CD56+bright and CD56+dim natural killer cells in patients with chronic myelogenous leukemia progressively decrease in number, respond less to stimuli that recruit clonogenic natural killer cells, and exhibit decreased proliferation on a per cell basis.

Human natural killer cells (NK) require accessory cell-derived contact and soluble factors for maximal expansion. However, it is unclear whether increased recruitment of clonogenic NK, increased proliferation on a per cell basis, or a combination of both is responsible for the increased expansion. We show that expansion of both CD56+dim and CD56+bright NK from normal donors is increased in the presence of M2-10B4 accessory cell-soluble factors. In contrast, the addition of M2-10B4 stromal ligands further augments only the expansion of CD56+bright NK. Using single-cell sorting of CD56+bright NK, M2-10B4-soluble and contact factors independently increase both recruitment of clonogenic NK and proliferation on a per cell basis. This well-defined M2-10B4 accessory cell system was used to investigate potential defects in NK from patients with CML. Although we have previously shown diminished interleukin-2 (IL-2)-activated NK outgrowth and function from patients with chronic myelogenous leukemia (CML) as their disease progresses, it has been unclear if this is due to a defect in an accessory cell function or an inherent abnormality of CML NK themselves. CD56+/CD3- NK purified by fluorescence-activated cell sorting from 21 patients (7 early chronic phase [ECP] patients, 10 late chronic and accelerated phase [LCP/AP], and 4 blast crisis [BC] patients) were studied. The proliferative capacity, clonogenic frequency, and cytotoxic capacity of CML NK were compared with NK from normal donors. The absolute number of circulating NK per milliliter of peripheral blood is significantly decreased in patients with CML compared with normal donors (normal, 63,700 +/- 6,400; ECP, 40,700 +/- 6,700; LCP/AP, 31,900 +/- 6,000; BC, 10,700 +/- 5,200). Additionally, the unique CD56+bright NK subset, analyzed as a percentage of the total circulating NK pool, is significantly reduced in all patients with CML (normal, 5.7% +/- 0.8% v CML [all stages combined], 2.5% +/- 0.5%, P = .001]. After purification of NK to correct for differences in circulating NK number, resting NK cytotoxicity against K562 tumor targets is significantly reduced in patients with CML on or recently on hydroxyurea therapy. However, this reduced cytotoxicity can be corrected by 18 hours of incubation with 1,000 U/mL recombinant IL-2. When plated in limiting dilution on viable M2-10B4, which maximally stimulates NK from normal donors, we show that both NK clonogenic frequency and proliferative capacity are significantly reduced as CML progresses, demonstrating an inherent defect in their ability to respond to normal NK stimuli. Although NK cloning efficiency between normal donors and ECP CML patients was the same, significant differences were observed in (1) the absolute number of circulating CD56+/CD3- NK, (2) the absolute number of circulating CD56+bright NK, and (3) proliferation on a per cell basis. Unlike resting NK function, prior cytotoxic therapy alone did not account for these observed abnormalities. These data suggest that, although NK are not derived from the malignant clone, they are inherently affected by their malignant microenvironment.