| Popis: |
Background: Therapeutic screening Pancreatic Ductal Adenocarcinoma (PDAC) relies on well-level assessment for high throughput response evaluation. Patient-derived cancer organoids (PCOs) model subclonal populations, however the significance of resistant populations is uncertain when characterized with well-level response. Using a high-throughput screening assay, we present an automated alignment algorithm to characterize populations of organoid growth as compared to validated well-level therapeutic response assays. Methods: High content imaging was performed in low volume (10uL), 96-well angiogenesis plating format (Ibidi, Inc) at 4x objective with 5-frame Z-stack (600um) with brightfield imaging captured at 0h and 72h. Images underwent processing using Gen5 suite (Biotek, Inc) including Z-projection to render organoids into single two-dimensional planes. Baseline objects were defined between 50µm and 750µm, and filtration based on circularity defined as >0.4. Object alignment was performed based on root-mean-square-deviation (RMSD) between all combinations of objects to optimize match determination. This analysis was performed in drug screen of 80 independent agents in early clinical trials in combination with CDK7 inhibitor, SY-5609. Well level viability was performed using standardized 3D CellTiterGlo (CTG, Promega Inc.) (33% v/v). Response was assessed using descriptive statistics, effect size (Glass’s Δ), and Therapeutic Sensitivity Index (TSI) defined as the weighted average between elements with growth from media control versus treated population. Results: Z projection of 600um in Low-volume plating (10µL) of matrix suspension yielded 1.68 organoids per µL relative to the traditional hanging drop design (50µL) 0.75 organoids per µL (p5% reduction in of matched objects, while a decrease in circularity by 0.1 yielded a Conclusion: We provide a method for high fidelity alignment of PCOs in low-volume format for matrix-based screening applications. These techniques can be adapted to existing staining protocols to characterize subclonal response in the context of both molecular heterogeneity and clinical outcomes. Citation Format: Ethan Samuel Lin, Md Shahadat Hossan, Austin Stram, Eleanor E. Riedl, Luke J. Koeppel, Jaimie M. Warner, Jeremy D. Kratz. Automated organoid alignment for clonal response characterization in pancreatic ductal adenocarcinoma. [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 5327. |
