Combined loss of HLA I and HLA II expression is more common in the non-GCB type of diffuse large B cell lymphoma

A super-SILAC based proteomics analysis of diffuse large B-cell lymphoma-NOS patient samples to identify new proteins that discriminate GCB and non-GCB lymphomas

Diffuse large B-cell lymphoma—not otherwise specified (DLBCL-NOS) is a large and heterogeneous subgroup of non-Hodgkin lymphoma. DLBCL can be subdivided into germinal centre B-cell like (GCB) and activated B-cell like (ABC or non-GCB) using a gene-expression based or an immunohistochemical approach. In this study we aimed to identify additional proteins that are differentially expressed between GCB and non-GCB DLBCL. A reference super-SILAC mix, including proteins of eight B-cell lymphoma cell lines, was mixed with proteins isolated from seven non-GCB DLBCL and five GCB DLBCL patient tissue samples to quantify protein levels. Protein identification and quantification was performed by LC-MS. We identified a total of 4289 proteins, with a four-fold significant difference in expression between non-GCB and GCB DLBCL for 37 proteins. Four proteins were selected for validation in the same cases and replication in an independent cohort of 47 DLBCL patients by immunohistochemistry. In the validation cohort, we observed a non-significant trend towards the same differential expression pattern as observed in the proteomics. The replication study showed significant and consistent differences for two of the proteins: expression of glomulin (GLMN) was higher in GCB DLBCL, while expression of ribosomal protein L23 (RPL23) was higher in non-GCB DLBCL. These proteins are functionally linked to important pathways involving MYC, p53 and angiogenesis. In summary, we showed increased expression of RPL23 and decreased expression of GLMN in non-GCB compared to GCB DLBCL on purified primary DLBCL patient samples and replicated these results in an independent patient cohort.

CD47 Expression Defines Efficacy of Rituximab with CHOP in Non-Germinal Center B-cell (Non-GCB) Diffuse Large B-cell Lymphoma Patients (DLBCL), but Not in GCB DLBCL

Addition of rituximab (R) to "CHOP" (cyclophosphamide, doxorubicin, vincristine, and prednisone) chemotherapy improved outcome for diffuse large B-cell lymphoma (DLBCL) patients. Approximately 40% of patients who receive R-CHOP still succumb to disease due to intrinsic resistance or relapse. A potential negative regulator of DLBCL treatment outcome is the CD47 "don't eat me" immune checkpoint. To delineate the impact of CD47, we used a clinically and molecularly well-annotated cohort of 939 DLBCL patients, comprising both germinal center B-cell (GCB) and non-GCB DLBCL subtypes, treated with either CHOP or R-CHOP. High (above median) CD47 mRNA expression correlated with a detrimental effect on overall survival (OS) when DLBCL patients received R-CHOP therapy (P = 0.001), but not CHOP therapy (P = 0.645). Accordingly, patients with low CD47 expression benefited most from the addition of rituximab to CHOP [HR, 0.32; confidence interval (CI), 0.21-0.50; P < 0.001]. This negative impact of high CD47 expression on OS after R-CHOP treatment was only evident in cancers of non-GCB origin (HR, 2.09; CI, 1.26-3.47; P = 0.004) and not in the GCB subtype (HR, 1.16; CI, 0.68-1.99; P = 0.58). This differential impact of CD47 in non-GCB and GCB was confirmed in vitro, as macrophage-mediated phagocytosis stimulated by rituximab was augmented by CD47-blocking antibody only in non-GCB cell lines. Thus, high expression of CD47 mRNA limited the benefit of addition of rituximab to CHOP in non-GCB patients, and CD47-blockade only augmented rituximab-mediated phagocytosis in non-GCB cell lines. Patients with non-GCB DLBCL may benefit from CD47-targeted therapy in addition to rituximab.

The role of HLA variation in lymphoma aetiology and survival

Epidemiologic and laboratory evidence has consistently supported a strong inflammatory and immune component for lymphoma aetiology. These studies have consistently implicated variation in the immune gene, human leucocyte antigen (HLA), to be associated with lymphoma risk. In this review, we summarize the historical and recent evidence of HLA in both lymphoma aetiology and survival. The recent momentum in uncovering HLA associations has been propelled by the conduct of genome-wide association studies (GWAS), which has permitted the evaluation of imputed HLA alleles in much larger sample sizes than historically feasible with allelotyping studies. Based on the culmination of smaller HLA typing studies and larger GWAS, we now recognize several HLA associations with Hodgkin (HL) and non-Hodgkin lymphomas (NHLs) and their subtypes. Although other genetic variants have also been implicated with lymphoma risk, it is notable that HLA associations have been reported in every NHL and HL subtype evaluated to date. Both HLA class I and class II alleles have been linked with NHL and HL risk. It is notable that the associations identified are largely specific to each lymphoma subtype. However, pleiotropic HLA associations have also been observed. For example, rs10484561, which is in linkage disequilibrium with HLA-DRB1*01:01˜DQA1*01:01˜DQB1*05:01, has been implicated in increased FL and DLBCL risk. Opposing HLA associations across subtypes have also been reported, such as for HLA-A*01:01 which is associated with increased risk of EBV-positive cHL but decreased risk of EBV-negative cHL and chronic lymphocytic leukaemia/small cell lymphoma. Due to extensive linkage disequilibrium and allele/haplotypic variation across race/ethnicities, identification of causal alleles/haplotypes remains challenging. Follow-up functional studies are needed to identify the specific immunological pathways responsible in the multifactorial aetiology of HL and NHL. Correlative studies linking HLA alleles with known molecular subtypes and HLA expression in the tumours are also needed. Finally, additional association studies investigating HLA diversity and lymphoma survival are also required to replicate initial associations reported to date.