CRISPR screens in cancer spheroids identify 3D growth specific vulnerabilities

Cancer genomics studies have nominated thousands of putative cancer driver genes(1); a major challenge is to develop high-throughput and accurate models to define their functions. Here we devised a scalable cancer spheroid model and performed genome-wide CRISPR screens in 2D-monolayers and 3D lung cancer spheroids. CRISPR phenotypes in 3D more accurately recapitulate those of in vivo tumors, and genes with differential sensitivities between 2D and 3D are strongly enriched for significant mutations in lung cancers. These analyses also revealed novel drivers essential for cancer growth in 3D and in vivo, but not in 2D. Notably, we discovered that CPD (Carboxypeptidase D) is responsible for removal of a c-terminal RKRR motif(2) of IGF1R α-chain, critical for receptor activity. CPD expression correlates with patient outcomes in lung cancer, and loss of CPD reduced tumor growth. Our results reveal key differences between 2D and 3D cancer models, and establish a generalizable strategy to perform CRISPR screens in spheroids to uncover cancer vulnerabilities.