Researchers built a publicly available tumor organoid biobank comprising 256 clinically annotated organoids spanning colorectal, oesophageal, ovarian, pancreatic and gastric cancers, paired with whole-genome and transcriptome data and genome-wide CRISPR–Cas9 screens across 162 organoids. The resource provides gene dependency mapping intended to overcome limitations of conventional cell lines, including adaptation to culture and lack of patient-matched clinical context. Organoid evolution was analyzed using paired pre- and post-treatment samples to capture changing vulnerabilities. In colorectal cancer, functional work on the EGFR–RAS–MAPK pathway showed differential effects across KRAS variant alleles, illustrating how genotype can shape drug response profiles even within shared pathway frameworks. The study’s value for drug discovery lies in the integration of multi-omics and large-scale perturbation results within patient-derived, genetically stable 3D models. It expands the set of mechanistic starting points for precision oncology programs, including target selection and biomarker development.