Cancer Genetics 207 (2014) 12e18
Lymph node hyperplasia: clonal chromosomal and genomic rearrangements. Report of two new cases and literature review Nicoletta Villa a,*, Serena Redaelli b, Sara Lissoni a, Fabiana Saccheri a, a,b, Valentina Carrino c, Elena Sala a, Francesca Crosti a, Leda Dalpra Pietro Pioltelli d, Giuseppe Isimbaldi e b a
Medical Genetics Lab, San Gerardo Hospital, Monza, Italy; Department of Surgery and Interdisciplinary Medicine, University of Milano-Bicocca, Milan, Italy; c Tettamanti Research Center, Pediatric Clinic, University of Milano-Bicocca, Milan, Italy; d Department of Haematology, San Gerardo Hospital, Monza, Italy; e Department of Surgical Pathology, San Gerardo Hospital, Monza, Italy Cytogenetic analysis is not routinely performed on lymph node hyperplasia (LH). We describe clonal chromosomal rearrangements in two unrelated cases of LH. Lymph nodes of both patients showed typical morphologic features of benign follicular hyperplasia. Cytogenetic analysis revealed clonal chromosomal rearrangements in both cases. Patient 1 showed interstitial 14q and 6q mosaic deletions, whereas patient 2 showed a terminal 14q mosaic deletion. Fluorescence in situ hybridization with IGH break-apart probes identified a partial deletion of IGH in both cases, but the loss of the LSI IGH in patient 2 and loss of the LSI IGHV in patient 1 were observed on the morphologically normal chromosome 14. In the latter case, the finding of two morphologically normal chromosomes 14 with the IGHV deletion in one of the chromosomes suggested that the first mutational event was the IGH deletion and the second event was the interstitial deletion of chromosome 14 with the IGH intact. Array comparative genomic hybridization performed on both biopsies confirmed the IGH deletion at mosaic, but not the chromosomal deletion. Patient 1 was re-biopsied after 9 months and a marginal zone lymphoma was diagnosed. The finding of clonal cytogenetic abnormalities in LH highlighted the difficulties in interpretation of results and clinical follow-up. Keywords Array CGH, clonal chromosomal rearrangements, immunoglobulin heavy chain, lymph node hyperplasia, fluorescence in situ hybridization ª 2014 Elsevier Inc. All rights reserved.
Lymphoproliferative disorders are characterized by specific chromosomal rearrangements that could have diagnostic and prognostic significance and are an important contribution to the morphological and clinical evaluations. Non-Hodgkin lymphomas show a large number of recurrent chromosomal rearrangements: t(14;18) in follicular lymphoma, t(8;14) in Burkitt lymphoma, t(11;14) in mantle cell lymphoma, and t(2;5) in anaplastic large cell lymphoma. Deletion of the long arm of chromosome 6 is observed
Received June 27, 2013; received in revised form November 26, 2013; accepted January 9, 2014. * Corresponding author. E-mail address: [email protected] 2210-7762/$ - see front matter ª 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.cancergen.2014.01.003
in hematological malignancies, but is more frequently described in lymphoproliferative disorders. Rarely reported as the only cytogenetic abnormality, the 6q deletion is usually associated with other cytogenetic abnormalities, suggesting a role in disease progression. Deletion of band 6q21q23 is the most recurrent cytogenetic abnormality in B cell lymphoma. Taborelli et al. (1) found 6q deletions by fluorescence in situ hybridization (FISH) in 94.2% of malignant lymphomas (Hodgkin and non-Hodgkin) and suggested a correlation between chromosome band deletion and lymphoma subtypes, as previously reported by Offit et al. (2). Thelander et al. (3) confirmed the 6q deletion in lymphoma and restricted the common deleted region in 6q21 to 3 Mb, where three candidate genes are mapped (FOXO3A, PRDM1, and HACE1). On the contrary, lymph node hyperplasia (LH) is
Clonal genomic rearrangements in LH
13
a benign lymphadenopathy of unexplained origin without morphological or immunophenotypic characteristics of malignancy. To our knowledge, in the last 10 years, only 3 reports (261 total cases) in the literature focused on the cytogenetic analysis of LH, but without a complete molecular characterization. Here we describe two cases with histological diagnoses of reactive LH and clonal chromosomal rearrangements. The findings of clonal chromosomal rearrangements in LH, which have been linked to subsequent development of malignancy, should prompt a thorough morphologic, phenotypic, and clinical examination as well as a careful follow-up observation.
Case reports Patient 1 was a 64-year-old woman with an unremarkable previous personal history. After a persistent lymphoadenomegaly, she underwent a first biopsy of a left supraclavicular lymph node that was 2 cm in diameter. A diagnosis of reactive LH was rendered. Nine months later, a second biopsy on persistent lymph nodes was performed at the same site. At this time, a diagnosis of marginal zone lymphoma (MZL) was made on the basis of morphological and immunophenotypical results, which was confirmed by a clonal rearrangement of the immunoglobulin heavy chain gene by PCR and FISH analyses. Based on the results of histopathological and genetic reports, the patient underwent chemotherapy (6 cycles of the CHOP-cyclophosphamide, doxorubicin, vincristine, and prednisone-scheme) and has been in complete remission for 4 years. Patient 2 was a 55-year-old man who developed a follicular lymphoma when he was 39. During a programmed checkup, a 1.5-cm laterocervical lymph node was removed. After 5 years, the patient is alive and well. In both cases, lymph nodes showed typical morphologic and phenotypic features of benign follicular hyperplasia (Figure 1A). The germinal centers (GCs) were polymorphous in shape, with a follicular appearance, expansion of the cortical zone, and an attenuated but preserved mantle zone. The interfollicular zone was well evident. In particular, in patient 1, the paracortex displayed a moderate proliferation of the high endothelial venules. Peripheral sinuses were patent. At high power, tingible body that contained macrophages were easily identified in the GC. The immunophenotypic study confirmed the topographic preservation of the lymph node cytoarchitecture. The GCs demonstrated a reactive immunoprofile: CD10/ Bcl6þ; Bcl2 negative (Figure 1B,C). A homogeneous high proliferation rate was evident with the Ki67 antibody (data not shown). Occasional CD30-positive immunoblastic-like cells were noted mostly in perigerminal areas and, more rarely, inside the GC (Figure 1D).
Materials and methods Cytogenetic analysis Lymph node biopsies were mechanically disaggregated for cytogenetic analyses. Representative parts of the biopsies were fixed in buffered formalin and paraffin, and were
Figure 1 (A) Lymph node histology after H&E stain (100x, patient 1 case); (B) immunohistochemical assay with BCL2 negative in germinal centers (100x, patient 1 case); (C) immunohistochemical assay with BCL2 antibody negative in germinal centers (100x, patient 2 case); (D) immunohistochemical assay with CD30 antibody (100x, patient 2 case) showing occasional CD30 positive immunoblastic-like cells mostly perigerminal areas.
routinely processed. Slides were stained with H&E. Immunohistochemical analysis was performed by a panel of the following primary antibodies: CD20, CD3, CD5, CD23, CD43, CD30, BCL2, BCL6, CD10, cyclin D1, and Ki67. Following the protocol suggested by The AGT Cytogenetics Laboratory Manual (7), the cell suspensions were spun, and the pellets were washed with Hank’s balanced solution and cultured in RPMI1640 medium supplemented with 20% fetal calf serum. The cell suspensions were subsequently processed after a 24-hour culture (37 C) and 30 minutes of colcemid (N-Methyl-N-deacetyl-cholchicine, Roche Diagnostics). The karyotypes were described following the International System of Cytogenetic Nomenclature 2013 (ISCN 2013).
PCR analysis PCR was performed for the immunoglobulin heavy chain (IGH ) and T cell receptor rearrangement analyses (8,9).
FISH analysis FISH experiments were performed according to the manufacturer’s protocol. FISH was performed on the cell suspension and/or on paraffin embedded tissue. The probes used were: 14q telomere (Cytocell Ltd., Cambridge, UK) (locus D14S1420, 200 kb from the telomere); LSI IGH dualcolor break-apart (14q32.1) (LSI IGHV 900-kb spectrum
14 green probe covering the entire IGH variable region at the 50 of the J locus or switch regions of the immunoglobulin heavy chain and LSI IGH spectrum orange 250-kb probe at the 30 region) (Abbott Molecular, Des Plaines, IL); LSI MYC/IGH CEP8 (8q24;14q32) (Abbott Molecular); LSI AP12/MALT1 (11q21;18q21) (Abbott Molecular); and LSI IGH/BCL2 (14q32;18q21) (Abbott Molecular).
Array comparative genomic hybridization (aCGH) analysis Genomic copy number analysis was performed with aCGH using the SurePrint G3 Human Genome CGHþSNP Microarray Kit, 4 180 K (Agilent Technologies, Santa Clara, CA), following the manufacturer’s recommendations. The microarray contained 110,712 CGH probes and 59,647 single nucleotide polymorphism (SNP) probes with a 25.3-kb overall median probe spacing (5 kb in International Standards for Cytogenomics Array Consortium (ISCA) regions). The target DNA was extracted from the lymph node biopsies that had been stored in liquid nitrogen from both patients (the first biopsy of patient 2 was not available for investigation) using proteinase K digestion and phenol-choloroform purification. The extracted DNA was quantified using the Nanodrop 1000 Spectrophotometer (Thermo Fisher Scientific, Waltham, MA) and the quality was verified using agarose gel electrophoresis and the Nanodrop readings. The arrays were scanned at 3-m m resolution using an Agilent microarray scanner and analyzed using feature extraction v10.7 and DNA analytics v6.5 software (Agilent Technologies). The aberration detection method 2 (ADM-2) algorithm was used to compute and assist in the identification of aberrations for a given sample. Significant chromosomal aberrations were determined using the algorithm ADM-2 with a threshold of 5 and a minimum absolute average log2 ratio of 0.25. Putative chromosome copy number changes were defined by intervals of 3 or more adjacent probes and were considered as being duplicated or deleted when results exceeded 0.25. All nucleotide positions were based on the Human Reference Sequence Assembly, February 2009 GRCh37/hg19 of the UCSC Genome Browser (http://genome.ucsc.edu/).
Results Cytogenetics and FISH analyses In the 24-hour cell suspension culture of patient 1, a total of 24 cells were analyzed, 14 of which showed a normal female karyotype and 6 showed an interstitial 14q22q32.1 deletion (Figure 2A). Moreover, in 4 of these 24 cells, a 6q21 interstitial deletion was observed (Figure 2B); in 2 of these 24 cells, a del(14)(q22q32.1) was present (Table 1). A specific 14q telomere probe hybridized to both chromosomes 14dthe deleted and the normal one (Figure 2C)dwhich indicated the presence of an interstitial deletion. FISH performed with the aim to define the lost 14q region (LSI IGH break-apart probes; Abbott Molecular) highlighted two normal signals (red signal of the LSI IGH 30 flanking region and green signal of the LSI IGHV 50 probe) on the
N. Villa et al. cytogenetically deleted chromosome as well as the presence of the red signal and loss of the green signal on the cytogenetically normal homologous chromosome 14 (Figure 2D). Few metaphases with two cytogenetically normal chromosomes 14 were detected, which indicated the deletion of the green signal on one chromosome 14 (Figure 2E). The same pattern of IGH hybridization signals was observed on 14.3% of the interphase large nuclei on paraffin embedded tissue (Figure 2F). Unfortunately, because of the poor preservation of tissue architecture during the FISH hybridization protocol, it was impossible to establish with certainty if these large nuclei were in the GCs. Moreover, in 100 nuclei of the cell suspension, which was screened for the t(8;14), a normal signal pattern was observed. After 9 months, a diagnosis of low-grade marginal zone B cell lymphoma was determined by a second lymph node biopsy. The 11;18 and 14;18 translocations were absent on the paraffin-embedded second lymph node that was biopsied after 9 months (202 nuclei analyzed by FISH), which was requested on the basis of the MZL histological diagnosis. Moreover, in order to detect rearrangements of chromosome 14, FISH for the t(8;14) and for the IGH locus (using breakapart probes) was set up. The last experiment highlighted the presence of 21 of 200 nuclei (10.5%) that showed one green signal and two red signals per cell which indicated a mosaic deletion of the 50 region of the IGH locus. The morphological aspect of these nuclei counterstained with DAPI suggested that the cells were interfollicular T-zone monocytoid cells. These findings were similar to the deletion detected in the first biopsy, which took place 9 months earlier. FISH for the t(8;14) showed a complex picture interpreted as trisomy/tetrasomy of chromosome 14 and MYC amplification. All results are summarized in Table 1. In a 24-hour cell culture of patient 2, only eight metaphases were detected: four with a normal male karyotype and four with a 14q32.1qter deletion (Figure 2G, Table 1). In metaphase cells, LSI IGH break-apart probes demonstrated one homologue with a normal pattern of hybridization as well as the presence of the green signal (LSI IGHV 50 probe) and the loss of the red signal on the other homologous chromosome 14 (Figure 2H). The same pattern of hybridization signals was also observed on large interphase nuclei (Figure 2I). Immunoglobulin heavy chain gene analysis by PCR amplification showed the absence of clonal products in patient 2 and in the first patient 1 biopsy, whereas the second lymph node biopsy showed clonal rearrangement. The FISH results demonstrating a partial deletion of the IGH locus deletion were confirmed by aCGH analysis: The loss of the 50 flanking region in patient 1 (583 kb, from nt 106636701 to nt 107219763 in 14q32.33) was found in 49% of cells, whereas the loss of the 30 region was found in 61% of cells for patient 2 (402 kb from nt 106371690 to nt 106774672 in 14q32.33) (Figure 3, Table 1). Cytogenetic chromosomal 14 and 6 deletions, found respectively in 33.3% and 16.6% of metaphases in patient 1, were not identified by aCGH. In addition, the terminal 14q deletion found in four of eight metaphases in patient 2 was not evidenced by aCGH. We did not consider copy number variations (CNVs) reported as “benign” in the Database of
Clonal genomic rearrangements in LH
15
Figure 2 Patient 1: (A) Chromosome 14q22q32.1 partial deletion; (B) chromosome 6q21 partial deletion; (C) FISH with specific 14q telomeric probe; the arrow indicates the deleted chromosome 14; (D) FISH with IGH break-apart probes; the white arrow shows cytogenetically deleted chromosome 14, whereas the red arrow indicates chromosome 14 with partial deletion of the IGH probe (green signal); (E) cytogenetically normal chromosome 14 with a partial FISH deletion of the IGH probe (arrow) (green signal); (F) tissue FISH shows the IGH probe deletion on a nucleus (arrow). Patient 2: (G) Chromosome 14q32.1qter partial deletion; (H) FISH with IGH breakapart probes on metaphase; (I) interphase nuclei; the arrow shows the red signal loss.
Genomic Variants (http://projects.tcag.ca/variation/), and the remaining CNVs identified in both samples are reported in Supplementary Table 1S. Briefly, the other 17 different CNVs were seen in patient 1, with a size range from 234 nt424 kb, whereas 46 CNVs were seen in patient 2, with a size range from 831 nt797 kb (Supplementary Table 1).
Discussion The finding of clonal cytogenetic abnormalities in cases of LH raises difficulties with results interpretation and clinical management. We cannot know the fate of abnormal cells in vivo. They could gain a proliferative advantage, go toward apoptotic pathways, or remain dormant. In fact, cells with the typical chromosomal translocation t(14;18)(q32;q21), which is associated with follicular lymphoma, were described in normal tonsils or peripheral blood lymphocytes without progression to
follicular lymphoma (10). As a consequence, it is likely that more than one genetic event is needed to determine the evolution toward lymphoma (11). Somatic hypermutation is a physiological process occurring in B cell development and involving the IgH variable region in GC B cells. Pasqualucci et al. (12) describe an aberrant hyper-mutation activity that targets multiple loci, including proto-oncogenes in cases of diffuse large-cell lymphoma. Hypermutable genes are susceptible to chromosomal translocation, which generate DNA breaks and could contribute to lymphomagenesis. The detailed data from the cytogenetic analyses of LH described in the literature are reported in Table 2 with only a few of the reports demonstrating FISH and genomic characterization of the cytogenetic abnormalitites. Reports published before 2000 was not considered because of the lack of clinical follow-up. For a total of 261 cases analyzed, a high percentage of culture failure (2146%) suggests that molecular techniques such as FISH, aCGH, and SNP arrays together with
16 Table 1
N. Villa et al. Summary of histological and genetic results in patient 1 and patient 2 FISH results Histological diagnosis
Karyotype
14q Subtel probe 14 Break-apart
Patient 1 Lymph node first biopsy
Lymph node mos 46,XX,del(6) N hyperplasia (q21q21),del(14) (q22q32.1)[2]/ 46,XX,del(14) (q22q32.1)[6]/ 46,XX,del(6) (q21q21)[2]/ 46,XX[14] Lymph node Marginal zone nd nd second biopsy lymphoma
Patient 2 Lymph node biopsy
Lymph node mos 46,XY,del(14) nd hyperplasia (q32.1)[4]/ 46,XY[4]
Loss of IGHV green signal on normal chromosome 14 and in 14.3% of interphase nuclei
t(14;18) t(11;18) t(8;14)
aCGH
nd
583 kb 50 IGH locus deletion (49%)
nd
N
Loss of IGHV N green signal in 10.5% of interphase tissue nuclei
N
nd MYC amplification trisomy/ tetrasomy chromosome 14
Loss of IGH red nd signal on normal chromosome 14 and in interphase nuclei
nd
nd
402 kb 30 IGH locus deletion (61%)
Abbreviations: nd, not done; N, normal hybridization signal pattern.
conventional cytogenetics should be adopted. Cases with clonal chromosomal rearrangements were observed, which ranged from 2.233.3%, and 1 breakpoint fell in 14q32 in a total of 8 of 25 cases with clonal abnormalities (Table 2). Clinical follow-up of clonal patients reported by Sevilla et al. showed a “free of disease” condition in all cases (6). All 10 patients with clonal genetic abnormalities reported by Au et al. (4) had a diagnosis of malignant lymphoma after a second lymph node biopsy performed in a range of 0.3-37.1 months from the first one (Table 2). We had the opportunity to study two LH cases that showed clonal chromosomal deletions. The cytogenetic regions involved are different: The patient 1 deletion is larger and interstitial (14q22q32.1), and the patient 2 deletion is smaller and terminal (14q32.1qter). Interestingly, the partial IGH deletion, which was observed in both patients, is present in the cytogenetically normal chromosome 14. Furthermore, the finding of metaphases with two structurally normal chromosomes 14 and the IGH deletion (Figure 2E) suggests that the first hit could be the IGH deletion and the second hit the chromosomal deletion of the homologous chromosome. On the other hand, Gandhi et al. (13) recently demonstrated that homologous chromosomes make contact at the sites of double-strand breaks (DSBs) within gene sequences. DSBs induce a complex response involving various repair pathways: A misrepair mediated by ATM kinase activity and DNA-PK could generate a chromosomal rearrangement as observed in the chromosomes 14 in both patient 1 and patient 2. In both cases, the partial IGH deletion was also observed on interphase nuclei and only in the largest ones. Small
Figure 3 aCGH combined results of chromosome 14 and graphic representation of the IGH region with patient 1 and patient 2 deletions evidenced.
Clonal genomic rearrangements in LH Table 2
Literature reports on LH cases
Number Age (y), Reference of cases range (4) (5) (6) a b
17
30 84 147
Cases with 14q32 involvement/ Cases with total cases Cytogenetic clonal karyotype/ with clonal failure, % total cases (%) karyotype IGH FISH
2088 30 Not reported 46 282 21
10/30 (33.3) 1/45 (2.2) 14/116 (7)
4/10 0/1 4/14
d d 3/4
Follow-up: number of cases/clonal cases
Range (months)
Free of disease Malignancy
0.1116.9 2/10a d d 5168 13/14b
10/10 d d
Remission after lymphoma. One patient lost at follow-up.
nuclei presented the two normal fusion signals. We could suppose that the rearrangement only involves centroblasts and not small lymphocytes. The aCGH results were consistent with the FISH data and allow for the definition of breakpoints at the molecular level. In both cases, one breakpoint was in the immunoglobulin heavy chain coding region and the second was in the flanking region that contained RNA coding genes (KIAA0125, ADAM6, LOC100133469). We emphasize that patient 2 had developed a follicular lymphoma 16 years earlier, and the diagnosis was only histological because the tissue was not available for molecular investigation. Patient 2 is free of disease after 5 years from the LH diagnosis. The role of cytogenetic translocations in the pathogenesis of lymphomas leading to the activation of oncogene expression is well known. In our cases, no translocations were identified, but deletions of a region of 14q were observed. The partial 14q chromosomal deletion and IGH locus involvement are recurrently observed in B cell malignancies such as chronic lymphocytic leukemia (CLL), acute lymphoblastic leukemia (ALL), and multiple myeloma (MM) (Mitelman Database of Chromosome Aberrations in Cancer, http://cgap.nci.nih.gov/Chromosomes/Mitelman)(14). Discordance was observed when the results obtained by cytogenetics, FISH, and aCGH were compared in the two cases presented here: Partial IGH locus deletion identified by FISH was confirmed by aCGH, whereas the chromosomal 14q deletion was not detected by aCGH. This technique is strongly limited by its inability to detect losses or gains that occur in
Lymph node hyperplasia: clonal chromosomal and genomic rearrangements. Report of two new cases and literature review Nicoletta Villa a,*, Serena Redaelli b, Sara Lissoni a, Fabiana Saccheri a, a,b, Valentina Carrino c, Elena Sala a, Francesca Crosti a, Leda Dalpra Pietro Pioltelli d, Giuseppe Isimbaldi e b a
Medical Genetics Lab, San Gerardo Hospital, Monza, Italy; Department of Surgery and Interdisciplinary Medicine, University of Milano-Bicocca, Milan, Italy; c Tettamanti Research Center, Pediatric Clinic, University of Milano-Bicocca, Milan, Italy; d Department of Haematology, San Gerardo Hospital, Monza, Italy; e Department of Surgical Pathology, San Gerardo Hospital, Monza, Italy Cytogenetic analysis is not routinely performed on lymph node hyperplasia (LH). We describe clonal chromosomal rearrangements in two unrelated cases of LH. Lymph nodes of both patients showed typical morphologic features of benign follicular hyperplasia. Cytogenetic analysis revealed clonal chromosomal rearrangements in both cases. Patient 1 showed interstitial 14q and 6q mosaic deletions, whereas patient 2 showed a terminal 14q mosaic deletion. Fluorescence in situ hybridization with IGH break-apart probes identified a partial deletion of IGH in both cases, but the loss of the LSI IGH in patient 2 and loss of the LSI IGHV in patient 1 were observed on the morphologically normal chromosome 14. In the latter case, the finding of two morphologically normal chromosomes 14 with the IGHV deletion in one of the chromosomes suggested that the first mutational event was the IGH deletion and the second event was the interstitial deletion of chromosome 14 with the IGH intact. Array comparative genomic hybridization performed on both biopsies confirmed the IGH deletion at mosaic, but not the chromosomal deletion. Patient 1 was re-biopsied after 9 months and a marginal zone lymphoma was diagnosed. The finding of clonal cytogenetic abnormalities in LH highlighted the difficulties in interpretation of results and clinical follow-up. Keywords Array CGH, clonal chromosomal rearrangements, immunoglobulin heavy chain, lymph node hyperplasia, fluorescence in situ hybridization ª 2014 Elsevier Inc. All rights reserved.
Lymphoproliferative disorders are characterized by specific chromosomal rearrangements that could have diagnostic and prognostic significance and are an important contribution to the morphological and clinical evaluations. Non-Hodgkin lymphomas show a large number of recurrent chromosomal rearrangements: t(14;18) in follicular lymphoma, t(8;14) in Burkitt lymphoma, t(11;14) in mantle cell lymphoma, and t(2;5) in anaplastic large cell lymphoma. Deletion of the long arm of chromosome 6 is observed
Received June 27, 2013; received in revised form November 26, 2013; accepted January 9, 2014. * Corresponding author. E-mail address: [email protected] 2210-7762/$ - see front matter ª 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.cancergen.2014.01.003
in hematological malignancies, but is more frequently described in lymphoproliferative disorders. Rarely reported as the only cytogenetic abnormality, the 6q deletion is usually associated with other cytogenetic abnormalities, suggesting a role in disease progression. Deletion of band 6q21q23 is the most recurrent cytogenetic abnormality in B cell lymphoma. Taborelli et al. (1) found 6q deletions by fluorescence in situ hybridization (FISH) in 94.2% of malignant lymphomas (Hodgkin and non-Hodgkin) and suggested a correlation between chromosome band deletion and lymphoma subtypes, as previously reported by Offit et al. (2). Thelander et al. (3) confirmed the 6q deletion in lymphoma and restricted the common deleted region in 6q21 to 3 Mb, where three candidate genes are mapped (FOXO3A, PRDM1, and HACE1). On the contrary, lymph node hyperplasia (LH) is
Clonal genomic rearrangements in LH
13
a benign lymphadenopathy of unexplained origin without morphological or immunophenotypic characteristics of malignancy. To our knowledge, in the last 10 years, only 3 reports (261 total cases) in the literature focused on the cytogenetic analysis of LH, but without a complete molecular characterization. Here we describe two cases with histological diagnoses of reactive LH and clonal chromosomal rearrangements. The findings of clonal chromosomal rearrangements in LH, which have been linked to subsequent development of malignancy, should prompt a thorough morphologic, phenotypic, and clinical examination as well as a careful follow-up observation.
Case reports Patient 1 was a 64-year-old woman with an unremarkable previous personal history. After a persistent lymphoadenomegaly, she underwent a first biopsy of a left supraclavicular lymph node that was 2 cm in diameter. A diagnosis of reactive LH was rendered. Nine months later, a second biopsy on persistent lymph nodes was performed at the same site. At this time, a diagnosis of marginal zone lymphoma (MZL) was made on the basis of morphological and immunophenotypical results, which was confirmed by a clonal rearrangement of the immunoglobulin heavy chain gene by PCR and FISH analyses. Based on the results of histopathological and genetic reports, the patient underwent chemotherapy (6 cycles of the CHOP-cyclophosphamide, doxorubicin, vincristine, and prednisone-scheme) and has been in complete remission for 4 years. Patient 2 was a 55-year-old man who developed a follicular lymphoma when he was 39. During a programmed checkup, a 1.5-cm laterocervical lymph node was removed. After 5 years, the patient is alive and well. In both cases, lymph nodes showed typical morphologic and phenotypic features of benign follicular hyperplasia (Figure 1A). The germinal centers (GCs) were polymorphous in shape, with a follicular appearance, expansion of the cortical zone, and an attenuated but preserved mantle zone. The interfollicular zone was well evident. In particular, in patient 1, the paracortex displayed a moderate proliferation of the high endothelial venules. Peripheral sinuses were patent. At high power, tingible body that contained macrophages were easily identified in the GC. The immunophenotypic study confirmed the topographic preservation of the lymph node cytoarchitecture. The GCs demonstrated a reactive immunoprofile: CD10/ Bcl6þ; Bcl2 negative (Figure 1B,C). A homogeneous high proliferation rate was evident with the Ki67 antibody (data not shown). Occasional CD30-positive immunoblastic-like cells were noted mostly in perigerminal areas and, more rarely, inside the GC (Figure 1D).
Materials and methods Cytogenetic analysis Lymph node biopsies were mechanically disaggregated for cytogenetic analyses. Representative parts of the biopsies were fixed in buffered formalin and paraffin, and were
Figure 1 (A) Lymph node histology after H&E stain (100x, patient 1 case); (B) immunohistochemical assay with BCL2 negative in germinal centers (100x, patient 1 case); (C) immunohistochemical assay with BCL2 antibody negative in germinal centers (100x, patient 2 case); (D) immunohistochemical assay with CD30 antibody (100x, patient 2 case) showing occasional CD30 positive immunoblastic-like cells mostly perigerminal areas.
routinely processed. Slides were stained with H&E. Immunohistochemical analysis was performed by a panel of the following primary antibodies: CD20, CD3, CD5, CD23, CD43, CD30, BCL2, BCL6, CD10, cyclin D1, and Ki67. Following the protocol suggested by The AGT Cytogenetics Laboratory Manual (7), the cell suspensions were spun, and the pellets were washed with Hank’s balanced solution and cultured in RPMI1640 medium supplemented with 20% fetal calf serum. The cell suspensions were subsequently processed after a 24-hour culture (37 C) and 30 minutes of colcemid (N-Methyl-N-deacetyl-cholchicine, Roche Diagnostics). The karyotypes were described following the International System of Cytogenetic Nomenclature 2013 (ISCN 2013).
PCR analysis PCR was performed for the immunoglobulin heavy chain (IGH ) and T cell receptor rearrangement analyses (8,9).
FISH analysis FISH experiments were performed according to the manufacturer’s protocol. FISH was performed on the cell suspension and/or on paraffin embedded tissue. The probes used were: 14q telomere (Cytocell Ltd., Cambridge, UK) (locus D14S1420, 200 kb from the telomere); LSI IGH dualcolor break-apart (14q32.1) (LSI IGHV 900-kb spectrum
14 green probe covering the entire IGH variable region at the 50 of the J locus or switch regions of the immunoglobulin heavy chain and LSI IGH spectrum orange 250-kb probe at the 30 region) (Abbott Molecular, Des Plaines, IL); LSI MYC/IGH CEP8 (8q24;14q32) (Abbott Molecular); LSI AP12/MALT1 (11q21;18q21) (Abbott Molecular); and LSI IGH/BCL2 (14q32;18q21) (Abbott Molecular).
Array comparative genomic hybridization (aCGH) analysis Genomic copy number analysis was performed with aCGH using the SurePrint G3 Human Genome CGHþSNP Microarray Kit, 4 180 K (Agilent Technologies, Santa Clara, CA), following the manufacturer’s recommendations. The microarray contained 110,712 CGH probes and 59,647 single nucleotide polymorphism (SNP) probes with a 25.3-kb overall median probe spacing (5 kb in International Standards for Cytogenomics Array Consortium (ISCA) regions). The target DNA was extracted from the lymph node biopsies that had been stored in liquid nitrogen from both patients (the first biopsy of patient 2 was not available for investigation) using proteinase K digestion and phenol-choloroform purification. The extracted DNA was quantified using the Nanodrop 1000 Spectrophotometer (Thermo Fisher Scientific, Waltham, MA) and the quality was verified using agarose gel electrophoresis and the Nanodrop readings. The arrays were scanned at 3-m m resolution using an Agilent microarray scanner and analyzed using feature extraction v10.7 and DNA analytics v6.5 software (Agilent Technologies). The aberration detection method 2 (ADM-2) algorithm was used to compute and assist in the identification of aberrations for a given sample. Significant chromosomal aberrations were determined using the algorithm ADM-2 with a threshold of 5 and a minimum absolute average log2 ratio of 0.25. Putative chromosome copy number changes were defined by intervals of 3 or more adjacent probes and were considered as being duplicated or deleted when results exceeded 0.25. All nucleotide positions were based on the Human Reference Sequence Assembly, February 2009 GRCh37/hg19 of the UCSC Genome Browser (http://genome.ucsc.edu/).
Results Cytogenetics and FISH analyses In the 24-hour cell suspension culture of patient 1, a total of 24 cells were analyzed, 14 of which showed a normal female karyotype and 6 showed an interstitial 14q22q32.1 deletion (Figure 2A). Moreover, in 4 of these 24 cells, a 6q21 interstitial deletion was observed (Figure 2B); in 2 of these 24 cells, a del(14)(q22q32.1) was present (Table 1). A specific 14q telomere probe hybridized to both chromosomes 14dthe deleted and the normal one (Figure 2C)dwhich indicated the presence of an interstitial deletion. FISH performed with the aim to define the lost 14q region (LSI IGH break-apart probes; Abbott Molecular) highlighted two normal signals (red signal of the LSI IGH 30 flanking region and green signal of the LSI IGHV 50 probe) on the
N. Villa et al. cytogenetically deleted chromosome as well as the presence of the red signal and loss of the green signal on the cytogenetically normal homologous chromosome 14 (Figure 2D). Few metaphases with two cytogenetically normal chromosomes 14 were detected, which indicated the deletion of the green signal on one chromosome 14 (Figure 2E). The same pattern of IGH hybridization signals was observed on 14.3% of the interphase large nuclei on paraffin embedded tissue (Figure 2F). Unfortunately, because of the poor preservation of tissue architecture during the FISH hybridization protocol, it was impossible to establish with certainty if these large nuclei were in the GCs. Moreover, in 100 nuclei of the cell suspension, which was screened for the t(8;14), a normal signal pattern was observed. After 9 months, a diagnosis of low-grade marginal zone B cell lymphoma was determined by a second lymph node biopsy. The 11;18 and 14;18 translocations were absent on the paraffin-embedded second lymph node that was biopsied after 9 months (202 nuclei analyzed by FISH), which was requested on the basis of the MZL histological diagnosis. Moreover, in order to detect rearrangements of chromosome 14, FISH for the t(8;14) and for the IGH locus (using breakapart probes) was set up. The last experiment highlighted the presence of 21 of 200 nuclei (10.5%) that showed one green signal and two red signals per cell which indicated a mosaic deletion of the 50 region of the IGH locus. The morphological aspect of these nuclei counterstained with DAPI suggested that the cells were interfollicular T-zone monocytoid cells. These findings were similar to the deletion detected in the first biopsy, which took place 9 months earlier. FISH for the t(8;14) showed a complex picture interpreted as trisomy/tetrasomy of chromosome 14 and MYC amplification. All results are summarized in Table 1. In a 24-hour cell culture of patient 2, only eight metaphases were detected: four with a normal male karyotype and four with a 14q32.1qter deletion (Figure 2G, Table 1). In metaphase cells, LSI IGH break-apart probes demonstrated one homologue with a normal pattern of hybridization as well as the presence of the green signal (LSI IGHV 50 probe) and the loss of the red signal on the other homologous chromosome 14 (Figure 2H). The same pattern of hybridization signals was also observed on large interphase nuclei (Figure 2I). Immunoglobulin heavy chain gene analysis by PCR amplification showed the absence of clonal products in patient 2 and in the first patient 1 biopsy, whereas the second lymph node biopsy showed clonal rearrangement. The FISH results demonstrating a partial deletion of the IGH locus deletion were confirmed by aCGH analysis: The loss of the 50 flanking region in patient 1 (583 kb, from nt 106636701 to nt 107219763 in 14q32.33) was found in 49% of cells, whereas the loss of the 30 region was found in 61% of cells for patient 2 (402 kb from nt 106371690 to nt 106774672 in 14q32.33) (Figure 3, Table 1). Cytogenetic chromosomal 14 and 6 deletions, found respectively in 33.3% and 16.6% of metaphases in patient 1, were not identified by aCGH. In addition, the terminal 14q deletion found in four of eight metaphases in patient 2 was not evidenced by aCGH. We did not consider copy number variations (CNVs) reported as “benign” in the Database of
Clonal genomic rearrangements in LH
15
Figure 2 Patient 1: (A) Chromosome 14q22q32.1 partial deletion; (B) chromosome 6q21 partial deletion; (C) FISH with specific 14q telomeric probe; the arrow indicates the deleted chromosome 14; (D) FISH with IGH break-apart probes; the white arrow shows cytogenetically deleted chromosome 14, whereas the red arrow indicates chromosome 14 with partial deletion of the IGH probe (green signal); (E) cytogenetically normal chromosome 14 with a partial FISH deletion of the IGH probe (arrow) (green signal); (F) tissue FISH shows the IGH probe deletion on a nucleus (arrow). Patient 2: (G) Chromosome 14q32.1qter partial deletion; (H) FISH with IGH breakapart probes on metaphase; (I) interphase nuclei; the arrow shows the red signal loss.
Genomic Variants (http://projects.tcag.ca/variation/), and the remaining CNVs identified in both samples are reported in Supplementary Table 1S. Briefly, the other 17 different CNVs were seen in patient 1, with a size range from 234 nt424 kb, whereas 46 CNVs were seen in patient 2, with a size range from 831 nt797 kb (Supplementary Table 1).
Discussion The finding of clonal cytogenetic abnormalities in cases of LH raises difficulties with results interpretation and clinical management. We cannot know the fate of abnormal cells in vivo. They could gain a proliferative advantage, go toward apoptotic pathways, or remain dormant. In fact, cells with the typical chromosomal translocation t(14;18)(q32;q21), which is associated with follicular lymphoma, were described in normal tonsils or peripheral blood lymphocytes without progression to
follicular lymphoma (10). As a consequence, it is likely that more than one genetic event is needed to determine the evolution toward lymphoma (11). Somatic hypermutation is a physiological process occurring in B cell development and involving the IgH variable region in GC B cells. Pasqualucci et al. (12) describe an aberrant hyper-mutation activity that targets multiple loci, including proto-oncogenes in cases of diffuse large-cell lymphoma. Hypermutable genes are susceptible to chromosomal translocation, which generate DNA breaks and could contribute to lymphomagenesis. The detailed data from the cytogenetic analyses of LH described in the literature are reported in Table 2 with only a few of the reports demonstrating FISH and genomic characterization of the cytogenetic abnormalitites. Reports published before 2000 was not considered because of the lack of clinical follow-up. For a total of 261 cases analyzed, a high percentage of culture failure (2146%) suggests that molecular techniques such as FISH, aCGH, and SNP arrays together with
16 Table 1
N. Villa et al. Summary of histological and genetic results in patient 1 and patient 2 FISH results Histological diagnosis
Karyotype
14q Subtel probe 14 Break-apart
Patient 1 Lymph node first biopsy
Lymph node mos 46,XX,del(6) N hyperplasia (q21q21),del(14) (q22q32.1)[2]/ 46,XX,del(14) (q22q32.1)[6]/ 46,XX,del(6) (q21q21)[2]/ 46,XX[14] Lymph node Marginal zone nd nd second biopsy lymphoma
Patient 2 Lymph node biopsy
Lymph node mos 46,XY,del(14) nd hyperplasia (q32.1)[4]/ 46,XY[4]
Loss of IGHV green signal on normal chromosome 14 and in 14.3% of interphase nuclei
t(14;18) t(11;18) t(8;14)
aCGH
nd
583 kb 50 IGH locus deletion (49%)
nd
N
Loss of IGHV N green signal in 10.5% of interphase tissue nuclei
N
nd MYC amplification trisomy/ tetrasomy chromosome 14
Loss of IGH red nd signal on normal chromosome 14 and in interphase nuclei
nd
nd
402 kb 30 IGH locus deletion (61%)
Abbreviations: nd, not done; N, normal hybridization signal pattern.
conventional cytogenetics should be adopted. Cases with clonal chromosomal rearrangements were observed, which ranged from 2.233.3%, and 1 breakpoint fell in 14q32 in a total of 8 of 25 cases with clonal abnormalities (Table 2). Clinical follow-up of clonal patients reported by Sevilla et al. showed a “free of disease” condition in all cases (6). All 10 patients with clonal genetic abnormalities reported by Au et al. (4) had a diagnosis of malignant lymphoma after a second lymph node biopsy performed in a range of 0.3-37.1 months from the first one (Table 2). We had the opportunity to study two LH cases that showed clonal chromosomal deletions. The cytogenetic regions involved are different: The patient 1 deletion is larger and interstitial (14q22q32.1), and the patient 2 deletion is smaller and terminal (14q32.1qter). Interestingly, the partial IGH deletion, which was observed in both patients, is present in the cytogenetically normal chromosome 14. Furthermore, the finding of metaphases with two structurally normal chromosomes 14 and the IGH deletion (Figure 2E) suggests that the first hit could be the IGH deletion and the second hit the chromosomal deletion of the homologous chromosome. On the other hand, Gandhi et al. (13) recently demonstrated that homologous chromosomes make contact at the sites of double-strand breaks (DSBs) within gene sequences. DSBs induce a complex response involving various repair pathways: A misrepair mediated by ATM kinase activity and DNA-PK could generate a chromosomal rearrangement as observed in the chromosomes 14 in both patient 1 and patient 2. In both cases, the partial IGH deletion was also observed on interphase nuclei and only in the largest ones. Small
Figure 3 aCGH combined results of chromosome 14 and graphic representation of the IGH region with patient 1 and patient 2 deletions evidenced.
Clonal genomic rearrangements in LH Table 2
Literature reports on LH cases
Number Age (y), Reference of cases range (4) (5) (6) a b
17
30 84 147
Cases with 14q32 involvement/ Cases with total cases Cytogenetic clonal karyotype/ with clonal failure, % total cases (%) karyotype IGH FISH
2088 30 Not reported 46 282 21
10/30 (33.3) 1/45 (2.2) 14/116 (7)
4/10 0/1 4/14
d d 3/4
Follow-up: number of cases/clonal cases
Range (months)
Free of disease Malignancy
0.1116.9 2/10a d d 5168 13/14b
10/10 d d
Remission after lymphoma. One patient lost at follow-up.
nuclei presented the two normal fusion signals. We could suppose that the rearrangement only involves centroblasts and not small lymphocytes. The aCGH results were consistent with the FISH data and allow for the definition of breakpoints at the molecular level. In both cases, one breakpoint was in the immunoglobulin heavy chain coding region and the second was in the flanking region that contained RNA coding genes (KIAA0125, ADAM6, LOC100133469). We emphasize that patient 2 had developed a follicular lymphoma 16 years earlier, and the diagnosis was only histological because the tissue was not available for molecular investigation. Patient 2 is free of disease after 5 years from the LH diagnosis. The role of cytogenetic translocations in the pathogenesis of lymphomas leading to the activation of oncogene expression is well known. In our cases, no translocations were identified, but deletions of a region of 14q were observed. The partial 14q chromosomal deletion and IGH locus involvement are recurrently observed in B cell malignancies such as chronic lymphocytic leukemia (CLL), acute lymphoblastic leukemia (ALL), and multiple myeloma (MM) (Mitelman Database of Chromosome Aberrations in Cancer, http://cgap.nci.nih.gov/Chromosomes/Mitelman)(14). Discordance was observed when the results obtained by cytogenetics, FISH, and aCGH were compared in the two cases presented here: Partial IGH locus deletion identified by FISH was confirmed by aCGH, whereas the chromosomal 14q deletion was not detected by aCGH. This technique is strongly limited by its inability to detect losses or gains that occur in
