Background

With the expansion of personalized precision medicine, the analysis of individual tumor's genetic characteristics has become increasingly important. The Broad Institute in the United States has been leading the Cancer Dependency Map (DepMap) project for the past 10 years, which aims to identify cancer cell survival genes. This research provides a foundation for the development of targeted therapeutics by screening for genetic vulnerabilities in cancer cell lines using gene editing technology.

However, 2D cell culture models have limitations in replicating the tumor environment in vivo. They fail to mimic the cell-cell interactions or physical stimuli of tumors, which have a 3D structure. This has been a cause of failure in clinical trials for effective substances. 3D culture of patient-derived tissues into organoids and spheroids has emerged as an alternative, but large-scale genomic screening data has been lacking.

Key Findings

The research team performed genome-wide CRISPR screening on 148 next-generation 3D cancer models from 10 cancer types. This data was integrated with data from more than 1,000 existing 2D cell lines to analyze changes in gene dependency according to the culture environment. 3D suspension-cultured neurospheres and organoids grown in gel realistically reflected the actual cancer state of patients.

The analysis revealed that 3D models exhibited unique genetic vulnerabilities that were not detected in 2D cultures. For example, in glioblastoma organoids, cells with loss of the tumor suppressor gene CDKN2A were extremely sensitive to CDK6 inhibition compared to normal control cells. This provides evidence to introduce CDKN2A deletion as a precision diagnostic biomarker when applying existing CDK6 inhibitors to the treatment of glioblastoma patients.

Unique vulnerabilities were also identified in digestive system organoids, such as pancreatic cancer. The specific gene expression patterns of patient tumors were maintained in 3D organoids but were lost in 2D cell lines. Cancer cells with this pattern strongly depended on the WNT signaling pathway for survival. This vulnerability is only manifested in the 3D environment, making it a new milestone for future targeted therapy research.

The specific causes of dependency changes were also analyzed. Genes involved in cell adhesion and cytoskeleton formation were sensitive to physical culture forms, while genes related to lipid metabolism were regulated by the culture medium components. This demonstrates the importance of selecting an appropriate culture method that matches the experimental purpose.

Significance and Prospects

The large-scale 3D dependency data established in this study is an asset that will expand the horizons of precision medicine. The research data is publicly available on the DepMap portal and Cell Model Passports for anyone to use. It is expected to be used as a standard material to increase the success rate of clinical prediction in the drug candidate screening stage.

This joint study, which involved the Wellcome Sanger Institute in the United Kingdom and the National Cancer Institute (NCI) in the United States, as well as the Human Cancer Models Initiative (HCMI), has raised the level of global cancer treatment research. The integration of organoid bank information has strengthened the ability to analyze rare cancers. However, the high cost of maintaining and screening 3D models is a barrier. The research team plans to focus on follow-up studies aimed at standardizing culture and reducing costs.