TIM-3 Is a Promising Target to Eradicate Acute Myeloid Leukemia Stem Cells Sparing Normal Hematopoietic Stem Cells

Abstract 559

Acute myeloid leukemia (AML) originates from self-renewing leukemic stem cells (LSCs), an ultimate therapeutic target for permanent cure. To selectively kill AML LSCs sparing normal hematopoietic stem cells (HSCs), one of the most practical approaches is to target the AML LSCs-specific surface or functionally indispensable molecules. Based on differential transcriptome analysis of prospectively-purified CD34⁺CD38⁻ LSCs from AML patient samples and normal HSCs, we found that T-cell immunoglobulin mucin-3 (TIM-3) was highly expressed in AML LSCs but not in normal HSCs (Kikushige et al., Cell Stem Cell, 2010). In normal hematopoiesis, TIM-3 is mainly expressed in mature monocytes and a fraction of NK cells, but not in granulocytes, T cells or B cells. In the bone marrow, TIM-3 is expressed only in a fraction of granulocyte/macrophage progenitors (GMPs) at a low level, but not in HSCs, common myeloid progenitors, or megakaryocyte/erythrocyte progenitors. In contrast, in human AML, TIM-3 was expressed on cell surface of the vast majority of CD34⁺CD38⁻ LSCs and CD34⁺CD38⁺ leukemic progenitors in AML of most FAB types, except for acute promyelocytic leukemia (M3). FACS-sorted TIM-3⁺ but not TIM-3⁻ AML cells reconstituted human AML in the immunodeficient mice, indicating that the TIM-3⁺ population contains most of functional LSCs. To selectively eradicate TIM-3-expressing AML LSCs, we established an anti-human TIM-3 mouse IgG2a antibody, ATIK2a, possessing antibody-dependent cellular cytotoxic and complement-dependent cytotoxic activities in leukemia cell lines transfected with TIM-3. We first tested the effect of ATIK2a treatment on reconstitution of normal HSCs in a xenograft model. ATIK2a was intraperitoneally injected to the mice once a week after 12 hours of transplantation of human CD34⁺ cells. Injection of ATIK2a did not affect reconstitution of normal human hematopoiesis except removing TIM-3-expressing mature monocytes. In contrast, injection of TIM-3 to the mice transplanted with human AML samples markedly reduced leukemic repopulation. In some mice transplanted with AML bone marrow, only normal hematopoiesis was reconstituted after anti-TIM-3 antibody treatment, suggesting that the antibody selectively killed AML cells, sparing residual normal HSCs. To further test the inhibitory effect of ATIK2a on established human AML, eight weeks after transplantation of human AML cells, engraftment of human AML cells was confirmed by blood sampling and thereafter ATIK2a was injected to these mice. In all cases tested, ATIK2a treatment significantly reduced human TIM-3⁺ AML fraction as well as the CD34⁺CD38⁻ LSCs fraction. In addition, to verify the anti-AML LSCs effect of ATIK2a treatment, human CD45⁺AML cells from the primary recipients were re-transplanted into secondary recipients. All mice transplanted from primary recipients treated with control IgG developed AML, whereas none of mice transplanted with cells from ATIK2a-treated primary recipients developed AML, suggesting that functional LSCs were effectively eliminated by ATIK2a treatment in primary recipients.

Thus, TIM-3 is a promising surface molecule to target AML LSCs. Our experiments strongly suggest that targeting this molecule by monoclonal antibody treatment is a practical approach to eradicate human AML.