RG6333 (CD19-CD28), a CD19-Targeted Affinity-Optimized CD28 Bispecific Antibody, Enhances and Prolongs the Anti-Tumor Activity of Glofitamab (CD20-TCB) in Preclinical Models

Potent T cell activation and effector function not only requires T cell receptor (TCR) signaling ("signal I"), provided by a TCR interaction with its appropriate peptide-MHC ("major-histocompatibility") complex or via a CD3e engagement with T cell bispecific antibodies (TCBs), but also an additional signal ("signal II"). This can be delivered via costimulatory receptors, like the CD28 Ig superfamily member, when interacting with their associated ligand(s). Glofitamab (CD20-TCB), a CD20-targeted TCB, which recently demonstrated remarkable single agent activity in relapsed/refractory non-Hodgkin-Lymphoma (r/r NHL) patients, delivers a strong "signal I" to T cells resulting in T cell activation and subsequent tumor cell killing. Despite the 39.4% of complete Response (CR) in the heavily pretreated 3L+ DLBCL patient population (Dickinson M, ASCO 2022), a proportion of patients fails to respond to glofitamab monotherapy. Therefore, strategies that increase durability of response to TCBs by, for example, providing an optimal costimulation to intratumoral T cells further strengthening and deepening TCB activity are needed. RG6333 (CD19-CD28) is a heterodimeric 1+1 IgG1-based CD19-targeted bispecific antibody containing a low affinity, monovalent anti-CD28 binding moiety and a silent Fc-domain aiming to provide a strong but safe "signal II'' to activated T cells. Unlike CD28 superagonists that induce T cell activation and strong toxicity in human in the absence of a "signal I", RG6333 strictly depends on the presence of a "signal I" for activity due to its unique properties. It is designed to neither exhibit single-agent activity nor induce cytokine-release syndrome (CRS) but to further boost intratumoral T cell effector function and tumor cell killing once "signal I" (i.e. glofitamab) is provided. In preclinical experiments we show that affinity reduction of the CD28-binder resulted in lower binding to human CD28 positive T cells in vitro and in vivo in humanized NSG mice (huNSG). However, its in vitro potency in the presence of glofitamab was maintained and studies in huNSG showed an enhanced combination effect with glofitamab when CD28 affinity was reduced. Importantly, no single agent activity, as assessed by T cell activation and cytokine secretion, was observed neither in vitro nor in huNSG. In contrast, TGN1412, a CD28 superagonist antibody, induced strong T cell activation, proliferation and cytokine secretion in vitro and in huNSG. Scheduling studies with glofitamab and RG6333 in huNSG suggest a safe and potent treatment regimen by using Gazyva pre-treatment followed by a staggered infusion of glofitmab and RG6333 applying an interval of three days at the first treatment cycle. Due to enhanced potency, concomitant administration of glofitamab and RG6333 resulted in increased CRS and body weight loss in huNSG. In huNSG, bearing a difficult to treat disseminated WSU-DLCL2 DLBCL tumor model, the combination treatment of glofitamab and RG6333 led to the formation of tumor-free animals as compared to the respective monotherapy groups. Mode of action studies in vivo revealed a strong boosting of glofitamab-mediated T cell infiltration in tumor tissue, both for CD4+ and CD8+ T cells subsets without any evidence of increased regulatory T cell activity. RG6333 second line treatment prolonged the duration of glofitamab response and delayed tumor relapse in vivo in the subcutaneous OCI-Ly18 DLBCL model. Interestingly, the alternation of RG6333 with an alternative 4-1BB costimulatory agent (RG6076; CD19-4-1BBL) completely prevented tumor relapse during glofitamab treatment for more than 120 days when RG6333 was given for the first treatment cycles followed by RG6076 at later cycles. Finally, RG6333 boosted glofitamab-mediated cytokine secretion and T cell activation in DLBCL patient samples ex vivo validating its activity not only on T cells derived from healthy donors but also from patients. Taken together, the reported preclinical data suggest a strong rationale for combining RG6333 with glofitamab in r/r NHL patients to deepen and further prolong the treatment response. Optimal scheduling including alternation of costimulatory bispecific antibodies suggest a powerful off-the-shelf T cell redirection approach as an alternative to CAR-T cell therapies. RG6333 is currently in a phase I, open-label, dose-escalation study in combination with glofitamab (NCT05219513).