A next-generation sequencing (NGS) assay has accurately identified cancer-related genetic variants from different tumor tissues, with near identical results, at four different laboratories, the NCI has announced.
It marks the first time results of an NGS assay have been validated and reproduced across laboratories, and clears the way for the study's next step: NGS testing of thousands of adults with advanced solid tumors, lymphomas, or myelomas who fail standard therapy and are candidates for available personalized treatments.
Results from the phase II NCI-Molecular Analysis for Therapy Choice, or NCI-MATCH trial, demonstrated that the assay is extremely sensitive—almost 97 percent for 265 known mutations—with 99.99 percent specificity and inter-operator agreement at the four laboratories (J Mol Diagn 2017;19:313-327).
NCI-MATCH seeks to determine whether treating cancers with drugs that target specific molecular abnormalities are effective. Eventually the trial will include 10 separate research arms, each testing a different molecular-targeted therapy. The goal is to screen thousands of patients who have relapsed or have refractory solid tumors and lymphomas after standard systemic treatment. They are being recruited from up to 2,400 National Clinical Trial Network (NCTN) sites.
Validation of the tool was necessary to ensure its performance was acceptable for screening patients, and that standardization and reproducibility were possible at different molecular diagnostic laboratories that will process and analyze biopsy specimens. NCI-MATCH researchers also developed standard operating procedures.
Led by the Eastern Cooperative Oncology Group-American College of Radiology Imaging Network Cancer Research Group, the next phase will involve patients enrolled at NCTN sites, and as many as 2,400 research centers. The trial will evaluate tumor biopsy specimens collected from about 6,000 patients. The tumor types analyzed in the new study represented tissues likely to be submitted for analysis as the trial moves forward.
From biopsy collection to reporting, the assay system was found “fully fit for purpose,” the researchers said, adding that the validation process used can serve as a template for validating other NGS assays.
The researchers use an assay that is able to detect 4,066 genomic variations, including “actionable” molecules of interest, or MOIs. The test was able to identify more than 4,000 pre-defined genomic variations across 143 genes, including those associated with cancers of the pancreas, bone, skin, and melanoma. These variations included single nucleotide variants (SNVs), insertions/deletions (indels), copy number variations (CNVs), and gene fusions.
The laboratories sequenced 186 samples and 12 cell lines and the results were evaluated for uniform sensitivity, specificity, reproducibility, and limit-of-detection—the lowest quantity of a genetic variant needed for reliable detection. The limit of detection for each variant type was 2.8 percent for SNVs, 10.5 percent for indels, 6.8 percent for large indels (gap ≥4 bp), and four copies needed for gene amplification. Only six samples failed to pass quality metrics, but all passed when sequenced again.
“The validation study is another step moving the field closer to the time when precision medicine will generate the expected benefits in improved clinical outcomes,” said Elizabeth R. Unger, MD, PhD, of the Division of High-Consequence Pathogens and Pathology at the CDC.
“We really do not know if we can treat patients with a drug reasonably likely to address a specific mutation regardless of the tumor type, and NCI MATCH allows us to study that issue,” said Barbara Conley, MD, Associate Director of the Cancer Diagnosis Program at NCI.
“In terms of clinical results, MATCH is a signal finding trial,” she told Oncology Times. “We have put in place levels of evidence before we include a drug and before we include molecular variants in the study. Any drug included in NCI MATCH must have at least shown some response or benefit in a patient with a similar molecular variant to those under study.”
Molecular variants that are included must at least have evidence in cell or animal models that the variant can be impacted by a drug in the same class as the drug included in NCI MATCH, she explained.
“Assessing benefit for patients will likely require additional clinical trials in for example studies that include just one tumor histology that has the mutation of interest.”
The study focuses on refractory cancers because the investigators did not want to potentially harm patients in a signal finding trial by allowing them to be treated with investigational treatments when there are proven treatments that can provide benefit.
“Such testing is already available in a number of laboratories, both in academic centers and in commercial laboratories,” added Keith T. Flaherty, MD, Director of the Termeer Center for Targeted Therapies at Massachusetts General Hospital Cancer Center.
“In routine clinical practice, a test of this sort is typically used to identify genetic alterations for which there are approved therapies. However, we are not attempting to validate this platform for its clinical utility.”
Instead, the analytical question is whether such a test can be used to empower a large-scale precision medicine trial exploring multiple investigational therapies.
“Here, in the context of this phase II clinical trial setting, we have validation that the diagnostic platform we selected for the trial can be used simultaneously to find genetic alterations for which there are approved therapies and, more commonly, for which there are investigational therapies that can be assigned based on the genetic finding,” Flaherty noted.
The findings begin a long-overdue process of establishing a framework for uniform NGS testing, said David Neil Hayes, MD, Associate Professor of Clinical Research in Hematology and Oncology at the University of North Carolina at Chapel Hill's Lineberger Comprehensive Cancer Center.
He said current NGS testing is being offered under governing laboratory developed testing rules, in vitro diagnostic tests designed, manufactured, and used within a single laboratory.
The statute was intended for much less complex testing, such as those for blood clots—not whole genome screening, he told Oncology Times, adding that the FDA has lagged behind in developing regulations applicable to NGS assays.
“The statute has very specific rules, but over the past 10 years it has been used to allow labs and hospitals to offer genetic testing. This was never the law's intent, certainly not with regard to testing by laboratories at the national level,” Hayes said.
“This study at last has begun addressing a long-overdue regulatory gap. This type of sequencing has become very routine using commercial products from several vendors, but there has been very little oversight. The problem is that this has created a kind of Wild West scenario. No one really knows how good the available tests really are,” he noted.
“The new data begins a process of standardization and democratization of NGS across multiple laboratories using a test that has now been validated for sensitivity and specificity.”
Kurt Samson is a contributing writer.