Anti-KIR (1-7F9): A Fully Human Monoclonal Antibody (mAb) That Blocks KIR2DL1, −2 and −3, Promoting Natural Killer (NK) Cell-Mediated Lysis of Tumor Cells In Vitro and In Vivo.

In patients with acute myeloid leukemia (AML), haplo-identical stem cell transplantation (SCT) can lead to expansion and activation of Killer Immunoglobulin-like Receptor (KIR)-HLA class I mismatched NK cells, resulting in reduced rates of leukemia relapse and no graft-versus-host disease (Ruggeri et al. Science 2002). However, this SCT is not available to the majority of AML patients who are elderly. To explore the feasibility of achieving similar NK-mediated anti-leukemia activity by a pharmacological approach, we generated fully human anti-KIR mAbs that block the interactions of inhibitory KIR2DL receptors with their HLA-C ligands, thereby enhancing NK activity. Here we describe one such therapeutic candidate anti-KIR mAb, designated 1-7F9. As distinct HLA-C allotypes are recognized by KIR2DL1 or −2/3, only mAbs that cross-react with these KIRs would be expected to work in the entire population. Hence, 1-7F9 was initially selected based on its ability to bind soluble, recombinant KIR2L1, −2 and −3. By Biacore analysis, the bivalent affinities for KIR2DL1 and −3 were 0.43 × 10⁻⁹ M and 0.025 × 10⁻⁹ M, respectively. In experimental systems and in normal human blood, 1-7F9 bound KIR2DL1, −2 and −3, and −2DS1 and −2, but not to KIR2DS3 or −4. 1-7F9 dose-dependently inhibited the binding of soluble KIR2DL1-Fc to cell surface HLA-Cw4. 1-7F9 augmented the lysis of 721.221-Cw4 B-EBV cells by an NK cell line transfected with KIR2DL1 (YTS-2DL1) from 5% lysis in absence of mAb to a maximal 55% lysis at 5 ug/ml of mAb, but did not affect lysis by KIR-negative NK cells. Lysis of PHA-stimulated blasts and primary AML blasts by autologous IL-2 activated NK cells (E:T=6:1) was 10 and 15%, respectively, in absence of mAb vs 80% and 55% in presence of 1-7F9. Incubation of IL-2 activated blood mononuclear cells with 1-7F9 resulted in expression of the activation marker CD107 on about 10% of KIR2D-positive NK cells, which increased to 20% upon addition of HLA-C-positive B-EBV targets, suggesting that 1-7F9 preferentially induces activation of NK cells in presence of transformed cells. The isotype of 1-7F9 is IgG4; accordingly, it did not cause depletion of KIR positive cells in vitro or in vivo in KIR-transgenic mice despite long-lived KIR-occupancy. As KIR are not found in mice, in vivo activity was tested in a NOD-SCID mouse model where inoculation of in vitro-expanded NK cells (80% of NK cells KIR2D-positive) and autologous human B-EBV cells (E:T=1:3) resulted in death of all mice by day 26. A single injection of 1-7F9 (125 ug/mouse) resulted in long-term survival, with 100% of treated mice alive beyond day 60; in contrast, 60 ug/mouse of the mAb was ineffective. Similarly, ex vivo pre-incubation of NK cells with 1-7F9 (37,3 ug/10⁶ NK cells) prior to inoculation in mice resulted in elimination of the autologous transformed B cells in vivo and survival of 100% of the treated animals. These data show that 1-7F9 augments NK-mediated tumor killing in vitro and in vivo, and that it exhibits long-lived KIR binding in vivo, providing a preclinical basis for initiating phase 1 clinical trials with the mAb.