Role of regulatory T cells in acute myeloid leukemia patients undergoing relapse-preventive immunotherapy

Abstract

Regulatory T cells (Tregs) have been proposed to dampen functions of anti-neoplastic immune cells and thus promote cancer progression. In a phase IV trial (Re:Mission Trial, NCT01347996, http://www.clinicaltrials.gov ) 84 patients (age 18-79) with acute myeloid leukemia (AML) in first complete remission (CR) received ten consecutive 3-week cycles of immunotherapy with histamine dihydrochloride (HDC) and low-dose interleukin-2 (IL-2) to prevent relapse of leukemia in the post-consolidation phase. This study aimed at defining the features, function and dynamics of Foxp3+CD25highCD4+ Tregs during immunotherapy and to determine the potential impact of Tregs on relapse risk and survival. We observed a pronounced increase in Treg counts in peripheral blood during initial cycles of HDC/IL-2. The accumulating Tregs resembled thymic-derived natural Tregs (nTregs), showed augmented expression of CTLA-4 and suppressed the cell cycle proliferation of conventional T cells ex vivo. Relapse of AML was not prognosticated by Treg counts at onset of treatment or after the first cycle of immunotherapy. However, the magnitude of Treg induction was diminished in subsequent treatment cycles. Exploratory analyses implied that a reduced expansion of Tregs in later treatment cycles and a short Treg telomere length were significantly associated with a favorable clinical outcome. Our results suggest that immunotherapy with HDC/IL-2 in AML entails induction of immunosuppressive Tregs that may be targeted for improved anti-leukemic efficiency.

Keywords: Acute myeloid leukemia; IL-2; Immunotherapy; Regulatory T cells.

Conflict of interest statement

Authors Mats Brune and Kristoffer Hellstrand are past or present consultants to the study sponsor (Meda Pharma). Authors Frida Ewald Sander, Kristoffer Hellstrand, Fredrik B. Thorén and Anna Martner hold issued or pending patents protecting the use of histamine dihydrochloride in cancer immunotherapy. Authors Anna Martner, Robin Foà and Fredrik B. Thorén have received honoraria and/or travel grants from the study sponsor. Author Anders Ståhlberg declares stock ownership in TATAA Biocenter. The other authors declare no conflict of interest.

Figures

Fig. 1

Tregs expand during immunotherapy with HDC/IL-2. a Representative dot plots of Tregs (defined as Foxp3+CD25highCD4+) before (cycle 1, day 1, C1D1) and after (C1D21) the first HDC/IL-2 treatment cycle. b Box plots represent blood counts of Tregs before (D1) and after (D21) cycles 1 and 3 of immunotherapy (C1D1 n = 59, C1D21 n = 53, C3D1 n = 51, C3D21 n = 50, Student’s paired t test). c Patients were dichotomized by the median for low number of Tregs in black and high number of Tregs in red, at onset of immunotherapy (C1D1; left panel) or end of cycle 1 (C1D21; mid panel). The right panel shows the LFS of patients with low or high induction of Treg cell numbers during the first treatment cycle as analyzed by the log-rank test

Fig. 2

Expanded Tregs resemble thymic-derived nTregs. a Methylation pattern of 15 CpG islands in the TSDR, located in the FOXP3 gene locus, for sorted Tcons from healthy donors (n = 2), sorted Tregs from healthy donors (n = 8) and sorted Tregs from Re:Mission patients (n = 9) with samples collected after treatment cycle 3 (C3D21). The color code indicates percentage methylation of each CpG island with yellow representing absence of methylation and blue 100% methylation. NA not analyzed. bBars show the mean methylation of each CpG-site for healthy donors (n = 8) and Re:Mission trial patients (n = 9). Error bars display standard error of the mean (SEM). c Expression of Helios in Tregs before and after cycle 1 and cycle 3 of treatment with HDC/IL-2 (C1D1 n = 16, C1D21 n = 22, C3D1 n = 13, C3D21 n = 14). Statistical analyses were performed by Student’s paired t test

Fig. 3

Expanded Tregs from Re:Mission trial patients are immunosuppressive. Median fluorescence intensity (MFI) of CTLA-4 on Tregs (a) and Tcons (b) in patient blood before and after treatment cycles 1 (C1D1 n = 19, C1D21 n = 25) and 3 (C3D1 n = 16, C3D21 n = 17). c Representative histograms of Tcon proliferation from a healthy donor and a Re:Mission patient. Black lines show the proliferation of Tcons in wells without Tregs and red shaded areas show proliferation of Tcons when Tregs were added in a ratio of 1:1. Division index (d) and proliferation index (e) are shown for Tcons from healthy donors (n = 5) and Re:Mission trial patients (n = 4) at the end of treatment cycle 3. Statistical analyses were performed by Student’s paired t test

Fig. 4

Expansion of Tregs is reduced in later cycles of immunotherapy. Box plots display (a) the frequency of Tregs within the CD4+ compartment (C1D1 n = 59, C1D21 n = 63, C3D1 n = 52, C3D21 n = 53), and (b) frequency of NK cells as percentage of lymphocytes (C1D1 n = 62, C1D21 n = 63, C3D1 n = 53, C3D21 n = 53), before (D1) and after (D21) the first and third HDC/IL-2 treatment cycle. Statistical analyses were performed by Student’s paired t test. c Patients were dichotomized by the median for low (black) or high (red) reduction in Treg percentage from the end of cycle 1 to the end of cycle 3, and LFS was analyzed by the log-rank test. d Relative telomere length of Tregs FACS-sorted from patient blood obtained before and after the third treatment cycle or from healthy blood donors (Ctrl). e Kaplan–Meier plot comparing the LFS of patients with Treg telomere lengths on C3D21 below (black) and above (red) the median (log-rank test)