Low Counts Of Natural Killer Cells CD56 bright CD16 negative After Engraftment Are Associated With Worse Survival In Patients Receiving Allogeneic Hematopoietic Stem Cell Transplantation

Natural Killer Cells are innate immune system cells important in host defenses against viruses and tumor cells. Two subpopulations are well described: NK CD56bright CD16neg (NK56++16-, lower frequency on peripheral blood-PB, high cytokine production) and NK56dim16pos (NK56+16+, higher frequency on PB, high cytotoxic activity). They are activated through a balance between signals given from activating and inhibitory receptors (KAR and KIR, respectively). The ligands of KIRs are the MHC molecules and in the absence of compatible MHC, NK cells are activated. In the allogeneic hematopoietic stem cell transplantation (HSCT), recent studies showed that NK cells recovery is important on infection control and, in the presence of a KIR-MHC mismatch, they may be important on graft versus host disease (GVHD) and graft versus leukemia effects. However few studies evaluated NK subpopulations recovery and HSCT endpoints.To evaluate the impact of NK subpopulations recovery on HSCT endpoints: relapse, GVHD, non-relapse mortality and overall survival.NK (CD3-, CD56+) subpopulations (NK56++16- and NK56+16+) were quantified by multiparametric flow cytometry at 9 sequential time points (before conditioning, at engraftment, and at days 3, 7, 14, 21, 60, 100 and 180 after engraftment). Overall, 111 patients, from 4 HSCT centers (65% male, median age 17 years, range 1-74), receiving bone marrow (BM, 46%), umbilical cord (UCB, 32%) or peripheral blood (PB, 22%) from unrelated (n=90) or related donors (n=21) were studied. The most common diagnosis was acute leukemia (AML 36%, ALL 31%, MDS 9%, CML 9%, Aplastic anemia 8%). Most patients received myeloablative conditioning (MAC) regimens (60%). Antithymocyte globulin (ATG) was used in 44 patients (40%) and total body irradiation (TBI) in 56 (51%). Median follow up time was 14 months (range 4-35).Eighty-six patients presented sustained allogeneic recovery (no differences among sources). Of these, median time to neutrophil engraftment was 18 days (range: 8-52). The cumulative incidence (CI) of non-relapse-related mortality (NRM) was significantly higher in those with lower counts of NK56++16– during first 3 weeks after HSCT (34% at 1 year for patients with less than 30 cells/uL at day 21 vs 11% for patients with higher counts, p=0.03). Overall survival was significantly worse in patients with lower counts of NK56++16- subpopulations in the day 21 after engraftment (86% at 1 year vs 54% for patients with less than 30 cells/uL – p=0.003). CI of grade II-IV acute GVHD and relapse were not significantly affected by NK counts. The number of NK56+16+ cells did not affect any endpoint studied. Cell source, age and conditioning regimen did not affect any of the NK subpopulations counts. In multivariate analysis, NK56++16- counts lower than 30 cells/uL at 21 and 60 days after engraftment remain an independent risk factor for non relapse mortality [HR: 4.8, CI (95%): 1.2-18.8].Low NK56++16- counts in the first weeks after HSCT are associated with increased non relapse related mortality, but not acute GVHD or relapse. The mechanisms that rules the NK56++16- role on immunity deserve further investigations.No relevant conflicts of interest to declare.