Cytokine-induced killer cells are in vitro expanded cells used in adoptive therapy for the treatment of solid and hematopoietic tumors. They express the T cell markers CD3 and CD56 and lack the cytotoxicity machinery of natural killer (NK) cells. Several papers have described the interactions between cytokine-induced killer cells and tumor targets. We focused our attention on the specific interplay between cytokine-induced killer cells and peripheral blood mononuclear cells in order to elucidate the possible consequences of this interaction. Cytokine induced killer cells were generated from peripheral blood mononuclear cells of 5 healthy donors by 21 days ex vivo expansion after priming with interferon gamma and monoclonal anti-CD3 antibody in medium supplemented with interleukin-2. After the expansion, cells were stained with CFSE and co-cultured with autologous peripheral blood mononuclear cells for 5 hours. The co-cultures were then stained with anti-CD14 antibody for monocytes, anti-CD19 antibody for B lymphocytes, anti-CD56 antibody as activation marker, anti-CD3, anti-CD4 and anti-CD8 for lymphocytes subpopulations, and analyzed by means of Imagestream and cytofluorimetric techniques. We observed that after the 5 hours incubation, the percentage of CFSE-CD14 double positive cells increased in a ratio dependent manner. The physical interaction was confirmed by Imagestream technique. We detected a small number of double positive CFSE-CD19 cells and we noticed a decrease in the total number of CD19 cells suggesting a possible specific cytotoxic activity against B cells. We confirmed this in vivo noticing that two patients undergoing CIK therapy showed a transient reduction in number of circulating CD19. The direct toxicity versus healthy circulating B cells could explain why this therapy leads to the maximum therapeutic benefit in the treatment of hematological malignancies. We noticed as well in vitro a CD56 increased expression on CIK cells that could be explained by the increased activation after monocyte contact. Further investigation is required to elucidate whether this activation could be different in the presence of suppressive monocytes. Our results could be of crucial relevance especially in the setting of solid tumor where the strong immunosuppressive context could negatively influence the activation status of the adoptively infused cells.