Early Administration of Low-Dose IL-2 Intensify Graft-Versus-Leukemia Effect without Worsening GVHD through Sequential Enhancement of Effector T Cell and CD62L+ Regulatory T Cell Subset

Allogeneic hematopoietic stem cell transplantation (HSCT) is the curative treatment of many hematological malignancies, but the relapse is one of the major causes of treatment failure and mortality. Previous clinical studies have shown that low-dose IL-2 was well tolerated and could lower the relapse rate in human T cell depleted allogeneic BMT. On the other hand, we also demonstrated that low-dose IL-2 could preferentially enhance regulatory T cell (Treg) and improve the symptoms of chronic GVHD (NEJM 2011,Sci Trans Med 2013). In murine bone marrow transplantation (BMT) model showed that Treg can protect from lethal GVHD while preserving graft-versus-leukemia (GVL) effect. Especially, CD62L+ Tregs has a higher capacity in suppressing GVHD than their CD62L- counterpart. Our recent study reported that exogenous low-dose IL-2 induced the expression of PD-1 on Treg preferentially and it was most evident in CD44+CD62L+central-memory Treg (Blood 2016). However, the impact of CD62L+ Treg enhanced by low-dose IL-2 on GVHD and GVL has not been well investigated. In this study, we studied the effect of exogenous IL-2 on the lymphocyte phenotypic profiles including naïve- versus memory-type balance and also examined donor-derived immunity, especially GVHD and GVL, by using leukemia-bearing BMT model. Lethally irradiated B6D2F1 recipients were transplanted spleen cells and BM cells from B6 donors and 5000 IU Teceleukin (recombinant IL-2) was injected once a day for 14 days. The number of CD4+CD25+Foxp3+ Treg, CD4+CD25-Foxp3- conventional T cell (Tcon) and CD8+ T cell was compared between IL2- and vehicle-treated groups. The expressions of CD44, CD62L, CCR4, CCR7, Ki-67, PD-1, CTLA-4, LAG3, GITR, ICOS and TIM-3 in each subset were also examined. Lymphocytes profiling analysis showed that IL-2 treatment in the first 7 days resulted in the dominant expansion of effector T cells without Treg increase (Treg 0.10 vs 0.16,p=0.35, Tcon 2.83 vs 4.20, p=0.01, CD8 T cell 5.52 vs 10.4, p=0.01, million). On the contrary, the extended IL-2 treatment in the next 7 days increased CD62L+ Treg (21.5 vs 48.2 % of total Treg, p<0.001, 0.02 vs 0.07 million, p=0.006) without increasing effector T cells. PD-1 on Treg was very high in BMT-recipients even without receiving IL-2, so further enhancement of PD-1 expression by IL-2 was not observed. However, the expression of other inhibitory molecules including CTLA-4 (MFI, 6.61 vs 9.92, p=<0.001), LAG-3 (1.73 vs 2.17, p=0.04) and ICOS (2.41 vs 3.84, P<0.001) was enhanced by the IL-2 treatment. To elucidate the mechanisms of the increase in CD62L+ Treg, the in vivo violet dye dilution assay was performed. We first sorted out CD62L+ T cells by micro-beads from spleen cells, and then adaptively transferred into recipients. Five days treatment of low-dose IL-2 resulted in the enhancement of the division of CD62L+ Treg (% divided cells; 19.5 vs 29.8 % of Treg, p=0.005, 62.3 vs 108 in number, p=0.002). Interestingly, when CD62L- T cells were transferred, IL-2 treatment also induced the division of the cells and, of note, induced the phenotypic change of CD62L- into CD62L+ Treg (7.63 vs 19.8 % of Treg, p=0.009, 91.3 vs 361, p=0.01). These results suggested that IL-2 have the possibility to enhance the proliferative activity of Treg with the up-regulation of CD62L expression. Lastly, we transplanted lucipherase+P815 leukemia cells into this BMT model and evaluated the clinical impact of low-dose IL-2 on GVL by using in vivo bioluminescence system. To assesse the GVL and GVHD independently, we used two different tumor burden models. In the high tumor burden model, IL-2 injection decreased bioluminescence intensity (BLI) of P815 (BLI at week3, 8.3x10E7 vs 8.1x10E6 photon/second, p<0.001) and significantly extends overall survival. On the other hand, in the low tumor burden model, both treatment groups did not die from leukemia but vehicle-treated control group showed a significant higher clinical GVHD score and died from GVHD. These data demonstrated the clinical benefits on anti-tumor effect of IL-2, which could enhance the GVL activity without the detrimental effect on GVHD. In conclusion, our data indicate multiple effects of IL-2 on post-transplant immunity can coordinate to intensify GVL separately from GVHD on the mechanism of sequential enhancement of effector T cell and CD62L+ Treg in the different phases. These data provide the important information for developing Treg-targeted therapy.