ALBERT

Description: Splenic marginal-zone B cells, marginal-zone B cells of Peyer’s patches in the gut, and nodal marginal-zone B cells (also identified as monocytoid B cells) share a similar morphology and immunophenotype. These cells likely represent a distinct subset of B cells in humans and rodents, but their precise ontogenetic relationship as well as their origin from B cells of the germinal center is still debated. To study this, we performed a mutation analysis of the rearranged immunoglobulin variable genes (VH) of microdissected single nodal and splenic marginal-zone cells. In addition, we investigated the presence of proliferating cells and B-cell clones in the human splenic and nodal marginal zone as well as adjacent germinal centers. This was performed by immunohistochemical staining for the Ki-67 antigen and denaturing gradient gel analysis of amplified immunoglobulin heavy chain genes’ complementarity determining region 3 of microdissected cell clusters. A variable subset of nodal and splenic marginal-zone B cells showed somatic mutations in their rearranged VH genes, indicating that both virgin and memory B cells are present in the nodal and splenic marginal zone. Nodal and splenic marginal-zone B cells preferentially rearranged VH3 family genes such as DP47, DP49, DP54, and DP58. A preferential rearrangement of the same VH genes has been shown by others in the peripheral CD5− IgM+ B cells. These data suggest that the splenic and nodal marginal-zone B cells are closely related B-cell subsets. We also showed that marginal-zone B cells may cycle and that clones of B cells are frequently detected in the nodal as well as the splenic marginal zone. These clones are not related to those present in adjacent germinal centers. These data favor the hypothesis that clonal expansion occurs in the marginal zone. Whether the somatic hypermutation mechanism is activated during the clonal expansion in the marginal zone and which type of immune response triggers the clonal expansion need to be elucidated.

Description: Single-cell polymerase chain reaction (PCR) has been used as a tool to demonstrate clonality and B-cell origin of Reed-Sternberg (RS) cells in Hodgkin disease (HD). An analogous approach was used to investigate genomic imbalances in a (cyto)genetically poorly characterized subentity: lymphocyte predominance Hodgkin disease (LPHD). Nineteen cases of LPHD were selected for a comparative genomic hybridization (CGH) study. CGH was performed with degenerate oligonucleotide primed–PCR (DOP-PCR)–amplified DNA from 4-5 microdissected CD20+ malignant cells. All analyzed cases revealed a high number of genomic imbalances (average 10.8 per case), involving all chromosomes but the excluded 19, 22, and Y, indicating a high complexity of LPHD. The majority of detected aberrations were recurrent. Gain of 1, 2q, 3, 4q, 5q, 6, 8q, 11q, 12q, and X, and loss of chromosome 17 were identified in 36.8% to 68.4% of the analyzed cases. Some of them have also been found in non-Hodgkin lymphoma (NHL), and possibly represent secondary changes associated with disease progression. Gain of 2q, 4q, 5q, 6, 11q, however, are much more rarely observed in NHL and could be more specifically associated with LPHD. Particularly interesting is a frequent overrepresentation of chromosome arm 6q, a region usually deleted in NHL. Rearrangement of theBCL6 gene (3q27) demonstrated by cytogenetics and fluorescence in situ hybridization in 2 cases in this study suggests its contribution in pathogenesis of LPHD. In conclusion, the data show a consistent occurrence of genomic alterations in LPHD and highlight genomic regions that might be relevant for development and/or progression of this lymphoma entity.

Description: We studied the genomic status of BCL6 in 23 cases of nodular lymphocyte predominance Hodgkin lymphoma (NLPHL) and 40 cases of classical Hodgkin lymphoma (cHL), using dual-color interphase fluorescence in situ hybridization (FISH). The BCL6rearrangement was identified in 48% of NLPHL cases and was not detected in cHL cases. As a confirmation, sequential or simultaneous immunohistochemistry (IHC) and FISH using CD20 or BCL6 antibodies and BCL6 DNA probes was performed in 8 NLPHL cases. The BCL6-associated translocations, t(3;22)(q27;q11), t(3;7)(q27;p12), and the most probable t(3;9)(q27;p13), were identified in 3 cases. A consistent expression of BCL6 protein in popcorn cells with the highest number of intensely stained cells in cases with a genomic BCL6rearrangement was shown by IHC. These findings support the hypothesis of a germinal center B cell–derived origin of NLPHL, indicate a significant role of BCL6 in the pathogenesis of NLPHL, and provide further evidence of the genetic diversity underlying the pathogenesis of NLPHL and cHL.

Description: Mantle cell lymphoma (MCL) is hallmarked by IGH-mediated t(11;14)(q13;q32) resulting in deregulation of CCND1. This gene encodes cyclin D1 that together with cyclin D2 and -D3, plays a key role in cell cycle progression. Given that cyclin D1 is not expressed by normal B lymphocytes, its aberrant expression in lymphoma has a diagnostic value. Recently, it has been shown that rare cyclin D1-negative MCL cases express either cyclin D2 or cyclin D3. The molecular mechanisms underlying these events are unknown. We report here the genetic characteristics of 12 MCL cases without cytogenetic and FISH evidence of CCND1 rearrangement. These cases were collected over the last 15 years; they displayed the typical morphology and gene expression profile (determined by comparative expressed sequence hybridization) of MCL. Of interest, 5 of these lymphomas revealed cerebral or meningeal lesions at diagnosis or during a course of disease. Immunohistochemistry and RT-PCR studies showed a lack of expression of D-type cyclins in 2 cases. One of these cases was documented by t(2;14)(p24;q32) targeting NMYC, as shown by FISH and qRT-PCR. In this case upregulation of NMYC possibly led to inactivation of Rb and an aberrant progression of G1/S transition. The second case was characterized by a complex karyotype including loss of 17p. In the remaining 10 cases expression of cyclin D1, -D2 and -D3 in respectively 3, 2 and 5 cases was found. Cytogenetics and FISH identified t(2;12)(p11;p13) resulting in the IGK-CCND2 rearrangement in 1 of the 2 cases expressing cyclin D2 and der(14)t(6;14)(p21;q32) and der(20)t(14;20)(q32;p13) in one of the cases expressing cyclin D3. The breakpoint at 6p21 however was mapped distal to CCND3. FISH studies of p16, p27, p53 and Rb performed on all cases except one, identified loss of one copy of 1 to 4 of the analyzed genes in 7 cases. These data support the existence of t(11;14)-negative MCL that frequently express one of the three D-type cyclins. In one of the cases a new IG translocation affecting NMYC was found. Other genetic aberrations possibly involved in pathogenesis of t(11;14)-negative MCL included chromosomal translocations affecting CCND2 and unidentified genes at 6p21 and 20p13, and loss of tumor suppressor genes involved in cell cycle regulation.

Description: Abstract 1099 Poster Board I-121 The 10p11 chromosomal site is recurrently affected by genomic abnormalities in B cell leukemia/lymphoma, but the molecular consequences of these aberrations remain unknown. We have collected three cases (2 female, 1 male; 67-69 years) of chronic lymphocytic leukemia (CLL) with a novel IGH-mediated t(10;14)(p11;q32) (#1 and 2) or its IGL variant t(10;22)(p11;q11) (#3), and one case of blastic mantle cell lymphoma (MCL) with a complex t(11;14)-positive karyotype including add(10)(p12) (#4). Case 1 showed unmutated VH genes and usage of VH3-21/D1-26/JH6. Cases 2 and 3 revealed mutated VH genes and usage of VH3-21/D1-14/JH4 and VH3-11/D4-4/JH6, respectively. In cases 1 and 2 with a monoallelic loss of RB1, the t(10;14)(p11,q32) was found as a secondary change at time of CLL progression/Richter transformation observed 148 and 50 months after diagnosis, respectively. In case 3, the sole t(10;22)(p11;q11) was identified at time of CLL diagnosis. All three patients died from disease progression 37-149 months after diagnosis. Of note, both patients with t(10;14) died from refractory disease 1 and 3 months after CLL progression/transformation and genetic detection f t(10;14). The patient #3 did not require therapy, but died from septic shock 37 months after diagnosis, and autopsy revealed diffuse and massive lymph node and spleen involvement. The patient with MCL (#4), 67 years old man with a previous history of prostatic adenocarcinoma and CLL, developed secondary central nervous system liesions and died 40 months after diagnosis of MCL. Extensive FISH studies of t(10;14) and t(10;22) mapped the 10p11 breakpoint in the region of the BMI1 locus. SNP array analysis of case 4 complemented by FISH identified amplification of the 10p11 region covering BMI1. Immunostaining with anti-BMI1 serum performed in 2 available cases (#2 and 3) showed expression of the BMI1 protein by neoplastic cells. This approach, however, did not allow to estimate whether the level of BMI1 expression in these cases was higher than in control CLL cases. BMI1 (B-cell-specific Moloney murine leukemia virus integration site 1) is a member of the Polycomb group of genes that are important epigenetic chromatin modifiers with an essential role in embryogenesis and the maintenance of adult stem cells. BMI1 was first identified as a protooncogene that cooperates with c-myc in the generation and development of mouse pre B-cell lymphomas. The BMI1 protein is a component of Polycomb repressive complex1 (PRC1), one of two multiprotein PRCs that maintain cell identity by gene suppression. The gene plays an important role in self-renewal of hematopoietic and neural stem cells. Multiple lines of evidence implicate BMI1 and other Polycomb genes in tumorigenesis. Among others, an aberrant expression of BMI1 has been observed in a wide spectrum of cancers and usually correlated with unfavourable clinical outcome. BMI1 is one of the genes marking the ABC-type signature of DLBCL, associated with a poor prognosis. High levels of BMI1 mRNA/protein were also detected in CLL, MCL, MM and HL. In some MCLs, overexpression of BMI1 correlated with genomic amplification of the BMI1 gene detected by aCGH in 12-24% of analysed cases (Jares and Campo, 2008). BMI1 has been shown to contribute to cell cycle regulation and senescence by acting as transcriptional repressor of the INK4a-ARF tumor suppressors. Despite the wealthy information on BMI1 contributing to a variety of tumors, molecular mechanisms underlying deregulation of BMI1 in neoplasms and functional consequences of these events remain elusive. In summary, we show for the first time that BMI1, a postulated lymphoma-related oncogene, is recurrently targeted by IG-mediated chromosomal translocations in CLL and confirm its amplification in MCL. We found that in both cases with t(10;14)/IGH-BMI1, the translocation was acquired during clinical transformation of CLL initially harboring the del(13q)/RB1. The adverse prognostic impact of BMI1 aberrations in all four collected cases is reflected by a constantly aggressive clinical course of disease and short survival of the affected patients (1-50 months). In this regard, further molecular studies of the BMI1 pathway and its downstream partners are necessary to develop novel therapeutic approaches targeting this oncogene in human cancers. Disclosures No relevant conflicts of interest to declare.

Description: We present 3 cases of large B-cell lymphoma (LBCL) with a granular cytoplasmic staining for anaplastic lymphoma kinase (ALK). All of the cases showed striking similarities in morphology and immunohistochemical profile characterized by a massive monomorphic proliferation of CD20-/CD138+ plasmablast-like cells. In one of the cases, initially diagnosed as a null-type anaplastic large cell lymphoma (ALCL), the B-cell phenotype became evident only at recurrence. Fluorescent in situ hybridization (FISH) and molecular studies led to the detection of a CLTC-ALK rearrangement in all 3 cases, without any evidence of full-length ALK receptor expression. The associated t(2;17)(p23;q23) was demonstrated in the karyotype of 2 cases. Although a similar CLTC-ALK aberration was previously identified in ALK-positive T-/null cell ALCL and inflammatory myofibroblastic tumor, its association with ALK-positive LBCL seems to be specific and intriguing. (Blood. 2003;102:2638-2641)

Description: Recently, a distinctive entity characterized by expression of the anaplastic lymphoma kinase (ALK) protein [most frequently due to the t(2;5)(p23;q35)-associated NPM-ALK fusion] has emerged within the heterogenous group of non-Hodgkin’s lymphomas (NHL) classified as anaplastic large-cell lymphoma (ALCL). Sporadic variant 2p23/ALK abnormalities identified in ALK-positive ALCL indicate that genes other than NPM may also be involved in the deregulation of ALK and lymphomagenesis. We report here three cases with an inv(2)(p23q35) detected by fluorescence in situ hybridization (FISH) in young male patients with ALK-positive ALCL. In contrast to ALCL cases with the classical t(2;5)(p23;q35) that usually show both cytoplasmic and nuclear or predominantly nuclear alone localization of the NPM-ALK chimeric product, in all three cases with an inv(2)(p23q35) the ALK protein accumulated in the cytoplasm only, supporting the previous assumption that the oncogenic potential of ALK may not be dependent on its nuclear localization. As the first step to identify theALK partner gene involved in the inv(2)(p23q35), we performed extensive FISH studies and demonstrated that the 2q35 breakpoint occurred within the 1,750-kb region contained within the 914E7 YAC. Moreover, a striking association of the inv(2)(p23q35) with a secondary chromosomal change, viz, ider(2)(q10)inv(2)(p23q35), carrying two additional copies of the putative ALK-related fusion gene, was found in all three patients, suggesting that, in contrast to the standard t(2;5)/NPM-ALK fusion, multiple copies of the putative 2q35-ALK chimeric gene may be required for efficient tumor development. In summary, we demonstrate that the inv(2)(p23q35), a variant of the t(2;5)(p23;q35), is a recurrent chromosomal abnormality in ALK-positive ALCL, the further characterization of which should provide new insight into the pathogenesis of these lymphomas. © 1998 by The American Society of Hematology.

Description: The genetics of t(11;14)(q13;q32)/cyclin D1–negative mantle cell lymphoma (MCL) is poorly understood. We report here 8 MCL cases lacking t(11;14) or variant CCND1 rearrangement that showed expression of cyclin D1 (2 cases), D2 (2 cases), and D3 (3 cases). One case was cyclin D negative. Cytogenetics and fluorescence in situ hybridization detected t(2;12)(p11;p13)/IGK-CCND2 in one of the cyclin D2-positive cases and t(6;14)(p21;q32)/IGH-CCND3 in one of the cyclin D3-positive cases. Moreover, we identified a novel cryptic t(2;14)(p24;q32) targeting MYCN in 2 blastoid MCLs: one negative for cyclin D and one expressing cyclin D3. Interestingly, both cases showed expression of cyclin E. Notably, all 3 blastoid MCLs showed a monoallelic deletion of RB1 associated with a lack of expression of RB1 protein and monoallelic loss of p16. In sum-mary, this study confirms frequent aberrant expression of cyclin D2 and D3 in t(11;14)-negative MCLs and shows a t(11;14)-independent expression of cy-clin D1 in 25% of present cases. Novel findings include cyclin E expression in 2 t(11;14)-negative MCLs characterized by a cryptic t(2;14)(p24;q32) and identification of MYCN as a new lymphoma oncogene associated with a blastoid MCL. Clinically important is a predisposition of t(11;14)-negative MCLs to the central nervous system involvement.

Description: NOTCH1 signaling is required for normal T cell development. Its aberrant activation by either the rare t(7;9) translocation or frequent mutation(s) in T-ALL points to an important role of this gene in T cell lymphomagenesis. So far, a pathogenetic role for NOTCH1 in B-cell malignancies is unknown. We report here evidence for NOTCH1 rearrangements in B-cell lymphoma. Two novel 9q34 translocations involving either the 14q32/IGH or the 22q11/IGL locus were identified in 2 patients with respectively, Richter syndrome (RS) and follicular lymphoma (FL). The first case showed a clone with a sole trisomy 12 detected by FISH in 25% of interphase cells and a subclone with an additional dic(9;14)(q34;q32) in 60% of cells from a lymph node sample. The karyotype of the FL case was characterized by the typical t(14;18)(q32;q21)/IGH-BCL2 accompanied by additional t(8;14)(q24;q32)/IGH-CMYC and t(9;22)(q34;q11). FISH analysis of dic(9;14) and t(9;22) using a set of BAC clones mapped both 9q34 breakpoints to the 5′end of NOTCH1. These breakpoints were different from the breakpoint of t(7;9)(q35;q34) (intron 24), as demonstrated by FISH analysis of the T-ALL SUPT1 cell line. NOTCH1 rearrangements showed to be rare in B-NHL. Any of the additional 30 lymphoma cases with structural aberrations of 9q34 analyzed by FISH showed rearrangement of this gene. So far, aberrant activation of NOTCH1 in the reported cases with dic(9;14) and t(9;22) could not be documented: both lymphomas but also 9 control CLL and FL cases without 9q34 aberrations revealed expression of NOTCH1 mRNA by RT-PCR, its ligand JAGGED2 and its transcriptional target HES1. This expression pattern indicates a ligand-driven NOTCH1 signaling in all analyzed cases, possibly reflecting its physiological activation in progenitor B cells and thus, masking the presumed aberrant activation of NOTCH1 in present lymphomas. Whether IGH/L-NOTCH1 translocations in B-NHL lead to the generation of truncated/activated NOTCH1 proteins, similar to these found in T-ALL, remains to be determined. Particularly interesting is the association of the dic(9;14) with Richter transformation in the first case suggesting that rearrangement of NOTCH1 could be responsible for a rapid progression of the underlied CLL. In the second case of FL, t(9;22) seemed to be associated with the evolution of t(14;18)-positive karyotype. These findings contrast with t(7;9), considered as a primary oncogenic event in development of T-ALL. Further molecular and immunohistochemical investigations of the reported cases are in progress.

Description: Chromosomal translocations involving the immunoglobulin heavy chain genes cluster (IGH) at 14q32 have been found in up to 70% of B-non Hodgkin’s lymphoma. These aberrations lead to deregulation of putative oncogenes by their juxtaposition with IGH enhancer elements. So far, more than 20 genes affected by t(14q32) have been identified in B-cell malignancies. The most frequently rearranged genes include BCL1/11q13, BCL2/18q21, BCL6/3q27 and CMYC/8q24. We report here a novel t(3;14)(p13;q32) involving IGH, as shown by FISH, in three lymphoma cases. One of these cases was diagnosed as gastric MALT-type lymphoma, the remaining two cases as diffuse large B-cell lymphoma with a focal nodular growth pattern reminiscent of follicular lymphoma. In order to identify the gene targeted by the t(3;14)(p13;q32) we performed FISH analysis of the first case and narrowed down the breakpoint to RP11-154H23 at 3p13 that showed a split signal on both derivative chromosomes. This BAC clone covers the FOXP1 gene. Using a pair of probes selected for the 5′ and 3′ ends of FOXP1, we demonstrated rearrangement of the gene by dual color FISH in the remaining two cases. An analysis of FOXP1 expression in present cases is being performed. FOXP1 (Forkhead box-P1) is a winged-helix transcription factor that acts as a transcriptional repressor. It is expressed in a wide variety of normal and neoplastic tissues, including lymphomas. FOXP1 has been shown to be expressed in normal activated B-cells using genomic-scale expression profiling, and in B-cells within and outside the germinal center by immunohistochemistry. Its physiological role in lymphocytes, however, is unclear. Recent studies showed that FOXP1 is strongly expressed in a subset of DLBCL. Interestingly, high expression of FOXP1 carries an independent prognostic significance, what suggests a possible role of the gene in the biology of this group of lymphoma. Although genomic rearrangements of FOXP1 have not been demonstrated so far in cancer, rearrangements of other members of FOX gene family in various pathological conditions have been reported. The involvement of FOXP1 in t(3;14)(p13;q32) found in two DLBCL cases indicates that this translocation may underlie a strong expression of FOXP1 in at least part of these lymphomas. Its rearrangement in a case of gastric lymphoma is intriguing, but additional investigations are required to find out whether t(3;14)(p13;q32) is a recurrent aberration in MALT lymphoma. Further molecular, immunophenotypic and clinical studies of the three cases with t(3;14)(p13;q32) are under progress.