Genome-wide microarray-based comparative genomic hybridization analysis of lymphoplasmacytic lymphomas reveals heterogeneous aberrations.

Lymphoplasmacytic lymphoma (LPL) is not a sharply delineated lymphoma entity, either morphologically, phenotypically, or clinically. The diagnosis is often made by excluding other small cell lymphomas with plasmacytic differentiation, thus a genetic diagnostic marker would be of great benefit. Conventional cytogenetic techniques have previously demonstrated a deletion of 6q in a proportion of cases, varying from 7 to 55%. In this report, we apply array-based comparative genomic hybridization on 11 LPL samples. Genomic aberrations were detected in 9 of 11 cases, and included gains and losses. In general, the number of genetic aberrations was relatively low (two to three abnormalities per case). Recurrent aberrations detected were deletion of 6q (two cases), deletion of chromosome 17 (two cases), gain of 3q (two cases), and gain of chromosome 7 (two cases). This report not only confirms the reported loss of 6q in a proportion of cases but also highlights the genetic heterogeneity of LPL, in accordance with the known immunophenotypical, morphological, and clinical diversity of the disease.

Telomere length and correlation with histopathogenesis in B-cell leukemias/lymphomas.

Telomere length was recently reported to correlate with cellular origin of B-cell malignancies in relation to the germinal center (GC). In this report, we measured telomere length by quantitative-PCR in 223 B-cell lymphomas/leukemias and correlated results with immunoglobulin (Ig) mutation status and immunostainings for GC/non-GC subtypes of diffuse large B-cell lymphoma (DLBCL). Shortest telomeres were found in Ig-unmutated chronic lymphocytic leukemia (CLL) [median telomere to single copy gene value (T/S) 0.33], differing significantly to Ig-mutated CLL (0.63). Contrary to this, mantle cell lymphomas (MCLs) exhibited similar telomere lengths regardless of Ig mutation status (0.47). Telomere length differed significantly between GC-like (0.73) and non-GC-like DLBCLs (0.43), and follicular lymphomas (FLs) had shorter telomeres (0.53) than GC-DLBCL. Hairy cell leukemias, which display Ig gene intraclonal heterogeneity, had longer telomeres (0.62) than FLs and non-GC-DLBCL, but shorter than GC-DLBCL. We conclude that although DLBCL and CLL subsets can be clearly distinguished, telomere length reflects many parameters and may not simply correlate with GC-related origin.

Detailed assessment of copy number alterations revealing homozygous deletions in 1p and 13q in mantle cell lymphoma.

Mantle cell lymphoma (MCL) is characterized by over-expression of cyclin Dl as a result of the characteristic t(11;14)(q13;q32). However, this translocation alone has proven not to be sufficient for lymphomagenesis, suggesting the involvement of additional alterations. We have characterized 35 cases of MCL by array comparative genomic hybridization with an average resolution of 0.97 Mb distributed over the complete human genome. The most common alterations were losses in 1p13.2-p31.1, 6q16.2-q27, 8p21.3, 9p13.2-p24.3, 9q13-q31.3, 11q14.3-q23.3, 13q14.13-q21.31, 13q33.1-q34, and 22q11.23-q13.33 and gains involving 3q21.2-q29, 7p12.1-p22.3, 8q24.13-q24.23, and 18q21.33-q22.3. Four homozygous deletions were identified in totally three patients; two overlapping at 1p32.3, and two adjacent at 13q32.3. The homozygous deletions at 1p32.3 cover the CDKN2C locus (coding for p18), while the region at 13q32.3 does not encompass any known tumor suppressor genes. A gain in 3q was significantly associated with shorter survival (P=0.047).

Immunoglobulin gene analysis of mature B-cell malignancies: reconsideration of cellular origin and potential antigen involvement in pathogenesis.

Mature B-cell malignancies stem from B cells transformed at various developmental stages, accounting for the wide range of heterogeneous features observed in the different disease entities. Analysis of the immunoglobulin (Ig) genes can facilitate the identification of the normal B-cell counterpart of lymphomas and leukemias, as Ig genes acquire somatic hypermutation in germinal centers during the immune response to antigen. Therefore, lymphomas that derive from a naïve, pregerminal center B cell lack somatic hypermutation in the clonal Ig gene, whereas germinal center-derived lymphomas, such as diffuse large B-cell lymphoma and follicular lymphoma, display somatic hypermutation of their Ig genes. Furthermore, biases in the Ig variable heavy chain gene repertoire in B-cell malignancies can indicate a possible antigenic influence in pathogenesis. Much work has been accomplished in the past decade to characterize the Ig genes in different lymphoma entities, and the separation of chronic lymphocytic leukemia into two prognostic subgroups in the late 1990s based on the presence or absence of somatic hypermutation led to investigations of Ig genes in larger cohorts of previously uncharacterized entities, such as mantle cell lymphoma. This review will briefly discuss relevant aspects of normal B-cell development, and then focus on what can be ascertained from Ig studies of newly characterized entities, mantle cell lymphoma, hairy cell leukemia, lymphoplasmacytic lymphoma/ Waldenström's macroglobulinemia, and splenic marginal zone lymphoma, from the point of view of cellular origin and variable heavy chain gene restrictions as a sign of antigen involvement. Correlations with gene expression profiling data and the clinical implications of Ig gene studies, when relevant, will be mentioned. The recent evidence that an alternative pathway of gaining somatic hypermutation might exist is also considered, and the implications this has for understanding the cellular origin of B-cell malignancies.

Lymphoplasmacytic lymphoma/Waldenström's macroglobulinemia derives from an extensively hypermutated B cell that lacks ongoing somatic hypermutation.

Lymphoplasmacytic lymphoma (LPL) is a rare lymphoma thought to originate from a B cell stimulated to differentiate to a plasma cell, and which is usually accompanied by clonal IgM secretion, defining the diagnosis of Waldenstrom's macroglobulinemia (WM). However, the immunoglobulin variable heavy chain (VH) gene usage and the somatic hypermutation status have not been widely investigated in LPL. LPL biopsies (CD19+/CD20+/CD22+/CD5-/CD10-/CD23-/kappa+) from 14 patients were included most of whom had a serum IgM component of variable magnitude (two cases with IgG). Highly mutated VH genes (mean mutation rate 8%) were revealed in 13 of 14 LPLs, whereas one case displayed a germline VH gene configuration. Cloning of the VH gene rearrangements in nine cases showed homogeneous sequences without intraclonal heterogeneity. Furthermore, no bias in the VH gene usage was shown, with VH3, VH4, and VH1 gene family members represented. These data confirm that the LPL precursor cell has been exposed to the germinal centre environment, as indicated by extensive hypermutation, but also that the transformation event occurred after affinity maturation, since there was a lack of intraclonal variation. Additionally, the VH gene repertoire was not skewed in LPL/WM as has been demonstrated in other B cell malignancies.

Heterogeneous somatic hypermutation status confounds the cell of origin in hairy cell leukemia.

Hairy cell leukemia (HCL) is thought to arise from a post-germinal center (GC) B-cell, however the exact normal counterpart remains unclear. We performed VH gene analysis of 32 HCL cases, revealing somatically mutated VH genes (<98% homology) in 27 cases and unmutated VH genes in five cases, four of which displayed germline VH genes. Intraclonal heterogeneity was evident in the majority of eight mutated HCLs investigated, although at a lower level compared to GC-derived lymphomas. A novel finding of preferential VH3-30 gene usage was detected (19% of HCLs). Our data confounds the postulated post-GC origin in HCL considering (1) the finding of unmutated HCLs, generally correlating with a pre-GC origin, and (2) the presence of intraclonal variation in mutated HCLs. The latter suggests that the transformed B-cell was frozen when it still had an active mutation process, implying a closer relation to the GC than previously assumed. Furthermore, restricted VH3-30 usage indicates that antigen selection could be a promoting factor in HCL development.

Mantle cell lymphomas with clonal immunoglobulin V(H)3-21 gene rearrangements exhibit fewer genomic imbalances than mantle cell lymphomas utilizing other immunoglobulin V(H) genes.

A preferential use of one particular immunoglobulin variable heavy chain gene, V(H)3-21, has recently been reported in mantle cell lymphoma, where almost all of these V(H)3-21+ mantle cell lymphomas showed usage of the same light chain Vlambda gene (Vlambda3-19) and also had a tendency towards improved prognosis. These findings suggested that V(H)3-21+ mantle cell lymphomas constitute a distinct subgroup, possibly with antigen stimulation involved in disease pathogenesis. In this study, we applied the comparative genomic hybridization (CGH) method on 37 mantle cell lymphoma tumors in order to investigate if the V(H)3-21+ tumors are different at the genomic level. Interestingly, V(H)3-21+ mantle cell lymphomas (n=14) showed significantly fewer genomic aberrations (mean 2.4) compared to non-V(H)3-21 mantle cell lymphomas (n=23) (mean 4.9). The chromosomal aberrations identified in our study were generally in accordance with previous CGH studies of mantle cell lymphoma; the most frequent aberration was complete or partial loss of chromosome 13, followed by recurrent losses within 6q, 9p, 9q and 11q and frequent gains in 3q, 7p, 8q and 15q. Deletions within 8p and 9p as well as gains in 7p and 15q were found exclusively in the non-V(H)3-21-utilizing tumors. In summary, V(H)3-21+ mantle cell lymphomas demonstrated both a lower number and a different spectrum of genomic aberrations than mantle cell lymphoma in general, thus supporting the hypothesis that V(H)3-21+ mantle cell lymphomas constitute a new subgroup. The findings presented in this report may explain the tendency for a better clinical outcome for patients whose tumors utilize V(H)3-21.

Mutated VH genes and preferential VH3-21 use define new subsets of mantle cell lymphoma.

Mantle cell lymphoma (MCL) is believed to originate from a naive B cell. However, we recently demonstrated that a subset of MCL displayed mutated V(H) genes. We also reported restricted use of certain V(H) genes. To assess the prognostic impact of these new findings, we performed V(H) gene analysis of 110 patients, revealing that 18 (16%) patients had mutated and 92 (84%) patients had unmutated V(H) genes. Because the mutation rate was low in the mutated group (2.2%-6.7%), further investigation of the germline V(H) gene in T cells from 5 patients with mutated V(H) genes was carried out; results showed that the unrearranged V(H) gene was identical to the published sequence. These data confirm that the base pair substitutions within the rearranged V(H) genes represent hypermutations, and indicate germinal center exposure. However, V(H) gene mutation status did not correlate with prognosis because there was no difference in clinical outcome between the unmutated and mutated groups. The most frequently used V(H) genes were V(H)3-21 (21 patients) and V(H)4-34 (19 patients). A novel finding was that V(H)3-21(+) MCL almost exclusively expressed lambda light chains and displayed highly restricted use of the V(lambda)3-19 gene. V(H)3-21(+) patients had longer median survival than the remaining patients (53 vs 34 months; P =.03), but they tended to be younger at diagnosis. The combined use of V(H)3-21/V(lambda)3-19 suggests a possible role for antigen(s) in the pathogenesis of these tumors and indicates that V(H)3-21(+) patients constitute a new MCL entity.