NEW YORK – New research has added to growing evidence that liquid biopsy methods, specifically circulating tumor DNA analysis, could help oncologists prognose, monitor, and hopefully treat patients with glioblastoma brain tumors.
The results, published today in Clinical Cancer Research, showed that patients with a higher concentration of cell-free DNA in their blood had shorter progression-free survival than those with less cfDNA, and that spikes in circulating DNA seemed to correlate with or even predict disease progression.
The researchers from the University of Pennsylvania's Perelman School of Medicine noted that cfDNA testing is particularly attractive for brain tumors because of its potential to represent disease heterogeneity and because of the highly invasive nature of the brain surgeries necessary for tissue testing.
However, because of the blood-brain barrier, it has been viewed as a challenge to obtain information linked to a brain tumor from circulating DNA in the blood, and some studies have demonstrated disappointingly low yields of circulating tumor mutations in brain tumor patients compared to other solid tumors.
"Doctors have begun using liquid biopsies more frequently to monitor certain cancers — particularly lung cancer — in recent years as research has shown their effectiveness in other disease sites. But until now, there has been little focus on the clinical utility of liquid biopsy in brain tumors," Erica Carpenter, senior author of the study and director of the Liquid Biopsy Laboratory at Penn, said in a statement.
However, recent studies have suggested that in at least some patient groups, potentially actionable information like the presence of IDH1 and other mutations can be gleaned from blood frequently enough to suggest utility.
With their new study, the Penn investigators wrote that they hoped to fill some gaps by exploring patterns in cell-free and circulating tumor DNA from patients with comprehensive clinical outcome data, and in a specific clinical scenario: the pre- and post-surgery timepoint, when the amount of cell-free DNA coming from tumors and entering the blood would be expected to be highest.
The group collected samples under the umbrella of a larger institutional cfDNA banking protocol for central nervous system tumors at Penn, analyzing blood draws from a total of 42 patients — first at diagnosis, and then before surgery, and at regular intervals throughout standard of care chemotherapy and radiation.
The investigators hoped to explore several different questions: whether overall levels of cfDNA were associated with patient outcomes; whether they correlated with radiographic measures of tumor burden and could potentially serve as a surrogate or an aid to distinguishing true progress from pseudoprogression; and whether next-generation sequencing of this DNA could uncover informative mutations in patients' tumors.
According to the authors, the results did show that there was a prognostic association between overall cfDNA levels and outcome. Using PCR to quantify levels of cell-free DNA, the researchers found that the 28 patients with a lower concentration of cfDNA before surgery — defined as a level that was below the average of the total group — had almost double the progression-free survival (a median of 9.5 months) compared with the 14 patients with higher concentrations (median 4.9 months).
This association between high baseline plasma DNA and prognosis remained even after adjusting for other known factors, including age, IDH mutational status, MGMT promoter methylation, KPS, and extent of surgical resection. Although less than half of the subjects in the cohort had died at the time of the data cut-off for the study, the group also saw a trend toward inferior overall survival in subjects with high baseline cfDNA, though a non-statistically significant one.
The authors also reported that plasma cfDNA concentration was correlated with radiographic tumor burden on patients' first post-radiation magnetic resonance imaging scan, tending to rise prior to or concurrently with tumor progression becoming evident in imaging.
When the researchers sequenced the DNA samples from a subset of patients using Guardant Health's Guardant360 test, they found that 55 percent (11 individuals) had at least one somatic mutation present. Interestingly, none of the mutations detected in plasma were seen in matched tumor tissue, and vice versa, although two patients had different variants in the same gene detected by the two platforms.
The group speculated that the divergence may be explained in part by tumor heterogeneity, which has been noted to be especially robust in GBM "even on a single-cell level within a single focus of tumor."
As they stand, the results suggest that plasma DNA most likely can't serve as a surrogate for tumor tissue testing, but that that doesn't mean it couldn't have added value, the authors noted. Several of the plasma-detected mutations were in potentially therapeutically relevant genes, including TP53, EGFR, ERBB2, PDGFRA, and NF1, challenging "the current paradigm that detection of ctDNA in the plasma of patients with GBM is futile."
"The discovery of mutations by plasma NGS that would have gone undetected by tissue sequencing in over half of our patients suggests that combined use of tissue and plasma NGS may provide a more comprehensive assessment of tumor molecular heterogeneity in GBM, which is considered one of the most significant barriers to precision medicine in this disease," the group added.
The authors caution that the current work is more hypothesis-generating than practice-changing, considering the small size of the study. However, they reported that they continue to enroll patients and plan to perform a larger analysis in the future.
"If our findings are validated by further studies, it would mean that these patients may be able to get a simple blood test that would give us a more accurate assessment than imaging of whether their disease has progressed or not, as well as more data on the mutations in their tumors," lead author Stephen Bagley, an assistant professor of Hematology-Oncology in Penn's Perelman School of Medicine, said in a statement.
Other groups have turned to cerebral spinal fluid as a way to potentially capture more tumor-originating DNA. But the Penn authors argued that the relative invasiveness of CSF procurement may be a problem in bringing such approaches to the clinic.