High frequency of exon 20 S768I EGFR mutation detected in malignant pleural effusions: A poor prognosticator of NSCLC

Abstract Background Lung cancer is the cause of a fourth of all cancer‐related deaths. About a third of all lung adenocarcinoma tumours harbour mutations on exons 18 to 21 of the epidermal growth factor receptor ( EGFR ) gene. Detection of these mutations allows for targeted therapies in the form of EGFR Tyrosine kinase inhibitors. Recently, “liquid biopsies” have emerged as an alternative to conventional tissue mutation detection. Aim In this pilot study, we attempted to optimize EGFR mutation detection from malignant pleural effusions (MPEs) as “liquid biopsies” when tissue biopsies were unavailable. Resulting mutations were then to be mapped on the EGFR gene and explored using cBioPortal, a public cancer genomic database. Methods and results We first attempted a direct sequencing approach and showed that single nucleotide variants (SNVs) were likely to be missed in MPEs. We then switched to and optimized an EGFR mutant‐specific quantitative polymerase chain reaction‐based assay. This assay was piloted on n = 10 pleural effusion samples (one non‐malignant pleural effusion as a negative control). 5/9 (55.55%) samples harboured EGFR mutations with 2/9 (22.22%) being exon 19 deletions and 3/9 (33.33%) the S768I mutation. The frequency of the S768I SNV in our study was significantly higher than that observed in other studies (~0.2%). Utilizing cBioPortal data, we report that patients with S768I have a shorter median survival time (6 months vs 38 months), progression‐free survival time (8 months vs 44 months) and lower tumor mutation count compared to patients with other EGFR mutations. Conclusions The shorter survival of patients with the S768I SNV predicts aggressive disease and poor prognosis as a result of this mutation. Studies in larger cohorts and/or animal models are necessary to confirm these findings.