CHICAGO – A novel technique for ferreting out DNA mutations in circulating plasma appears to be more sensitive than is conventional pathology at detecting drug-resistant mutations in patients with gastrointestinal stromal tumors, said an investigator at the annual meeting of the American Society of Clinical Oncology.
Secondary KIT mutations associated with resistance to tyrosine kinase inhibitors (TKIs) such as imatinib (Gleevec) were detected in 47% of plasma samples using "BEAMing" (beads, emulsions, amplification, magnetics) technology, compared with only 12% of tissue biopsy specimens, said Dr. George D. Demetri, director of the center for sarcoma and bone oncology at the Dana-Farber Cancer Institute in Boston.
"We know that tumor cells are constantly turning over, apoptosing, dying, and releasing free DNA into the bloodstream. I want to emphasize that this is not looking at circulating tumor cells; this is looking at free circulating DNA in the plasma, and by looking at that circulating DNA, we may be able to get a more comprehensive assessment of all the mutations from across all tumor burden in any given patient," Dr. Demetri said.
The technique, which some investigators have dubbed "liquid biopsy," involves treating plasma with beads coated with DNA sequences that are complementary to target mutational sequences to create an emulsion polymerase chain reaction (PCR). The PCR amplifies the circulating DNA, which can then be identified with flow cytometry. The test can detect one circulating DNA molecule per 10,000 in plasma, according to the American Association for Cancer Research.
Dr. Demetri and his colleagues used the technology to retrospectively study mutations from patient samples in the GRID (GIST-Regorafenib in Progressive Disease) phase III study. In that trial, regorafenib (Stivarga), a multikinase inhibitor, significantly improved progression-free survival (PFS) compared with placebo (hazard ratio, 0.27; P less than .0001) in patients with metastatic GIST after failure of both standard and targeted therapies, including imatinib and sunitinib (Sutent).
The investigators looked for primary and secondary mutations in KIT in both plasma samples (available for 163 of 199 patients in GRID) and tumor tissue (from 102 patients), and found that BEAMing detected mutations in 58% of plasma samples, compared with 66% of tumor samples analyzed by the Sanger DNA sequencing method.
In addition, BEAMing detected mutations in PDGFRA in 1% of samples, and in KRAS in one of two samples, compared with 3% and 1%, respectively, for sequencing. Neither analysis method detected any BRAF mutations, and the numbers of PDGFRA and KRAS mutations were too small for researchers to draw meaningful conclusions.
When it came to KIT, however, there was a high degree of concordance between the tests where both types of samples from individual patients were available, including 100% agreement for primary KIT exon 9 mutations, 79% for primary KIT exon 11 mutations, and 84% overall for primary KIT exon 9 and 11 mutations.
The BEAMing technology can help determine prognosis of patients with GIST, Dr. Demetri said, noting that in the GRID trial, patients in the placebo arm who had secondary KIT mutations had shorter PFS than did patients without KIT mutations (HR, 1.82; P = .05). Patients with KIT exon 9 mutations had received a shorter course of imatinib than did the rest of the study population (HR, 2.02; P = .002) and a longer course of sunitinib (HR, 0.54; P = .005).
BEAMing analysis also showed that patients with secondary KIT mutations who received regorafenib had significantly better PFS than did patients in the placebo arm (HR, 0.22; P less than .001).
The data Dr. Demetri presented were "very exciting," said invited discussant Dr. Shreyaskumar R. Patel, professor of sarcoma oncology at the University of Texas MD Anderson Cancer Center in Houston.
"The important take-home messages are that there is some very good concordance between the tumor tissue analysis and the plasma analysis," and that regorafenib "appears to be totally agnostic to any of the mutational subsets and seems to equally benefit all patients," Dr. Patel said.
The study was supported in part by Bayer HealthCare. Dr. Demetri disclosed serving as a scientific consultant to the company. Dr. Patel reported having no disclosures relevant to the study.