NEW YORK (GenomeWeb) – Large-scale mutational tumor profiling of advanced cancer patients treated at Memorial Sloan Kettering Cancer Center has allowed a significant number of patients to enroll in trials of targeted treatments and may help predict patients' response to immunotherapy, according to a new study.
In the paper, published in Nature Medicine yesterday, a large team of clinicians and researchers at the center reported that about a third of the first 10,336 patients profiled with its targeted next-generation sequencing test, MSK-IMPACT, harbored actionable mutations and 11 percent of patients participated in a genomically matched clinical trial based on their results. The data, some of which were presented at a meeting last year, and which the center is sharing with the scientific and clinical community, can also help with research, particularly of rare cancer types.
"It's a milestone for us in terms of the number of patients for who we've been able to offer comprehensive testing," said Michael Berger, associate director of the Marie-Josée and Henry R. Kravis Center for Molecular Oncology at Memorial Sloan Kettering and the senior author of the study.
The study includes results from patients who received MSK-IMPACT between January 2014 and May 2016, and it analyzed enrollment in clinical trials for the first 5,009 patients. As of today, close to 17,000 patients have been profiled by the test.
"As the number gets larger and larger, it gives us the ability to better understand rare tumor types and rare mutations that only become significant with large numbers of cases," Berger said.
Also, building a large database of patients with mutational tumor profiles enables the center to find good candidates for clinical trials. "Any time a new study opens, it gives the oncologists leading that study the opportunity to find patients who are being treated at Sloan Kettering that might be the best patients to enroll," Berger said. The program "has intangible benefits in that sense but it's also something that we as a cancer center feel benefits the patients that are being treated."
MSK-IMPACT, which is designed for solid tumors and analyzes both tumor and matched normal tissue, detects protein-coding mutations, copy number alterations, and certain promoter mutations and structural rearrangements in cancer-associated genes. Over time, the panel has grown in size from 341 genes to 410 genes, and most recently to 468 genes. The center has implemented a similar test for hematological cancers, called MSK-IMPACT Heme, that was approved by the New York State Department of Health a few months ago.
The center has also added a custom RNA-based gene fusion assay, approved by the health department last year, which it only runs for patients where MSK-IMPACT does not reveal a clinically relevant mutation. The RNA assay, which uses the ArcherDx technology, addresses fusions that MSK-IMPACT might have missed.
Testing with MSK-IMPACT is prioritized for patients with advanced cancers who are likely candidates for clinical trials. MSK has invested heavily in building the infrastructure for the test, and a lot of institutional and philanthropic funding supports the cost of testing, which is not considered standard practice and therefore not reimbursed by health insurance companies or Medicare for many tumor types.
MSK's pathology department, which is responsible for analyzing and reporting the test results, created a team of 16 computational biologists and software engineers just to manage the data for MSK-IMPACT, Berger said, and about a dozen pathologists, as well as many laboratory staff members and clinical fellows, are primarily involved with the initiative.
The study found that almost 37 percent of patients tested had clinically actionable mutations in their cancers, a number he said is maybe a little lower than those reported by other studies because the team used a fairly conservative definition of "actionable."
To determine the clinical utility of a mutation, the researchers used a curated knowledgebase of somatic mutations called OncoKB that was developed with input from nearly 100 oncologists and clinical experts. A description of OncoKB, which groups mutations into tiers of clinical actionability, will be published separately in the near future, Berger said.
Eleven percent of patients enrolled in a clinical trial for a targeted therapy based on a mutation found by MSK-IMPACT, which is a higher percentage than what other studies have reported, he said, although it is still considerably fewer than the fraction of patients with actionable results.
One reason for the relatively good enrollment is access to clinical trials, which is better at a cancer center like MSK with a large clinical trial portfolio. Memorial Sloan Kettering has also had a big effort in opening so-called basket studies, Berger said, which enroll patients on the basis of specific molecular alterations across many different cancer types, allowing even patients with rare tumor types to join. Moreover, the MSK-IMPACT gene panel is fairly comprehensive, so if there is a targetable mutation, it is expected to be found, Berger said. Finally, because most patients profiled with MSK-IMPACT have advanced or metastatic cancer, they are more likely to be eligible for trials, and more willing to participate.
Enrollment in clinical trials may also increase over time, as patients with test results who were originally treated conventionally may see their disease progress in the future. "You have to let some time accrue before you can close the book on whether patients' treatment is influenced by their MSK-IMPACT result," he said.
Going forward, results from MSK-IMPACT might actually become even more actionable, because they may be able to predict which patients benefit from immunotherapy. The test identified 102 patients with a mismatch repair signature and evidence of microsatellite instability, for example, and MSI status is increasingly used as a biomarker for response to immune checkpoint inhibitors. One prostate cancer patient, for instance, had an unanticipated MMR signature and received immunotherapy as a result, with good response.
"I think there is a lot of justified excitement around immunotherapy, and there are a lot of patients who had MSK-IMPACT testing done and it was deemed that immunotherapy was the most promising option for them," Berger said, "and there are patients who have received both targeted therapy and immunotherapy, either separately or together."
In patients who received MSK-IMPACT testing and also underwent immunotherapy, the researchers have looked retrospectively whether their test results harbored biomarkers predicting their immunotherapy response, he said, for example complex mutation signatures, and the results are currently under review for publication.
Because immunotherapy has been so successful in a number of tumor types, it may become the default treatment for many patients, he said, even in the absence of high mutation burden or a mutation signature. However, "I don't think this necessarily means that [patients] go straight to immunotherapy and would not consider a targeted therapy — at least not here," he added.
Another way MSK-IMPACT has benefited patients that is not mentioned in the study is by detecting inherited mutations that expose a cancer risk in their family. Because the test analyzes both tumor and matched normal tissue — a feature that differentiates MSK-IMPACT from many other targeted cancer tests — somatic and germline mutations can be clearly distinguished. Memorial Sloan Kettering developed a program for communicating such results to patients, which involves a separate informed consent and genetic counseling. The findings often lead to follow-up visits and monitoring of patient family members, Berger said, and an analysis of data for patients who received germline results is currently under review for publication.
In 2015, the researchers published an analysis of pathogenic germline alterations in an anonymous patient cohort, which showed that such mutations are more prevalent than expected and show up in patients where clinical guidelines would not have recommended genetic testing.
Whether or not tumor panel testing improves the long-term survival of patients, however, is not clear yet. "That's the most important next step, to see if overall outcomes are being improved on the basis of the mutation profiling," Berger said. "We certainly have many individual cases of patients who received therapies that they benefited from that would not have been considered without the MSK-IMPACT results. But there are many patients who don't, where there isn't a therapy that's available today." This might change, he said, as new drugs are developed and proven to be effective through clinical trials.
Part of the problem is that not enough time has elapsed since patients have entered trials, so outcome studies have not been published yet. The researchers are also working on improving the accuracy of outcome data that can be extracted from medical records, which Berger said is a highly manual process at the moment.
To enable other researchers to study data from MSK-IMPACT, and find correlations between genomic mutations and patients' therapy response, Memorial Sloan Kettering is making the data available through the cBioPortal for Cancer Genomics, including basic de-identified clinical data. The center also participates in the American Association for Cancer Research's Project GENIE.
In addition, MSK has a number of initiatives to make MSK-IMPACT available, on a limited basis, to patients outside of the institution, for example through partnerships with community hospitals. One program, called "Make an Impact", allows patients with very rare tumor types to have their tumors sent in and analyzed by MSK-IMPACT.