BackgroundCD137 (4-1BB) represents a costimulatory pathway that promotes T, NK, and dendritic cell effector functions favorable for antitumor immunity. The extracellular domain of CD137, comprised of four cysteine-rich domains (CRD-I, CRD-II, CRD-III, CRD-IV), trimerizes upon binding to CD137 ligand (CD137L) to induce cell stimulatory transcriptional and epigenetic changes.1 2 The investigation of CD137-targeting agonist antibody, urelumab (CRD-I-binding, IgG4), in human subjects showed immunologic and pharmacodynamic effects, but poor efficacy due to dose-limiting liver toxicity.3 Preclinical studies using a murine surrogate antibody, clone 3H3 (CRD-I-binding, rIgG2a), also demonstrated hepatotoxicity that correlated with activation of CD137-expressing myeloid cells and memory CD8+ T cells.4 5 In contrast, utomilumab (CRD-II/III-binding, IgG2) showed acceptable tolerability, but limited clinical efficacy.6 7 These and more recent findings implicate epitope and Fc gamma receptor (FcγR)-dependent antibody cross-linking as critical factors for CD137 therapeutic antibody design.MethodsWe investigated the molecular and cellular effects of AGEN2373 (CRD-IV-binding, IgG1), a conditionally active CD137-targeting agonist antibody designed to bind and induce CD137 signaling upon FcγR cross-linking while permitting ligand binding to CD137. The role of epitope and FcγR binding as critical factors for anti-CD137 therapeutic activity were elucidated in primary cell-based assays and syngeneic tumor-bearing mouse models using anti-mouse antibody clones S3B1 (CRD-IV-binding) and 3H3, surrogates of AGEN2373 and urelumab, respectively. In an ongoing phase 1 trial (NCT04121676), we evaluated the safety and tolerability of AGEN2373.ResultsAGEN2373 bound with high-affinity to CD137 CRD-IV and promoted potent agonist activity of CD137 that was conditionally dependent on Fc-dependent antibody cross-linking. AGEN2373 surrogate, S3B1, showed comparable binding and cross-link dependent agonist activity. In CT26 tumor-bearing mice, S3B1 and 3H3 demonstrated complete tumor control that was not reproducible with a Fc-silent S3B1 antibody. The Fc-dependent activity of S3B1 correlated with induced immunologic changes in the TME including CD8 T cell expansion, NK cell activation, and Treg depletion. Patients with advanced solid cancers, treated with AGEN2373 up to 1 mg/kg every 4 weeks, demonstrate clinical activity with no evidence of hepatotoxicity.ConclusionsConditional and potent agonist activity of AGEN2373 is dependent on binding to CD137 CRD-IV and FcγR. Preclinically, our data demonstrate that AGEN2373-like murine surrogate antibodies promote potent immune activation and anti-tumor immunity. Phase 1 clinical trials investigating the safety and efficacy of AGEN2373, alone or combination with balstilimab (anti-PD-1), are underway.Trial RegistrationNCT04121676ReferencesWen TJ, Bukczynski and Watts TH. 4-1BB ligand-mediated costimulation of human T cells induces CD4 and CD8 T cell expansion, cytokine production, and the development of cytolytic effector function. J Immunol 2002;168(10): p. 4897–906.Bitra A, et al. Crystal structures of the human 4-1BB receptor bound to its ligand 4-1BBL reveal covalent receptor dimerization as a potential signaling amplifier. J Biol Chem 2018;293(26): p. 9958–9969.Segal NH, et al., Results from an integrated safety analysis of urelumab, an agonist anti-CD137 monoclonal antibody. Clin Cancer Res 2017;23(8): p. 1929–1936.Bartkowiak T, et al., Activation of 4-1BB on liver myeloid cells triggers hepatitis via an interleukin-27-dependent pathway. Clin Cancer Res 2018;24(5): p. 1138–1151.Lin GH, et al., GITR-dependent regulation of 4-1BB expression: implications for T cell memory and anti-4-1BB-induced pathology. J Immunol 2013;190(9): p. 4627–39.Segal, N.H., et al., Phase I study of single-agent utomilumab (PF-05082566), a 4-1BB/CD137 agonist, in patients with advanced cancer. Clin Cancer Res 2018;24(8): p. 1816–1823.Li Y, et al., Limited Cross-Linking of 4-1BB by 4-1BB ligand and the agonist monoclonal antibody utomilumab. Cell Rep 2018;25(4): p. 909–920 e4.
Regulatory risk disclosure in the banking industry: a scoring model approach