MYD88, CD79B, and CD79A Gene Mutations in CD5-Positive Diffuse Large B-Cell Lymphoma

INTRODUCTION

Diffuse large B-cell lymphoma (DLBCL) is regarded as a clinicopathologically heterogeneous group of lymphomas and is classified as activated B-cell-like (ABC) DLBCL or germinal center B-cell-like (GCB) DLBCL based on the cell-of-origin (COO). Nuclear factor-kappa B (NF-kB) activation is indispensable for ABC DLBCL cell survival. Genes in the NF-kB signaling pathway, such as myeloid differentiation factor 88 (MYD88) and CD79B, are mutated in 20-40% of ABC DLBCLs. CD5 is expressed in approximately 10% of DLBCLs. CD5-positive (CD5⁺) diffuse large B-cell lymphoma (DLBCL) is characterized by frequent central nervous system (CNS) relapse and a predominant activated B-cell-like nature. Primary DLBCL in sanctuary sites (DLBCL-SS) also shows these features, and more than 70% of the patients exhibit MYD88 (L265P) and CD79B mutations. This study aimed to ascertain the frequencies of MYD88 (L265P), CD79B, and CD79A mutations in CD5⁺ DLBCL and to determine whether CD5⁺DLBCL shows the same features as DLBCL-SS in terms of gene mutations.

PATIENTS AND METHODS

This study included 40 patients with CD5⁺DLBCL. All the patients were diagnosed with DLBCL, not otherwise specified, according to the 2008 WHO classification between 1993 and 2014 at Mie University Hospital. CD5 expression in tumor cells was analyzed by immunohistochemistry using frozen sections or by flow cytometry. Mutation analysis was performed by direct sequencing.

RESULTS

Direct sequencing was successful with samples from all 40 patients with CD5⁺ DLBCL. The median age was 64 years (range: 15 to 91 years). The COO classification was determined by gene expression profiling (GEP) in 29 patients. In the other 11 patients, the COO was identified by immunohistochemistry according to Hans' criteria. Thirty-six (90%) cases were confirmed as ABC/non-germinal center DLBCL. Two patients fulfilled the criteria for primary testicular DLBCL. One patient had systemic CD5⁺ DLBCL with bone and brain involvement. None of the other patients had CNS involvement at diagnosis. The MYD88 L265P mutation was identified in 13 (33%) of the cases, and no other MYD88 mutations were found. CD79B mutations were detected in 15 (38%) cases, and 10 of these cases overlapped with the MYD88-mutated group. One of the two patients with testicular involvement had double mutations. The other had MYD88 mutation alone. One patient with brain involvement had double mutations. Only one case (3%) had a CD79A mutation as well as CD79B and MYD88 mutations. The incidence of MYD88 L265P, CD79B, and CD79A mutations in ABC/non-GCB DLBCLs was 13/36 (36%), 14/36 (39%), and 1/36 (3%), respectively. Of the 15 cases with a CD79B mutation, 14 had missense mutations in an immunoreceptor tyrosine-based activation motif (ITAM) domain; all of these mutations occurred in the first tyrosine of the ITAM (Y196H [7/14], Y196S [4/14], Y196N [2/14], and Y196C [1/14]), and two cases showed double mutations (Y196H/H225Y and Y196S/K219R). One case had a 13-base deletion in exon 5 before the first tyrosine of the ITAM. MYD88 and CD79B mutations were associated with localized disease (P = 0.038 and P = 0.003, respectively). Clinical information on first-line treatment was available for 38 patients. Before 2002, anthracycline-containing chemotherapies without rituximab were selected as first-line treatment. Since 2002, CHOP chemotherapy with rituximab (R-CHOP) has predominated. For this reason, 18 patients who were treated with R-CHOP were included in the survival analysis in this study. The COO of these 18 patients was confirmed as ABC/non-GCB DLBCL. Two patients with primary testicular lymphoma and one with CD5⁺ DLBCL with brain involvement were not included in this cohort of 18 patients. With a median follow-up of 7.1 years, there was no significant difference in overall survival based on MYD88 mutation status (P = 0.98) or CD79B mutation status (P= 0.69).

CONCLUSIONS

To the best of our knowledge, this is the first study toreport the frequency of MYD88, CD79B, and CD79A mutations in the largest cohort of CD5⁺ DLBCLpatients. The incidence of MYD88 and CD79B mutations in CD5⁺ DLBCL is lower than that in DLBCL-SS, suggesting that CD5⁺DLBCL is not the same disease as DLBCL-SS.