Plasma next generation sequencing and droplet digital PCR-based detection of epidermal growth factor receptor (EGFR) mutations in patients with advanced lung cancer treated with subsequent-line osimertinib

Abstract

Background: Gene mutation analysis from plasma circulating tumor DNA (ctDNA) can provide timely information regarding the mechanism of resistance that could translate to personalised treatment. We compared concordance rate of next generation sequencing (NGS) and droplet digital polymerase chain reaction (ddPCR) in the detection of the EGFR activating and T790M mutation from plasma ctDNA with diagnostic tissue biopsy-based assays. The second objective was to test whether putative osimertinib resistance associated mutations were detectable from plasma using NGS.

Methods: From January 2016 to December 2017, we prospectively collected plasma samples from patients prior to commencement of second- or third-line osimertinib therapy and upon disease progression, in a single tertiary hospital in South Western Sydney, Australia. Amplicon-based NGS and ddPCR assays were used to detect activating epidermal growth factor receptor (EGFR) and T790M mutations in 18 plasma samples from nine patients; all patients were required to have tissue biopsies with known EGFR status.

Results: High concordance of allelic fractions were seen in matched plasma NGS and ddPCR for activating EGFR mutations and T790M mutations (R² = 0.92, P < 0.0001). Using tissue biopsies as reference standard, sensitivity was 100% for NGS and 94% for ddPCR. Several possible osimertinib resistance associated mutations, including PIK3CA, BRAF and TP53 mutations, were detected by NGS in samples upon progression on osimertinib therapy.

Conclusion: ddPCR assays for EGFR mutations appear to be as sensitive and highly concordant as amplicon-based NGS. NGS has the ability to detect novel resistance mutations.

Keywords: Circulating tumor DNA; droplet digital PCR; next generation sequencing; plasma genotyping; resistance mechanisms.

Figure 1

Treatment duration and survival outcome of patients. Swimmer plot indicating treatment types and duration that patients received and survival outcome. All patients were started on osimertinib after disease progression with first‐generation EGFR TKI or chemotherapy. () Osimertinib, () Chemotherapy (carboplatin/gemcitabine), () Nivolumab, () BSC, () Disease progression, () Death and () Ongoing therapy.

Figure 2

(a) High concordance between Qubit reading and WT allele concentration (R² = 0.81). WT, wild type. (b) Concordance EGFR mutation detection ddPCR versus NGS. A total of 18 samples from nine patients were analysed with both ddPCR and NGS to assess EGFR mutations. There was high quantitative concordance (R² = 0.92) of the mutant alleic frequency (MAF) across 36 EGFR variants (activating: L858R, exon 19 deletion, L861Q, S768I and T790M resistance mutation) from 18 samples detected both with plasma NGS and plasma ddPCR. NGS, next generation sequencing; ddPCR, droplet digital polymerase chain reaction; MAF, mutant allelic frequency. () EGFR activating mutations, and () EGFR T790M.