Simultaneous Targeting of Tumor and Stroma Cells to Enhance Solid Tumor CAR-T Cell Therapy

摘要 利用工程T细胞表达特定的嵌合抗原受体(CARS)来治疗癌症 产生了对多种癌症的持久治愈，并导致FDA批准的第一种CAR-T细胞疗法 2017年治疗儿童急性淋巴细胞性白血病。尽管取得了这一成功，CAR-T免疫疗法一直是 在靶向实体肿瘤方面效果要差得多。这一有限的成功部分源于实体肿瘤 微环境，它形成了免疫细胞渗透的物理屏障，并产生可溶的因子 下调T细胞活性，加速T细胞耗竭。而针对实体瘤的免疫疗法是 最初有效的是，肿瘤微环境对T细胞的抑制阻止了这些治疗方法的产生 持久的反应。在这个应用中，我们提出了一种新的CAR-T细胞疗法，旨在改善患者的预后 克服困扰现有CAR-T细胞缺陷的晚期胰腺癌患者 治疗。为此，我们将设计T细胞来表达多个CARS，使这些细胞能够靶向肿瘤 免疫抑制肿瘤微环境中的细胞和细胞。具体地说，我们将利用非病毒、 TC Buster DNA转座子系统插入包含多个CARS的大型多顺反子基因构建物 和一个选择标记进入T细胞。利用这个平台，我们将利用CARS产生针对间皮质素的T细胞 80%-85%的胰腺癌肿瘤表达的蛋白(MSLN)和成纤维细胞激活蛋白(FAP)，a 肿瘤微环境中肿瘤相关成纤维细胞的标记物。然后，我们将选择一个纯种群 表达MSLN-和FAP-CARS的T细胞，并在体外测定这些细胞的活性和特异性。我们 希望经过改造的T细胞将产生一种特殊而强大的反应，仅引发细胞毒功能 对抗表达其目标抗原的细胞。然后我们将确定工程T细胞在体内的有效性 应用异种移植小鼠模型建立MSLN和FAP阳性胰腺癌模型 T细胞的转移。我们期待表达FAP-CARS和MSLN-1的双特异性T细胞的免疫治疗。 CARS将诱导针对MSLN+/FAP+实体瘤的强大和持久的T细胞反应，导致肿瘤 缩小和增加存活率。此外，我们期待着开发出一种灵活、高效、可靠的 用单一的、非病毒基因传递方法产生双特异性T细胞的过程将促进出现 以克服当今工程T细胞疗法面临的许多问题，包括抗原逃逸 和靶标特异性。

Abstract The use of T cells engineered to express specific chimeric antigen receptors (CARs) to treat cancer has generated durable cures for many types of cancer and resulted in the first FDA approved CAR-T cell therapy to treat childhood acute lymphoblastic leukemia in 2017. Despite this success, CAR-T immunotherapies have been much less effective at targeting solid tumors. Part of this limited success stems from the solid tumor microenvironment, which forms a physical barrier to immune cell infiltration and produces soluble factors that downregulate T cell activity and accelerate T cell exhaustion. While immunotherapies targeting solid tumors are initially effective, the tumor microenvironment’s inhibition of T cells prevents these treatments from producing durable responses. In this application, we propose a novel CAR-T cell therapy aimed to improve outcomes for patients with advanced stage pancreatic cancer by overcoming the deficiencies that plague current CAR-T cell therapies. To this end, we will engineer T cells to express multiple CARs, enabling these cells to target tumor cells and cells in the immune-suppressive tumor microenvironment. Specifically, we will leverage the non-viral, Tc Buster DNA transposon system to insert a large multicistronic genetic construct containing multiple CARs and a selection marker into T cells. Using this platform, we will generate T cells with CARs targeting mesothelin (MSLN), a protein expressed by 80-85% of pancreatic cancer tumors, and fibroblast activation protein (FAP), a marker of cancer associated fibroblasts in the tumor microenvironment. We will then select a pure population of T cells expressing MSLN- and FAP-CARs and determine the activity and specificity of these cells in vitro. We expect that engineered T cells will generate a specific and robust response, eliciting cytotoxic functions only against cells expressing their target antigen. We will then determine the efficacy of engineered T cells in vivo using a xenograft mouse model to generate MSLN and FAP positive pancreatic carcinomas followed by adoptive transfer of T cells. We expect immunotherapeutic delivery of bispecific T cells expressing FAP-CARs and MSLN- CARs will elicit a robust and long-lasting T cell response against MSLN+/FAP+ solid tumors resulting in tumor shrinkage and increased survival. Furthermore, we expect the development of a flexible, efficient, and reliable process to generate bispecific T cells with a single, non-viral gene delivery approach will facilitate the emergence of novel therapies to overcome many issues facing engineered T cell therapy today, including antigen escape and target specificity.