Abstract Background: After initial responses to tyrosine kinase inhibitors (TKIs), NSCLC patients (pts) harboring EGFR activating mutations inevitably progress, with the “gatekeeper” EGFR T790M resistance mutation accounting for approximately 60% of cases of acquired resistance (AR) to TKIs. EGFR activating and T790M resistance mutations can be found in plasma on both circulating free tumor DNA (ctDNA) and on RNA contained within exosomes. While ctDNA is thought to be primarily released by dying cells, exosome RNA is actively released by many living cells (Jahr et al. Cancer Res 2001; Thery et al. Nat Rev Immunol 2009). Some pts, particularly those with either early stage or intra-thoracic disease, do not seem to release mutations on ctDNA into circulation that is detectable by any current method. Here we present data demonstrating the detection of activating and AR EGFR mutations using a combined single-step exosomal RNA (exoRNA) and ctDNA approach to maximize sensitivity and demonstrate the ability to detect mutations using exosomal RNA on pts previously described as negative by ctDNA methods alone. Methods: Matched pretreatment tumor tissue and plasma were collected from 81 NSCLC pts enrolled in TIGER-X, a Ph1/2 study of rociletinib in previously treated mutant EGFR pts with advanced NSCLC. Among the 81 pts (all enrolled before Dec 2014), 56 cases were randomly chosen from the clinical patient population (including 35 cases ≥10 mutant copies/mL and 21 cases <10 mutant copies/mL) and an additional 25 cases were previously determined to have low T790M levels in plasma (<10 copies/mL) using ctDNA-based approaches. We applied a column-based method (EXO52) to co-isolate both exoRNA and ctDNA from up to 6 mL of patient plasma, where available, and analyzed the mutations with a custom procedure for next generation sequencing (EXO1000). The targeted EXO1000 sequencing panel covers EGFR mutations on exon 19, 20 and 21. A custom library preparation method and bioinformatics pipeline was used to identify rare mutations in a qualitative and quantitative manner. Results: For the 56 cases randomly chosen from the clinical patient population, 54 had valid tumor tissue results. The positive percent agreement (PPA) between plasma and tumor was 96% (52/54) for activating mutations and 86% (42/49) for T790M with tumor as the reference method. For most cases analyzed, the combined mutation signal from exoRNA and ctDNA was greater than the signal from ctDNA alone. Furthermore, we detect mutations in the circulation of some pts who were previously called negative by analysis of ctDNA alone, suggesting improved sensitivity from addition of exoRNA to the analysis. In the subset of pts with low or undetectable levels of ctDNA and valid tumor results (N = 45), we detect activating mutations in 38 of 45 cases (PPA 84%) compared to 27 of 45 (PPA 60%) by ctDNA alone, as well as T790M in 22 of 35 evaluable cases (PPA 63%) compared to 19 of 35 in ctDNA alone (PPA 54%). Conclusions: Our data demonstrate the ability to detect low copy numbers of activating and AR mutations in plasma of lung cancer pts by combining the mutation signal from exoRNA and ctDNA isolated by EXO52 and using the EXO1000 targeted NGS gene panel. By combining these two analytes, a higher sensitivity of mutation detection may be possible compared to analysis by ctDNA alone. Citation Format: Anne K. Krug, Chris Karlovich, Tina Koestler, Kay Brinkmann, Alexandra Spiel, Jennifer Emenegger, Mikkel Noerholm, Vince O'Neill, Lecia V. Sequist, Jean-Charles Soria, Jonathan W. Goldman, D. Ross Camidge, Heather A. Wakelee, Shirish M. Gadgeel, Elaina Mann, Shannon Matheny, Lindsey Rolfe, Mitch Raponi, Daniel Enderle, Johan Skog. Plasma EGFR mutation detection using a combined exosomal RNA and circulating tumor DNA approach in patients with acquired resistance to first-generation EGFR-TKIs. [abstract]. In: Proceedings of the AACR-NCI-EORTC International Conference: Molecular Targets and Cancer Therapeutics; 2015 Nov 5-9; Boston, MA. Philadelphia (PA): AACR; Mol Cancer Ther 2015;14(12 Suppl 2):Abstract nr B136.
Read full abstract
Aug 14, 2016
Edison T Liu + 9
Open Access