Natural killer (NK) cells and bone marrow transplant (BMT) in severe combine combined immunodeficiency (SCID)

Abstract

Rationale Natural killer (NK) cells are thought to have a possible role in marrow graft acceptance; little has been published about NK effects in SCID patients before and after BMT. Methods SCID patients receiving T cell-depleted haplo-identical or HLA-identical BMT without pre-BMT chemotherapy at Duke from May, 1982 through July, 2003 were eligible. NK and T cells were enumerated by flow cytometry, and T cell function was assessed by determining responses to mitogens. Stimulation indices (SI) (PHA cpm/medium cpm) >/=100 were normal. Demographic and survival data were collected. Statistical analyses were performed using SAS (SAS institute, Cary, NC). Results BMT was performed in 132 patients (mean age 178 days, 83% male, 47% γc deficiency, 77% survive). Sixty-four patients had sufficient data for pre-BMT analyses, 61 for post-BMT. The median pre-BMT NK count didn't differ significantly (p=0.81, Wilcoxon rank sum) between patients who developed an SI >100 (N=48, 61.5 cells/cmm, interquartile range (IQR) 134) and those who did not (N=16, 33.8 cells/cmm, IQR 141). More patients with a post-BMT NK peak and T cells (CD3 >/=2000/cmm) developed SI >/=100 (17/17, 100%) than patients with no NK peak (17/23, 74%) or an NK peak but few T cells (12/21, 57%) (p=0.0037, Fisher's exact). The time to an SI >/=100 did not differ significantly among the three groups (p=0.37, Kruskal-Wallis), nor did survival (p=0.18, Fisher's exact). Conclusions With no pre-BMT chemotherapy, pre-BMT NK cells were not associated with graft failure. More patients with post-BMT NK peaks and T-cells developed SI >/=100, but survival was not significantly improved.