The Sanger Institute has opened a bank of 256 tumor organoids and mapped the genes that cancers need to grow
The Sanger Institute has opened a bank of 256 tumor organoids and mapped the genes that cancers need to grow
On August 5, the Wellcome Sanger Institute published an open collection of organoids representing five cancer types in Nature. An organoid is a three-dimensional culture grown from tissue taken from a specific tumor. In 162 of these cultures, the researchers disabled genes one at a time and measured how each intervention affected cell growth.
DNA sequencing identifies changes in a tumor, but experiments in living cultures are needed to determine how those changes affect its growth. Conventional cell lines grow in laboratories for years and adapt to laboratory conditions. An organoid is grown from an individual patient’s tumor tissue, so experimental results can be compared with the genome of the original tumor, a matched normal sample from the patient, and information about the patient’s treatment.
The team collected material through a network of five hospitals. From 907 samples provided by 878 donors, the researchers established 256 renewable cultures of colorectal, esophageal, ovarian, pancreatic, and gastric cancers. A matched original tumor was available for 171 cultures. In 76 pairs, at least 75% of somatic mutations were shared. This level of agreement links an experimental result in a culture to the tumor from which that culture was derived.
The team used CRISPR-Cas9 in 162 organoids. This method disables a selected gene. If the culture grows more slowly after the gene is disabled, its cells need that gene for growth. The researchers then compared these dependencies with mutations, tumor subtype, and treatment information. This gives the mutation map a functional interpretation by showing which tumor variant depends on a specific gene.
KRAS provides a clear example of how the map works. This gene is frequently mutated in colorectal cancer. The EGFR receptor protein on the cell surface transmits a growth signal through a signaling pathway that includes KRAS. Across 85 organoids, cultures with KRAS mutations depended more strongly on KRAS itself, while cultures without these mutations depended more strongly on EGFR. In tests involving 21 organoids, models with high EGFR dependency in the CRISPR screen were among the most sensitive to afatinib and gefitinib, which are drugs that inhibit EGFR. The map can therefore be used to test why different tumor variants depend on different genes.
The open bank gives other laboratories access to cultures linked to the original tumors and patient data, allowing them to test these dependencies in a reproducible model.