A new assay dubbed Allele Specific Locked Nucleic Acid quantitative PCR, or ASLNAqPCR, can identify and quantitate KRAS and BRAF mutations from a variety of sample types with 100 percent specificity, and greater accuracy and higher analytical sensitivity than Sanger sequencing, according to a group of Italian researchers.
The team, from the University of Bologna and the University Hospital Santa Maria della Misericordia in Udine, published data from a 300-sample validation of the ASLNAqPCR method in PLoS One last month.
The group has also filed for two patents related to the method with the European Patent Office — one for KRAS testing and the other for detecting BRAF V600E mutations.
In their paper, the researchers describe the assay, which uses LNA-modified allele-specific primers and LNA-modified beacon probes specific for either codon 12 and 13 KRAS mutations or BRAF V600E to increase sensitivity and specificity of qPCR. According to the team, ASLNAqPCR can be performed in any laboratory with real-time PCR equipment, is "cost-effective," and the assay procedure can be completed in about three hours.
Both KRAS and BRAF mutations have become important markers for response to anti-EGFR kinase inhibitors, and also have the potential to offer diagnostic and prognostic prediction, according to the ASLNAqPCR group. With the growth of targeted therapies, the need for rapid and sensitive mutational testing is growing.
According to the study authors, the group's allele-specific LNA qPCR method overcomes several limitations of Sanger sequencing, including its low throughput, dependence on higher quality DNA, and low analytical sensitivity. The assay also allows measurement of the ratio of mutant and wild-type alleles, unlike other methods, the researchers wrote. While the team focused initially on codon 12 and 13 KRAS mutations and BRAF V600E, the test "can be easily adapted to detect hot spot mutations in other oncogenes," they said.
In its report, the ASLNA team wrote that the choice to use LNA was in part based on the potential of conventional allele-specific PCR using DNA primers to "miss-anneal the target sequence, particularly when PCR conditional are suboptimal … causing false positive results that may have unwanted consequences for [tyrosine kinase inhibitor] patient treatments."
To evaluate ASLNAqPCR against Sanger sequencing, the group analyzed three hundred tumor samples from 281 patients, including samples from primary tumors from the lung, colon, pancreas, and thyroid, as well as metastases at various sites. Most samples were from formalin-fixed paraffin embedded sections and 21 were fine needle aspirates.
The researchers designed primers and molecular beacon probes, modifying forward mutation-specific primers with LNA nucleotides at the 3'-end oligonucleotide terminal, and creating internal LNA-modified beacon probes for both KRAS and BRAF real-time analysis, they reported.
The group tested all 300 samples for KRAS mutations, and 201 of the samples for BRAF using both Sanger sequencing and the ASLNAqPCR method. For 21 samples with discordant results between Sanger and ASLNAqPCR, the team re-tested using pyrosequencing, and characterized each method by its number of true positive, false positive, true negative, and false negative results to establish sensitivity and specificity.
By serially diluting DNA from mutated cell lines, the researchers also tested the analytical sensitivity of both the allele-specific LNA assay and Sanger sequencing.
In the dilution experiments, Sanger sequencing required at least 20 percent mutated DNA to identify KRAS mutations and 10 percent mutated DNA to identify BRAF V600E. ASLNAqPCR, meanwhile, could identify both KRAS and BRAF mutations at a dilution of 0.1 percent with PCR efficiency of 111 percent, and 116 percent respectively.
Overall, the group found that the ASLNA method outmatched Sanger, with 95 percent sensitivity compared to the sequencing method's 81 percent sensitivity, and higher positive and negative predictive values.
Sanger and ASLNAqPCR generated conflicting results in 22 of the 300 samples for KRAS mutations, and in five of 201 samples for BRAF V600E, the researchers reported.
The group retested 18 of the discordant KRAS samples and three of the discordant BRAF samples using pyrosequencing, while six remaining discordant samples did not have enough material left for retesting.
The researchers found that two samples with a KRAS Q61H mutation and one with a KRAS G12F mutation were ASLNAqPCR false negatives. Meanwhile all the KRAS mutations identified by ASLNAqPCR, but missed by Sanger, were confirmed as false negative for the sequencing approach.
Importantly, the researchers found that for all these Sanger-negative samples, the mutated to wild-type ratio was below the 10 percent threshold they had measured for KRAS Sanger sequencing analytical sensitivity. "In the majority of samples the discrepant result was simply explained by the low tumor to non-neoplastic cell ratio and the higher analytical sensitivity of ASLNAqPCR," the group wrote.
In the two cases of KRAS Q61H false negatives by ASLNAqPCR, the authors explained that the failure could be explained by the fact that the mutations were not targets of the primers they chose. In the case of the G12F mutation, the ASLNA method misidentified a rare double nucleotide substitution, which the ASLNA primers were not designed to identify.
"A limitation of ASLNAqPCR, common to all hot spot mutation assays, is that it identifies — by definition — only the targeted mutation … Had our study been limited to codon 12 and 13 KRAS mutations, ASLNAqPCR would have been 100 percent sensitive," the authors wrote. The design of the method allows for easy addition of other RAS allele-specific primers as well as hot spot mutations in other genes, they argued.
According to the study, the team has since utilized the assay to indentify IDH1-R132H mutations with "high specificity and sensitivity" in more than 100 glioma samples.
Because of the assay's high analytical sensitivity relative to Sanger sequencing and other methods, the group highlighted ASLNAqPCR as having especially strong potential for sample analysis in cases where the ratio of tumor to non-neoplastic cells is below the recommended Sanger sequencing threshold, common in metastatic deposit samples or fine needle aspirates.
Meanwhile, the authors wrote, "when tumor cells are abundant, the use of ASLNAqPCR [could make] the dissection of tumor material unnecessary," removing a laborious step that "increases the potential for sample contamination."
Additionally, they wrote, the ASLNA assay allows for quantification of mutated alleles, via the LNA-modified beacon probe used for real-time analysis. This can allow the test to discriminate samples in which a mutation is widespread from those where it is present in only small neoplastic cell clones. According to the authors, this was the case in several of the groups' samples.
"Had the DNA from these patients been analyzed with a method that has high analytical sensitivity (e.g. pyrosequencing), but that does not allow precise quantification of the mutated allele, the tumors would have been diagnosed as mutated. This can, at least in the case of KRAS, deny a potentially beneficial treatment to CRC patients whose response to [tyrosine kinase inhibitors] may not be affected by the presence of small mutated clones," the researchers wrote.
"On the other hand," they added, "Sanger sequencing would have scored the case as negative and failed to predict a possible limited response to TKI … Although the impact of tumor heterogeneity in deciding patient management is a matter of debate, quantitative mutational data may help to clarify the issue, while providing the oncologist with accurate data to manage the patient."