Abstract 2047: Scale up and scale down approaches for screening of 3D patient-derived cell models

Abstract 3D cell culture is becoming more prevalent as scientists work to better model human physiology using in vitro systems. However, differences in handling 3D cell cultures compared to traditional 2D cell lines have limited adoption of 3D techniques in compound and cell therapy screening experiments. To facilitate use of patient-derived 3D tumoroid models, we developed protocols for both scale up and scale down of these models in Gibco™ OncoPro™ Tumoroid Culture Medium. OncoPro Tumoroid Culture Medium is a serum-free, conditioned medium-free culture medium designed to promote the growth of patient-derived 3D cancer models (tumoroids, cancer organoids). OncoPro medium is compatible with a suspension culture approach for easier handling compared to traditional embedded culture formats. We studied the growth of tumoroids in different flask sizes during suspension culture to optimize scale up for downstream assays. Tumoroid growth rates and morphologies were consistent across non-tissue culture treated flask sizes, and viable cell yield increased with surface area during scale up. In some cases, as many as 100 × 106 dissociated tumoroid cells can be recovered from a single Nunc™ TripleFlask™ after a week of growth. After scaling up, tumoroids were dissociated and seeded in multiwell plates to test multiple screening conditions (compound identity, compound concentration) in parallel. Dissociated cells were seeded in 96-well plates using both manual and automated liquid handling (Tecan Fluent 780). Due to the tendency of 3D cell models to aggregate and fall out of solution, initial seeding in a 96-well flat bottom plate resulted in high well-to-well variability. Optimization of cell seeding protocols led to standard deviations in tumoroid metabolic activity, number, and size that were comparable to manual seeding, with a coefficient of variation of less than 3% between wells for multiple tumoroid lines when automation was implemented. Generally, variability in cell seeding was lower when using the optimized, automated cell seeding protocol compared to manual methods. After seeding, cells were grown for three days to form tumoroids prior to the addition of compounds. At this time point, there was low well-to-well variability in tumoroid metabolic activity (CV<2%). Compounds and reagents to analyze cell health could also be added using automated liquid handling prior to analysis on plate readers or high-content imaging platforms. Citation Format: Colin Paul, Anthony Chatman, Amber Bullock, Xiaoyu Jenny Yang, Garrett Wong, Brittany Balhouse, Chris Yankaskas, Erik Willems, Matt Dallas, David Kuninger. Scale up and scale down approaches for screening of 3D patient-derived cell models [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 2047.