealed by comparative genomic hybridization Claus L. Andersena,h, Alicja Gruszka-Westwoodb,i, Shayne Atkinsonb, Estella Matutesb, Daniel Catovskyb, Rikke K. Pedersenc, Bjarne B. Pedersend, Stanislaw Pulczynskie, Peter Hoklandf, Elisa Jacobseng, Jørn Kocha,j,* Department of Hematology, Laboratory of Cancer Cytogenetics, A˚rhus Sygehus, Tage Hansens Gade 2, DK-8000 A˚rhus C, Denmark b Academic Department of Haematology and Cytogenetics, Institute of Cancer Research/Royal Marsden NHS Trust, London, UK c Institute of Pathology, Chromosome Laboratory, Odense University Hospital, Odense C, Denmark d Medical Department, Viborg Hospital, Viborg, Denmark e Medical Department, Holstebro Centralsygehus, Holstebro, Denmark f Department of Hematology, Laboratory of Immunohematology, A˚rhus Sygehus, A˚rhus C, Denmark g Department of Hematology, A˚rhus Sygehus, A˚rhus C, Denmark h Department of Clinical Biochemistry, Molecular Diagnostics Laboratory, Skejby Sygehus, A˚rhus C, Denmark i Leukaemia Research Fund Centre for Cell and Molecular Biology of Childhood Leukaemia, Institute of Cancer Research, London, UK j Institute of Pathology, A˚rhus Sygehus, A˚rhus C, Danmark a

Received 18 November 2003; received in revised form 27 April 2004; accepted 28 April 2004

Abstract

The cytogenetics of splenic marginal zone lymphoma (SMZL) is less well characterized than the cytogenetics of other non-Hodgkin B-cell lymphomas. The aim of this study was to address this issue by identifying characteristic copy number imbalances in SMZL, for which purpose we analyzed 20 SMZL cases by comparative genomic hybridization (CGH), adding chromosome banding and fluorescence in situ hybridization (FISH) in some cases. CGH identified copy number imbalances in 70% of the cases. Imbalances were recurrently observed for chromosomes 3 (20%), 6 (20%), 7 (25%), 12 (20%), and 14 (10%). The minimally involved regions of these chromosomes were gains of 3q25~qter and 12q13~q15, and loss of 6q23, 7q31, and 14q22~q24. A compilation of our data with data from 3 previous SMZL CGH studies revealed a significant heterogeneity between the studies. Eleven imbalances were recurrently observed in the compiled data set, as opposed to only 5 in our data set. The most frequently observed imbalances in the 73 SMZL cases of the compiled data set were gains of 3q (27%) and 12q (15%), and loss of 7q (18%). Our data suggest that SMZL constitute a genetically heterogeneous disease where gain of 3q25 and loss of 7q31 are the most likely imbalances to be involved in the pathogenesis of the disease. 쑖 2005 Elsevier Inc. All rights reserved.

1. Introduction In 1987 Melo et al. described a group of patients diagnosed with splenomegaly and moderate lymphocytosis. In all patients, abnormal circulating lymphocytes characterized by the presence of thin and unevenly distributed cytoplasmic villi were observed. The condition was termed “splenic lymphoma with circulating villous lymphocytes” (SLVL) [1]. Subsequently, the condition has been included as a part of the entity splenic marginal-zone lymphoma (SMZL) with * Corresponding author. Tel.: ⫹45-8949-3671; fax: ⫹45-8949-3690. E-mail address: [email protected] (J. Koch). 0165-4608/05/$ – see front matter 쑖 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.cancergencyto.2004.04.026

or without villous lymphocytes in the R.E.A.L. [2] and the WHO classifications [3]. B-cell malignancies besides SMZL where splenomegaly is a prominent feature include hairy cell leukemia (HCL) and its variant (HCL-V), some cases of chronic lymphocytic leukemia (CLL), and B-cell prolymphocytic leukemia (PLL). These disorders may also show morphologic and immunophenotypic similarities to each other and to SMZL, but both the prognosis and the optimal therapy vary significantly between diseases, so discrimination between them is essential. Unfortunately, an unequivocal diagnosis of SMZL is in many cases difficult because spleen histology is not available.

C.L. Andersen et al. / Cancer Genetics and Cytogenetics 156 (2005) 122–128

The cytogenetic characteristics associated with the diseases are expected to be helpful in the differential diagnosis. However, classical chromosome banding has not provided a uniform set of cytogenetic characteristics for SMZL with an early report indicating the most frequently involved chromosomes to be 2, 7, 11, and 17 [4], and later studies reporting the most frequently involved chromosomes to be 3, 7, 12 [5], 1, 3, 7, 8 [6], 3q, 14q32, and chromosome 17 [7]. The variation across studies may reflect that the limited proliferative capacity of the SMZL-cells generally were overcome by in vitro stimulation of B-cell division, possibly leading to sub-clone selection. The present study, therefore, aims to identify the most important cytogenetic abnormalities in SMZL without the need for B-cell stimulation, which is done by using comparative genomic hybridization (CGH), a dual-color hybridization procedure providing a global overview of copy number imbalances in the tumor genome in a single experiment [8] for this purpose. To identify the most frequent copy number imbalances of SMZL, we furthermore compared the results obtained from our 20 patients with the CGH-results of the 40 patients reported in the literature [9,10]. The pinpointed regions represent positions in the genome that may harbor genes of importance for the pathogenesis of SMZL and, therefore, may serve as starting points for subsequent searches for such genes.

2. Materials and methods 2.1. Patients Twenty patients were studied, 13 women and 7 men, with a median age of 66 years (range 42–89). The diagnosis of SMZL/SLVL was based on peripheral blood morphology, bone marrow histology, and immunophenotyping. All patients had splenomegaly, and for 5 of them spleen histology was available and compatible with SMZL. Patients with unknown spleen histology were diagnosed as SLVL, but are in the remainder of the paper termed SMZL to comply with the WHO classification. The median lymphocyte count in peripheral blood was 22.0 × 109/L (range 2.6–87.6 × 109/L), hemoglobin 11.7 g/dL (range 7.8–16.0 g/dL), and the median platelet count was 168 × 109/L (range 61–538 × 109/L). All patients but one (case 10) had presented villous lymphocytes in the peripheral blood. A mature B-cell immunophenotype with expression of CD22 and FMC7 and strong surface immunoglobulin expression was found in all patients tested. 2.2. Comparative genomic hybridization Tumor DNA was isolated from the spleen (4 cases) or peripheral blood (16 cases) according to standard methods. The fraction of B-cells in each sample was estimated by flow cytometry prior to DNA isolation, and in the cases where the B-cell fraction was too low for safe CGH analysis

123

(⬍ 50%), B-cell enrichment was performed, using immunomagnetic beads conjugated with CD20 antibody (MACS; Miltenyi Biotec, Gladbach, Germany). Reference DNA was obtained from peripheral blood lymphocytes of sex-matched healthy donors. CGH analysis was performed essentially according to Kallioniemi et al. [8]. Briefly, tumor DNA (test DNA) was labeled with biotin-16dUTP (Roche Molecular Biochemicals, Basel, Switzerland) and normal DNA (reference DNA) was labeled with digoxigenin-11-dUTP (Roche Molecular Biochemicals) by nick translation (Nick Translation Kit; Roche Molecular Biochemicals). Equal amounts (~300 ng each) of labeled tumor and normal DNA, and 10 µg of unlabeled human Cot-1 DNA (Invitrogen, Life Technologies, Carlsbad, CA) were cohybridized to slides with human metaphase chromosome spreads prepared from a healthy male. Tumor and normal DNA were detected by avidin-fluorescein isothiocyanate (Roche Molecular Biochemicals) and rhodamine-conjugated anti-digoxigenin (Roche Molecular Biochemicals), respectively. Only metaphases with high signal intensity and uniform hybridization, effective suppression of repetitive sequences, and low background were used for the CGH analysis. For each case 10 metaphases were analyzed using Quips software (Applied Imaging, Newcastle upon Tyne, UK). The quality of the reagents and the normal metaphase slides was validated in control CGH hybridizations with differentially labeled male and female DNA. To minimize false positive identifications control hybridizations with interchanged digoxigenin-dUTP and biotin-dUTP labels for normal and tumor DNA were performed for all cases. Chromosomal regions were considered gained or lost when the mean ratioprofile was above or below thresholds of 1.20 or 0.80, respectively. The centromeric regions, heterochromatic blocks of chromosomes 1, 9, and 16, the satellite regions of the acrocentric chromosomes, and the Y chromosome were excluded from evaluation because of the abundance of highly repetitive DNA sequences. 2.3. Cell cultures and G-banding Cells were seeded at 1–2 ×106 lymphocytes per ml of medium (RMPI 1640 with 25 mm HEPES buffer, 20% fetal calf serum (Harlan Seralabs, Loughborough, UK), glutamine (0.15 mg/mL), and antibiotics (penicillin and streptomycin). Stimulated [TPA (tumor promoting agent): phorbol 12-myristate 13-acetate; 20 mg/L; Sigma, St. Louis, MO) and pokeweed mitogen (Gibco BRL)] 3- and 5-day cultures and non-stimulated overnight cultures were set up. Prior to harvest, the cultures were incubated with colcemid at a final concentration of 1 µg/ml at 37⬚C for 1 hour. The cells were fixed and spread according to standard procedures. Slides for G-banding were aged at room temperature for 24 hours, and stained with Giemsa according to standard procedures. Chromosome analysis was done with the Cytovision software (Applied Imaging).

124

C.L. Andersen et al. / Cancer Genetics and Cytogenetics 156 (2005) 122–128

2.4. Fluorescence in situ hybridization Single and dual-color FISH were performed according to standard procedures with the following probes either alone or in combination: whole chromosome painting probes 2, 12, X (Cambio Ltd., Cambridge, UK) and 7 (Oncor Appligene, Illkirch, France), a centromere 7 probe (Cambio Ltd.), and a locus specific 7q31 probe [custom made from a bacterial artificial chromosome (BAC) clone (GS1-80F7; NCBI Human BAC Resource [11])]. Dual-color labeling was obtained by combining FITC labeled probes and biotin labeled probes, where the biotin label was detected with avidin conjugated Texas Red. Hybridized slides were mounted in antifade (Vectashield; Vector Laboratories, Burlingame, CA) with counter-stain (DAPI). Image acquisition was performed using Smart Capture software (Digital Scientific, Cambridge, UK) connected to an epifluorescence microscope (Zeiss, Oberkochen, Germany) equipped with a CCD-camera (Photometrics SenSys; Roper Scientific, Trenton, NJ).

3. Results 3.1. Comparative genomic hybridization The CGH data are summarized in Fig. 1. DNA copy number changes were observed in 14 of the 20 cases analyzed (70%). A total of 30 different DNA copy number changes were detected, 13 gains and 17 losses. Most cases showed more than one chromosomal imbalance. Importantly, several of the identified copy number changes were observed in multiple patients. The most frequently recurring copy number imbalances covered regions of chromosomes 3, 6, 7, and 12. Partial loss of 14q was seen in 2 cases. Imbalances of 6q and 14q have not previously been reported associated with SMZL. By aligning the copy number changes observed for all patients, the minimally involved chromosome regions were delineated (Fig. 1). The partial loss of 14q was delineated to 14q22~q24. For the most frequent genetic imbalances the regions delineated were gain of 3q25~qter (observed in 4 of 20 cases, 20%), loss of 6q23 (observed in 4 of 20 cases, 20%), loss of 7q31 (observed in 5 of 20 cases, 25%), and gain of 12q13~q15 (observed in 4 of 20 cases, 20%). 3.2. Compilation of CGH results from 3 studies Three CGH studies on SMZL were recently published [9,10,12]. We generated a large SMZL CGH data set by compiling the data from these studies and from our own study (Fig. 2). This new data set covers a total of 73 SMZL patients. While our data showed roughly equal proportions of gains and losses, the 3 other CGH studies predominantly showed gains (Fig. 2). The compiled data set contains a total of 213 DNA copy number changes, covering 154 gains and 59 losses. Copy number imbalances were observed for all chromosomes except the Y chromosome.

Eleven imbalances were observed in ⭓10% of the patients. The 4 most frequently observed imbalances involved gains of 3q (27%) and 12q (15%), and loss of 7q (18%). Alignment of the observed imbalances enabled the delineation of the following commonly involved regions for the 11 most frequently observed imbalances: gains of 1q31, 3q25, 5q15, 12q15, 12q21, 18q12~qter, 20q12~qter, and Xq21, and losses of 6q23~q24, 7q31, and 17p11~p12. 3.3. Chromosome banding The cases showing copy number imbalances by CGH were also investigated by chromosome banding, when specimens were available (8 cases; 5 displayed an abnormal and 3 a normal karyotype at the banding analysis). In support of our assumption of clonal selection as a result of the stimulation needed to produce metaphase chromosomes, we found that only four of the karyotypes correlated with the findings by CGH. Three of the 4 correlating karyotypes contained chromosome 7 translocations. The chromosome 7 breakpoints in the translocations were located in the q-arm, albeit in 4 different bands (q21, q22, q31, and q36), and no recurrent partner was observed. Of the 2 cases showing partial loss of 14q by CGH, only 1 showed the expected 14q deletion by chromosome banding, the other showed a normal karyotype. 3.4. Fluorescence in situ hybridization Four of the cases showing abnormal chromosomes 7 by CGH and/or chromosome banding were further investigated by FISH analysis. In all 4 cases, FISH analysis identified abnormalities of chromosome 7. In cases 5 and 14, a single malignant clone was identified, which in both instances contained an unbalanced chromosome 7 translocation, a t(7;12) in case 5 (Figs. 3A and B), and a t(X;7) in case 14 (Figs. 3C and D). In the third case (case 10), 2 clones were identified, the most prevalent involving a balanced t(2;7) (Fig. 3E). In the other clone, an unbalanced translocation, add(7)(q36), was observed in addition to the balanced t(2;7) (Fig. 3F). In case 16, whole chromosome paint 7 showed no recognizable rearrangement, but 1 chromosome 7 homologue appeared slightly shorter than the other (data not shown). Interphase FISH with a locus-specific probe for 7q31 (BAC GS1-80F7) showed deletion of 7q31 in 87% of the cells, and thus confirmed the loss identified by CGH.

4. Discussion In the present study, we have identified by CGH chromosomal imbalances in 70% of SMZL patients. This frequency is comparable to the previously reported prevalence of approximately 80% for SMZL [9,10,12]. Importantly, the imbalances we observed cluster around a few chromosome regions (Fig. 1): gains of parts of chromosomes 3 and 12,

C.L. Andersen et al. / Cancer Genetics and Cytogenetics 156 (2005) 122–128

1

19 15 14 4

6

13

19

2

15 15 19

16 14 11 5 4

7

7 6

14

3

10

8

11 14 15 17

19

9

15

20

19 5

10

16

21

22

10

125

4

5

11

12

5 10 12 19

1

18

1

18

X

17

Y

Fig. 1. Summary of genomic imbalances in 20 patients with SMZL. Gains are shown on the right side of the chromosomal ideogram and losses on the left side. The number at the bottom of each line is the case number. Green bars for gains and red bars for losses mark the minimally involved chromosome regions.

and losses of parts of chromosomes 6, 7, and 14. Imbalances of 6q and 14q have not previously been reported associated with SMZL. We delineated the minimally involved regions of the recurrently involved chromosomes to gains of 3q25~qter and 12q13~q15, and loss of 6q23, 7q31, and 14q22~q24 (Fig. 1). To investigate how our results compare with previous SMZL CGH-results, we performed a literature search and found 3 studies [9,10,12] covering a total of 53 SMZLpatients. A compilation of the results from all 4 studies revealed substantial discrepancies in the reported findings (summarized in Fig. 2). In contrast to our study the 3 previous studies identified a much higher number of gains than of losses. The compiled data also revealed that the imbalances identified in the various studies were distributed differently (Fig. 2). Of the 11 most frequently occurring imbalances only 4 were observed in all 3 studies. There are several possible explanations for the discrepancies, such as technical problems in relation to the CGHanalysis (false negative and positive identifications), misdiagnosis of patients, and general genetic heterogeneity within

the disease entity SMZL. Boonstra et al. [12] used imbalance thresholds of 1.15 and 0.85—making it likely that they have a higher proportion of false positives than the other studies, which used either 1.20 and 0.80 (this study and Dierlamm et al. [9]) or 1.25 and 0.75 (Hernandez et al. [10]). The 2 most frequently occurring imbalances of the compiled dataset were gain of 3q (27%) and loss of 7q (18%). These were also the 2 most frequently observed imbalances in our own data set. The high number of patients in the compiled data set enabled us to narrow down the minimal regions of involvement for these 2 imbalances to gain of 3q25 and loss of 7q31. The frequency of copy number imbalances of chromosomes 3 and 12 in SMZL has previously been investigated by interphase FISH [6,13,14]. The reported frequencies of 17–22% and 0–5% of cases are significantly lower than the 27% and 15%, respectively, we found in the compiled data set. However, since the FISH studies were mostly done with centromeric probes, which are located far from the minimally involved regions, this discrepancy merely illustrates that the optimal design of FISH probes for evaluation of copy

126

C.L. Andersen et al. / Cancer Genetics and Cytogenetics 156 (2005) 122–128

1

2

3

6

7

8

13

14

15

19

20

9

21

4

5

10

11

12

16

17

18

22

X

Y

Fig. 2. Summary of genomic imbalances from the 73 patients of the compiled dataset. Imbalances identified in this study are presented as black lines, whereas imbalances identified by Dierlamm et al. 1997 [9] as dashed lines, Hernandez et al. 2001 [10] as grey lines, and Boonstra et al. 2003 [12] as dash dot lines. Gains are on the right side of the chromosomal ideogram and losses on the left side. Green bars for gains and red bars for losses mark the minimally involved chromosome regions.

number imbalances requires that the minimal genomic extent of the imbalance has already been mapped, e.g., by CGH. The BCL6 gene located at 3q27 has been found rearranged in 9% of marginal zone B-cell lymphoma (MZBCL) cases [15], and with the frequent gains of 3q in this disease, this has led to the suggestion that deregulation of BCL6 might play a role in the pathogenesis of MZBCL [15]. The minimally gained 3q region of SMZL, delineated in this study, does not include 3q27, indicating that BCL6 deregulation is less likely to play an important role in SMZL. Previous studies using chromosome banding, interphase FISH, and LOH have shown that partial deletion of the long arm of chromosome 7 is a recurrent abnormality in SMZL [6,14,16,17]. Our CGH results confirm that partial loss of 7q is among the most frequent imbalances of SMZL and furthermore maps the minimally lost region to 7q31. A potentially pathogenic deregulation of the CDK6 gene was recently found in a small group of SMZL patients with translocations involving band 7q21 [18]. Our chromosome banding and FISH investigations also revealed several cases with 7q translocations. However, there was no consistency in neither the chromosome 7 breakpoint, nor in the choice

of translocation partner, and only one of these involved 7q21 (case 10, which also showed deregulated CDK6 expression, as reported elsewhere [19]). Instead, with one exception, a common feature of all the observed translocations was that they were unbalanced, and thus associated with loss of 7q material. In line with this it has been reported that partial losses of 7q occur more frequently than 7q-translocations [14]. Together, these data indicate that even though 7q translocations are observed recurrently in SMZL, it is unlikely that the formation of a fusion gene plays a pivotal role in the disease. Instead, we hypothesize that a common pathogenic deregulation of one or more genes is reached through the observed abnormalities, allelic losses as well as genomic rearrangements. While CDK6 might be one of these genes, the CGH data suggest that the more important genes may be located in the vicinity of band 7q31. It has recently been suggested that cases with loss of 7q material and cases with gain of 3q material make up 2 distinct cytogenetic subtypes of SMZL [14]. Our data is not supportive of such subtypes, as we find 10% (n ⫽ 2) of our cases presenting with both abnormalities.

C.L. Andersen et al. / Cancer Genetics and Cytogenetics 156 (2005) 122–128

A

B

127

der(7)t(7;12) 12

7 der(7)t(7;?) 12

wcp 7 cen 7

C

wcp 12 cen 7

7

D

der(7)t(X;7) X

7

mar X

7

der(7)t(7;?)

der(?)t(X;?)

wcp 7

wcp X cen 7

E

der(7)t(7;?)

F der(7)t(2;7)

2 7 der(2)t(2;7) der(7)t(2;7)

der(2)t(2;7)

wcp 2 wcp 7

wcp 7

Fig. 3. Chromosome 7 rearrangement status of 3 cases evaluated by FISH. Case 5 (A, B); case 14 (C, D); and case 10 (E, F). (A) Identification of an unbalanced translocation involving chromosome 7 by hybridization with a whole chromosome painting (WCP) 7 (green) and a centromere 7 (red) probes. (B) Identification of the translocation partner der(7)(7;12) by hybridization with a WCP 12 (green) and centromere 7 (red) probes. (C) Identification of an unbalanced translocation involving chromosome 7 by hybridization with WCP 7. (D) Identification of the translocation partner der(7)(X;7) by hybridization with a WCP X (green) and centromere 7 (red) probes. (E) Identification of a clone with a balanced t(2;7) as the only chromosome 7 rearrangement by hybridization with WCP 2 (green) and 7 (red). (F) Identification of a second clone that in addition to the balanced t(2;7) contains an unbalanced t(7;?) by hybridization with WCP 7.

In summary, the compiled data set revealed significant heterogeneity among the analyzed SMZL cases. The cause of this heterogeneity remains unclear. Analysis of the compiled data revealed gain of 3q25 and loss of 7q31 as the most the

frequently occurring imbalances. These 2 imbalances pinpoint regions of the genome that are likely to harbor genes involved in the pathogenesis of SMZL. Identification of these genes may not only provide important information

128

C.L. Andersen et al. / Cancer Genetics and Cytogenetics 156 (2005) 122–128

on the disease etiology and pathogenesis, but may also provide new targets for therapeutical intervention. The imbalances are furthermore genetic markers, which potentially have both prognostic and diagnostic value. Our investigations moreover revealed that even though 7q translocations occur recurrently in SMZL, the formation of a 7q fusion gene is not likely to play a pivotal role in the disease.

Acknowledgments The authors thank Dr. Bendt Nielsen for his review of the manuscript. This work was supported by the Danish ˚ rhus University Hospital Research IniResearch Agency, A tiative, the Danish Cancer Society (grant no. 98 216 54), and the John and Birthe Meyer Foundation. The cell bank at the Academic Department of Haematology and Cytogenetics is supported by a grant from the Kay Kendal Leukaemia Fund.

References [1] Melo JV, Hegde U, Parreira A, Thompson I, Lampert IA, Catovsky D. Splenic B cell lymphoma with circulating villous lymphocytes: differential diagnosis of B cell leukaemias with large spleens. J Clin Pathol 1987;40:642–51. [2] Harris NL, Jaffe ES, Stein H, Banks PM, Chan JK, Cleary ML, Delsol G, De Wolf-Peeters C, Falini B, Gatter KC. A revised European-American classification of lymphoid neoplasms: a proposal from the international lymphoma study group. Blood 1994;84:1361–92. [3] Jaffe ES, Harris NL, Stein H, Banks PM, Vardiman J. Pathology and genetics of tumours of haematopoietic and lymphoid tissues. In: Kleihues P, Sobin LH, editors. World health organization classification of tumours 1st ed. Lyon: International Agency for Research on Cancer Press, 2001; pp. 135–7. [4] Oscier DG, Matutes E, Gardiner A, Glide S, Mould S, Brito-Babapulle V, Ellis J, Catovsky D. Cytogenetic studies in splenic lymphoma with villous lymphocytes. Br J Haematol 1993;85:487–91. [5] Brynes RK, Almaguer PD, Leathery KE, McCourty A, Arber DA, Medeiros LJ, Nathwani BN. Numerical cytogenetic abnormalities of chromosomes 3, 7, and 12 in marginal zone B-cell lymphomas. Mod Pathol 1996;9:995–1000. [6] Sole F, Woessner S, Florensa L, Espinet B, Mollejo M, Martin P, Piris MA. Frequent involvement of chromosomes 1, 3, 7 and 8 in splenic marginal zone B-cell lymphoma. Br J Haematol 1997;98:446–9. [7] Troussard X, Mauvieux L, Radford-Weiss I, Rack K, Valensi F, Garand R, Vekemans M, Flandrin G, Macintyre EA. Genetic analysis of splenic lymphoma with villous lymphocytes: a Groupe Francais d’Hematologie Cellulaire (GFHC) study. Br J Haematol 1998;101: 712–21. [8] Kallioniemi A, Kallioniemi OP, Sudar D, Rutovitz D, Gray JW, Waldman F, Pinkel D. Comparative genomic hybridization for molecular cytogenetic analysis of solid tumors. Science 1992;258:818–21.

[9] Dierlamm J, Rosenberg C, Stul M, Pittaluga S, Wlodarska I, Michaux L, Dehaen M, Verhoef G, Thomas J, de Kelver W, Bakker-Schut T, Cassiman JJ, Raap AK, De Wolf-Peeters C, Van den Berghe H, Hagemeijer A. Characteristic pattern of chromosomal gains and losses in marginal zone B cell lymphoma detected by comparative genomic hybridization. Leukemia 1997;11:747–58. [10] Hernandez JM, Garcia JL, Gutierrez NC, Mollejo M, MartinezCliment JA, Flores T, Gonzalez MB, Piris MA, San Miguel JF. Novel genomic imbalances in B-cell splenic marginal zone lymphomas revealed by comparative genomic hybridization and cytogenetics. Am J Pathol 2001;158:1843–50. [11] Cheung VG, Nowak N, Jang W, Kirsch IR, Zhao S, Chen XN, Furey TS, Kim UJ, Kuo WL, Olivier M, Conroy J, Kasprzyk A, Massa H, Yonescu R, Sait S, Thoreen C, Snijders A, Lemyre E, Bailey JA, Bruzel A, Burrill WD, Clegg SM, Collins S, Dhami P, Friedman C, Han CS, Herrick S, Lee J, Ligon AH, Lowry S, Morley M, Narasimhan S, Osoegawa K, Peng Z, Plajzer-Frick I, Quade BJ, Scott D, Sirotkin K, Thorpe AA, Gray JW, Hudson J, Pinkel D, Ried T, Rowen L, Shen-Ong GL, Strausberg RL, Birney E, Callen DF, Cheng JF, Cox DR, Doggett NA, Carter NP, Eichler EE, Haussler D, Korenberg JR, Morton CC, Albertson D, Schuler G, de Jong PJ, Trask BJ. Integration of cytogenetic landmarks into the draft sequence of the human genome. Nature 2001;409:953–8. [12] Boonstra R, Bosga-Bouwer A, van Imhoff GW, Krause V, Palmer M, Coupland RW, Dabbagh L, van den Berg E, van den Berg A, Poppema S. Splenic marginal zone lymphomas presenting with splenomegaly and typical immunophenotype are characterized by allelic loss in 7q31-32. Mod Pathol 2003;16:1210–7. [13] Gruszka-Westwood AM, Matutes E, Coignet LJ, Wotherspoon A, Catovsky D. The incidence of trisomy 3 in splenic lymphoma with villous lymphocytes: a study by FISH. Br J Haematol 1999;104:600–4. [14] Sole F, Salido M, Espinet B, Garcia JL, Martinez Climent JA, Granada I, Hernandez JM, Benet I, Piris MA, Mollejo M, Martinez P, Vallespi T, Domingo A, Serrano S, Woessner S, Florensa L. Splenic marginal zone B-cell lymphomas: two cytogenetic subtypes, one with gain of 3q and the other with loss of 7q. Haematologica 2001;86:71–7. [15] Dierlamm J, Pittaluga S, Stul M, Wlodarska I, Michaux L, Thomas J, Verhoef G, Verhest A, Depardieu C, Cassiman JJ, Hagemeijer A, De Wolf-Peeters C, Van den Berghe H. BCL6 gene rearrangements also occur in marginal zone B-cell lymphoma. Br J Haematol 1997;98: 719–25. [16] Oscier DG, Gardiner A, Mould S. Structural abnormalities of chromosome 7q in chronic lymphoproliferative disorders. Cancer Genet Cytogenet 1996;92:24–7. [17] Mateo M, Mollejo M, Villuendas R, Algara P, Sanchez-Beato M, Martinez P, Piris MA. 7q31-32 allelic loss is a frequent finding in splenic marginal zone lymphoma. Am J Pathol 1999;154:1583–9. [18] Corcoran MM, Mould SJ, Orchard JA, Ibbotson RE, Chapman RM, Boright AP, Platt C, Tsui LC, Scherer SW, Oscier DG. Dysregulation of cyclin dependent kinase 6 expression in splenic marginal zone lymphoma through chromosome 7q translocations. Oncogene 1999; 18:6271–7. [19] Brito-Babapulle V, Gruszka-Westwood AM, Platt G, Andersen CL, Elnenaei MO, Matutes E, Wotherspoon AC, Weston-Smith SG, Catovsky D. Translocation t(2;7)(p12;q21-22) with dysregulation of the CDK6 gene mapping to 7q21-22 in a non-Hodgkin’s lymphoma with leukemia. Haematologica 2002;87:357–62.