Abstract B091: A LAG-3/PD-L1 bispecific antibody inhibits tumor growth in two syngeneic colon carcinoma models

Combining immunotherapeutic antibodies for treatment of cancer patients has shown benefits over single agent treatment. For example, the addition of anti-PD-1 antibodies to anti-CTLA-4 treatment in advanced melanoma patients has increased objective response rates from 19% to 57.6% (Larkin,J et al. 2015. NEJM 373:23-34). An alternative to combining two antibodies is the development of bispecific antibodies that not only bring two biologies together but may result in novel mechanisms of action that are impossible to attain with combinations. Lymphocyte Activation Gene-3 (LAG-3) is a member of the Ig superfamily expressed on activated T cells, NK cells, pDCs, B cells, γδ T cells and participates in immune suppression, particularly through persistent strong expression in a percentage of regulatory T cells (Tregs). Programmed Cell Death receptor (PD-1) binds to its ligand PD-L1, expressed not only on activated immune cells to inhibit cellular immune responses but also on tumor cells. Expression of LAG-3 and PD-1 leads to T cell exhaustion, allowing tumor escape from immune surveillance. Combining inhibitory antibodies to PD-1 and LAG-3 reactivates T cells leading to efficacy in murine models over single treatment (Woo,S-R et al. 2012. Cancer Res. 15: 917-27) which prompted interest in the development of a LAG-3 and PD-L1 bispecific antibody for inhibiting tumor growth. A murine-specific anti-LAG-3 and PD-L1 bispecific antibody was engineered which binds both antigens simultaneously and with nanomolar affinities. The anti-LAG-3/PD-L1 bispecific antibody inhibits LAG-3 binding to MHCII and PD-L1 binding to PD-1 and CD80, resulting in T cell activation in an in vitro assay. This potency translates into in vivo efficacy, where the anti-LAG-3/PD-L1 bispecific antibody decreased tumor burden in the MC38 colon carcinoma tumor model, both in early-established or well-established tumors. At the end of the study tumor-free animals were more numerous in the LAG-3/PD-L1 bispecific group than in the group given a combination of individual anti-LAG-3 and PD-L1 antibodies. The results were recapitulated in the CT26 murine colon cancer model, where the LAG-3/PD-L1 bispecific showed an increase of anti-tumor activity as compared to the combination of anti-LAG-3 and anti-PD-L1 antibodies. Thus, the preclinical data supports developing an anti-human LAG-3/PD-L1 bispecific for the treatment of cancer patients. Citation Format: Jacqueline Doody, Matthew Kraman, Katy Everett, Carlo