Cell-free DNA (cfDNA) testing of epidermal growth factor receptor mutations (EGFRmut) is being investigated as an adjunct for diagnosis and monitoring in non-small cell lung cancer (NSCLC) patients. The performance of various amplicon-based targeted next-generation sequencing (NGS) methods, both with and without error correction, is of high interest. Outcomes of error-corrected NGS in plasma EGFRmut testing have not been previously independently reported. We deployed an in-house amplification-refractory mutation system PCR (ARMS-PCR) assay in a prospective study, benchmarking its performance against two NGS platforms in a patient subset.
An ultrasensitive ARMS-PCR assay for hotspot EGFRmut was established, with detection limits between 0.02% and 0.1%. A total of 134 plasma samples were prospectively analysed from 68 patients with metastatic lung adenocarcinoma at diagnosis or progression, recruited between Jan 13-Apr 17 from 5 centres, with serial monitoring of plasma EGFRmut till radiologic progression in one centre. We further evaluated the performance of ARMS-qPCR assay, AmpliSeq Lung and Colon NGS assay and Oncomine Lung cfDNA NGS assay in 29 NSCLC and 20 healthy plasma controls.
Concordance rate between cfDNA and tumor was 83.8%, with sensitivity 80.0%, specificity 94.4%, positive predictive value 97.6%, and negative predictive value 63.0%. Dynamic monitoring of plasma EGFRmut levels demonstrated rising levels a median of 2.1 months [0.9-3.9] before radiological progression. This detection also held true for tissue EGFRmut positive patients negative for plasma EGFRmut at study entry. 20 of 49 patients at progression were plasma T790M-positive, and clinical benefit rates were 91.0% for osimertinib-treated patients. Evaluation of ARMS-PCR and NGS platforms yielded an average concordance rate, sensitivity and specificity was 85.9%, 63.4%, 92.3% (ARMS-qPCR), 87.2%, 47.8%, 100% (Ampliseq) and 84.1%, 83.1%, 87.4% (Oncomine).
ARMS-PCR provides a useful diagnostic and monitoring adjunct for NSCLC EGFRmut patients. Amplicon-based targeted next-generation sequencing approaches with error correction is a promising approach requiring additional validation.
Clinical trial identification
Legal entity responsible for the study
Institute of Bioengineering and Nanotechnology
Agency for Science, Technology and Research (A*STAR)
All authors have declared no conflicts of interest.