Salvage therapy with azacitidine increases regulatory T cells in peripheral blood of patients with AML or MDS and early relapse after allogeneic blood stem cell transplantation

Article metrics

Relapse after allogeneic blood stem cell transplantation (allo-SCT) is a major cause of treatment failure in patients with acute myeloid leukemia (AML) or myelodysplastic syndromes (MDL), with most recurrences occurring within the first year after transplantation.1 Donor lymphocyte infusions (DLI) are a well-established salvage therapy in this situation. In contrast to the success seen in chronic myeloid leukemia, their efficacy in AML or MDS is restricted to some patients. In addition, this benefit is also hampered by the substantial risk to induce severe graft-versus-host disease (GVHD), especially when DLI are administered early after transplantation.2 Recently, the combination of azacitidine (Aza) and DLI has proven to induce sustained remissions in about one-third of the patients with AML or MDS relapsing after allo-SCT.3, 4, 5 The incidence and severity of GVHD following this combined approach has been reported to be lower in comparison with historical data using DLI alone.3, 4, 5, 6, 7 As a potential mechanism, murine models have suggested that Aza upregulates the transcription factor FoxP3, thereby expanding CD4+ regulatory T cells (Tregs), which play an important role in the control of GVHD.8, 9, 10 This has also been recently shown in 17 patients with AML receiving Aza maintenance therapy following allo-SCT.11 Data on Tregs in patients who receive Aza and DLI to treat relapse are lacking so far.

We serially monitored CD3+CD4+CD25+FoxP3+ Tregs and lymphocyte subsets in the peripheral blood (PB) of 13 patients with AML (n=8) or MDS (n=5) who received Aza and DLI as first salvage therapy for relapse after allo-hematopoeitic SCT.

Detailed demographics as well as relapse characteristics are summarized in Table 1. Seven of these patients were treated within a prospective multicenter phase II study (AZARELA, NCT-00795548) and their clinical results have recently been published,5 while the other six patients were treated accordingly at our institution. All patients gave written informed consent and the study was approved by the Institutional Review Board of The Heinrich Heine University Dusseldorf. The treatment schedule contained up to eight cycles of Aza (Vidaza, Celgene Corporation, Summit, NJ, USA) either 100 mg/m2/day on days 1–5 or 75 mg/m2/day subcutaneously on days 1–7 repeated every 28 days and DLI envisaged after every second Aza cycle (day 34/90/146) with increasing numbers of CD3+ cells. Additional DLI were permitted according to the individual physician's decision.5

Table 1 Patients characteristics and clinical course of patients

Hematological relapse occurred after a median of 446 days (range: 19–1688 days) following allo-SCT. The median number of Aza cycles was 6 (range: 4–8). DLI were administered in all patients, with a median number of 2 DLI per patient (range: 1–4 DLI). Five patients received one DLI, three patients two DLI, four patients three DLI and one patient four DLI, resulting in a median total T-cell dose of 5.0 × 106 CD3+ cells/kg per patient (range: 1–119 CD3+ cells/kg) (Table 1).

Following this treatment 8 of these 13 patients achieved a complete remission (CR). Seven of eight patients remained in remission for a median of 15 months (range: 7–42 months) without additional antileukemic treatment, while one patient relapsed again after 16 months. The CR rate of 62% is higher than CR rates that were previously reported (15–33%).3, 5, 6 This is due to a selection bias, as we included only patients in this immune monitoring study who had received at least four cycles of Aza in order to assure a serial measurement in individual patients.

During the course of treatment, Tregs (CD3+CD4+CD25+FoxP3+) and lymphocyte subpopulations including T cells (CD3+), T helper cells (CD3+/CD4+), cytotoxic T cells (CD3+/CD8+), NK cells (CD3-/CD56+) and B cells (CD20+) were measured by flow cytometry using a FACSCalibur (BD Biosciences, Heidelberg, Germany). Data were analyzed with FCS Express V3 software (De Novo Software, LA, CA, USA). Peripheral blood samples were routinely obtained prior to treatment, after the first (day 6), second (day 34), fourth (day 90) and sixth Aza cycle (day 146). For staining of Tregs and lymphocyte subsets the FoxP3 Staining Kit and the Multitest IMK Kit (both from BD Biosciences, Heidelberg, Germany) were used according to the manufacturer’s recommendations. The gating strategy to measure Tregs (CD3+CD4+CD25+FoxP3+) is shown in Figure 1a.

Figure 1

(a) Gating strategy to identify regulatory Tcells. Peripheral blood samples of patients were subjected to Ficoll density centrifugation and then analyzed by FACS. First, leukocytes were identified by their position in the SSC/FCS plot. Following gating on CD3+/CD4+ cells these cells were analyzed for CD25 and FoxP3 co-expression. The frequency of Tregs was calculated as the number of CD4+ cells that were CD4+CD25+FoxP3+. (b) Representative FACS plots showing FoxP3 expression on CD4+ cells of a patient with early relapse prior treatment and after four cycles of Aza. (c) Bars representing mean and s.e.m. for the absolute number of regulatory T cells in patients with early and late relapse. Wilcoxon signed rank test was used to detect statistically significant differences (Prism 5.01 GraphPad Prism Software Inc., La Jolla, USA). Asterisks display p-values. *P<0.05.

Looking at all the patients, we observed a 1.5-fold increase in the absolute number of Tregs in the PB after four Aza cycles (day 0: 8.23/μl vs day 90: 13.26/μl, P=0.0479), but this increase varied between the individual patients. By grouping the patients on the basis of the median time to relapse (day 446), we found a 3.2-fold increase in the absolute number (day 0: 4.7/μl vs day 90: 14.8/μl, P=0.031) as well as a 1.9-fold increase in the frequency of Tregs (day 0: 6.7% vs day 90: 12.9% of CD3+CD4+ cells, P=0.06) during treatment with Aza in the group of patients who relapsed early (Figures 1b and c). On the other hand, in those patients who relapsed late the absolute number (day 0: 12.2/μl vs day 90: 11.9/μl, NS) and frequency (day 0: 4.7% vs day 90: 3.9%, NS) of Tregs in PB was already higher and remained unchanged during treatment (Figure 1c). With regard to other lymphocyte subpopulations no significant changes were observed (Supplementary Figure S1).

The finding that Aza-induced expansion of Tregs is apparently restricted to patients relapsing early after allo-SCT is in line with data recently published by Goodyear et al.11 They reported a significant increase of the Treg numbers (>3-fold) in patients treated with Aza as maintenance therapy in comparison to time-matched controls. These patients commenced Aza therapy at 1–7 months following allo-SCT. Similar to the time dependency in our study, Goodyear et al.11 observed the greatest effect on Treg expansion after three cycles of Aza, reflecting an early period following allo-SCT, while no difference was observed after six or nine cycles.11

The expansion of Tregs during Aza treatment might also explain the relatively low rate and mild presentation of GVHD in our patients despite a dose-escalating DLI schedule. Acute GVHD (aGVHD) occurred in five patients (38%, grade I 3 patients, grade III 2 patients), beginning at a median of 129 days (range: 20–253 days) following the first DLI, while chronic GVHD (cGVHD) developed in six patients (46%, limited 5 patients, extensive 1 patient). Additionally, at the beginning of Aza treatment three patients were still on immunosuppression, which could be tapered in all cases without GVHD flare (Table 1).

Correlating with the time dependency of the Aza-induced Treg expansion, only one of the patients with early relapse (8%) developed aGVHD (grade I) in contrast to four patients with aGVHD in those with late relapse (30%).

In conclusion, our data support preliminary results from murine and human studies that Aza has the capacity to increase circulating Tregs, especially in patients relapsing early after allo-SCT. As Tregs are important to control GVHD, this phenomenon could be responsible for the low rate of GVHD seen after DLI that were given along with Aza to treat relapse after allo-SCT. However, recent results obtained from MDS patients treated with Aza in the non-transplant setting suggest that FoxP3+ T cells expanded during Aza treatment might exhibit different functional characteristics in comparison with naturally occurring Tregs.12 In the study of Costantini et al.12 the absolute number and frequency of Tregs at diagnosis correlated with the likelihood to respond to Aza, a finding that was not evident in our patients after allo-SCT (data not shown). Therefore additional studies are required to further characterize the immunomodulatory functions of Aza in the context of post-allo-SCT maintenance and salvage treatment.


  1. 1

    Pavletic SZ, Kumar S, Mohty M, de Lima M, Foran JM, Pasquini M et al. NCI First International Workshop on the Biology, Prevention, and Treatment of Relapse after Allogeneic Hematopoietic Stem Cell Transplantation: Report from the Committee on the Epidemiology and Natural History of Relapse following Allogeneic Cell Transplantation. Biol Blood Marrow Transplant J Am Soc Blood Marrow Transplant 2010; 16: 871–890.

  2. 2

    Roddie C, Peggs KS . Donor lymphocyte infusion following allogeneic hematopoietic stem cell transplantation. Expert Opin Biol Ther 2011; 11: 473–487.

  3. 3

    Czibere A, Bruns I, Kroger N, Platzbecker U, Lind J, Zohren F et al. 5-Azacytidine for the treatment of patients with acute myeloid leukemia or myelodysplastic syndrome who relapse after allo-SCT: a retrospective analysis. Bone Marrow Transplant 2010; 45: 872–876.

  4. 4

    Graef T, Kuendgen A, Fenk R, Zohren F, Haas R, Kobbe G . Successful treatment of relapsed AML after allogeneic stem cell transplantation with azacitidine. Leuk Res 2007; 31: 257–259.

  5. 5

    Schroeder T, Czibere A, Platzbecker U, Bug G, Uharek L, Luft T et al. Azacitidine and donor lymphocyte infusions as first salvage therapy for relapse of AML or MDS after allogeneic stem cell transplantation. Leukemia 2013; 27: 1229–1235.

  6. 6

    Lubbert M, Bertz H, Wasch R, Marks R, Ruter B, Claus R et al. Efficacy of a 3-day, low-dose treatment with 5-azacytidine followed by donor lymphocyte infusions in older patients with acute myeloid leukemia or chronic myelomonocytic leukemia relapsed after allografting. Bone Marrow Transplant 2010; 45: 627–632.

  7. 7

    Schmid C, Labopin M, Nagler A, Bornhauser M, Finke J, Fassas A et al. Donor lymphocyte infusion in the treatment of first hematological relapse after allogeneic stem-cell transplantation in adults with acute myeloid leukemia: a retrospective risk factors analysis and comparison with other strategies by the EBMT Acute Leukemia Working Party. J Clin Oncol 2007; 25: 4938–4945.

  8. 8

    Choi J, Ritchey J, Prior JL, Holt M, Shannon WD, Deych E et al. In vivo administration of hypomethylating agents mitigate graft-versus-host disease without sacrificing graft-versus-leukemia. Blood 2010; 116: 129–139.

  9. 9

    Edinger M, Hoffmann P, Ermann J, Drago K, Fathman CG, Strober S et al. CD4+CD25+ regulatory T cells preserve graft-versus-tumor activity while inhibiting graft-versus-host disease after bone marrow transplantation. Nat Med 2003; 9: 1144–1150.

  10. 10

    Sanchez-Abarca LI, Gutierrez-Cosio S, Santamaria C, Caballero-Velazquez T, Blanco B, Herrero-Sanchez C et al. Immunomodulatory effect of 5-azacytidine (5-azaC): potential role in the transplantation setting. Blood 2010; 115: 107–121.

  11. 11

    Goodyear OC, Dennis M, Jilani NY, Loke J, Siddique S, Ryan G et al. Azacitidine augments expansion of regulatory T cells after allogeneic stem cell transplantation in patients with acute myeloid leukemia. Blood. 2012; 119: 3361–3369.

  12. 12

    Costantini B, Kordasti S, Kulasekararaj AG, Jiang J, Seidl T, Perez Abellan P et al. The effects of 5-azacytidine on the function and number of regulatory T-cells and T-effectors in myelodysplastic syndrome. Haematologica 2012; e-pub ahead of print 24 December 2012.

Download references


We would like to thank the staff of the transplant unit of the Department of Hematology, Oncology and Clinical Immunology for excellent patient care, as well as Annemarie Koch for excellent laboratory work. This work was supported by Celgene Corporation, Germany, Leukämie Lymphom Hilfe Duesseldorf e.V., and Forschungskommission of the Heinrich Heine University Duesseldorf.

Author information

Correspondence to T Schroeder.

Ethics declarations

Competing interests

TS received financial travel support and honoraria from Celgene Corporation, Germany. UP received advisory fees, honoraria and research funding from Celgene Corporation, Germany. GB and RF received advisory fees, honoraria, travel grants and research funding from Celgene Corporation, Germany. UG received honoraria and research funding from Celgene Corporation, Germany. NK and GK received research funding from Celgene Corporation, Germany. The remaining authors declare no conflict of interest.

Additional information

Supplementary Information accompanies this paper on the Leukemia website

Supplementary information

Rights and permissions

Reprints and Permissions

About this article

Further reading