In June, the FDA approved the first blood-based genetic test for clinical decision making, the cobas EGFR Mutation Test v2. The assay is indicated for the detection of EGFR
exon 19 deletions or exon 21 (L858R) substitution mutations as a companion diagnostic for first-line erlotinib (Tarceva) therapy. The test is approved for use with plasma or tumor tissue, according to Roche, which developed the assay.2
Mack said ctDNA is most available in aggressive malignancies. “Fragments of shed DNA in the bloodstream are stable enough to collect and interrogate as long as you’re using a highly sensitive assay,” he said. “Here you have to be cautious because with highly sensitive assays comes an increased risk of false-positives.
“The best assays that are available, and there are several commercially available, are ones that have gone through rigorous validations to make sure that they are as close to 100% specificity as possible,” Mack added. “In other words, that they simply do not present false data.”
At the 2016 ASCO Annual Meeting, Mack and colleagues presented findings from a genomic analysis of ctDNA in blood samples from more than 15,000 patients representing multiple cancer types; approximately 37% of participants had lung cancer.3
Researchers said it was the largest-ever liquid biopsy study.
The Guardant360 next-generation sequencing (NGS) test, which analyzes 70 cancer-associated genes through digital polymerase chain reaction (PCR) sequencing, was used to detect somatic mutations. Most of the results were compared with data from The Cancer Genome Atlas (TCGA) and some were correlated with results of genetic testing on tumor tissue.
“The major finding of the study was that ctDNA mutation patterns were highly concordant with tissue analysis as reported by the TCGA and plasma analysis as reported in this study,” Mack said during an ASCO press briefing.
Overall, the positive predictive value for ctDNA sequencing ranged from 94% to 100% across 6 key biomarkers. For patients with NSCLC, the plasma testing detected the presence of truncal mutations in about 85% of cases, Mack said.
Notably, ctDNA testing for the EGFR
T790M resistance mutation did not correlate with the TCGA, probably because patients in the tissue-based population had not yet received the anti-EGFR inhibitor therapy that promotes the mutation, researchers indicated.
The study also demonstrated that NGS testing identified more biomarkers than did tissue testing. In a subset of 362 nonsquamous NSCLC cases, 63% had insufficient tissue for full genotyping. When ctDNA testing was added, an additional 51 cases with actionable biomarkers were detected for an increase in biomarker yield of 42%.
In another study, researchers from the Dana-Farber Cancer Institute tested ctDNA using droplet digital PCR for plasma samples from 180 patients with advanced NSCLC to determine whether the technique could be used for rapid genotyping.4
They found the technology had a positive predictive value of 100% for EGFR exon 19 deletions, L858R
mutations, and KRAS
mutations. The predictive value dipped to 79% for T790M
mutations. Overall sensitivity ranged from 64% to 82%, depending on the mutation. In addition, the turnaround time was 2 to 3 business days, as opposed to the 27 days typically needed for a patient obtaining a new tumor biopsy, the researchers said.
- NCCN clinical practice guidelines in oncology: non–small cell lung cancer version 4.2016. www.nccn.org. Updated February 29, 2016. Accessed August 3, 2016.
- FDA grants first liquid biopsy approval to the Roche cobas EGFR Mutation Test v2 [press release]. Pleasanton, CA: Roche Molecular Diagnostics; June 1, 2016. https://goo.gl/M7WAiE.
- Zill OA, Mortimer S, Banks KC, et al. Somatic genomic landscape of over 15,000 patients with advanced-stage cancer from clinical next-generation sequencing analysis of circulating tumor DNA. J Clin Oncol 34, 2016 (suppl; abstr LBA11501).
- Sacher AG, Paweletz C, Dahlberg SE, et al. Prospective validation of rapid plasma genotyping for the detection of EGFR and KRAS mutations in advanced lung cancer [published online April 7, 2016]. JAMA Oncol. doi:10.1001/jamaoncol.2016.0173.