Lack of Major Histocompatibility Class II (MHCII) Protein and Decreased Immunosurveillance in Plasmablastic Lymphoma (PBL)

Abstract 5186

PBL is a distinct clinicopathological entity classified separately from diffuse large B-cell lymphoma (DLBCL). A recent immunohistochemical (IHC) study described the plasmablastic phenotype of PBL including co-expression of PRDM1/BLIMP1 and XBP1 with lack of the B cell markers PAX5 and CD20. This protein expression profile is unusual for DLBCL and therefore helps to differentiate PBL tumors from conventional DLBCL. In a minority of DLBCL cases, the acquisition of a partial plasmablastic phenotype (Blimp1 positive) is associated with a worse outcome. Given that Blimp1 and MHCII expression are inversely related as normal B cells enter the terminal differentiation program towards plasma cells, and that loss of MHCII mRNA and protein expression correlates with poor outcome DLBCL (likely due to a loss of immunosurveillance), we hypothesized that PBL cases would also lack HLA-DR and have correspondingly low numbers of tumor infiltrating T cells.

Twenty-three cases of PBL, which were part of a previously published case series (S. Montes-Moreno et al, Haematologica 2010) from the Spanish National Cancer Research Center were studied with approval of the Carlos III Institutional Review Board. Cases were stained with antibodies specific to HLA-DR and CD8. Three consecutive 60x fields with a minimum of 950 cells were counted per case. As previously (L.M. Rimsza et al, Blood 2004) described, the area of tumor with the lowest frequency of CD8(+) cells was chosen for counting. The IHC results were quantified by counting the number of HLA-DR(+) cells and CD8(+) cells in the total number of malignant cells or lymphoid-appearing cells respectively (obvious stromal and histiocytic cells excluded). HLA-DR staining intensity was also scored as followed: 0 = no staining; 1+ = faint partial staining; 2+ = complete or partial moderate staining; 3+ = complete strong staining. Additionally, HLA-DR staining was characterized as surface membrane, cytoplasmic, or negative.

Only three PBL cases (13%) showed the typical B-cell pattern of HLA-DR cell surface membrane expression in a few (2% ± 2) of tumor cells with a faint to partial expression (median intensity of 1.8 ± 0.7), and average of 14% (±10) CD8(+) cells. Cytoplasmic HLA-DR expression in the absence of membrane expression was observed in 10 cases (43.5%) in a minority of cells (9% ±18) with a median intensity of 1.9+ (±0.6), and an average of 10.3% (±6) CD8(+) cells. Ten cases (43.5%) were completely negative for HLA-DR and showed only 7% (± 6) CD8(+) cells.

In summary, this study demonstrates the lack of MHCII protein expression on the surface membrane of most cases of PBL, which is associated with a decrease in CD8(+) tumor infiltrating T-cells cells, likely indicative of decreased immunosurveillance. These results are in agreement with our previous studies, in which DLBCL cases showed an average of 11% CD8(+) T-cells in the presence of MHCII cell surface expression, but only 2.8% CD8(+) T-cells in the absence of MHCII protein cell surface expression. The significance of the cytoplasmic localization is not clear, however may represent a stage of partial expression, which is associated with an intermediate level of T cells. Decreased immunosurveillance has long been correlated with deficient host response and tumor containment. The absence of MHCII protein expression may provide a reason for the poor outcome in PBL patients as well as serve as additional tool for diagnosis.