P65: Development of allele-specific PCR assays for detection of mutations in KRAS gene in colorectal cancer in Russian patients

KRAS is component of Ras/MAPK signaling cascade that regulates cell proliferation and cell survival. Somatic mutations in KRAS gene are often found in tumors and affect the sensitivity of tumors to target therapy. Mutations in codons 12 or 13 of KRAS gene in colorectal cancer (CRC) are associated with resistance to anti-EGFR antibodies Cetuximab and Panitumumab. The objectives of this work were to develop PCR tests for detection mutations in KRAS gene and analyze the frequency of mutations in KRAS gene in CRC in Russia. DNA sequencing by Sanger is the most common method for mutation analysis. However, the method has a sensitivity of 20% mutant allele, which is often not sufficient for the analysis of somatic mutations in tumors. One of the most sensitive mutation analysis methods is allele-specific real-time PCR. This method allows to detect 1% of mutated DNA in the sample. This sensitivity is sufficient for analysis of mutations in tumor samples containing 2–5% or more of tumor cells in normal tissue. In this study we developed and compared 3 new KRAS assays (1) real-time PCR with allele-specific primers; (2) real-time wild-type blocking PCR with LNA (locked nucleic acid) blocker; and (3) Sanger sequencing with LNA-blocker. First assay is a PCR test with seven reactions using allele-specific primers for detection and genotyping 7 mutations in 12 and 13 codons of KRAS gene. Second assay is real-time PCR with only a single pair of primers and LNA oligonucleotide blocker. LNA-blocker is an oligonucleotide which has a wild-type sequence of codons 12 and 13 of KRAS gene. LNA-blocker binds strongly to wild-type KRAS DNA and suppresses its amplification, but does not block amplification of mutant DNA. The real-time PCR with LNA -blocker can detect mutant DNA but does not genotype mutation. Such assay can be used as a simple and sensitive screening test for mutant KRAS cases if exact genotyping of mutation is not required. We also used LNA-blocker to increase sensitivity of Sanger sequencing. To evaluate sensitivity and specificity of new tests DNA standards were prepared with different ratios of normal and mutant alleles using normal human DNA without mutation and recombinant plasmids with mutations in KRAS (G12C, G12S, G12R, G12V, G12D, G12A, G13D). After optimization all three assays had sensitivity 5% of mutant alleles for the detection of mutations in KRAS gene, using 2.5–40 ng of human DNA. Performance of new assays for KRAS mutations was compared using 81 colorectal tumor samples. Before analysis relative content of tumor cells in the samples was evaluated by pathologist. If tumor content was less than 20% in the sample then regions with a maximum number of tumor cells were manually macrodissected before the DNA extraction. DNA was purified from formalin fixed paraffin embedded (FFPE) tissue using “FFPE-DNA Kit” (Biolink). All three assays had high sensitivity (95–100%) and specificity (100%) for detection of KRAS mutations in clinical tumor samples. Mutations of the KRAS gene were found in 37 of 81 cases (46%) of CRC including 12 cases of mutation G13D, 11 cases of G12D, 5 cases of mutation G12V, 4 cases G12C, 3 cases of mutation G12A, 2 cases G12S, 1 case G13R. A single case with mutation G13R was missed by allele-specific PCR but was detected by real-time PCR with LNA-blocker and confirmed by sequencing. New assays have high sensitivity and specificity and are suitable for detection of KRAS mutations in clinical FFPE tumor samples.