SAN FRANCISCO (GenomeWeb) – Despite being touted for several years now as a way to monitor leukemia and lymphoma patients for disease recurrence, the uptake of next-generation sequencing for minimal residual testing has been slow, according to researchers in the field.
However, with growing clinical evidence, the recent US Food and Drug Administration approval of Adaptive Biotechnologies' ClonoSeq assay, and improved sequencing technology, experts anticipate that will change in the coming years.
MRD is used to determine whether a patient is eligible for bone marrow transplantation or to assess patients after transplantation or throughout treatment to see whether they have disease recurrence or are responding to treatment. Flow cytometry has been the main way of measuring MRD for leukemia and lymphoma patients.
However, when sequencing costs began to drop, researchers hypothesized that NGS might be able to more sensitively detect MRD, leading to better prognostics and management. This hypothesis seemed to bear fruit in initial research by companies such as Adaptive, Sequenta (which was acquired by Adaptive), and Chronix, as well as by scientists from leukemia diagnostic laboratory Münchner Leukämielabor in Germany and Washington University in St. Louis.
Nonetheless, according to Jeffery Klco, a hematopathologist who studies the genomics of acute myeloid leukemia at St. Jude Children's Research Hospital, although "it holds great promise; to date, it's been slow to get up to speed. Not many places are doing clinical deep sequencing for MRD purposes."
Despite the proliferation of sequencing to "characterize every alteration in the tumor at the time of diagnosis or relapse, our methods for deep and sensitive sequencing for MRD purposes are less developed," he added.
Klco said he thinks NGS-based MRD testing hasn't seen greater uptake due to the challenges of setting up a new clinical assay capable of the type of deep and sensitive sequencing that would be needed, and because the gold-standard method of testing, flow cytometry, is so entrenched in the clinical paradigm. "There's a comfort level there," he said. In addition, flow cytometry tests and NGS-based MRD tests "measure two different things," so the results are not always directly comparable, making it harder for clinicians to adjust to the newer technology.
"It's feasible, but just a matter of translating that into a clinical lab and making sure that it's done in appropriate ways in order to base clinical decisions on it," Klco said.
According to Todd Druley, associate professor of pediatrics, genetics, and developmental biology at Washington University in St. Louis, another reason NGS has not been adopted rapidly for clinical MRD detection is because sequencing error rates have been too high, particularly for detecting low-frequency mutations.
"If you need to find something that's one mutation in a thousand molecules, standard sequencing isn't going to see that," he said. "It's only been in the last few years that we've been able to use methods to mitigate the noise and identify the true mutations that are at very low frequencies."
Klco agreed that sequencing error rates have been an issue. NGS error rates are around 1 percent, but MRD assays need to detect mutations at or below that level. Increasingly though, there are methods to get around the noise, such as better bioinformatics techniques and barcoding methods.
Druley's lab has been developing error correction methods that make use of unique molecular identifiers to get around the issue and detect low-frequency variants, a technique his group described in a study published in Leukemia in 2015. The group has since developed a targeted gene panel using the approach and is collaborating on several retrospective pediatric AML studies in collaboration with the Children's Oncology Group and ArcherDX. For the collaboration with ArcherDX, the Wash U team will test its in-house designed panel, as well as two of ArcherDX's hematological malignancy assays.
Druley noted that because he primarily focuses on pediatric disease, Adpative's ClonoSeq assay is not as relevant for his purposes.
Klco added that there are differences between adult and pediatric blood cancers so MRD assays cannot always be used interchangeably. Pediatric AML is more commonly driven by fusions, which cannot be readily monitored by a lot of targeted DNA sequencing panels, for instance. An RNA-based panel may be preferable or very deep sequencing on a well-designed DNA panel, Klco said.
Jeffrey Wolf, director of the myeloma program at the University of California, San Francisco, said that "the field is changing very rapidly" and that his team at UCSF has been using Adaptive's ClonoSeq assay both for clinical purposes and in research. For instance, in one clinical trial, he said the group is using it to make decisions about whether to stop treatment in patients who have been on maintenance therapy, but have had serial MRD measurements of zero.
Wolf said that the assay and other NGS methods for measuring MRD are "certainly very sensitive, but one reason it's not being taken up that fast is because physicians need to know that it's clinically relevant — that they can make a decision based on it." Even if a test can measure the amount of disease in a patient, that's not necessarily relevant if "there's not a therapy or if the physician doesn't know whether intervention at that moment would make a difference versus waiting until a more conventional test is abnormal." Physicians may be reluctant to start patients on what are often toxic therapies based on an NGS MRD test if the patient is feeling good. "There's limited data that responding to a change in MRD makes a difference in outcomes," Wolf said.
Charles Sang, Adaptive's executive vice president of diagnostics, said that the company has seen an uptick in interest since receiving FDA approval of ClonoSeq. He also noted that the test is different from some targeted gene panels in that it is specific to the patient, evaluating the B and T cells identified at diagnosis as being cancerous and tracking them over time. It currently has approval for the detection and monitoring of MRD in multiple myeloma and acute lymphocytic leukemia. And, Sang said that the firm has seen the most interest in using it to determine whether a patient is responding to a therapy. In the near future, Sang said Adaptive aims to expand the FDA's approval to include other blood-based cancers and is also working on a version that will use blood as the analyte as opposed to bone marrow.
"I think this is the future and will be taken up," Wolf said, "it's just a matter of time before we prove its value."
In addition, a number of researchers have begun to evaluate single-cell sequencing technologies, and startup Mission Bio has commercialized a single-cell platform and assay targeting AML that researchers at MD Anderson Cancer Center and elsewhere have been testing for its ability to look at clonal heterogeneity, but also MRD.
Druley said that while his lab has also used the Mission Bio technology, it was "not yet sensitive enough for clinical use" and was also not yet suitable for analyzing formalin-fixed paraffin-embedded samples.
"Single-cell sequencing is great if you want to understand the nature of the cells that are left, but it's not a great way to quantify MRD," Wolf added.
Klco agreed that single-cell sequencing is at the moment a useful research technology, but "the jury is still out as to whether it will be robust enough to be a clinical assay."