3.4. Marginal Zone Lymphomas
Marginal zone B-cell lymphomas (MZLs) are currently divided into three distinct subtypes (extranodal MZLs of mucosa-associated lymphoid tissue (MALT) type, nodal MZLs, and splenic MZLs) according to the WHO Classification. However, the pathobiological differences as well as the relationship between the subtypes are still not well defined. In 2011, Rinaldi et al. [74] performed an extensive genomic analysis of DNA copy number changes in a large series of MZL cases, aiming to identify differences among subtypes. To do this, the Authors applied the Affymetrix Human Mapping 250K SNP array to 218 MZL samples (25 nodal, 57 MALT, 134 splenic, and two not better specified MZLs). First, it was found that MALT lymphoma presented significantly more frequently gains at 3p, 6p, 18p, and del(6q23) (TNFAIP3/A20). The latter finding, as A20 is a negative regulator of the NFκB pathway, was consistent with the activation of this pathway that is a frequent event in MALT lymphoma, more often due to chromosomal translocations. On the other hand, splenic MZLs was associated with del(7q31) and del(8p). Nodal MZLs lacked the splenic MZLs-related 7q losses but it did not show any statistically significant difference when compared with MALT lymphoma. Gains of 3q and 18q were common to all subtypes, confirming the existence of common pathogenetic pathways in the three variants (Table 4). From a clinical perspective, though del(17p), affecting TP53, did not determine a significantly worse outcome and del(8p) was only of borderline significance, the presence of both deletions (indeed often associated) had a highly significant negative impact on the outcome of splenic MZLs [74].
microarrays-04-00551-t004_Table 4 Table 4  The most important recurrent genetic aberrations in Marginal zone B-cell lymphomas (MZLs).   *MALT: mucosa-associated lymphoid tissue.   More recently, another group focused on MALT lymphomas, aiming to identify by SNPs array karyotyping, possible genomic changes associated with disease progression [75]. Indeed, they included seven cases characterized by small cell morphology, eight composite lymphomas (i.e., cases with small cell morphology and areas constituted by large cells) and 13 large cell variants using the Affymetrix Genome-Wide Human SNP Array 6.0 array. Consistent with the initial hypothesis, the Authors found an increase of genomic complexity with lymphoma progression from small to large cytology, and identified gains of well known oncogenes, including REL, BCL11A, ETS1, PTPN1, PTEN and KRAS, which were found exclusively in the large cell variants (Table 4). Copy numbers of ADAM3A, SCAPER and SIRPB1 also varied among the three cytological subtypes, again indicating the presence of aberrations associated with progression from small to large cell lymphoma. Interestingly, the observation that the number of aberrations was slightly higher in the large cell part of composite lymphomas as compared to large cell lymphomas raised the hypothesis that large cells areas within composite lymphomas may represent a transition state of clonal selection during the progression from small cell to large cell variants. In line with this, when the cytologically different portions of two cases were analyzed in depth, an increase of genomic complexity from the small cell to the more blastoid part of the lymphoma was observed [75]. Interestingly, SNPs array analysis highlighted the frequency of acquired UPD in MZL, a previously uncovered phenomenon [74,75]. Among other loci, aUPD interested regions containing PECAM1/CD31, PRDX1, E2F2, AKT3, and TRAF3IP2 (Table 4) [75]. Furthermore, a very recent analysis of 29 ocular adnexal MALT lymphomas, as reported by Takahashi et al. [76], revealed a higher occurrence rate of gains, as compared to losses and UPDs. Indeed, the authors reported gains including trisomy 3 in 31% of cases, trisomy 18 identified in 17% of samples, and 6p and 21q in 14% of tumors. On the other hand, losses of 6q and 9p were detected in 7% of all cases, along with UPD of 6q in 14% of cases.