Background: Immune editing is recognized as important in various pathogenic processes. We previously showed that regulatory T cell (Treg) levels were elevated in MDS with a poor prognosis and that Treg levels decreased with Azacitidine (AZA) treatment and became equivalent to that in the untreated group. In this study, we examined changes in the number of circulating Tregs, T cells, and NK cells in the peripheral blood of patients with MDS treated with AZA, and we evaluated the clinical significance of these findings with survival.

Patients and Methods: Patients with MDS and with acute myeloid leukemia with myelodysplasia-related changes (AML-MRC) according to the WHO 2008 classification who were initially treated between January 2015 and December 2016 and followed up for more than 6 months were enrolled in the study. AZA was given intravenously at a dose of 75 mg/m2 daily for either 5 or 7 consecutive days every 4 weeks. The response was assessed using the IWG response criteria for MDS. A total of 50 patients with AML-MRC (n=11), refractory anemia with excess blasts (RAEB) (n=8), refractory cytopenia with multilineage dysplasia (RCMD) (n=28), myelodysplastic/myeloproliferative neoplasms (MDS/MPN) (n=12), and refractory anemia with ring sideroblastis (RARS) (n=1) were included in the study. The IPSS was classified as low (n=15) intermediate-1 (n=13), intermediate-2 (n=10), and high (n=13). Twenty-six patients did not require treatment, including a transfusion (untreated group). Twenty-two patients were treated with AZA (AZA group). The remaining patients received conventional chemotherapy. After obtaining informed consent, blood was drawn from all of the patients at initial diagnosis. In the 22 patients treated with AZA, blood was also drawn after 6 completed treatment cycles. CD4+CD25+Foxp3+ Treg cells, CD3+CD4+ T cells, CD3+CD8+ T cells, and CD3-CD56+ NK cells were analyzed using flow cytometry, and the absolute numbers of these cells were calculated. Overall survival (OS) probabilities were estimated using the Kaplan-Meier method and compared by the log-rank test for univariate analysis.

Results: The median age was 74 years (range: 44-93 years), and 35 of the patients were male. The median period of observation was 391 days. In the AZA group, 10 patients died (AZA-ineffective group), and 12 had either a CR, PR, or HI (AZA-effective group). The 2-year OS of the AZA group was 43.7%. The absolute number of Tregs and Treg% values before treatment were not significantly different between the AZA group and the untreated group. The Treg levels in peripheral blood before treatment in the AZA-effective (0.80±0.39%) tended to be lower than that in the AZA-ineffective (1.29±0.82%) group (p=0.076). The CD8+ T cell counts before treatment were moderately higher in the AZA-effective group compared to that in the AZA-ineffective group (385.5 ± 129.0 vs. 195.5 ± 70.0 /μL, p<0.01). Similarly, the CD3-CD56+ NK cell counts before treatment differed significantly between the AZA-effective group (197.4 ± 138.8 /μL) and the AZA-ineffective group (82.3 ± 36.7 /μL) (p=0.015). In a univariate analysis, the MDS patients with elevated NK cells survived longer than the group with reduced NK cells (p=0.013). In a multivariate analysis, the lower percentage of Treg cells and the higher percentage of CD8+T cells before treatment were independent factors for OS (HR, 4.80; p=0.085, HR, 4.13; p=0.040, respectively). In the AZA group, the percentage of CD8+T cells before treatment associated with OS (HR, 39.24; p=0.018). After treatment, the AZA group had an increase in the NK cell counts, which tended to associate with survival (HR 4.29; p=0.067).

Conclusions: The number of Tregs in MDS cases requiring treatment was higher than that in MDS cases for which treatment was unnecessary. The prognosis of the patients with high CD8+ T cells counts was better. In contrast, there was an increase in the number of NK cells after AZA treatment, which was found to be associated with an improved prognosis. These results suggest that an immunological mechanism associated with the prognosis of MDS may be enhanced by treatment with AZA. Further examination of this mechanism and of the influence of AZA on the immunological effect are required.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.