CD3+/CD56+ Cells, but Not Natural Killer T Cells, Are Increased in Peripheral Blood of Untreated Patients with Leukemia.

Despite recent interest in the use of natural killer T (NKT) cells for cancer immunotherapy, there remains confusion over the phenotype and function of these cells. We observed a population of CD3+/CD56+ cells in the peripheral blood (PB) of patients with acute myeloid leukemia (AML) at diagnosis. We therefore performed more detailed multiparametric flow cytometric analysis on PB from previously untreated patients with AML (n=30), chronic lymphocytic leukemia (CLL) (n=12) and healthy normal volunteers (n=17) to characterize these cells in more detail. Since we were assessing rare cell populations, at least 200,000 events were always acquired from the mononuclear and live cell gates. In healthy individuals CD3+/CD56+ cells comprise 1% to 8.6% (median 2.74%) of PB T cells, but these cells are significantly increased both in percentage of T cells as well as in absolute numbers in the PB of patients with AML (p=0.0009) and CLL (p<0.0001). The percentage of these cells expressing the activation markers CD25 and CD69, is significantly higher in patients with AML (p=0.0007), but not CLL (p=0.67). We next assessed whether there was an increase in true NKT cells identified using anti-CD3 mAb and CD1d molecule loaded with alpha-galactosylceramide tetramer (kindly provided by the NIH tetramer facility). A tetramer consisting of an unloaded CD1d molecule was used as a negative control. Results were confirmed using fluorochrome-labelled antibodies to the T cell receptor chains Valpha24 and Vbeta11. Consistent with previous reports, true NKT cells ranged from undetectable to 1.59% (median 0.03%) of PB T cells in normal individuals. These NKT cells express CD4 or CD8alpha but few express the NK markers CD16 or CD56. There was no significant increase in true NKT cells in percentage or absolute numbers in AML (p=0.56) or CLL (p=0.4), however there are significant differences in subsets of these NKT cells in patients with AML compared with normal individuals. Taken together, these results demonstrate an increase in CD3+/CD56+ cells, but not true NKT cells, in PB of patients with acute and chronic leukemias. CD3+CD56+ cells do not represent NKT cells in health or in disease and true NKT cells require specific characterization. Functional analysis of both CD3+/CD56+ cells as well as true NKT cells in their ability to recognize and lyze AML and CLL cells is underway.