The hypomethylating agent azacytidine (AZA) represents the standard treatment for many high-risk MDS and AML patients. While the clinical efficacy has been confirmed in several studies, the precise molecular mechanism of action has not been fully understood yet.
Human NK-cells play an important role in the regulation of immune responses against malignant cells. Their function is controlled by a complex interplay of activating and inhibitory receptors - some of them being regulated by methylation of the respective genes. We, therefore explored, whether AZA modulates in vitro NK-cell function as well as in vivo during minimal-residual disease (MRD)-guided treatment of imminent relapse in MDS and AML patients treated within the prospective RELAZA trial (NCT00422890).
After purifying NK-cells of healthy donors by MACS (magnetic cell sorting), NK-cells were exposed in vitro to different concentrations of AZA (100nM, 1μM, 3μM) with or without IL-2. In parallel, the NK-cell phenotype of patients (n=12) with AML or MDS, undergoing MRD-guided treatment with AZA after stem cell transplantation was monitored by FACS from peripheral blood samples on day 1, 5 and 7 of the first and second AZA cycle. All patients were still in complete haematological remission at the time of therapy.
In vitro, we observed a significant reduction (3,1% to 1,8% p=0.028) of the immature and cytokine-regulating CD56bright NK-cell subpopulation with increasing concentrations of AZA. There was a trend towards a reduced expression of the death-ligand TRAIL, the activating receptors NKG2D and NKp46 and for an increased expression of the inhibitory KIR CD158b1/b2, whereas we could not detect any changes in the expression of FAS-L, Perforin, Granzyme B, NKp30, NKp44, CD69, CD57, DNAM-1, CD16, and NKG2A-CD94. Confirmatory, we observed a significant decrease in the expression of TRAIL (p=0.003), NKG2D (p=0.03) and NKp46 (p=0.006) during AZA treatment in-vivo. Interestingly, these changes appeared to be reversible.
The observed reduction of NK-cell activating receptors and TRAIL during AZA treatment correlated with a reduction or stable course of MRD in all analyzed patients.
In summary these data suggest that the clinical effects of AZA are not mediated by enhancing NK-cell activity. In fact, the drug may have inhibitory effects on NK-cell function which should be considered when applying AZA in the post-transplant setting.
Platzbecker:Celgene: Honoraria, Membership on an entity's Board of Directors or advisory committees, Research Funding.
Asterisk with author names denotes non-ASH members.