First-in-human evaluation of IL2RG blockade in patients with severe aplastic anemia that is refractory to or relapsed on immunosuppressive therapy

Abstract Background: Aplastic anemia (AA) is considered a T cell-mediated autoimmune disease where hematopoietic stem and progenitor cells are destroyed by cytotoxic T cells, resulting in peripheral cytopenias (neutropenia, anemia, thrombocytopenia). For patients in whom hematologic stem cell transplantation is not available or suitable as a treatment option, T cell-targeted immunosuppressive therapy (IST) is standard of care. However, current IST options (eg, anti-thymocyte globulin and cyclosporine) are limited by toxicities. Given that lymphocyte development, proliferation, and activity are mediated in large part by gamma chain (gc) cytokines (IL2, IL4, IL7, IL9, IL15, IL21), which share a common receptor subunit (interleukin 2 receptor subunit gamma; IL2RG), IL2RG blockade may provide an novel means to suppress pathogenic T cell responses in T cell-mediated diseases, like AA. In nonhuman primates, IL2RG blockade profoundly decreases NK cells and reduces T cells by ~50%, with no impact on neutrophils or platelets, supporting its potential utility in AA. Methods: REGN7257, a human monoclonal antibody that binds IL2RG, inhibiting signaling of all gc cytokines, was evaluated in a phase 1, multinational, open-label, first-in-human study involving patients with severe aplastic anemia (SAA) that is refractory to or relapsed on standard of care IST. Seventeen participants were enrolled across 4 dose levels: 5 mg/kg (n=3), 10 mg/kg (n=6), 20 mg/kg (n=4), and 40 mg/kg (n=4), administered as a single IV infusion. The primary endpoint was the incidence of adverse events (AEs) and serious adverse events (SAEs) up to 12 months post-treatment. Adverse events of interest (AEIs) were defined as any non-hematologic grade 3 or above AE (excluding febrile neutropenia or typical infections that occur in the neutropenic patient), opportunistic infections, or evidence of potential liver impairment. Secondary endpoints included changes in blood lymphocyte counts, serum pharmacokinetics (PK), and hematologic response rates. Exploratory analyses included changes in serum cytokines. Results: Observed AEs were generally consistent with underlying SAA. Eighty-eight AEs were reported across 16 of the 17 participants; most (85%) were non-serious. Seven participants (41%) experienced at least 1 SAE; most (86%) were related to febrile neutropenia/infectious events or hospitalizations for transfusions to address anemia or thrombocytopenia. All SAEs were considered by the investigator as not related to study drug. No AEIs were reported. Two participants with profound neutropenia (neutrophils ≤20 cells/µL) at baseline experienced multiple infectious AEs and died after early termination from the study. Both deaths were assessed by the investigator as not related to study drug. Two participants experienced infusion-related reactions (1 in the 5 mg/kg dose group, 1 in the 40 mg/kg dose group); both received the full dose of REGN7257 and both events were considered non-serious. REGN7257 exhibited nonlinear PK, with greater than dose-proportional increases in exposure with increasing dose. Treatment induced transient increases in NK and CD8 T cell counts across all dose levels, followed by reductions to below baseline at doses ≥10 mg/kg in some patients. Transient increases in gc cytokines in serum were observed, consistent with target engagement. Transient reductions in inflammatory/effector molecules in serum were also observed. One patient with IST-relapsed SAA who received a single dose of 20 mg/kg achieved a partial hematologic response. Improvement in peripheral cytopenias correlated contemporaneously with reduction in lymphocyte counts and serum hematopoietic growth factors. The hematologic response was lost as drug cleared but returned with a repeat dose (40 mg/kg). Conclusion: Data from this first-in-human evaluation of REGN7257 in patients with IST-refractory or IST-relapsed SAA show that IL2RG blockade was generally well-tolerated with observed AEs being generally consistent with the underlying disease. Reductions in lymphocytes and changes in serum cytokines and inflammatory/effector molecules were consistent with drug activity. One patient achieved a partial hematologic response that waned as drug cleared, but returned with retreatment. Taken together, REGN7257's acceptable safety profile and biologic activity in AA support its continued development, as well as its potential utility in less severe T cell-mediated diseases.