Increased Tumor-Associated CD66b+ Myeloid-Derived Suppressor Cells in Waldenstrom Macroglobulinemia Inhibit T-Cell Immune Function

Waldenstrom macroglobulinemia (WM) is a discrete clinicopathologic entity resulting from the accumulation of lymphoplasmacytic lymphoma (LPL) cells in the bone marrow (BM) that secrete a monoclonal immunoglobulin (Ig)M protein. MYD88 (>90%) and CXCR4 (30-35%) mutations are the most commonly identified mutations in this disease. While much is known about the genomics of WM, far less is known about the role of the BM microenvironment in this disease. Emerging evidence has attested to the role of myeloid-derived suppressor cells (MDSCs) and their potent immunosuppressive activity, however, there is limited information about these cells in WM and monoclonal gammopathy of undetermined significance (MGUS). This study was therefore designed to explore the phenotype of MDSCs and their interaction with other cells in the BM in WM. A multiparametric approach was combined with single-cell genomic and proteomic tools to better understand MDSC immunology and the antitumor immune response in WM. In the present study, using flow cytometry, we first analyzed the number of MDSCs (by gating on Lin- HLA-DR- CD11b+ CD33+) in normal BM (NBM; n=11) and WM symptomatic /MGUS (n=18) and found an increased number of MDSCs in WM specimens compared to controls (p=0.005). We also found that most MDSCs in WM specimens had a granulocytic phenotype (Lin- HLA-DR- CD11b+ CD33+ CD15+). A detailed phenotypic study using CyTOF (mass spectrometry) was performed on MDSCs from NBM (n=4) and WM (n=8). The CyTOF analysis identified 3 subtypes of MDSCs including CD66b+ MDSCs with a granulocytic phenotype (G-MDSCs) and confirmed the expansion of CD66b+ MDSCs in WM symptomatic patients when compared to smoldering WM and normal controls. Our previous studies have shown that granulocyte colony-stimulating factor (G-CSF) expression is increased in the bone marrow of WM (Jalali et al., 2018). Therefore, to determine whether G-CSF accounted for the expansion of CD66b+ G-MDSCs, we treated WM-BM MDSCs with G-CSF and found that the CD66b+ G-MDSC population substantially increased (p=0.0017). Further, to establish a better understanding of the transcriptome profile and a unique phenotype of CD66b+ G-MDSCs, we performed Cellular Indexing of Transcriptomes and Epitopes by Sequencing (CITE-seq) on WM (n=3) and NBM (n=2) specimens. A significantly high expression of NLRP12, IGFBP7, CXCR4, IL2, CD55, SOX4, and FoxP1 genes was observed in CD66b+ MDSCs as compared to other MDSC populations. This gene expression signature suggested upregulation of inflammatory pathways potentially affecting immune function. Therefore, to evaluate the effect of CD66b+ G-MDSCs on the immune function of other cells in the BM, CD66b+ and CD66b- MDSCs were co-cultured with activated autologous T-cells. We found that while all populations of MDSCs suppressed T-cell activation, CD66b+ G-MDSCs had the greatest suppressive effect. In conclusion, this study has shown that CD66b+ G-MDSCs are expanded in WM patients, exhibit a unique inflammatory transcriptome, and substantially inhibit T-cell activation and proliferation. Furthermore, CD66b+ G-MDSCs are expanded due to increased G-CSF in the BM microenvironment in WM. Clinical strategies to inhibit the expansion of CD66b+ G-MDSCs may therefore enhance the efficacy of immunological therapies in patients with WM.