Comparison of the SuperARMS and ARMS for detecting EGFR mutations in liquid-based cytology specimens from NSCLC patients

BACKGROUND: Non-surgical cytological specimens are adequate not only for accurate histological subtyping but also for molecular profiling. A modified amplification refractory mutation system polymerase chain reaction (ARMS PCR), known as SuperARMS PCR, was improved by optimizing the primers designation, which provides a higher sensitivity and specificity approach for free plasma DNA detection. It is unclear whether SuperARMS PCR detects epidermal growth factor receptor (EGFR) mutations in cytology samples. The aim of this study was to compare the EGFR mutations detected by ARMS PCR and SuperARMS PCR in cytology samples derived from advanced non-small cell lung cancer (NSCLC) patients. METHODS: From March 2016 to March 2018, a total of 234 cytological samples were obtained from primary or metastatic lesions of NSCLC, including 144 fine-needle aspirations (FNAs), 36 endobroncheal ultrasonography (EBUS) FNAs, 36 transbronchial needle aspirations (TBNAs) and 18 pleural effusion (PLEs). EGFR mutations were simultaneously detected using an ADx-ARMS EGFR kit (Amoy Diagnostics CO., ltd., Xiamen, China) and an ADx-SuperARMS EGFR kit (Amoy Diagnostics CO., ltd., Xiamen, China). Digital droplet PCR (ddPCR) and next-generation sequencing (NGS) were further used to verify the EGFR mutant inconsistent samples. RESULTS: All of the 234 patients with advanced or recurrent NSCLC were diagnosed and assessed by two cytopathologists, and their EGFR mutation statuses were successfully detected by ARMS and SuperARMS. Importantly, the SuperARMS and ARMS methods showed a highly concordant result of 94.0% (220/234) (95%CI: 85.0, 95.0%). The positive rate of the SuperARMS was higher than the ARMS in the cytology samples for EGFR detection (46.2% vs. 40.2%). The specific EGFR mutation sites in 16 samples (6.8%) were not completely consistent between the SuperARMS and ARMS. A total of 14 patients showed EGFR mutations when detected by SuperARMS, but by ARMS there were EGFR wild-type. Two patients were detected as having one more EGFR mutation site by SuperARMS than by ARMS. ddPCR and NGS were used to further confirm the EGFR mutations in these inconsistent samples. Eight samples had the same mutation results as the SuperARMS, and 6 samples were not verified because the remaining DNA was insufficient. A total of 78 EGFR mutation patients received Tyrosine Kinase Inhibitor (TKI) treatment. The overall objective response rate (ORR) was 88.5% (69/78) for EGFR TKI treatment. CONCLUSION: SuperARMS showed a high sensitivity and specificity for EGFR detection and thus, is expected to become a routine test in the clinic to be used as a widely available, easy-to-operate and sensitive method for EGFR mutation detection in liquid-based cytology samples.