Researchers at Memorial Sloan Kettering Cancer Center (MSK) have reported the results of an initiative to characterize the genetic mutations in tumors from more than 10,000 patients with advanced cancer who were treated at the center.
The research team used a DNA sequencing test called MSK-IMPACT to characterize the genetic mutations in the patients’ tumors and used that information to match hundreds of patients to existing targeted therapies or to clinical trials. The researchers are continuing to follow up with participants to assess their long-term outcomes.
They also identified new cancer-related genetic mutations and found well-known mutations in cancer types where the mutations had not been observed previously. All of the data from the study have been made publicly availableExit Disclaimer in a way that protects patient privacy, allowing researchers across the world to access and analyze them.
The results demonstrate that a comprehensive, enterprise-wide effort to map the genetic alterations of patients’ tumors is feasible, can provide important information for timely patient care, and may help shape the future of cancer therapy, the researchers wrote.
Already “the data have been used by the research labs here at Sloan Kettering and are utilized and discussed as part of routine clinical care,” said the study’s lead investigator, Michael Berger, Ph.D.
The results of the study were reported May 8 in Nature Medicine.
DNA sequencing tests can identify genetic mutations that drive an individual patient’s cancer growth or development, which can help clinicians determine the most appropriate treatment.
The MSK-IMPACT test uses next-generation sequencing, a technology that allows several portions of DNA to be sequenced hundreds of times simultaneously, producing a genetic mutation profile quickly and with high accuracy. The researchers designed the test to identify different types of genetic alterations, including point mutations inside and outside of genes, alterations in the number of copies of a gene, and DNA rearrangements.
During the study period, MSK-IMPACT originally tested for alterations in 341 genes and was later expanded to include a total of 410 genes. The alterations covered by the test are constantly reviewed and revised by a team of clinicians, researchers, and clinical bioinformatics experts, explained Dr. Berger, allowing them to add newly discovered cancer-related genes or genes that are under lab or clinical study at MSK.
Some sequencing tests only focus on alterations typically found in one cancer type. But because the MSK-IMPACT test covers alterations in hundreds of genes, it is a broader approach that can be applied to any person with a solid tumor.
The researchers tested DNA from a biopsy sample of each patient’s primary or metastatic tumor and, for comparison, normal DNA from the patient’s blood cells. This allowed them to separate inherited from noninherited, or somatic, mutations. Testing and analysis were typically completed in less than 3 weeks.
Over a period of 2 years, the investigators tested 10,945 tumor samples from 10,336 patients with 62 types of advanced or metastatic cancer who had typically undergone several rounds of cancer treatment.
Overall they found 78,066 mutations, 22,989 alterations in gene copy number, and 1,875 DNA rearrangements. The majority of these alterations had not been previously identified. They estimated that 81% of the alterations would have been missed by other sequencing tests that only read frequently-mutated areas of genes, known as hotspots.
The researchers compared the mutations identified by MSK-IMPACT with those found via The Cancer Genome Atlas (TCGA), an initiative supported by NCI and the National Human Genome Research Institute that sequenced and analyzed untreated primary tumors from more than 11,000 patients with several types of cancer.
In general, the results of MSK-IMPACT and TCGA were highly consistent. However, they found many mutations at a higher frequency in the MSK study, which may reflect genetic differences between primary and advanced or metastatic tumors, Dr. Berger explained.
They also found some mutations that have been linked to treatment resistance that were not seen—and not expected to be seen—in the TCGA study, such as mutations in the androgen receptor gene that drive resistance to drugs that block the androgen receptor.
“Overall, things are pretty similar, but we do see expected differences,” Dr. Berger said. “We will continue to mine the data to look for acquired mutations associated with tumor progression.”
In addition, the investigators found several well-studied genetic alterations in unexpected cancer types. For example, certain DNA rearrangements that lead to fusion genes, including those involving the ALK gene, are typically present in lung tumors. However, the MSK researchers found these same fusion genes in 11 other cancer types.
Knowledge of genetic alterations that drive a patient’s cancer growth can be used to guide their treatment. In nearly 90% of patients, the researchers identified at least one genetic alteration that drives the cancer’s growth or development.
However, not every “driver” alteration is clinically actionable, meaning predictive of patient response to a Food and Drug Administration-approved or investigational therapy. An alteration may be a cancer “driver,” but it may not be actionable if there is no existing therapy that targets it. Using a databaseExit Disclaimer that catalogs clinically actionable mutations, they found that 37% of patients had at least one clinically actionable mutation.
In addition, some patients did not have a single actionable mutation, but rather, a group of mutations that are clinically actionable when present together. These groups, known as mutation signatures, can also be predictive of patient response to an approved or investigational therapy.
For example, a mutation signature known as microsatellite instability is predictive of patient response to treatment with immune checkpoint inhibitors. Hundreds of patients with this signature were subsequently enrolled in clinical trials and responded well to treatment with an immune checkpoint inhibitor.
The Impact of MSK-IMPACT
The MSK-IMPACT program “has been a huge team effort,” Dr. Berger said. “Hundreds of people are involved,” including teams of clinicians, researchers, clinical bioinformatics specialists, and software engineers. These teams also had to maintain tight coordination with the hospital’s medical records system, he added.
In addition to serving as a resource to guide treatment for individual patients, the data accrued from these tests also serve as an immediate investigative resource for the cancer research community. Researchers can carry out experiments to explore the biological functions of mutations identified by the study, which could potentially help clinicians interpret patients’ genetic sequencing results, Dr. Berger explained.
The data have also helped clinical researchers identify patients who qualify for open clinical trials at MSK. Some trials, called basket trials, recruit patients based on their tumor’s genetic alterations, rather than the organ where the patient’s cancer originated.
“If a new clinical trial were to open, we can identify the patients most likely to benefit all at once and enroll them very quickly,” Dr. Berger explained.
MSK-IMPACT is considered a “targeted panel” test because it sequences selected areas of DNA. Comprehensive sequencing approaches like whole-genome or whole-exome sequencing—which sequence the entire genome or large parts of it—may further enhance clinical sequencing efforts, said Jean Claude Zenklusen, Ph.D., director of NCI’s TCGA program office.
“Panels are [biased] by default because you are choosing” what DNA areas to sequence, he explained. “We can tell so much more from whole-exome and whole-genome sequencing.”
But given the limits to our current understanding of cancer genetics, the additional information gathered from these comprehensive tests may not be useful in the clinic right now, explained Dr. Berger. “By utilizing a targeted approach we were able to influence the treatment of a larger number of patients and identify more candidates for clinical trials,” he said.
Moving forward, Dr. Zenklusen believes that could change.
“I think at some point targeted panels are going to be a thing of the past because a more complete read gives you more information that helps you understand the picture,” he said. “Why use the CliffsNotes instead of reading the actual Shakespeare play?”
Zehir, A., Benayed, R., Shah, R. H., Syed, A., Middha, S., Kim, H. R., ... & Hellmann, M. D. (2017). Mutational landscape of metastatic cancer revealed from prospective clinical sequencing of 10,000 patients. Nature medicine.