Bottom-Up Programming of Cell States in Cancer Organoids with Defined Synthetic Adhesion Cues.

Despite advances in defined culture systems, current organoid models lack programmable control of transcriptomic states beyond fixed genetic constraints or broadly specific microenvironmental conditions. Here, a bottom-up biomaterial-based platform is introduced to program cell state changes in pancreatic cancer organoids by tuning minimal adhesion cues within a synthetic matrix. A Design of Experiments framework is used to systematically model the patient-specific transcriptome-wide impact of matrix-presented adhesion cues. Focusing on epithelial-mesenchymal transition (EMT) as a proof-of-concept cellular program, a multiobjective optimization approach is applied to identify patient-specific matrix compositions that enrich EMT-associated transcriptional programs. Organoids cultured in these optimized matrices exhibit transcriptomic signatures consistent with EMT enrichment and coordinated shift in EMT-associated regulatory signatures. Secretome profiling further reveals changes in cytokines previously linked to EMT-associated inflammatory, hypoxia, and TGF-β signaling. Together, these findings demonstrate that quantitative and targeted modulation of defined adhesion cues enables programmable control of transcriptomic states in pancreatic cancer organoids.