P-0175 - Simultaneous detection of KRAS, NRAS andBRAF hotspot mutations on Ion Torrent PGM platform

Date 28 June 2014
Event World GI 2014
Session Poster Session
Topics Biomarkers
Colon Cancer
Rectal Cancer
Personalised Medicine
Presenter Céline De Rop
Citation Annals of Oncology (2014) 25 (suppl_2): ii14-ii104. 10.1093/annonc/mdu165
Authors J. Dargent
  • IPG, Gosselies/BE



Colorectal cancer is the third most common cause of cancer and cancer death, and approximately 20% of all patients will present with metastatic disease. Treatment has become increasingly complex with the emergence of new chemotherapy drugs and targeted agents. To date, in metastatic colorectal cancer (mCRC), KRAS testing has been restricted to codons 12 and 13. However, phase II and III trials now suggest including both KRAS and NRAS codons 12, 13, 61, 117, and 146. Searching for these additional codons could help screen 20% more patients with mCRC for treatment with EGFR inhibitors. Other finding suggests that testing for the BRAF V600 mutation compliments KRAS and NRAS mutations analysis and may be as important for treatment decisions. Here we present a multiplexed and cost effective strategy enabling to test simultaneously these 11 hotspot mutations.


A custom panel (CP-KRAS), covering exons 2, 3 and 4 of KRAS and NRAS and exon 15 of BRAF, was designed using the Ion AmpliSeq Designer module. FFPE tissue sections were macro-dissected and DNA was extracted using the Maxwell FFPE Tissue LEV purification kit. Sequencing was performed according to the AmpliSeq protocol on the Ion Torrent PGM, starting from as less as 6 ng of DNA. Data were analysed with both Torrent Suite v4 and NextGENe software v2.3 with a hotspot positions-targeting bed file. Sensitivity of the technique was assigned from 2.5% at 500 x coverage to 10% at 100 x coverage, with a forward to reverse reads ratio >0.25.


After validation of the CP-KRAS panel on well-characterized EQC samples, more than 250 FFPE specimens (corresponding to a 8 months of CP-KRAS testing) were sequenced. 29% of the samples were mutated for KRAS codon 12 and 6% for codon 13. Of the wild-type samples tested for additional mutations, 3% of samples were mutated for KRAS codon Q61, 4% for KRAS codon A146, 4% for NRAS G12 and 4% for NRAS Q61. 9% additional samples were mutated for BRAF V600. Each additional mutation was confirmed by pyrosequencing, mini-sequencing (SNaPshot) or Sanger sequencing. In total, 24% tumours wild type for KRAS exon 2 harboured a mutation in another of the RAS pathway genes.


This next generation and high throughput sequencing workflow on a single-tube multiplexed amplification of those KRAS, NRAS and BRAF hotspot mutations could help to identify additional patients who possibly fail to respond to anti-EGFR treatment. Moreover, this strategy is cost-effective, less fastidious and requires minimal DNA quantity from FFPE tissue as compared to standardanalysis.