Predicting clinical response to anticancer drugs using an ex vivo platform that captures tumour heterogeneity

摘要: Predicting clinical response to anticancer drugs remains a major challenge in cancer treatment. Emerging reports indicate that the tumour microenvironment and heterogeneity can limit predictive power of current biomarker-guided strategies for chemotherapy. Here we report engineering personalized ecosystems contextually conserve heterogeneity, phenocopy using explants maintained defined grade-matched matrix support autologous patient serum. The functional ecosystems, engineered from 109 patients, drugs, together with corresponding outcomes, is used train machine learning algorithm; learned model then applied predict an independent validation group 55 where achieve 100% sensitivity predictions while keeping specificity desired high range. ecosystem algorithm, termed CANScript technology, emerge as powerful platform enabling medicine.