Abstract 2596
Background
Circulating tumour DNA (ctDNA) provides a non-invasive approach for gene mutation detection. The aim of this study is to evaluate the concordance of RAS/BRAF mutational status in tumour tissue and plasma in metastatic colorectal cancer (mCRC) patients.
Methods
Plasma samples were collected prospectively from previously untreated mCRC patients (TASCO1 NCT02743221) and analysed centrally by droplet digital PCR (ddPCR) with sensitivity down to 0.2% for KRAS exon 2, 0.5% for NRAS exon 2 and BRAF. Tumour RAS/BRAF status was determined locally from primary or metastatic tumours according to routine practice.
Results
Out of the 153 patients included, 121 had tissue and plasma mutation status available for KRAS exon 2, 129 patients for NRAS exon 2 and 70 patients for BRAF V600E. In these subgroups, the prevalence of KRAS exon 2, NRAS exon 2 and BRAF mutations detected in plasma was 30.6%, 0.8% and 11.4%, respectively vs. 47.1%, 1.6% and 12.9%, respectively in tumours. For KRAS, the concordance was 81.8% with a Kappa coefficient of 0.63. KRAS mutation was detected in tumour tissue and not in plasma for 21 patients (17.4%), potentially explained by low tumour burden or low tumour DNA shedding. KRAS mutation was detected in plasma but not in tumour tissue for just one patient. For NRAS, a concordance of 99.2% (kappa = 0.66) was observed between plasma and tumour tissue. One discordant case (0.8%) was observed for which NRAS was detected only in tissue. This case also presented KRAS mutation both in plasma and tumour tissue excluding an explanation related to tumour DNA shedding. For BRAF, a concordance of 95.7% (kappa = 0.80) was observed between plasma and tumour tissue: BRAF mutation was detected only in plasma in one case (1.4%) and only in tumour tissue in 2 cases (2.9%). One of the two cases displaying BRAF V600E in tumour but not in plasma also had an NRAS mutation in plasma and unknown NRAS status in tumours.
Conclusions
This study showed that RAS/BRAF mutations can be detected in plasma samples from mCRC patients by ddPCR. However, in the context of the study, analysis of the ctDNA did not allow detection of RAS/BRAF mutations in all patients where these mutations were present in the tumour.
Clinical trial identification
TASCO1 NCT02743221.
Legal entity responsible for the study
Servier Oncology.
Funding
Servier.
Editorial Acknowledgement
Disclosure
E. Van Cutsem: Research funding: Amgen, Bayer, Boehringer, Celgene, Ipsen, Lilly, Merck, Merck KgA, Novartis, Roche, Sanofi, Servier. M.P. Saunders: Honoraria: Roche, Merck, Amgen, Servier, Eisai. G. Argiles: Advisor: Hoffman la Roche, BMS, Bayer. C. Borg: Advisory boards: Roche, Servier, Sanofi; Research grant: Roche. P. Pfeiffer: Research funding: Amgen, Celgene, Lilly, Merck KgA, Roche, Taiho, Servier. V. Cattan, G. Desachy, N. Amellal: Employee: Servier. All other authors have declared no conflicts of interest.