Minimal residual disease (MRD) detection by circulating tumor DNA (ctDNA) predicted patients with stage I to III non-small cell lung cancer (NSCLC) who would relapse after surgical resection of their disease, according to results from the TRACERx study presented at the American Association for Cancer Research (AACR) Virtual Annual Meeting I 2020.
In the adjuvant setting of NSCLC, conventional trials use a heterogenous population with low relapse rates, resulting in the requirement for large recruitment numbers and long study duration. In addition, some patients are unnecessarily escalated to standard-of-care therapy with risk of toxicity. Identifying patients by MRD who are thought to be at risk of relapse can address these trial limitations.
“We think that the field is now ready for MRD-driven adjuvant trials,” Christopher Abbosh, MD, of the UCL Cancer Institute in London, and lead author and presenter of the study, said.
In the TRACERx study, plasma samples from 88 patients prior to the surgical resection of their stage I to III NSCLC tumors were analyzed. Patient-specific cfDNA enrichment was conducted with primary tumor tissue using multiregion exome sequencing that detected a median of 195 clonal and subclonal variants. ctDNA detection was maximized using a novel MRD-caller that controlled and estimated background sequencing error.
During validation of the assay, the sensitivity for mutant DNA was 89% with 25 ng of DNA. The low amount of DNA used was important because “we know that in a clinical setting, DNA input to assays can be limited,” Dr Abbosh said. However, he added that “sensitivity does scale with DNA input, as expected.”
The specificity was 100% experimentally and 99.9% when modelled in silico.
Among the 88 patients in the study, 72% experienced a relapse of their primary tumor that was detected by ctDNA. This included 49% of patients with adenocarcinoma, 100% with squamous cell carcinomas, and 75% of other NSCLCs, including large cell carcinoma, large cell neuroendocrine carcinomas, pleomorphic carcinomas, and adenosquamous disease.
ctDNDA was detectable among 91% of patients who shed ctDNA before their resection and in 64% of patients who did not shed viral particles. The median time from detection of ctDNA to clinical relapse was 164 days (range, 6-1022 days) among shedders and 22 days (range, 0-634 days) in non-shedders. The median disease-free survival was 362 days and 640 days for shedders and non-shedders, respectively.
Tracking of subclonal dynamics and copy number changes allowed for the prediction of neoantigens at relapse. This approach can “provide opportunity to track tumor biology in the prerelapse and relapse setting,” Dr Abbosh said.
Dr Abbosh concluded that “MRD surveillance can detect relapse in advance of standard-of-care imaging,” and the achievable MRD lead times were influenced by preoperative ctDNA shedding.
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. Presented at: American Association for Cancer Research (AACR) Virtual Annual Meeting I; April 27-28, 2020. Abstract CT023.
This article originally appeared on Cancer Therapy Advisor