NEW YORK(GenomeWeb) – The Dana-Farber Cancer Institute and Brigham and Women's Cancer Center have developed and started using a next-generation sequencing-based panel for the rapid diagnosis of certain blood-based cancers.
Launched in late August, the test called Rapid Heme Panel is currently used to detect genetic alterations associated with leukemia, myelodysplastic syndromes (MDS), and myeloproliferative disorders, cancers in which a rapid diagnosis and early treatment may be crucial for the health of the patient. While other test methods can take about two weeks to return results, the Rapid Heme Panel takes about five days for results, Dana-Farber and Brigham and Women's said.
The panel has been used on more than 100 patients so far, and as demand for the test ramps up, Dana-Farber and Brigham and Women's are eyeing possible expansion of their capacity — currently they can do about 40 Rapid Heme Panels per week — exploring additional genes to the panel, and building it out for additional blood-based cancers, Brigham and Women's Director of Hematopathology Jon Aster told GenomeWeb recently.
The test runs on Illumina's MiSeq platform and uses the company's TruSeq Custom Amplicon assay to cover 95 genes that Aster and his colleagues determined were high yield in the diagnosis, prognostication, and, in some cases, the stratification of patients for therapies.
The Rapid Heme Panel covers 730 exons and 175 kilobases of genomic sequence. Aster said that he and his colleagues routinely get about 1,000-fold coverage on average, and that they are "using it routinely now … in all patients who have or are suspected of having" leukemia, MDS, or myeloproliferative disorders.
Key genes on the panel include NPM1, CEBPA, and FLT3 for the prognosis of acute myeloid leukemia; and DNMT3A, TET2, and ASXL1 for diagnosing MDS or myeloproliferative neoplasm in patients with abnormal blood counts of uncertain cause.
The Rapid Heme Panel also contains IDH1 and IDH2 and the NOTCH1 genes. Patients for whom the test has detected mutations in those genes have been referred to clinical trials being run by Agios Pharmaceuticals, which is conducting a trial for a candidate therapy targeting IDH1/2, and Bristol-Myers Squibb, which has a trial targeting NOTCH1 mutations.
The test, Aster said, is highly targeted not only to specific genes, but also to regions of genes "where we know enough about pathogenic sequence variants that … we can say something relatively definitive about whether it's a driver, it's a SNP, etc."
About three years ago, Dana-Farber and Brigham and Women's launched a genotyping project to analyze tumor tissue for the discovery of cancer-related gene mutations. While there is some overlap in the genes that are on the Rapid Heme Panel and the genes being studied for the 2011 project called Profile, Aster noted that the Rapid Heme Panel is not derived from Profile, though it has benefitted from the project. Profile has resulted in a large database of sequence variants, and using that, as well as other public databases, including The Cancer Genome Atlas, has resulted in the identification of sequence variants that may be relevant for the Rapid Heme Panel. Conversely, it has also identified sequence variants that can be left off the panel.
"I think that arguably has been one of the biggest contributions of Profile," to the Rapid Heme Panel, he said. That project also provided lessons in how to analyze and report data, and "the reporting that we do on the Rapid Heme Panel, I would say, is sort of the next iteration of reporting.
"We've really tried to distill down the reporting information into the important bits that clinicians need to know with little thumbnail sketches that describe some of the data," Aster said, "and a little bit about the biology and what we know about the clinical significance of those particular sequence variants."
But, he noted, the Profile effort covers more than 300 genes and has a large discovery component, "so there are many sequence variants that are found with the test where frankly, it's unclear exactly what those variants mean," Aster said. The work with Profile also has focused largely on solid tumors, rather than blood-based cancers.
The Rapid Heme Panel, he said, is based mostly on genes that researchers at Brigham and Women's division of hematopathology, the hospital's Center for Advanced Molecular Diagnostics, and those working in hematology and oncology at both the hospital and Dana-Farber thought would be particularly informative about patients with acute leukemia, myelodysplasia, and myeloproliferative neoplasms.
The test was validated with control samples for the genes and variants on the test that Aster and his colleagues felt were the most actionable and most important in helping to guide clinicians in making treatment decisions. The steps included reproducibility steps and dilution sensitivity studies. "We were quite rigorous about setting thresholds," he said.
They choose the MiSeq platform as the basis for the Rapid Heme Panel, specifically because of the rapid turnaround time possible with the system. Bead capture-based methods have often been used for cancer-related DNA analysis, and the technology offers advantages for certain types of patient samples, such as paraffin-embedded tissues.
However, because the Rapid Heme Panel is for blood-based cancers and can be used on peripheral blood and/or aspirated bone marrow, the MiSeq is better suited for getting information to clinicians treating patients with aggressive cancers, for example, such as leukemia, and require rapid and early treatment, compared to solid cancers, such as breast or prostate cancer, which can play out over the course of years, Aster said.
Dana-Farber and Brigham and Women's declined to disclose the price of the Rapid Heme Panel, but Aster said that it is considerably lower than other gene-based tests that have been routinely used. He noted that three genes on the panel, CEBPA, NPM1, and FLT3, had been previously offered as single-gene tests at a cost of $1,000 per gene.
"So it used to cost us $3,000 to get information for three genes," and for less than that, the Rapid Heme Panel provides information on 95 genes, Aster said. Because the test has been available only since the summer, Aster and his colleagues have only just started submitting reimbursement claims for the panel.
The test is currently available only to patients of Dana-Farber and Brigham and Women's, though there is some talk about expanding its availability. Before such a decision is made, though, more analysis of the scalability of the operations around the Rapid Heme Panel, as well as the market size for it and strategies to capture the market, is needed, Aster said.
If the partners decide to broaden the commercialization of the Rapid Heme Panel, it would compete against a growing number of other organizations and firms offering blood-based cancer tests using NGS. They includeMassachusetts GeneralHospital, which offers a clinical, targeted fusion test; theUniversityofPennsylvania, which has a targeted NGS test; and Foundation Medicine, which launched FoundationOne Heme about a year ago.
Foundation Medicine, in particular, is finding traction in the market, and during the third quarter the company reported 1,036 FoundationOne Heme tests to ordering physicians, according to its Form 10-Q. In addition to providing a genomic profile for hematologic cancers, FoundationOne Heme, which covers 405 cancer-related genes, profiles many sarcomas and pediatric cancers, and Aster acknowledged that the test captures things not covered by the Rapid Heme Panel, which detects single nucleotide sequence variants and indels, while FoundationOne Heme can also detect fusion genes.
For such detection, Aster and his colleague must use other single-gene assays that rely on RT-PCR or FISH technologies. He said, however, that they are considering another NGS platform besides the MiSeq to add gene fusion detection capability.
"It's not that Foundation is identifying things that you can't test through other … modalities right now, but rather they have commercialized this into a next-gen sequencing-based test," Aster said.
As the Dana-Farber and Brigham and Women's team continues to improve the Rapid Heme Panel — it is on the second iteration of the test — it has tinkered with some of the sequences to capture regions that it originally wasn't capturing. The next version of the test will include additional genes currently not on the panel, while other genes that "don't seem to have very high yield" will be removed, Aster said.
The test could also be extended to include other hematologic cancers for which it doesn't currently test. A key consideration will be whether the rapid turnaround time enabled by the Rapid Heme Panel will be necessary for other cancers. In most cases, non-Hodgkin's lymphoma and myelomas, for example, "are not as acute and don't present as many diagnostic difficulties as some of the cancers that we're looking at with the Rapid Heme Panel," Aster said.
He and his colleagues are now deciding whether a bead-based capture method would be more appropriate than a MiSeq-based assay for those cancers.