Attempt to assess direct interactions between tumor burden, myeloid-derived suppressor cells and PD-1- and TIM-3-expressing T cells in multiple myeloma patients

The avoidance of immune surveillance by malignant plasma cells (PCs) in multiple myeloma (MM) is mediated by different mechanisms, among which an induction of T cell exhaustion and expansion of myeloid-derived suppressor cells (MDSCs) appear to play substantial roles, but it is still a lack of data on possible MDSC-mediated induction of T cell exhaustion. The aim of the present work was to evaluate possible relationship between frequencies of MM PCs, MDSCs and phenotypically exhausted PD-1+ and TIM-3+ T cells in bone marrow (BM) samples and peripheral blood (PB) of MM patients at various disease stages. Peripheral blood (n = 88) and BM samples (n = 56) were obtained from MM patients (newly diagnosed (n = 6), patients in remission (n = 71) and with progressive disease (n = 11)). Frequencies of T cells expressing checkpoint receptors PD-1 and TIM-3, polymorphonuclear MDSCs (PMN-MDSCs, Lin-CD14-HLA-DR- CD33+CD15+/CD66b+), monocyte MDSCs (M-MDSCs, CD14+HLA-DRlow/-), early MDSCs (E-MDSCs, Lin-HLA-DR-CD33+CD15-/CD66b-), and MM PCs (CD45dimCD38+CD138+CD56+CD19-CD117+CD27- CD81-) were assessed with flow cytometry. Circulating and BM-resident PD-1+/TIM-3+T cell subsets, BM E-MDSCs, as soon as MM PCs and serum beta2-microglobulin (B2-M) levels were gradually increased in patients at different stages. Despite that, there were no associations between the markers of tumor load and the studied cell subsets. In patients in remission, BM PMN-MDSCs negatively correlated with CD4+T cells, CD4+PD-1+ and CD8+TIM-3+T cell subsets; there were positive correlations between BM E-MDSCs and CD4+PD-1+TIM-3+ cells and PB M-MDSCs and CD8+PD-1+ and (as a trend) CD8+TIM-3+T cells. We found no associations for the samples of patients at diagnosis and with progression. We can conclude that a possible mutual influence of malignant PCs, MDSCs and PD-1+/TIM-3+T cells is nonlinear, especially during a manifest tumor growth at diagnosis and progression. The detected negative correlations between resident PMN- MDSCs and T cell subsets might be associated with MDSC suppressive function, affecting both predominantly activated PD-1+ cells and exhausted TIM-3+ subsets. The positive correlations between BM E-MDSCs and CD4+PD-1+TIM-3+ cell subset and circulating M-MDSCs and PD-1+ and TIM-3+ CD8+T cells might confirm an ability of MDSCs to induce T cell exhaustion.