Abstract AMG509, a novel STEAP1 x CD3 T cell-recruiting XmAb® 2+1 bispecific immune therapy for the treatment of prostate cancer and Ewing sarcoma. Metastatic castration resistant prostate cancer (mCRPC) and Ewing sarcoma (EWS) are both diseases with unmet medical needs. Six Transmembrane Epithelial Antigen of the Prostate 1 (STEAP1) is a cell surface antigen that is induced by androgens and highly expressed in prostate cancer. STEAP1 is also highly expressed in EWS, and it is a target of the EWS/FLI1 fusion, the primary oncogenic driver in this pediatric malignancy. AMG 509 is a novel, half-life extended, cynomolgus monkey cross-reactive, STEAP1 x CD3 T cell-recruiting XmAb® 2+1 bispecific antibody designed to treat STEAP1-expressing cancers. AMG 509 contains two identical humanized anti-STEAP1 Fab domains that bind STEAP1-expressing cells, an anti-CD3 scFv domain that binds T cells, and an Fc domain engineered to lack effector function and extend serum half-life. We found that membranous STEAP1 expression, as measured by IHC, was absent from most normal tissues (FDA normal organ tissue array, n = 72) except the prostate epithelium and a subset of lung macrophages. In contrast, we found membranous STEAP1 expression by IHC in 80% of primary prostate cancer tumors (n = 80) and moderate to high expression in 83% of metastatic prostate cancer lesions (n = 181) and 77% of bone metastases (n = 31). In addition, we found STEAP1 expression by IHC in all EWS cell lines (n=13) and xenograft models tested (n=12). AMG 509 bound to recombinant human and cynomolgus monkey CD3ϵ with a KD of 27.6 nM and 25.8 nM, respectively, and to 293T cells transfected with human or cynomolgus monkey STEAP1 with an EC50 of 3.8 nM and 3.1 nM, respectively. AMG 509 triggered potent T cell-redirected lysis of STEAP1-positive cancer cells, with a median EC50 of 18.9 pM across 17 different STEAP1-expressing prostate, EWS, gastric, and melanoma cancer cell lines. AMG 509 was 50-fold more potent in inducing the redirected lysis of C4-2B prostate cancer cells in vitro than an XmAb® molecule with a single anti-STEAP1 Fab domain. AMG 509 was also specific, as it showed no activity against C4-2B cancer cells engineered to lack STEAP1 expression. AMG 509 exhibited hallmarks of T-cell engagement in vitro, including T-cell proliferation and cytokine release. It also induced robust in vivo anti-tumor activity in prostate cancer and EWS xenograft models, with concomitant CD8+ T-cell activation and expansion in tumors. Data from a nonclinical safety study conducted in the cynomolgus monkey suggest that AMG 509 can be administered safely at doses that are predicted to be clinically active. AMG 509 is a potent and specific first-in-class T cell-recruiting antibody for the treatment of STEAP1-positive malignancies. We are currently evaluating the safety, tolerability, pharmacokinetics, and efficacy of AMG 509 in a phase 1, first-in-human study in patients with mCRPC (NCT04221542). Citation Format: Olivier Nolan-Stevaux. AMG 509: A novel, humanized, half-Life extended, bispecific STEAP1 × CD3 T cell recruiting XmAb® 2+1 antibody [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr DDT02-03.