Qualifying ADCs and multispecific therapeutic targets with a novel Agentic AI Insight platform.

e15179 Background: While genomic alterations are critical biomarkers for predicting response to targeted therapies, several novel classes of therapeutics may need to leverage additional contexts to qualify potential response and resistance mechanisms. Immunotherapies, antibody-drug conjugates and multispecifics may require biomarkers that localize drug targets within tumor and immune cells both in the tumor microenvironment as well as subcellular localization. Multiplex immunofluorescence (mIF) is an established technology capable of quantifying the subcellular localization of dozens of proteins within a single slide, enabling parallel assessment of a large number of drug targets. Here we describe a novel agentic AI platform for leveraging high-plex spatial proteomics to inform rationales for evaluating individual targets and combinations of targets across patient populations. Methods: We designed a 30-plex mIF panel to perform sequential immunofluorescence (seqIF) on the COMET platform to profile known tumor antigens and targets (n = 18) currently in development along with the tumor microenvironment (TME), immune cells, and key signaling cascades (n = 12). We tested this panel on 217 samples ( > 1.7M cells) from lung (n = 90), gastric (n = 11), and colorectal (n = 68) cancers, as well as normal tissue specimens (n = 48, 22 sites). Utilizing Nucleai’s deep-learning-based multiplex imaging analysis pipeline, we developed a novel agentic AI that can qualify protein abundance, localization, and interactions among 9 different cell types across the tumor core (TC), tumor-stromal interface (TSI), and within the adjacent stroma of the tissue. Results: We demonstrate that known drug targets with historical approvals are identified in each of these indications, which validated that the pipeline was able to correctly identify key targets based on subcellular localization, spatial proximity analysis, and analysis of off-target cell type expression. We also identified potential combination partners for multi-specifics which were found to have reductions in off-target cell expression. Conclusions: Nucleai’s Agentic AI platform enables researchers to natively engage with deep spatial proteomic profiling by capturing insights for actionable characterization of protein interactions and patient heterogeneity across tissues. We demonstrate that this framework can aid in the prioritization and derisking of potential drug targets used by key classes of therapeutics currently in development.