以最少的PDO保留预测能力：为转移性CRC个体化治疗加速药物检测

Patients with relapsed or metastatic colorectal cancer (CRC) face limited treatment options, significant side effects, and prolonged delays in identifying effective therapies. Patient-derived organoids (PDOs; HUB Organoids®) provide a clinically relevant platform that faithfully mirrors individual tumour biology, enabling personalised drug testing. However, conventional drug screening formats typically require several hundred organoids per well, limiting the feasibility of using PDOs to guide real-time treatment decisions at diagnosis or relapse. Translational speed is critical: in metastatic CRC, there is a narrow window to select effective therapy before disease progression or treatment-related toxicity occurs. Traditional preclinical models often take weeks to months, which is too slow to inform immediate patient care. To address this limitation, we developed an automated organoid-handling workflow using the Yamaha CELL HANDLER™ system, enabling precise transfer and image-based quantification while requiring far fewer organoids per well. Miniaturisation reduced input material by 96% (from 250 to 10 PDOs per well) compared to conventional screening. Drug sensitivity of PDOs measured using the miniaturised assay closely mirrored that of conventional screening (R=0.67-0.85, p<0.03). PDO responses in the miniaturised assay also correlated with patient outcomes, including progression-free survival (R = -0.85, p < 0.01).By combining automation, miniaturisation, and quantitative readouts, this platform preserves the predictive power of PDOs, drastically reduces the need for organoids, and shortens turnaround time.