Design and characterization of a tumor-retained interleukin-2 cytokine/antibody fusion protein for cancer immunotherapy

Abstract Interleukin-2 (IL-2) is a pleiotropic cytokine with clinical utility in stimulating anti-cancer immunity. However, its efficacy is limited by its short serum half-life, systemic toxicity, and concurrent activation of immunostimulatory and immunosuppressive immune cells. IL-2 can signal down an intermediate-affinity receptor consisting of IL-2 receptor β (IL-2Rβ) and the common gamma chain (γc) or a high-affinity receptor that also includes IL-2Rα. Regulatory T (Treg) cells express high levels of IL-2Rα making them more responsive to IL-2 than natural killer (NK), naïve/resting effector T (Teff), and memory T cells. To extend the half-life and selectively bias the immunostimulatory functions of IL-2 in a unimolecular format, we engineered a single-chain fusion protein, F10 immunocytokine (F10 IC), that intramolecularly fuses IL-2 to an anti-IL-2 antibody that sterically blocks IL-2Rα binding and allosterically enhances IL-2Rβ binding. However, preclinical data suggested a narrow therapeutic index (TI) for F10 IC. To improve the TI of F10 IC we designed a format of F10 IC poised for intratumoral delivery and retention. We fused collagen binding domains (CBD) to F10 IC in various topographies (F10 IC-CBD) to exploit the abundance and exposure of collagen in the tumor microenvironment. With intratumoral delivery, F10 IC-CBD is retained in the tumor and exhibits a therapeutically favorable pharmacokinetic profile. Furthermore, F10 IC-CBD increases and prolongs intratumoral immune cell infiltration, significantly slows tumor progression, minimizes F10 IC toxicity and synergizes with anti-PD-1 blockade in syngeneic mouse tumor models. We have illustrated the potential of F10 IC-CBD as a potent tumor-agnostic cancer therapy. Supported by grants from the NIH (R01 EB029341, T32 AI007417), Melanoma Research Alliance, Willowcroft Foundation and a Sanofi iAward.