Value of GCET1, HGAL (GCET2), and LMO2 in the Determination of Germinal Center Phenotype in Diffuse Large B-cell Lymphoma

Abstract

Diffuse large B-cell lymphoma (DLBCL) is a biologically heterogeneous disease that is classified into germinal center B-cell (GCB) and non-GCB subtypes, which are prognostically different. The Hans algorithm is the most widely used tool based on CD10, BCL6, and MUM1 expression, but some cases with the non-GCB phenotype are still known to be misclassified. In this study, we investigate the extent to which GCET1, HGAL, and LMO2 protein expressions reflect GCB phenotype together with their roles in determining the GCB phenotype of DLBCL and their contributions to the performance of the Hans algorithm. Sixty-five cases of DLBCL-not otherwise specified, 40 cases of follicular lymphoma (FL), and 19 non-GC-derived lymphoma cases were included in this study. The DLBCL cases were grouped as CD10+ (Group A) or only MUM1+ (Group B), and the remaining cases constituted the intermediate group (Group C). GCET1, HGAL, and LMO2 expressions were evaluated. In the FL group, GCET1, HGAL, and LMO2 were positive in 85%, 77.5%, and 100% of the cases, respectively. Among the non-GC-derived lymphoma cases, all three markers were negative in cases of small lymphocytic lymphoma, plasmablastic lymphoma, peripheral T-cell lymphoma, and anaplastic large cell lymphoma. GCET1 and HGAL were negative in cases of marginal zone lymphoma (MZL) and mantle cell lymphoma (MCL). Two of the 3 MZL and 2 of the 4 MCL cases were positive for LMO2. In the DLBCL group, the number of cases with GCET1, HGAL, and LMO2 positivity was 18 (90%), 17 (85%), and 20 (100%), respectively, in Group A and 0 (0%), 2 (13.3%), and 2 (13.3%), respectively, in Group B. Considering these rates, when the cases in the intermediate group were evaluated, it was concluded that 13 cases typed as non-GCB according to the Hans algorithm may have the GCB phenotype. GCET1, HGAL, and LMO2 are highly sensitive markers for determining the germinal center cell phenotype and can increase the accuracy of the subclassification of DLBCL cases, especially for cases that are negative for CD10.