We read with interest the recent report by Miglino et a1 (1992) on the clonality of plasma cells in plasma cell dyscrasias.These investigators found immunoglobulin rearrangements in only 2/15 cases of multiple myeloma, and observed no rearrangements in some samples with up to 80% plasma cells.Their results on the detection of clonality by immunoglobulin gene rearrangement were confirmed by examination of X-linked markers in four of the myeloma patients.We recently reported 36 patients with multiple myeloma and the unexpected finding that clonal immunoglobulin gene rearrangement were not uniformly present (Humphries et al, 199 1).However, we and others (Zehnbauer et a/.1986) have observed that digesting DNA with multiple enzymes and combinations of enzymes improved the sensitivity of the technique, with clonal rearrangements detected in 29/36 cases in our series.In 3/29 cases rearrangements were only detected in the samples simultaneously digested with BamHl and HindIII.These 29 included six patients with advanced myeloma and peripheral blood plasma cells.Even when clonal rearrangements were detected, the intensity of the rearranged bands was less than anticipated given the percentage of plasma cells in the sample.Miglino et a1 (1992) digested their DNA samples with only one or two restriction enzymes and no combinations of enzymes.This may have lowered to their lesser sensitivity and contributed to the low frequency of immunoglobulin gene rearrangements detected in their series.Since the publication of our report we have studied eight additional patients and have found immiinoglobulin gene rearrangement in 6/8 (Table I).Interestingly, one sample with 82% plasma cells showed no rearrangements.Though the apparent paradox of the production of a 'monoclonal' protein by a 'non-clonal' population of plasma cells remains
BCL1/PRAD1 gene rearrangements involving the cyclin D1 gene are a feature of about 70% of centrocytic/mantlecell lymphomas (CC/MCL) but are identified in only a small proportion of other B-cell non-Hodgkin's lymphomas. Of 37 lymphomas found to have BCL1/cyclin D1 (PRAD1, CCND1) gene rearrangements, 30 fit the morphologic and immunophenotypic criteria for typical CC/MCL. Seven cases with morphologic features atypical for CC/MCL were identified. CD5+ monoclonal B cells were documented in all these cases. Six cases were subsequently stained for cyclin D1 protein, and all showed nuclear positivity. Five cases had variably sized foci of cells with moderately abundant pale cytoplasm resembling parafollicular/monocytoid B cells, marginal zone cells, hairy cells, or even proliferation centers. Transformed-appearing cells were also present in some lymphomas. In one case, striking follicular colonization created a markedly nodular growth pattern mimicking a follicular lymphoma. A sixth case had a marked predominance of small, round lymphocytes at some sites, mimicking a small lymphocytic lymphoma. Five of these six cases also had areas more typical of CC/MCL. The seventh case was a CD5-positive splenic marginal zone-like lymphoma (SMZL) with plasmacytic differentiation and circulating villous lymphocytes consistent with a splenic lymphoma with villous lymphocytes (SLVL). These cases illustrate the morphologic spectrum of small B-cell lymphoid neoplasms that have BCL1/cyclin D1 gene rearrangements and overexpression of cyclin D1. Despite the BCL1 translocation and cyclin D1 overexpression, the splenic lymphoma with plasmacytic differentiation was definitely not a CC/MCL and fit the clinicopathologic entity of SMZL/SLVL. The other six cases are best considered CC/MCL variants based on a combined morphologic, immunophenotypic, and genotypic evaluation. Genotypic or immunophenotypic studies to identify cyclin D1 rearrangements and overexpression, although not-pathognomonic, are useful in recognizing these variant CC/MCL cases, which can mimic almost any of the other well-described but more indolent low-grade B-cell lymphomas and leukemias. Some of the variant CC/MCL cases had features in common with the CD5 + cyclin D1 + SMZL/SLVL, suggesting a possible relationship between these two otherwise distinct entities.
The chromosome 11q13 bcl-1 locus is rearranged in the majority of centrocytic lymphomas, a CD5-positive B-cell non-Hodgkins lymphoma, as a result of reciprocal translocation with the 14q32 immunoglobulin heavy chain genes. Although several 11q13 bcl-1 breakpoint sites have been characterized, a postulated bcl-1 oncogene was not identified. Recently, however, a gene encoding cyclin D1, designated PRAD1, was proposed as a candidate bcl-1 oncogene; accumulated evidence now indicates this gene is bcl-1. To further characterize 11q13 breakpoints in B-cell neoplasms, we analyzed 26 centrocytic lymphomas and 68 other B-cell cancers by Southern blot using a panel of breakpoint probes spanning 110 kilobases of the bcl-1 and PRAD1 loci. Nineteen centrocytic cases (73%) showed rearrangement, 15 at bcl-1 breakpoint sites and 5 at PRAD1 sites. One case was rearranged at both bcl-1 and PRAD1 loci. All but the latter case showed comigration of rearranged bcl-1 or PRAD1 bands and immunoglobulin heavy chain joining gene bands, consistent with the t(11;14). bcl-1 rearrangement was present in only one of 68 noncentrocytic B-cell neoplasms; none showed PRAD1 rearrangement. Thus, bcl-1 and PRAD1 rearrangement is strongly associated with centrocytic lymphoma, providing a useful molecular marker for classifying this subtype of lymphoma and suggesting an important role for PRAD1 cyclin D1 in the pathogenesis of this neoplasm.
We report a case of untreated non-Hodgkin's lymphoma with histologic progression over 1 yr from a low-grade, small cleaved follicular center cell lymphoma to a high-grade, small noncleaved follicular center cell lymphoma. Both lymphomas had identical immunoglobulin (Ig) heavy-chain joining gene (JH), kappa light-chain joining gene, and bcl-2 gene rearrangements, indicating the clonal identity of the two tumors. The Ig heavy chain locus on one chromosome 14 was involved in an initial t(14; 18) translocation as shown by comigrating JH and bcl-2 rearrangements. However, the oncogene c-myc was in the germline configuration in the initial lymphoma but had one allele rearranged near the 3' end of exon I in the high-grade tumor; DNA sequence analysis was consistent with a chromosomal breakpoint at that site. The presence of the c-myc rearrangement in the high-grade tumor suggest a role for c-myc in the clonal evolution of the low-grade tumor into a more aggressive lymphoma. The coexistence of both bcl-2 gene and c-myc oncogene rearrangements in the same tumor is unusual, with only a few cases reported. Furthermore, this case is unique in the direct demonstration of the histologic and clinical progression of a human lymphoma associated with the sequential rearrangement of the bcl-2 gene and the c-myc oncogene.
