ABSTRACT
BACKGROUND: Anti-epidermal growth factor receptor (EGFR) monoclonal antibodies are restricted to KRAS wild-type (WT) metastatic colorectal cancers (mCRCs), usually identified by direct sequencing, that may yield false negative results because of genetic heterogeneity within the tumour. We evaluated the efficiency of high-resolution melting analysis (HRMA) in identifying KRAS-mutant (MUT) tumours. METHODS: We considered 50 mCRC patients scored as KRAS-WT by direct sequencing and treated with cetuximab-containing chemotherapy, and tested the correlations between HRMA findings and response rate (RR), progression-free (PFS) and overall survival (OS). RESULTS: Aberrant melting curves were detected in four (8%) cases; gene cloning confirmed these mutations. Response rate (RR) of HRMA KRAS-WT patients was 28.3%. There was no response in HRMA KRAS-MUT patients. Disease control rate (responsive plus stable disease) was 58.7% in HRMA KRAS-WT patients and 25% in HRMA KRAS-MUT patients. There was no correlation between HRMA KRAS status and RR (P=0.287) or disease control (P=0.219). Median PFS (4.8 vs 2.3 months; hazard ratio (HR)=0.29, P=0.02) and OS (11.0 vs 2.7 months; HR=0.11, P=0.03) were significantly longer for the HRMA KRAS-WT than for HRMA KRAS-MUT patients. CONCLUSIONS: High-resolution melting analysis identified 8% more KRAS-MUT patients not responding to cetuximab-containing regimens, suggesting that HRMA may be more effective than direct sequencing in selecting patients for anti-EGFR antibodies.