513. Armed-Ad Gene Therapy Expressing PD-L1 Minibody Enhances the Anti-Tumor Effect of Adoptively Transferred Chimeric Antigen Receptor T-Cells for Solid Tumor Treatment

Abstract

Intratumoral treatment with oncolytic adenoviral vectors expressing an immunomodulatory molecule (Armed Onc.Ads) is safe and has shown some clinical benefit in patients with solid tumors. However, local treatment with Armed Onc.Ad has limited anti-tumor effect against metastasized tumors. T cells modified with tumor-directed chimeric antigen receptors (CARs) have shown promise for the systemic treatment of hematological malignancies, but have been less effective in treating solid tumors. Major reasons for this failure include lack of T-cell migration into solid tumors and the inhibitory microenvironments (e.g. PD-L1) at the tumor site. Recent clinical trials with immune-checkpoint inhibitors (e.g. anti-PD-L1 antibody) have broadly enhanced antitumor immunity by improving tumor-specific T cell responses. We therefore hypothesized that an Armed Onc.Ad expressing anti-PD-L1 antibody could enable the blockade of PD-1:PD-L1 interaction between CAR T-cells and cancer cells at the tumor site, and that combining these treatment modalities may have potent and synergistic anti-tumor effect in solid tumors. In this study, we confirmed that PD-L1 is upregulated on squamous cell carcinoma (Pre: 20%, Post: 98%), and prostate cancer cells (Pre: 60%, Post: 100%), in the presence of co-cultured IFNγ producing HER2.CAR T-cells, a population of effectors that we have safely administered to patients with solid tumors. Additionally, HER2.CAR T-cells also express PD-1 upon activation (Pre: 1%, Post: 30%) at 24 hr co-culture. Cancer cell killing by HER2.CAR T-cells was enhanced 2-fold in the presence of anti-PD-L1 IgG in vitro. We constructed a helper-dependent Ad (HDAd: no autonomous replication and no lytic effect) expressing PD-L1 minibody (fused anti-human PD-L1 scFv with human IgG1 constant region (HDPDL1 mini)) and confirmed by western blot that the minibody is secreted in a dose-dependent manner by cancer cells infected with HDPDL1 mini. The avidity of HDAd derived PD-L1 minibody for PD-L1 is similar to commercial anti-PD-L1 IgG, and the reagent enhanced HER2.CAR T-cell killing of target cells by 3-fold compared to HER2.CAR T-cells with control HDAd. Co-infection of Onc.Ad with HDPDL1 mini induces the same oncolytic effect as previously reported, but also induces replication of HDPDL1 mini and thereby markedly increases production of PD-L1 minibody in vitro and in vivo. Intratumoral co-administration of HDPDL1 mini with Onc.Ad induces 4-fold greater expansion of adoptively transferred HER2.CAR T-cells at the tumor site compared to combinatorial treatment of Onc.Ad (without HDPDL1 mini). Our therapy extended median survival from 35 to 70 days compared to HER2.CAR T-cell alone. Our “all-in-one” strategy also appears superior to infusion of anti-PD-L1 IgG, inducing 3-fold greater expansion of HER2.CAR T-cells at the tumor site. Hence, combining HDPDL1 mini with Onc.Ad overcomes a significant immune defense mechanism inhibiting CAR T-cell treatment of solid tumors, and we are now investigating its longer-term benefits.