Abstract LB171: Drug response profiling of patient-derived tumoroids and immortalized cancer cell lines using a novel automation workflow and AI-assisted image analysis

Abstract Three-dimensional in vitro cell models have gained popularity as they better recapitulated key aspects of human tissues and tumor microenvironments. Forming structures with patient-derived (PD) cancer cells show great promise for translational work. These 3D models are being used for drug discovery and development, but adoption has been limited in part due to challenges related to sample handling, assay development, and the need for optimized instrumentation. Here, we present a novel automated workflow for drug safety and efficacy studies using PD 3D tumor models and show utility for translational research. Individual tumoroids comprised of patient-derived colorectal cancer (CRC) cells or patient-derived breast cancer (BC) cells were isolated using the spheroONE into individual wells of the Pu·MA System flowchip. Isolated tumoroids were treated with varying concentrations of doxorubicin, oxaliplatin, Cetuximab and 5-Flourouricil using the integrated microfluidic architecture of the Pu·MA and iDOT systems. Treated tumoroids were incubated with compounds for 48 hours and stained with viability dyes. Tumoroids were imaged using widefield fluorescence microscopy and analyzed for viability using a custom AI-assisted analysis program. Patient-derived BCs, immortalized estrogen receptor positive breast cancer cells (MCF7) and triple negative breast cancer cells (MDA-MB-231) will be used to further validate the workflow we outlined. Viability of drug treated tumoroids showed a dose dependent increase in cell death, consistent with known cellular responses seen in humans. Furthermore, phenotypic differences and responses were observed between primary CRCs and spheroids generated from immortalized colorectal cells (HTC116). The mechanisms for these differences are being investigated. These data suggest that classical drug discovery using immortalized cell lines may not accurately reflect cellular behaviors seen in CRCs and potentially other patient-derived cancer cells. This novel automated workflow allows for the potential to use tumoroids from patients to be rapidly screened for drug effects, thus providing a promising platform for future clinical validation and personalized medicine. Citation Format: Anthony Stephen Eritano, Dylan Bryda, Andreea Merloiu, Alissa Schutte, Gerri Ortiz, Barbara McIntosh, Seyoum Ayehunie, Evan Cromwell. Drug response profiling of patient-derived tumoroids and immortalized cancer cell lines using a novel automation workflow and AI-assisted image analysis [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 2 (Late-Breaking, Clinical Trial, and Invited Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_2):Abstract nr LB171.