The Proportion of Natural Regulatory T-Cells CD4⁺CD25⁺Foxp3⁺ in Chronic Phase Chronic Myelogenous Leukemia Patients Correlates with Sokal Score at Diagnosis and Is Increased during Imatinib Mesylate Treatment.

Background. Natural regulatory T cells (nTregs) are key players in immune homeostasis and protection towards autoimmunity. There is also strong evidence that Tregs are involved in the immune recognition of tumor-related antigens. However, little is known about the role of nTregs in Chronic Myelogenous leukaemia (CML), an immunogenic tumor.

Aims. The aim of this study is to determine if the proportions of nTregs are significantly modified in CML patients at diagnosis or during imatinib mesylate (IM) treatment when compared to healthy counterparts, and if they correlate with prognosis factors and/or tumor burden.

Patients. Fourty-three CML patients were enrolled, 12 at diagnosis (median age 55 (range 32–77) years) and 31 during IM treatment (median age 56 (29–84) years) alone (n=21) or in combination with cytarabine (n=4) or Peg-interferon (n=6). Four patients received interferon before IM. They were compared to 13 healthy donors (median age 42 (21–56) years). We also analysed if the proportion of nTregs correlates with diagnosis (Sokal was low for 6, intermediate for 6 for the 12 patients analysed at diagnosis) or on-going prognosis factors (normalized bcr-abl/abl ratio using PB RQ-PCR).

Methods. Mononuclear cells were separated from PB by density gradient centrifugation and subsequently frozen until flow cytometry analysis. Before use, the CD4+ T-cell population was positively purified with immunomagnetic microbeads. Using a sequential gating strategy, nTregs were identified as CD4+CD25+Foxp3+ T-cells and proportions were expressed as a percentage of total CD4+ T-cells. Statistical analyses are based on at least 30,000 events gated on CD4+CD25+ T-cells. All comparisons were adjusted with patients or donors age at sample collection.

Results. We confirm that nTregs percentages increase with age both in patients and healthy donors (p=0.023). At diagnosis, whilst there is no significant difference in nTregs between patients and healthy donors, the percentage of nTregs is found to correlate with Sokal score with a marked increase of nTregs for patients with aggressive disease compared to patients with less aggressive disease at diagnosis (p<0.001). In patients under IM, a higher nTregs median percentage is found (6.88% +/− 2.56%) compared both with healthy donors (4.32% +/− 1.88%, p=0.035) or patients at diagnosis (4.06% +/− 2.23%, p=0.002). Lastly, while we do not find a correlation between nTregs and response to treatment assessed by RQ-PCR, we show that the percentage of nTregs decreases along the IM-treatment duration (p=0.024). No significant correlations were found between proportion of nTregs and association with cytarabine or peg-interferon, or with interferon prior IM. CML patients CD4+CD25+ nTregs functional analysis is pending.

Conclusions. In summary, we show that a high proportion of nTregs in CML patients PB at diagnosis is significantly associated with higher Sokal scores which predicts poorer outcome. However, we do not demonstrate an increase in nTregs at CML diagnosis compared to healthy donors, as found in other malignant diseases. On the opposite, patients with IM present a significant increase in nTregs, which slightly decrease over time regardless of their response to treatment. Further experiments and incremental numbers of patients with high risk Sokal score are needed to confirm these preliminary data.