Abstract 7686: Identification of a first-in-class therapeutic target using machine learning enabled translational-optimized proteomics system (TPS)

Abstract Background: At Foreseen, we built a translational-optimized proteomics system (TPS) and assembled a vast collection of well-characterized human clinical samples. Our platform can extract actionable insights from complex proteomics and clinical datasets to accelerate the discovery of clinically relevant targets for drug discovery and for diagnostic applications. Methods: Chronic hepatitis B (CHB) infection may lead to progressive liver diseases including fibrosis, cirrhosis, and ultimately hepatocellular carcinoma (HCC). An optimized plasma proteomics analysis using parallel accumulation-serial fragmentation combined with data-independent acquisition (DIA-PASEF) on a large liver cohort was performed. This cohort (N=340) consists of healthy individuals (n = 48), patients with either CHB (n = 39), liver fibrosis (n = 47), liver cirrhosis (LC, n = 56), and HCC (4 subgroups based on TNM staging system: HCC1-4; n = 36, 38, 39, and 37). Highly abundant proteins were removed from plasma prior to proteomic preparation. Samples were processed and analyzed using DIA-MS for high-throughput protein quantification. Our TPS can identify potential biomarkers and stage-specific molecular drivers associated with liver disease and liver cancer progression. Expression analysis using tissue microarrays (TMA), followed by validation through in vitro (cell lines) and in vivo (cell-derived xenograft models) studies were conducted on selected candidate targets. Results: Integrin alpha-2 (ITGA2) was among the seven tumor-associated antigens (TAAs) identified in hepatocellular carcinoma (HCC) we selected as final drug candidates after screening. It is a cell surface membrane receptor and may function as a regulatory factor in cancer, responsible for driving tumorigenesis, inducing chemoresistance, regulating genomic instability and remodeling the tumor microenvironment. In this study we demonstrated that the protein expression of ITGA2 is low in patients’ groups of CHB, fibrosis and cirrhosis, but significantly upregulated in HCC. Furthermore, ITGA2 is overexpressed in multiple solid tumors, including pancreatic, gastric, and colorectal cancers using TMA analyses. We discovered and developed an antibody targeting ITGA2 with nM binding affinity and good internalization properties. We advanced FS001(IPN60300) into the final stages of preclinical development. This a first-in-class antibody drug conjugate comprised of (i) Foreseen’s proprietary antibody targeting ITGA2 and (ii) linker-payload with excellent pharmaceutical features developed by Shanghai Escugen Biotechnology. Foreseen has granted an exclusive worldwide license of FS001(IPN60300) to Ipsen in July 2024. Conclusion: Our translational-optimized proteomics system (TPS) has led to the discovery of ITGA2, a novel tumor-associated antigen (TAA) and enabled the development of FS001(IPN60300). Citation Format: Catherine Wong, Ning Li, Shuaixin Gao, Nan Zhang, Jiming Yin, Yonghong Zhang. Identification of a first-in-class therapeutic target using machine learning enabled translational-optimized proteomics system (TPS) [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2026; Part 1 (Regular Abstracts); 2026 Apr 17-22; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2026;86(7 Suppl):Abstract nr 7686.