Abstract 4919: AI-enabled hit selection of drug screening on human pancreatic cancer organoids

Abstract

Abstract Cancer remains one of the leading causes of death in the 21st century. Despite the latest advances in oncology, most cancer patients lack tailored therapeutic approaches with lasting benefit. Measuring the impact of anticancer compounds and their combinations is only possible on ex vivo assays. To this end, patient-derived organoids (PDOs) have been proposed as viable and efficient models for ex vivo testing. PDOs show long-term expansion potential while retaining tumor histopathology as well as cancer gene mutations. However, the translation of organoids in screening applications has so far been hampered by the lack of homogeneity and difficulties in handling and automation. Moreover, organoids are typically randomly distributed across the culture which complicates imaging and images analyses. To overcome these challenges, we set up a compound screening workflow with PDOs using Gri3D platform, comprising plates with micro-cavities suitable for high-throughput and reproducible organoid culture. Based on a standard 96 microtiter plate, each well contains a microwell array patterned in a cell repellent hydrogel. On Gri3D®, organoids are robustly generated in the microwells and are located in the same imaging plane. This greatly facilitates quantitative analyses in high content image-based screens. Furthermore, the pipetting port enables automation of cell seeding, media exchange and compound incubation with liquid-handlers, thus increasing assay reproducibility. In the presented work, we exposed human pancreatic cancer PDOs to a panel of anti-cancer compounds at different doses and followed their response to the drugs using viability dyes Calcein AM (live stain) and Ethidium Homodimer-1 (dead stain) with high-content confocal imaging. Using an AI-based approach, we efficiently detected each single organoid and extracted phenotypic features from all three channels (TL, live stain and dead stain) which correlate with cytotoxicity. The extracted features (more than 100) were first dimensionally reduced to 3 components using either PCA or UMAP, the treatments were then visualized in 3D scatter plots, ranked and clustered using machine learning. The data suggests that the compound treatment Palbociclib 50uM has significant cytotoxicity effects similar to the positive control, which is consistent with the traditional live/dead analysis. The AI approach demonstrates feasibility to perform drug screening in a robust and un-biased data analysis approach. Citation Format: Zhisong Tong, Angeline Lim, Marine Meyer, Maria Clapes, Oksana Sirenko, Nathalie Brandenberg. AI-enabled hit selection of drug screening on human pancreatic cancer organoids [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 4919.