By Kara Nyberg, PhD
Posted: June 24, 2020
New findings from the TRACERx lung study underscore the value of using circulating tumor DNA (ctDNA) to predict NSCLC relapse postsurgery based on the detection of minimal residual disease (MRD). Moreover, the results appear sufficiently robust to begin to support use of ctDNA as a biomarker for MRD, and hence disease relapse, in the clinical setting.
“We now think that the technology is getting to the point at which we can start to leverage these biomarkers in clinical trials to direct or escalate adjuvant treatment,” remarked Chris Abbosh, MD, of the University College London, who presented the latest TRACERx findings.
The TRACERx research initiative aims to track the genetic evolution of lung cancer from diagnosis through to relapse in individual patients. “If we want to improve outcomes further in NSCLC, we need to focus on innovation in the early-stage space. … Driven by this need, we’ve been extensively investigating MRD in TRACERx,” Dr. Abbosh said.
The TRACERx team uses what they call “a tumor-informed personalized cell-free DNA enrichment approach” for MRD detection. In the current study, they first performed deep sequencing of primary tumors removed from 78 patients with stage I-III NSCLC. Variants for MRD tracking were prioritized to optimize MRD sensitivity at low mutant allele frequencies and enable phylogenetic tracking. Thereafter, anchored-multiplex PCR enrichment panels were individually generated for each patient targeting a median of 196 (range 72-482) clonal and subclonal variants of interest detected in the primary tumor.
The patient-specific ctDNA enrichment panels demonstrated impressive sensitivity and specificity, detecting variant fractions as low as 0.003% of the total DNA sample. Analysis of 271 postoperative ctDNA samples taken periodically from 37 individuals without disease relapse showed that 269 of the samples—more than 99%—were MRD negative. There was a single exception among the 26 patients with no evidence of clinical relapse, which was linked to an individual who had not yet undergone adjuvant radiotherapy, and that likely reflected delayed clearance of residual disease. The other exception was a single false-positive result that occurred among the group of 11 patients with second primary tumors in the lung or elsewhere, thus reflecting the specificity of the ctDNA enrichment panel toward the primary tumor.
Among the 53 patients with NSCLC relapse, the TRACERx team found that some individuals, particularly those with non-adenocarcinoma histology, shed tumor DNA in blood prior to surgery. This finding proved relevant. ctDNA could be detected at or before relapse in 91% (38/42) of preoperative shedders versus 64% (7/11) of nonshedders. Notably, the median lead time prior to relapse was 164 days in shedders versus 22 days in nonshedders.
“What these data suggest is that preoperative ctDNA detection status will be a proxy of the potential utility of ctDNA as an MRD biomarker in a clinical setting,” commented Dr. Abbash.
Importantly, tracking MRD using the ctDNA enrichment panel heralded relapse before disease recurrence showed up on standard imaging scans. For example, 26 patients had one or more surveillance scans showing no evidence of relapse or equivocal changes. All of these individuals had a preceding blood test positive for MRD, and 24 of 26 (92%) went on to develop NSCLC relapse.
“Does this assay simply enable us to detect relapse sooner and increase anxiety, or can we actually alter the trajectory and outcome in those who prove positive ahead of radiographic manifestations?” asked Corey Langer, MD, FACP, of the University of Pennsylvania Abramson Cancer Center, who critiqued the study findings. “One could posit a trial looking at MRD-positive, radiographically occult relapse or progression, comparing early immunotherapy or chemoimmunotherapy to standard observation,” he suggested. Indeed, there are clinical trials being planned in advanced NSCLC to put the TRACERx technology to the test to examine this very issue. ✦
“Does this assay simply enable us to detect relapse sooner and increase anxiety, or can we actually alter the trajectory and outcome in those who prove positive ahead of radiographic manifestations?” –Corey Langer, MD, FACP, ILCN Editor
1. Abbosh C, Frankell A, Garnett A, et al. Phylogenetic tracking and minimal residual disease detection using ctDNA in early-stage NSCLC: A lung TRACERx study. AACR Virtual Meeting; April 27-28, 2020. Abstract CT023.
cfDNA Analysis Suggests EML4-ALK Variant Does Not Influence Response to Lorlatinib