Abnormal Numbers Of Regulatory T Cell (Tregs) Subsets Are Sigificantly Associated With Adverse Disease Outcome In Lower Risk Myelodysplastic Syndromes (MDS) and Chronic Myelomonocytic Leukemia (CMML)

Expansion of naturally occurring regulatory T cells (rTregs) and in particular specific subsets including the highly suppressive effector CD4⁺CD45RA^-CD25^brightFoxp3^bright Tregs (eTregs) limit impaired immune responses. On the other hand, malignancies promote expansion of Tregs, which in turn play a central role to suppress antitumor immune responses. In line, increased numbers of Tregs were found in the bone marrow (BM) and peripheral blood (PB) in high risk MDS and AML (Kordasti; Blood 2007 110:847). It has recently been shown that an increase of highly suppressive effector memory Tregs was associated with anemia, increased bone marrow blasts and poor prognosis in lower risk MDS independently of IPSS, suggesting that expansion of a specific Tregs subset, rather than expansion of Tregs as a whole, characterizes higher risk patients (Mailloux; J Immunol 2012 189:3198). We screened a large cohort of 127 consecutive MDS and CMML patients for Tregs subsets and studied their influence on outcome.

85 MDS, 42 CMML patients and 34 sex and age matched controls were studied. PBMC were purified from blood samples on Ficoll-Hypaque gradients to determine populations of resting CD4⁺CD45RA⁺FoxP3⁺CD25⁺ Tregs (rTregs) and effector CD4⁺CD45RA^-CD25^brightFoxp3^brightTregs (eTregs) by flow cytometry and data was correlated with patient characteristics and outcome.

Median age of the 85 MDS and the 42 CMML patients was 74 years (range 31-91) and 75.5 years (range 47-91) respectively, with M/F: 75/52. Eight (6.3%) patients had RA, 8 (6.3%) RARS, 17 (13.4%) RCMD, 8 (6.3%) 5q-syndrome, 22 (17.3%) RAEB-1, 10 (7.9%) RAEB-2, 11 (8.7%) AML 20-30% blasts/RAEB-T, 37 (29.1%) CMML-1, 5 (3.9%) CMML-2 and 1 (0.8%) unclassified MDS. For patients evaluable by IPSS (MDS, AML 20-30% and CMML with WBC <13 G/l), IPSS was low in 71 (70.3%), int-1 in 9 (8.9%), int-2 in 10 (9.9%) and high in 11 (10.9%) patients. Median marrow blast count was 5% (range 1-26). Overall median numbers of eTregs did not significantly differ in patients and controls: 1.98% of lymphocytes (range 0.15-9.5%) for MDS, 2.03% (range 0-11.8%) for CMML and 1.84% (0.42-4.54%) for controls (p=0.88). Likewise rTregs median numbers did not differ between patients and controls: 0.7% (range 0-7.03%) for MDS, 0.74% (range 0-9.32%) for CMML and 0.80% (0.15-2.66%) for controls (p=0.90). However, eTregs numbers were significantly more frequently abnormal (increased or decreased) in MDS (37/85, 44%) and CMML (24/42, 57%) compared to controls (6/34,17%) (MDS vs Ctrl, p=0.011, CMML vs Ctrl, p=0.001). By contrast, frequency of abnormal rTregs numbers was similar across the groups: (57/85 (67%) in MDS, 30/42 (71%) in CMML, and 20/34 (59%) in controls. In lower risk patients, survival was significantly influenced by Tregs numbers: median survival from Treg analysis was 25 months in patients with abnormal eTregs numbers compared to 28 months in lower risk patients with normal eTregs numbers (p=0.02). On the other hand, abnormal eTreg numbers had no impact on survival in higher risk patients (median 19.5 months from Treg analysis vs 17.5 months in patients with normal numbers, p=0.65). In the whole MDS and CMML population, eTregs numbers had a significant impact on the risk of disease progression (to higher risk MDS or AML) or death: median 27 months from Tregs analysis in patients with normal eTregs numbers vs 14 months in patients with abnormal numbers (p=0.05).

eTregs numbers are significantly more often increased or decreased in MDS and CMML patients than in controls. In lower risk MDS and CMML abnormal eTregs numbers are significantly associated with poorer survival and disease progression. In addition to the pathophysiological questions raised by those results, analysis of Tregs subsets by flow cytometry may constitute an interesting tool to predict outcome of MDS and CMML patients, in addition to classical prognostic scores.

No relevant conflicts of interest to declare.