The Genomics and Pathology Services Laboratory at Washington University in St. Louis has expanded its next-generation sequencing clinical cancer panel and recently launched a cardiomyopathy panel, following success with its initial 28-gene cancer panel.
The CLIA certified and CAP accredited laboratory within Wash U's School of Medicine initially broke into the clinical sequencing field in 2011 with a 28-gene cancer panel that it ran on Illumina's HiSeq 2000.
Since then, it has taken orders for more than 1,500 tests and is seeing a reimbursement rate of between 80 and 90 percent, according to Hussam Al-Kateb, medical director at GPS. Additionally, the vast majority of samples are being run on the HiSeq 2500 for faster turnaround. The lab also has Illumina's MiSeq in-house.
Recently, GPS expanded its cancer panel to include 40 genes and is planning a third version for 2014. The 40-gene panel can detect SNVs and indels, and will soon also be validated for identifying some translocations. Version three of the cancer panel will encompass additional genes associated with hematological cancers, Catherine Cottrell, medical director at GPS, told Clinical Sequencing News.
While the laboratory currently does not do clinical RNA sequencing, its expanded panel is able to detect selected gene fusions
The cancer panel has a turnaround time of around three weeks and the laboratory bills insurance companies $7,500.
This month, GPS also launched CardioGene Set, a panel that assesses 69 genes associated with cardiomyopathies and arrhythmias, and is in the process of validating a panel for renal disease.
The CardioGene Set can be ordered as one comprehensive test, or physicians can order subsets of the test, choosing just the genes associated with Long QT Syndrome or Brugada Syndrome, for instance.
Last year the lab, along with the Rare Genomics Institute, offered a grant to rare disease advocacy groups for the sequencing of 99 exomes, and said at the time that it planned to launch a clinical exome sequencing service. However, Cottrell said that GPS is "in the process of evaluating whole exome sequencing to meet our clinicians' needs," but is currently not offering it as a diagnostic test.
Over the past two years, the GPS team has addressed a number of challenges related to offering clinical sequencing tests. For one thing, said Al-Kateb, they learned very early on that DNA quality of cancer samples was highly variable. "Some specimens are very old and the quality of DNA is compromised," he said. As such, the quantity of usable DNA in any given sample becomes an issue.
To overcome this challenge, Al-Kateb said that the lab has been able to reduce its input requirements from 750 ng of input DNA to 200 ng.
Additionally, added Cottrell, over the last two years, requirements for clinical laboratories offering next-gen based tests have become more straightforward. "When we first started developing our next-generation sequencing assays, there were not straightforward recommendations from the College of American Pathologists and other organizations," Cottrell said. "Now, many recommendations and guidelines have been issued from various organizations, which has assisted clinical labs in refining their clinical validation approach."
GPS recently published the clinical validation of its initial cancer panel in the Journal of Molecular Diagnostics.
Adding genes for the expanded panel required an additional validation of the test, Cottrell said. Because adding new genes required making changes to the target capture approach, the new test "required a fairly significant validation to ensure that the test maintained a high sensitivity and specificity and was capable of detecting the desired range of genetic alterations," she said. However, the "bioinformatics pipeline had been thoroughly vetted," she said, so validating that was less labor intensive.
The laboratory bases its decisions about what genes to add to its panel on evidence in the literature that a given gene is actionable, conversations with its physicians about what genes they would like to assess, and whether or not the gene is reimbursable.
"We're very selective in terms of what genes to add," Al-Kateb said.
The lab's reimbursement success is "not a coincidence," said Andrew Bredemeyer, director of business operations. By and large, genes that make the cut "have precedent for reimbursement as single-gene tests. That was certainly part of the calculus for the original gene set and was considered in the expansion."
Currently, around 40 percent of patients have actionable results, meaning the test could alter patient management. However, "even when a specific actionable change is not identified, that can also assist with making clinical decisions," Bredemeyer said, such as ruling out certain therapies or considering certain clinical trials.
GPS is purposefully highly selective in the number of genes it includes on its cancer panel, despite competitive offerings like Foundation Medicine's FoundationOne test that looks at more than 200 cancer-related genes.
Additionally, compared to some of the similarly sized cancer panels, such as Life Technologies Ion AmpliSeq cancer panel that has been validated as a clinical test by a number of academic laboratories and Laboratory Corporation of America's recently launched cancer panel on the MiSeq, the GPS panel "sequences the entire coding region of genes included in our test and select introns," Cottrell added. "It's not limited to hot spot regions and that's beneficial because then we can detect the full range of genetic alterations."