A team of researchers from the Wellcome Sanger Institute and Open Targets recently conducted one of the largest CRISPR screens of cancer genes to date (Nature 2019; https://dx.doi.org/10.1038/s41586-019-1103-9). The research is supported, in part, by Stand Up To Cancer (SU2C), specifically the SU2C-Dutch Cancer Society Tumor Organoids Dream Team.
The investigators utilized CRISPR technology to disrupt genes in over 300 cancer models from 30 cancer types and discovered thousands of key genes crucial for cancer's survival. A new system was then developed to prioritize and rank 600 drug targets that show the most promise for development into treatments.
"The development of new anti-cancer drugs remains extremely challenging. The vast majority of new therapies ultimately fail during clinical development and the range of new drug targets entering industry pipelines is actually on the decline," noted co-lead study author Mathew J. Garnett, PhD, from the Wellcome Sanger Institute. "This creates an extremely costly and inefficient process, which ultimately negatively impacts our ability to make the best medicines for patients.
"We reasoned that a more rigorous, systematic, and robust target identification at the beginning of drug development could help expand the range of targets, improve success rates, and accelerate the development of new therapies," he noted. "The use of functional genomics strategies, using technologies such as CRISPR screens, that effectively identify and prioritize candidate targets in tumors is one possible solution to this problem."
Study Details
To comprehensively catalogue genes required for cancer cell fitness, the researchers performed 941 CRISPR-Cas9 fitness screens in 339 cancer cell lines, targeting 18,009 genes. Following quality control, the final analysis set included 324 cell lines from 30 different cancer types across 19 different tissues, according to study authors.
The researchers focused on common cancers, including lung, colon, and breast, as well as cancers such as pancreatic where new treatments are urgently needed.
"We used CRISPR gene-editing technology to systematically 'deconstruct' cancer genomes by deleting each gene, one by one, in 324 cancer cell lines from 30 cancer types," explained Garnett. "This identified genes which were selectively required for the fitness of cancer cells in defined molecular contexts.
"These represent vulnerabilities in cancer cells which could be targeted therapeutically," he continued. "We integrated this information with genomic datasets from patients and information on the tractability of candidate targets for drug development to generate ranked lists of new targets for multiple cancer types.
"We identified over 600 candidate drug targets across multiple different cancer types. Some of these were associated with a genomic change in tumor cells, so called genomic biomarkers, that could be useful in selecting patients who would respond to a new therapy."
Amongst the targets, researchers identified and verified a striking dependency on the gene Werner syndrome helicase in cancers cells with a deficiency in a DNA repair process called mismatch repair, according to Garnett.
"This DNA repair defect occurs in a high proportion of colon, stomach, and endometrial cancers," he explained. "In contrast, loss of Werner had no effect in cells that do not have a mismatch repair defect. This is an extremely exciting finding and suggests that Werner could be a good drug target specifically in the setting of cancers which have a defect in this DNA repair process."
Garnett noted that the identification of Werner as a target in a defined subset of cancers is "likely to lead to drug development programs targeting this protein."
Implications, Next Steps
This research provides a rigorous and unbiased framework for the identification of candidate cancer drug targets, Garnett told Oncology Times. "We anticipate our work will lead to a broader range and more effective set of candidate drug targets in the future-which could help improve development success rates and bring patient benefit."
While the research lays the foundation for this approach, there is still more work to be done, Garnett noted. "We believe there could be benefits to expanding this approach across a larger and more diverse set of cancer types. For example, some specific tissues or histo-pathological subtypes are poorly represented in the current set of results. This means we can't identify new targets for these patients.
"There is also a huge amount of work to be done to follow up the candidate targets we have discovered, and we are very excited about the prospect of Werner as a new drug target. Overall, this study is part of a larger effort called the Cancer Dependency Map, which aims to systematically identify vulnerabilities in cancer cells which could be used to guide the development of new cancer medicines.
"We believe this will be important for translating our increasingly deep understanding of the genetics of cancer into improved treatments for patients," Garnett concluded.
Catlin Nalley is associate editor.