Activation of NK and CD8+ T-Cells with a Novel IL-15 and TGF-Beta Receptor Fusion Protein Confers Anti-Tumor Immunity

Abstract The use of cytokines as agents to augment immune responses against malignancies have been dealt setbacks due to immune selection of tumors, resulting in subpopulations that elaborate tumor-derived soluble factors, such as transforming growth factor-beta (TGF-β), which suppress immune effector functions. TGF-β is overexpressed by many solid and hematological malignancies and is well known to inhibit the proliferation and anti-tumor functions of lymphomyeloid cells. In order to maximize cytokine-based immunotherapy against tumors, we have designed a novel fusion protein consisting of proinflammatory murine interleukin-15 (IL-15) linked to the sushi domain of the IL-15Rα chain (IL15Rαsushi +IL15) fused in frame to the C'-terminus of a dimeric murine TGF-β-receptor (type II, TβRII) ectodomain-based ligand trap, termed FIST-15 (Fusion of Interleukin 15 with Sushi to TGF-β receptor). The rationale for the design of this protein is to prevent tumor-derived TGF-β from suppressing the immune response via the TGF-β ligand trap moiety, while simultaneously providing a potent stimulus for the activation of anti-tumor responses by an IL-15R agonist (IL-15Rαsushi +IL15). FIST-15 can neutralize TGF-β induced Smad signaling, and induce STAT3 and STAT5 phosphorylation by immunoblot and intracellular flow cytometric analysis of lymphocytes, suggesting that both protein domains are biochemically active. Functionally, FIST-15 is able to induce CD8+ T-cell proliferation at rates greater than IL-15 alone (CD8+ T-cell replicative index or fold-expansion of responding cells: 40, FIST-15, vs. 10, IL-15; p-value of unpaired T-test <0.05). The mitogenic effects of IL-15 are abrogated in CD8+ T-cells and NK cells in the presence of TGF-β. However, FIST-15 can overcome TGF-β mediated inhibition in both these cellular subsets (CD8+ T-cell replicative index: 20, FIST-15, vs. 5, IL-15, and NK cell replicative index: 40, FIST-15 vs. 5, IL-15; p-value <0.05). Rapid proliferation of the CD8+ central memory phenotype (CD62L+, CD44+) T-cells are seen with FIST-15 treatment. Compared to IL-15 expanded CD8+ T-cells, FIST-15 treatment also produced more IFN-γ, TNF-α, and IL-2 secreting CD8+ T-cells upon PMA/ionomycin stimulation. In addition to cytokines, production of anti-tumor effector molecules such as granzyme B is known to be inhibited by TGF-β. FIST-15 treated NK cells were superior to IL-15 treated NK cells in granzyme B production, even in the presence of TGF-β, as assayed by flow cytometric analysis (86.8% vs. 30.7% granzyme B expressing cells). Functionally, FIST-15 treated NK cells were also significantly more cytolytic against TGF-β secreting B16 murine melanoma cells in vitro compared to IL-15 treated NK cells (83.5% killing, FIST-15, vs. 24.4% killing, IL-15). C57Bl/6 mice with pre-established, syngeneic B16 melanoma tumors were treated with FIST-15 to assay the anti-tumor effects of the fusion protein in vivo. Mice receiving FIST-15 showed a significant delay in tumor growth (mean tumor volume: 345mm3) compared to control mice receiving conditioned media (mean tumor volume: 814.12mm3; p-value of paired T-test = 0.02) by day 21 post-tumor implantation. Furthermore, FIST-15 treated mice showed a significant survival advantage compared to control treated mice (80% vs 0%; p-value of log rank test = 0.0019) by day 27 post-tumor implantation. Mice immunized with B16 tumors transduced to express FIST-15 were also protected against subsequent wildtype B16 tumor challenge, suggesting that FIST-15 can trigger an adaptive immune response against tumor. Ongoing work utilizing FIST-15 in murine models of hematological malignancies, such as EL-4 lymphoma and C1498 AML, is currently underway. These models were selected due to their known overexpression of TGF-β isoforms that systemically inhibit endogenous anti-tumor responses, as well as the efficacy of immunotherapeutic agents. Indeed, many hematological malignancies acquire mutations that render them insensitive to the growth-inhibitory effects of TGF-β, where it may then be overexpressed as an oncogene to promote further tumor growth by inhibiting the immune system's anti-tumor capabilities (Dong et al Blood 2006). FIST-15 may present a viable immunotherapeutic strategy for hematological malignancies by combining the immune activating effects of IL-15 with the neutralization of immunosuppressive TGF-β. Disclosures Ng: Emory University: Patents & Royalties. Galipeau:Emory University: Patents & Royalties.