NEW YORK (GenomeWeb) – Data from a study released late last month has added to a growing evidence base for the use of liquid biopsy technologies in the clinical assessment of patients with multiple myeloma — a hematologic cancer.
The report, published in Nature Communications by authors from Dana Farber and the Broad Institute, found that in a subset of patients who have enough tumor DNA available — either circulating freely in their blood or from extracted circulating tumor cells — there was high concordance in the clonal somatic mutations and copy number alterations detected using whole-exome sequencing of blood versus a bone marrow biopsy.
Although the use of liquid biopsy technologies for longitudinal disease monitoring remains controversial overall, growing data in specific niches like multiple myeloma has begun to hint at which oncologic subsets may offer more immediate paths to the clinic.
Myeloma specifically has a need for longitudinal genomic analyses to track tumor dissemination and clonal evolution in clinical trials, and clinicians have begun to embrace the concept of minimal residual disease monitoring as an important prognostic indicator.
Currently, such monitoring requires painful bone marrow biopsies, which makes a non-invasive, blood-based methodology immediately attractive.
In the case of longitudinal genomics, being able to track the evolution of cancer cell populations is especially important in myeloma, because although the disease emerges out of known precursor conditions, it's still not fully understood what the molecular mechanisms of this transformation are.
"Until now, we haven't had a good way to measure how multiple myeloma cell populations evolve from precursor stages to diagnosed disease, and then respond to treatments," Dana Farber oncologist Irene Ghobrial, the senior author of the new study, said in a statement.
"Blood biopsies can make a huge difference — extending our understanding of multiple myeloma, and really giving us a timeline of how the disease progresses and responds to therapy," she added.
The Dana Farber team is not the only group that has been pushing for blood-based testing as a way to make these types of analyses easier and more widespread. Another group from Toronto's Princess Margaret Cancer Centre published a study last year in, also in Nature Communications, evaluating whether circulating cell-free DNA results would be comparable to bone-marrow testing.
Using a hybrid-capture-based targeted sequencing method, investigators analyzed the protein-coding exons of KRAS, NRAS, BRAF, EGFR and PIK3CA in 64 cfDNA specimens from 48 patients with matched bone-marrow data. According to the authors, the blood testing showed 96 percent concordance with bone marrow, and detected four additional mutations that were not present in the marrow samples.
The Dana Farber team's new study last month also builds on previous research that they conducted and published last year, analyzing the genomic landscape of CTCs from 29 myeloma patients, eight of which had paired bone marrow samples.
According to the authors, 100 percent of the clonal mutations in patients' bone marrow were also detected in CTCs, and 99 percent of clonal mutations seen in CTCs were present in matched bone marrow samples.
In the recently published study, the group extended this to cell-free DNA as well as CTCs, applying a novel technique developed recently by a Broad Institute-led team, which quantifies circulating tumor DNA in a blood sample, providing a quick and easy read on which patients have sufficient DNA levels to allow accurate genome-wide sequencing.
Viktor Adalsteinsson, who heads the Broad's Blood Biopsy Team and was also a lead author on the recent myeloma study, said that he and his colleagues' ichorCNA "ultra low-pass" WGS approach provides a way to identify the appropriate candidates for a follow-on comprehensive analysis of tumor DNA, thus avoiding a waste of sequencing resources on patients with too low a fraction.
Overall, the team examined cfDNA from 107 patients and CTCs from 56 patients, using the ichorCNA analysis first to identify patients who met a cutoff point of 10 percent tumor content sufficient for whole-exome sequencing.
This restricted the cohort to 24 patients, 9 of whom had both cfDNA results and matched bone marrow data, and another four in which the researchers could compare all three sample types.
Overall, the gene profiles overlapped closely, authors reported — demonstrating about 99 percent agreement between liquid and bone marrow biopsies for somatic mutations, and 81 percent for copy number alterations.
According to the authors, there was a wide range across patients, as well as significant discrepancies on a single-patient basis, in the levels of tumor DNA in CTCs versus cell-free DNA — with some patients showing much higher tumor fractions in cfDNA than enriched CTCs or vice versa. One subject, for example, had an 80 percent tumor fraction in enriched CTCs, but only about 7 percent in cfDNA. Another had a 91 percent cfDNA tumor fraction, and only 4 percent in their CTCs.
According to Adalsteinsson, this illustrates the fact that addressing both of these compartments offers a way to increase the number of individuals who can be analyzed. Using a cutoff point of 10 percent or more tumor fraction, 35 percent of the cohort had enough DNA for subsequent whole-exome sequencing. But this would not have been the case if the researchers had been profiling either cfDNA or CTCs alone.
One caveat to the findings is that if a 10 percent tumor fraction is necessary for a comprehensive analysis, the data so far indicate that only about 20 or 30 percent of patients would be able to be analyzed using blood. That leaves a majority stuck with bone marrow testing.
Adalsteinsson said, though, that it is encouraging that most patients had enough tumor DNA either in circulation or in CTCs to meet the 3 percent limit of detection for the ichorCNA method applied in the study.
Looking at 70 cfDNA and 39 CTC samples from newly diagnosed or relapsed patients, 76 percent had at least 3 percent tumor fraction in cfDNA, and 100 percent had at least 3 percent in captured CTCs.
Including patients with MM precursor conditions, the team saw that the percentages with at least 3 percent tumor fraction dropped for both cfDNA and CTCs, further cementing the observation that tumor fraction increases with disease progression and that a method like ichorCNA could be used to monitor disease progression and response to therapy.
Among their next steps, the researchers said they hope to develop methods that provide more detailed and clinically appropriate genetic information, and to work on new techniques that can allow deep and comprehensive sequencing of blood samples with lower fractions of cancer DNA than the 10 percent threshold discussed in the study.
"We chose that as the cutoff today based on the cost of sequencing right now, and how much is feasible to sequence," Adalsteinsson said. "But as the cost of sequencing declines, it will become more and feasible to sequence whole exomes from patients with less tumor DNA."
"Whether that becomes a routine part of clinical practice we have to see… but in the meantime it is giving us an opportunity to learn about patients whose tumors aren't routinely biopsied in clinical practice," he added.
Minimal residual disease monitoring is also an area where blood-based testing could play a valuable role in multiple myeloma, especially since some in the field are already pushing for bone-marrow based MRD to be incorporated into clinical practice.
Adalsteinsson said that his team is also working on that, but that the low-pass sequencing method used in ichorCNA is probably not sensitive enough for that application.
The important takeaway for MRD from the study is the finding that both CTCs and cfDNA give a complementary picture, and that using them together broadens the patient population that can be successfully analyzed.
"As we look deeper and deeper in the case of MRD, it looks like we should probably be doing both CTCs and cfDNA — at least in multiple myeloma," he said.