Abstract 3504: Evaluation of T cell-engaging bispecific antibody using perfusable three-dimensional model

Introduction The development of T cell-engaging bispecific antibody (T-BsAb) has been an attractive field in cancer research due to its clinical success, especially in the treatment of hematologic cancer. In recent years, several T-BsAbs have been investigated for treating solid cancers. To improve T-BsAbs in treating solid cancers, it would be important to develop a robust in vitro evaluation system that mimics in vivo tumors. A conventional two-dimensional (2D) model is often used to evaluate in vitro efficacy of T-BsAb. However, T cells can directly contact cancer cells without any obstacles in this assay, resulting in a limited correlation with their in vivo efficacy. Although a three-dimensional (3D) model using the extracellular matrix such as matrigel can also be applied for drug evaluation, the lack of a perfusable vascular network may be a limitation for evaluating some therapeutic modalities, including T-BsAb. Here, we established the novel perfusable 3D in vitro model and compared the efficacy of T-BsAb in several in vitro and in vivo models, in order to determine which in vitro model best mimics in vivo in the evaluation of T-BsAb. Material and Method We used three colorectal cancer cell lines, DiFi (EGFR+++), HT29 (EGFR++) and SW620 (EGFR-), and T-BsAb (named hEx3) targeting EGFR and CD3. As the conventional models, 2D and 3D in vitro models were adopted to evaluate the efficacy of hEx3. In the 3D model, T cells were applied on the matrigel with cancer cells. The perfusable 3D tissue is formed from a collagen gel composed of cancer cells and human umbilical vein endothelial cells (HUVECs). The tissue has a main blood vessel in the center through which the culture medium flows. The HUVECs in the gel have the self-organizing activity of a capillary-like structure from the main vascular vessel. To investigate in vivo efficacy, the cell line-derived xenograft (CDX) models were used. Results In the conventional 2D model, hEx3 exerted high cytotoxicity against DiFi and HT29 cells but not SW620 cells. On the other hand, in the conventional 3D model, the cytotoxicity of hEx3 was limited even for EGFR-positive cells. In addition, hEx3 treatment showed anti-tumor effects in only DiFi CDX model but not in HT29 and SW620 CDX models. These results suggested that the efficacy of T-BsAb in the CDX model is not reflected by the conventional 2D and 3D in vitro models. We attempted to perform T cell perfusion in the perfusable 3D in vitro model. T cells were distributed near the capillary-like blood vessels, indicating that T cells infiltrated into the tissue through the blood flow. We are currently using this model to evaluate the efficacy of hEx3 in vitro. Conclusion The conventional in vitro models would not be suitable to evaluate the efficacy of T-BsAb because their results were inconsistent with those in CDX models. Therefore, it would be necessary to develop an in vitro evaluation system that mimics in vivo blood vessel structures. Citation Format: Ryo Tsumura, Rinki Singh, Nobuhito Mori, Yasuyuki Kida, Ryutaro Asano, Hiroki Takashima, Hirobumi Fuchigami, Naoya Sakamoto, Masahiro Yasunaga, Yoshikatsu Koga. Evaluation of T cell-engaging bispecific antibody using perfusable three-dimensional model [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 3504.