A new study headed by teams at the Wellcome Sanger Institute, EMBL’s European Bioinformatics Institute (EMBL-EBI), and Open Targets has indicated how mutations that cause cancer drug resistance fall into one of four categories. The work detailed each of the four types of mutation, outlining how known mutations impact drug resistance, and also highlighting new DNA mutations that could be explored further. The results also indicate possible second-line cancer treatment options based on an individual’s genetic makeup.
Mathew Garnett, PhD, at the Wellcome Sanger Institute and Open Targets, said, “Before this study, it has been difficult to get a large-scale understanding of why and how drug resistance in cancer develops. This research brings us one step closer to being able to match combination or second-line therapies to a person’s genetic makeup, to try and ensure that treatments are as effective and personalized as possible. Additionally, we believe that our new systematic approach will be important for understanding genetic mechanisms of resistance to new drugs in the future. This could help even before the emergence of resistance in the clinic, and these early insights will improve developing cancer treatments.”
Garnett is senior author of the team’s published paper in Nature Genetics, titled “Base editing screens define the genetic landscape of cancer drug resistance mechanisms,” in which they concluded, “Our variant-to-function map has implications for patient stratification, therapy combinations and drug scheduling in cancer treatment.”
“Drug resistance is a principal limitation to the long-term efficacy of cancer therapies,” the authors wrote. Mutations in cancer cells mean that over time they become less responsive to therapies. “Drug resistance is frequently caused by DNA single nucleotide variants (SNVs) in the cancer genome, leading to point mutations in the drug target or proteins within the same signaling pathway.”
If a cancer has become resistant to initial treatment, options for second-line therapies can be limited. Understanding what molecular changes and mechanisms are causing the resistance, and what can be done to tackle this, can help uncover new options and inform clinical pathways for specific mutations. As well providing insight that could help scientists develop next-generation cancer drugs that avoid drug resistance emerging, such knowledge could point to new targets for personalized therapies, inform on potential treatments for individual patients based on their cancer’s genetic makeup, and suggest second-line treatment options for those patients who may currently have none. “The study of drug resistance can inform drug mechanism of action, the design of second-generation inhibitors targeting drug-resistant proteins, the development of combination therapies and patient stratification for second-line therapies,” Garnett and colleagues added.
However, current methods for identifying drug-resistance mutations require multiple samples from patients collected over a long time, making this a time-consuming and difficult process. “These are challenging samples to acquire, meaning it can take years to accrue enough to infer variant function,” the team noted. “These analyses are generally restricted to frequently observed variants and must be individually experimentally validated to establish a causal link to drug resistance.” This is “a slow process,” they continued, which also doesn’t allow for the direct comparison of different variant effects. “Rapid, prospective and systematic functional annotation of variants would accelerate the discovery of drug resistance mechanisms.”
For their newly reported work to gather large-scale information on cancer mutations the researchers used CRISPR gene editing and single-cell genomic techniques to investigate the impact of multiple drugs across human cancer cell lines and organoid cell models. By combining these techniques, they were able to create a map showing drug resistance across colon cancer, lung cancer, and Ewing sarcoma cell lines. These were chosen as they are prone to developing resistance and have limited second-line treatments available. The investigators tested 10 cancer drugs that are either currently prescribed or undergoing clinical trials, to investigate if any of these treatments could be repurposed or used in combination to address resistance, decreasing the time it would take to get any potential treatments to the clinic.
“We report a prospective genetic landscape of drug resistance mechanisms in cancer, one of the most comprehensive functional investigations of genetic drug resistance mechanisms to date, comprising ten drugs and profiling 11 cancer genes spanning common drug targets and oncogenic pathways,” they noted.
The map uncovered insights into the mechanisms of drug resistance, highlighting DNA changes that may be potential treatment biomarkers, and identifying promising combination or second-line therapies. “In summary, we systematically categorize variants that confer resistance to several inhibitors and highlight possible alternative inhibitors or treatment schedules that could be effective in treating drug-resistant cancers,” the scientists further commented. “We establish a framework for the functional classification of variants modulating drug sensitivity, which could inform clinical management.”
The team found that cancer mutations fall into four different categories dependent on the impact of the DNA change. “Single-cell transcriptomics enabled functional classification of drug resistance variants based on their mechanisms of action and transcriptional impact.”
Of these four types, drug resistance mutations, otherwise known as canonical drug resistance mutations, are genetic changes in the cancer cell that lead to the drug being less effective. For example, mutations that mean the drug can no longer bind to its target in the cancer cell.
Drug addiction mutations lead to some of the cancer cells using the drug to help them grow, instead of destroying them. For these types of mutations the study results support the use of “drug holidays,” which are periods without treatment. Such an approach could help to destroy the cancer cells with this type of mutation, as the cells become dependent on the treatment.
Driver mutations are gain-of-function genetic changes that allow cancer cells to use a different signaling pathway to grow, avoiding the pathway that the drug may have blocked.
The fourth category, sensitizing variants, comprises genetic mutations that make the cancer more sensitive to certain treatments, and could mean that patients with these genetic changes in their tumor would benefit from particular drugs.
First author Matthew Coelho, PhD, at the Wellcome Sanger Institute and Open Targets, said, “Cancer cells developing resistance to treatments is a huge problem, and having a rapid way to identify these mutations in patients and understand how to combat them is key to treating cancer. Our study details how mutations fall into four different groups, which might need different treatment plans. For example, if there are drug addiction mutations, taking a break from treatment may help. By using cutting-edge genetic techniques, we have started to build a large-scale and rapid way to understand drug resistance and hopefully find new targets for second-line treatments.”
Understanding more about the four different types of DNA changes may help support clinical decisions, explain why treatments are not working, support the idea of drug holidays in certain patients, and aid in the development of new treatments. This knowledge could also help research into next-generation cancer inhibitors that could better avoid drug resistance. “Our prospective and systematic approach could be important for understanding genetic mechanisms of acquired resistance to new molecules in the future, even before the emergence of resistance in the clinic, thereby providing early insights to improve cancer treatment efficacy,” the researchers concluded.
Magdalena Strauss, PhD, previously of EMBL’s European Bioinformatics Institute (EMBL-EBI) and now at the University of Exeter, stated, “By combining cutting-edge CRISPR gene editing and single-cell techniques with statistical machine learning, we have been able to obtain a detailed picture of the specific mechanisms by which each of the individual mutations that we studied impacts drug response. The functional framework that we have built allows researchers to start to piece together a complete map of common DNA changes seen during cancer treatment, adding to our collective knowledge. It also highlights mutations that could be used as biomarkers, highlighting cancer cells that are more sensitive to certain treatments, which could help inform future clinical trials.”