A new open resource of patient-derived organoid models has enabled researchers at the Wellcome Sanger Institute and clinical sites across the UK to construct the first large-scale map of cancer gene dependencies, according to a study published on August 5 in Nature. The biobank of 256 tumor models spans colorectal, esophageal, pancreatic, stomach, and ovarian cancers, providing scientists with a robust platform to identify critical genetic vulnerabilities and test potential treatments.
Building a Next-Generation Tumor Biobank
For decades, cancer laboratories relied heavily on two-dimensional cell lines grown on flat plates. While these traditional models offered foundational insights into tumor biology, they struggle to capture the complex diversity of actual patient tumors and often adapt to laboratory environments over time, according to the Wellcome Sanger Institute researchers. To bridge this gap, investigators partnered with clinical sites in Birmingham, Cambridge, Glasgow, London, and Southampton to collect fresh tumor samples from consenting patients.
The multidisciplinary team isolated cells from the fresh tissue and established specialized conditions that allowed them to grow into 3D organoid cultures. By sequencing DNA from the resulting organoids, patient blood samples, and original tumors, the research team created a deeply characterized reference resource designed to track genomic changes over extended periods of laboratory growth.
Mapping Genetic Dependencies With CRISPR Screening
To identify the specific weak points of these cancers, researchers applied CRISPR screening across 162 of the organoid models. By systematically deactivating genes one by one and observing whether the cells survived, the team mapped thousands of genetic dependencies, separating common survival genes from vulnerabilities unique to specific cancer types.
According to Dr. Carmen Herranz-Ors, first author at the Wellcome Sanger Institute, building this long-term resource enabled the team to study cancer in models resembling patient tumors much more closely. By combining these functional screenings with rich clinical and genomic data, the researchers identified 1,733 distinct links between genetic dependencies and specific features such as DNA alterations or prior treatment histories.
Tracking Treatment Resistance and Guiding Patient Care
Comparisons between organoids grown from the same patient before and after therapy revealed how specific tumors developed resistance to treatment while exposing new biological vulnerabilities. Dr. Andrew Beggs, Professor of Cancer Genetics and Surgery at the University of Birmingham, noted that this close collaboration between scientists, clinicians, and patients provides a clearer picture of how cancers behave in the clinic and why patient responses vary.
Dr. Catherine Elliott, Director of Research at Cancer Research UK, emphasized that studying cancer in such detailed models brings researchers closer to identifying new treatments that can improve patient survival. Dr. Mathew Garnett, senior author at the Wellcome Sanger Institute, added that bringing together patient-derived models, genomics, and functional screening at this scale offers a powerful new framework for uncovering cancer vulnerabilities across diverse tumor types.
Did You Know? The study evaluated 256 organoid models derived from five distinct cancer types where new therapeutic options are urgently needed, creating the most detailed reference resource of its kind to date.
Frequently Asked Questions
What are tumor organoids?
Tumor organoids are three-dimensional cultures grown from patient tissue samples that closely mirror the structural and genetic complexity of real human cancers.
How does CRISPR screening work in cancer research?
CRISPR screening allows scientists to systematically turn off individual genes in laboratory models to see which ones cancer cells rely on to survive and grow.
Why do researchers use organoids instead of traditional cell lines?
While 2D cell lines remain useful, organoids better preserve the diversity of patient tumors and do not adapt to laboratory conditions as quickly, making them more informative for studying drug resistance.