Metastatic CRCs KRAS and BRAF Genotyping in Routine Diagnosis: Results and Pitfalls
S Houlle, A Lamy, F Blanchard, J Bossut, E Fiant, F Le Pessot, F Di Fiore, R Sesboue, JC Sabourin. Rouen University Hospital, Rouen, France; Faculty of Medicine, Rouen, France
Background: KRAS and BRAF are both downstream in the EGFR signaling cascade. KRAS and BRAF activating mutations have respectively been reported in 30-40% and 5-10% of metastatic colorectal cancers (mCRCs). The discovery that KRAS activating mutations were predictive of resistance to monoclonal antibody EGFR-targeted therapies has led to restraining cetuximab and panitunimab in the treatment of wild-type KRAS mCRCs. Recently it has also been shown that BRAF V600E mutation had the same predictive effect on these therapies. KRAS and BRAF mutation screening of mCRCs is now performed in routine diagnosis but reliable guidelines are still to be defined.
Design: KRAS mutations in codons 12 and 13 were assessed in 910 formalin-fixed paraffin embedded mCRCs using a primer extension approach with SNaPshot® technology. BRAF V600E mutation was searched for in a subset of these patients (238). Tailed SNaPshot® primers were designed on both DNA strands. Extended primers were then separated and visualized in an automated DNA sequencer.
Results: KRAS mutational status was successfully assessed in 910 cases (772 primary tumors and 138 metastases). KRAS mutations were present in 40.4% and BRAF mutation in 6.3% of all analyzed mCRCs : as expected, KRAS and BRAF mutations were mutually exclusive. In the 138 metastases screened, KRAS mutations were found in 52 cases (37.7%) and BRAF mutations in only one. Among the 37 paired primary tumors/metastases analyzed, 4 (10%) presented a discordance. Variations in KRAS status were also observed due to tumoral heterogeneity in 5 patients. No discordance was found with BRAF mutations. Finally, artefactual (non reproducible in multiple analyses) KRAS mutations were observed in 29 patients (41 analyses). Fixation and paraffin embedding leading to DNA alterations might explain these results.
Conclusions: Our data show that variations of KRAS status occurred in a small subset of mCRCs during neoplastic dissemination, suggesting the assessment of KRAS mutation in metastasis when accessible. The mutual exclusiveness of KRAS/BRAF mutations is also a guarantee of the quality of analysis. Furthermore, KRAS mutations must be confirmed by two independent analyses in order to differentiate DNA changes due to carcinogenesis or formalin fixation and paraffin embedding.
Monday, March 22, 2010 8:45 AM
Platform Session: Section E, Monday Morning