JMB2403, a potential best-in-class PD-1-dependent IL2Rβγ-activating tri-specific antibody for safe and potent immunotherapy.

PD-1-based immunocytokines, such as interleukin (IL)-2 fused with anti-PD-1, have been designed to increase efficacy, but their use is hampered by dose-limiting toxicity. To overcome this, substantial efforts have focused on engineering attenuated IL-2 variants, albeit with limited success to date. Taking an alternative approach, we screened a naïve alpaca library for weak agonistic nanobodies of IL-2/15Rβ and the common γ chain and attached a potent anti-PD-1 IgG to generate a tri-specific antibody JMB2403. JMB2403 did not bind IL2Rα (CD25), but activated STAT5 phosphorylation in an engineered Jurkat cell reporter assay albeit far less potently compared to wildtype IL-2. This translated to a JMB2403-induced pSTAT5 increase in natural killer (NK) cells, but minimal STAT5 phosphorylation in Treg cells. Notably, JMB2403 retained PD-1 blocking activity and concentration-dependently induced phospho-STAT5 only in activated (PD-1high) but not in non-activated CD8+ T cells, indicating cis-action mediated by PD-1 engagement. In A375 (melanoma) and NCI-H292 (lung cancer) xenograft models, JMB2403 exhibited superior anti-tumor efficacy compared to its parental PD-1 antibody. In a cynomolgus monkey study, JMB2403 produced a dose-dependent increase in the proliferation of CD8+ T cells, PD-1+ CD8+ T cells, Treg cells, and NK cells after the first dose which returned to baseline before the second dose. No IL-2-related toxicities such as vascular leak syndrome or pulmonary edema were observed. In summary, JMB2403 can induce cis-activation of PD-1+ T cells and display enhanced anti-tumor efficacy with good tolerability. To our knowledge, this is the first tri-specific antibody of its kind that targets specifically IL2/15 receptor signaling subunits.