利用患者来源类器官研究实体瘤背景下抗体偶联药物的旁观者效应

Background: Antibody-drug conjugates (ADCs) have demonstrated significant clinical benefit, yet their efficacy can be influenced by heterogeneous target expression and bystander killing. Understanding these dynamics in physiologically relevant models is critical for optimizing ADC design and patient selection. Methods: We established and expanded patient-derived organoids (PDOs) from solid tumor backgrounds using our automated CX.ai platform, enabling high-throughput and standardized culture conditions. Target abundance for HER2 and c-MET was assessed via immunohistochemistry (IHC). Co-culture systems were optimized to include disease and healthy PDOs at defined ratios. Experimental assays incorporated mosaic PDO populations treated with ADCs (trastuzumab deruxtecan, telisotuzumab adizutecan, ADCs) and controls (antibody alone, warhead alone). Readouts included live-cell imaging, fixed-cell labelling, and high-content analysis to quantify PDO shrinkage, cell death, and target expression. Results: Preliminary data indicate feasibility of generating mosaic PDO systems and detecting differential responses to ADCs. Optimization of co-culture conditions and imaging-based metrics (nuclear count reduction, PDO object shrinkage) are underway to correlate target expression heterogeneity with bystander killing. Conclusions: PDO-based platforms combined with automated culture and advanced image analysis provide a robust system to study ADC bystander effects in heterogeneous tumor contexts. These findings will inform ADC development strategies and biomarker-driven patient selection.