Patient responses to therapeutic agents in hepatic malignancies are highly variable, yet the field lacks robust in vitro models for predicting individual drug responsiveness. Patient-derived organoids (PDOs) that faithfully recapitulate the biological characteristics of parental tumors offer a promising platform for personalized drug testing. We established PDOs from fresh tumor tissues of 29 patients with hepatic malignancies, achieving a success rate of 79.31% (23/29). Histological staining confirmed high morphological and phenotypic consistency between PDOs and original tumor tissues. Drug sensitivity testing was performed using a panel of standard-of-care therapeutics, and model stability was assessed through serial passaging and cryopreservation. Clinical correlation was evaluated by comparing PDO-predicted sensitivity with observed treatment responses in 16 patients followed prospectively. The established PDOs maintained structural and functional fidelity to parental tumors and demonstrated reproducible drug sensitivity profiles across serial passaging and cryopreservation. In a prospective clinical validation, patients whose treatment matched PDO-predicted sensitive agents showed significantly improved objective response rate (ORR: 36.36%, 95% CI: 10.93–69.21) and disease control rate (DCR: 90.91%, 95% CI: 58.72–99.77), compared to those receiving PDO-predicted resistant therapies (ORR: 0%, DCR: 20%). This study establishes a robust and clinically validated PDO platform for hepatic malignancies that accurately predicts patient-specific drug responses. The model demonstrates high reproducibility and stability, supporting its utility for guiding personalized therapy selection and improving outcomes in patients with liver cancer. Existing models for predicting drug response in hepatic malignancies, such as cell lines and animal models, have significant limitations in faithfully recapitulating human tumor biology and clinical outcomes, creating a barrier to personalized therapy. This study establishes a clinically validated patient-derived organoid (PDO) platform for hepatic malignancies that accurately predicts individual patient drug responses. It demonstrates that PDO-guided therapy selection is associated with significantly improved clinical outcomes, and confirms the model’s stability for practical biobanking and testing. This work provides a directly applicable tool for de-risking therapeutic decisions in clinical oncology. It can immediately inform personalized treatment strategies, serve as a robust platform for drug development and biomarker discovery, and promote a shift towards evidence-based, patient-specific therapy selection.
Zhou et al. (Fri,) studied this question.