Impact (in vitro and ex vivo) of CD14⁺HLA-DR^lo cells from CLL patients on T cells. (A) The dose-dependent suppressive activity of FACS-sorted CD14⁺HLA-DR^lo cells from CLL patients (n = 4) was evaluated in coculture experiments with autologous T cells activated by means of anti-CD2, -CD3, and -CD28 microbeads. Proliferation of T cells was assessed based on the VPD450 dye dilution after 5 days by FACS and compared with stimulated T cells alone (set as 100% T-cell proliferation). (B) Correlation of the T-cell count with the proportion of CD14⁺HLA-DR^lo cells in the peripheral blood of CLL patients. (C) Patients were grouped based on their MDSC (equals CD14⁺HLA-DR^lo cells) frequency (less than the median [n = 24] and greater than the median [n = 24]). Frequency of PD1⁺ cells among CD4⁺/CD8⁺ T cells was compared between the groups. (D) Time until doubling of the patients’ lymphocyte counts was compared in the 2 groups using a Gehan-Breslow-Wilcoxon test. (E) T-cell activation based on mean fluorescence index (MFI) levels of CD69, CD137, and HLA-DR, respectively, in the CD4⁺ and CD8⁺ T-cell compartment was compared between the groups. (F) Representative dot plot of a FACS-based T-cell maturation analysis. CD4⁺ and CD8⁺ T cells were divided into CD27⁺CD45RA⁺ naïve (NA) cells, CD27⁺CD45RA^neg central memory (CM), CD27^negCD45RA^neg effector memory cells, and CD27^negCD45RA⁺ effector memory RA⁺ cells (EMRA) cells. Proportions of these T-cell subsets were analyzed in both patient groups. The bars represent the standard error of the mean. *P < .05; **P < .01.