Precision-cut tumor slices for modeling hepatocellular carcinoma enable at-scale drug screening.

BACKGROUND: Disease modeling is vital for our understanding of disease mechanisms and for developing new therapeutic strategies. Accurately modeling the intact tumor microenvironment (TME) is increasingly recognized as essential for gaining insights into cancer biology and therapeutic response. Preclinical mouse models have provided utility for studying the evolving TME, but these models are costly and can lead to animal suffering and the discontinuation of drug investigations. To address these limitations, particularly in hepatocellular carcinoma (HCC), we have developed an ex vivo model using tumor precision-cut slices (TPCS) derived from orthotopic liver tumors. METHODS: Murine HCC tumors were generated via intrahepatic injection of Hep-53.4 cells, providing a source of tumor tissue for TPCS generation. Subsequent scaling to a 96-well format and modification to include a secreted luciferase enabled longitudinal ex vivo screening of 26 drugs applied at 2 doses over an 8-day period, using just 5 tumors. One drug identified in the screen, salinomycin, was then validated in vivo via intraperitoneal injection of mice with orthotopic liver tumors. RESULTS: Histological characterization determined that TPCS maintain the architecture, cellular complexity, and drug responsiveness of the original HCC-TME under simplified culture conditions that preserve viability and metabolic activity. In addition to typical HCC therapies, sorafenib and anti-PD1 immunotherapy, the screen identified 2 drugs as potent anticancer agents capable of impacting the viability of TPCS: salinomycin and rottlerin. Salinomycin was further validated in vivo, significantly reducing tumor burden without evidence of toxicity. CONCLUSIONS: We present a 3Rs (Reduction, Refinement, Replacement) approach for studying HCC biology and performing 96-well-scale drug screening within an intact, metabolically active TME, offering a more ethical and effective platform for drug discovery.