Rapid Detection of Epidermal Growth Factor Receptor Mutations in Lung Cancer by the SMart-Amplification Process

Purpose: A positive response to gefitinib in non–small cell lung cancer (NSCLC) has been correlated to mutations in epidermal growth factor receptor (EGFR) gene. Previous reports have been based mainly on diagnostic screening by sequencing. However, sequencing is a time-consuming and complicated procedure, not suitable for routine clinical use.

Experimental Design: We have developed rapid, simple, and sensitive mutation detection assays based on the SMart Amplification Process (SMAP) and applied it for analyzing EGFR gene mutations in clinical samples. By using SMAP, we can detect mutations within 30 min including sample preparation. To validate the assay system for potential use in clinical diagnostics, we examined 45 NSCLC patients for EGFR mutations using sequencing and SMAP.

Results: The outcomes of the SMAP assay perfectly matched the sequencing results, except in one case where SMAP was able to identify a mutation that was not detected by sequencing. We also evaluated the sensitivity and specificity of SMAP in mutation detection for EGFR. In a serial dilution study, SMAP was able to find a mutation in a sample containing only 0.1% of the mutant allele in a mixture of wild-type genomic DNA. We also could show amplification of mutated DNA with only 30 copies per reaction.

Conclusions: The SMAP method offers higher sensitivity and specificity than alternative technologies, while eliminating the need for sequencing to identify mutations in the EGFR gene of NSCLC. It provides a robust and point-of-care accessible approach for a rapid identification of most patients likely to respond to gefitinib.

Non–small cell lung cancer (NSCLC) is the most common cause of death by cancer worldwide (1). As the global burden of NSCLC continues to increase, new agents are being developed for more effective treatment within a wide range of modalities, including surgery, radiotherapy, and chemotherapy, as the first and second lines of treatment. However, many patients still experience a relapse in cancer growth after cytotoxic chemotherapy.

The epidermal growth factor receptor (EGFR) tyrosine kinase inhibitor, gefitinib (Iressa, ZD1839), has emerged as an effective therapeutic agent for some patients with advanced NSCLC (2). Clinical trials revealed that there is a significant variability in the response to gefitinib. Recently, two groups reported an association between somatic mutations in the EGFR gene and dramatic positive clinical responses to gefitinib in patients with NSCLC (3, 4). Among 848 cases of EGFR mutations in database,⁷

739 occurred in three “hotspots” associated with the response to gefitinib. Of these, 53.2% (393 of 739) were multinucleotide in-frame deletions encoded by exon 19, 44.0% (325) cases were point mutations in exon 21 that resulted in a specific amino acid substitution at position 858 (L858R), and 2.8% (21) were point mutations in exon 18 (G719S). The remaining mutations were nonsynonymous and scattered throughout exons 18 to 21 or in-frame duplication/insertions in exon 20.

Information about the EGFR mutation status in clinical samples is becoming incorporated into clinical decisions and designs for future clinical trials. The previous standard for experimental detection of genomic mutations is the sequencing of DNA amplified from tumor tissues. However, as long as EGFR mutation testing is based on sequencing, it is unlikely to be a widely adopted approach. Diagnostic sequencing techniques are too complex, time-consuming, and expensive for routine pretherapeutic screening programs. Adding to the complexity is that clinical samples often contain a subpopulation of mutant cells mixed within an excess of normal tissue, sometimes causing the mutations to be missed by sequencing due to technical limitations of the technology. For conclusive determination of a mutant subpopulation, the sequence needs to be present at a minimum of ∼25% of the total genetic content. At levels below this threshold, the signal may not be present or confused with background noise and not recognized. With the rapid development of mutation-detection techniques, many methods have been reported for detection of EGFR mutations (5–18). These techniques are based on PCR-related (5–14), sequence-related (15), and other methologies (16–18). Some are more accurate and sensitive than sequencing, but all are still time consuming, complicated, and unsuitable for routine clinical examination. We therefore were motivated to develop a fast, easy, inexpensive, and accurate EGFR-mutation detection method.

It was previously shown that the SMart Amplification Process (SMAP) method can detect single nucleotide polymorphisms with a high sensitivity, specificity, and within 30 min under isothermal conditions (19). We adapted the SMAP method to target the three EGFR hotspot mutations and showed the detection of these sequence changes in tumor samples within 30 min. The main goal of this study was to develop a reliable method for rapid screening of EGFR mutations that had potential for point-of-care testing (POCT).

The principle of SMAP used for EGFR mutation detection. The SMAP method is a unique genotyping technology that can detect a mutation within 30 min under isothermal conditions and in a single step (19). The basic principle of SMAP is that “DNA amplification equals detection;” no further analysis is required. The critical milestone in establishing the SMAP method was to suppress exponential background amplification and ensure single nucleotide precision for amplification of mutant-specific sequences.

The basic primer design of SMAP is presented in Fig. 1. A set of five specially designed primers that recognize a total of six distinct sequences on the target DNA enable the precise amplification of only mutant-specific sequences. The five primers are the folding primer (FP), turn-back primer (TP), boost primer (BP), and two outer primers (OP1 and OP2). One key advantage of SMAP is the flexibility of primer design; mutation detection primers can be designed on various positions at the 3′ end of FP, TP, or BP and at the 5′ end of TP.

For suppression of background (nonspecific) amplification, SMAP employs two strategies; use of an asymmetrical primer design, which minimizes potential misamplification pathways, and inclusion of Taq MutS in the assay mix. Taq MutS is a mismatch binding protein that recognizes mismatched primer-template pairs and prevents the complex from being an active template for DNA synthesis (20, 21). Through use of a strand-displacing DNA polymerase, amplification occurs to generate amplicons of two short intermediate products <200 bp. These two intermediates then are capable of self-priming through three possible pathways to synthesize concatenated products containing multiples of the amplicons up to several hundred units in length. Exponential amplification occurs continuously until the reaction components are exhausted and copious amounts of DNA are generated. The DNA can then be detected or its generation monitored in real time by real-time PCR instrumentation. Because background suppression leads to absolute precision in SMAP amplification, the presence of amplified DNA in a tube is the signal of mutation detection. The sequence of the primer set used in that tube identifies which particular mutation was present.

Cell lines and plasmids. To assist in the development of the SMAP method, positive control DNA was extracted from the lung cancer cell line NCl-H1975, which contains the exon 21 point mutation L858R, and from the lung cancer cell line NCl-H1650, which contains the exon 19 deletion E746-A750 (both were purchased from the American Type Culture Collection). Both cell lines are heterozygous for the respective mutations, each containing one normal allele. DNA was extracted and diluted to a concentration representing 30 genomic copies/μL. Genomic DNA extractions from cell lines were done according to standard procedures. The dilution needed for specificity studies were done by mixing the cell line DNA with human genomic DNA, which is wild-type homozygous for both exons 19 and 21 alleles. The dilution was prepared at 10% to 0.1%, with a constant amount of wild-type genomic DNA equivalent to ∼6,000 unit copies. All experiments using wild-type human genomic DNA in this study are referring to control DNA purchased from Promega. By examination of the scientific literature on EGFR mutations, we identified and classified many in-frame deletions found on exon 19. For this study, we focused on the most frequent seven types (Supplementary Table S1). To assist in developing SMAP primer sets, each of these seven deletions was engineered by site-directed mutagenesis on a wild-type EGFR clone. These mutants were then cloned into the plasmid pGEM-T, for use as template DNA in SMAP assays. These templates were heated at 95°C for 5 min and added to the assays.

The SMAP method and restriction enzyme analysis for substitution mutation detection. In Fig. 2A is an illustration of SMAP-based amplification and mutation typing using a mutant primer set on wild-type and mutant alleles. For both exons 18 and 21, the FPs were engineered to be the mutation detection primers, with homology to the mutant nucleotide at the 3′ terminal position or the (n − 1) 3′ terminal position. When the template was a mutant target, there was full homology with the mutant FP and extension from this primer occurred unencumbered. Amplification proceeded exponentially, and the mutant allele was recognized within 30 min. When the mutant SMAP primer hybridizes to a wild-type allele, however, the Taq MutS protein in the assay mix recognizes the single-base mismatch, binds it tightly, and obstructs further amplification through that site. As a consequence, the exponential amplification of mismatched primers is inhibited. The locations and sequences of the primers for SMAP-based mutation detection of exons 21 and 18 are shown in Fig. 2B and C. Note that for L858R mutation detection on exon 21, two different versions of mutant FP primers can be used; one containing a T > G substitution on the 3′ end position, and the other a C > A substitution on the (n − 1) 3′ end position. Both kinds of mutations have been observed in NSCLC. For the detection of the G719S mutation on exon 18, the FP primer is also the mutation detection primer, with the C > A substitution engineered on the (n − 1) 3′ end position.

SMAP reactions were assembled on ice and incubated at 60°C for 30 min. We used Mx3000P system (Stratagene) for maintaining isothermal conditions and monitoring the change in fluorescence intensity of intercalating SYBR Green I (Molecular Probes) dye during the reaction. The SMAP method for detecting the L858R mutation was carried out in a total of 25 μL reaction mixture containing 1.6 μmol/L of each FP and TP, 0.8 μmol/L BP, 0.1 μmol/L each OP1 and OP2 (primers), 0.5 mmol/L deoxynucleotide triphosphates (dNTP), 0.6 mol/L betaine, 10 mmol/L (NH₄)₂SO₄, 2.5 mmol/L MgCl₂, 10 mmol/L KCl, 2 mmol/L MgSO₄, 1:100,000 diluted SYBR Green I, 20 mmol/L Tris-HCl (pH 8.8), 0.1% Triton X-100, 8 units Bst DNA polymerase (New England Biolabs), 1 μg Taq MutS (Nippon Gene), and 1 μL of prepared template. After amplification, the products of the SMAP reaction were cleaved by MscI and run on 3% of Nusieve GTG agarose gel for sizing and comparison. For amplification of the G719S mutation, the components of reaction mix were identical to that of the L858R assay, except for changes in the concentration of the following components: 0.2 μmol/L each of OP1 and OP2, 0.4 mol/L betaine, and 5 mmol/L MgCl₂.

The SMAP method for deletion detection in the EGFR gene on exon 19. We designed EGFR mutation detection primer sets that were specific for SMAP amplification of the seven highest frequency deletions observed in NSCLC. The location and sequences of primers for SMAP-based deletion typing of EGFR are shown in Fig. 3. In this case, we chose to use the BP as the mutation detection primer. The two outside primers (OP1, OP2), the FP, and the TP are all commonly used among the exon 19 deletion detection assays. Specificity for detecting each deletion mutant resides exclusively in the design of the BPs. Each BP was made to anneal perfectly to a different deletion mutant with a primer sequence that spans across the actual deletion site and has perfect homology on either side. The wild-type BP [BP(Wt)] sequence is shown in Fig. 3D, and the sequence of all Boost deletion primers [BP(DE-A) to BP(DE-G)] are also shown. The precise locations of the seven high-frequency deletions are shown as underlined nucleotides in Fig. 3C. They are also described in more detail in Supplementary Data S1.

The SMAP reactions for detecting the exon 19 deletion alleles were done with the same reaction mixture used for detecting the exon 19 wild-type allele, except that BP(DE-X) was used instead of BP(Wt). Reactions were done in 25 μL of a SMAP reaction mixture containing 10 mmol/L (NH₄)₂SO₄, 2.5 mmol/L MgCl₂, 10 mmol/L KCl, 2 mmol/L MgSO₄, 1:100,000 diluted SYBR Green I, 20 mmol/L Tris-HCl (pH 8.8), 0.1% Triton X-100, 8 units Bst DNA polymerase (New England Biolabs), 1 μg Taq MutS (Nippon Gene), 0.5 mol/L betaine, 0.45 mmol/L each dNTPs, 0.5 μg Taq MutS, 2.0 μmol/L each of FP and TP, 0.25 μmol/L each of OP1 and OP2, 0.8 μmol/L BP, and 1 μL of prepared template. The reactions were incubated at 60°C for 30 min and monitored by real-time PCR as described in the previous section for substitution mutation detection.

Patients and tissues. Tumor samples were obtained from 45 consecutive NSCLC patients surgically treated at the Kanagawa Cardiovascular and Respiratory Center between November 1995 and December 2005. Institutional approval and informed consent from all patients were obtained in writing. Of these 45 patients, 15 cases were subsequently treated with gefitinib. The entire group consisted of 32 males and 13 females with an age at diagnosis ranging from 32 to 83 years (median, 65.6 years). Among the 45 lung cancer patients, 28 (62.2%) were diagnosed as having adenocarcinomas, 10 (22.2%) as having squamous cell carcinomas, 3 (6.7%) as having adenosquamous carcinomas, and 4 (8.9%) as having other types of cancer. After surgical removal, all tumor samples were immediately frozen and stored at −80°C until assayed.

Sample preparation and EGFR gene analysis. The EGFR mutations were analyzed by DNA sequencing of the relevant regions of exons 18 to 21. Genomic DNA was extracted from tumor tissues using QIAmp DNA Mini Kit (Qiagen). For the PCR-RFLP analysis and sequencing, four sets of previously reported primers (5) were used to amplify the EGFR gene. PCR products were cycle sequenced using the Big Dye Terminator v1.1 cycle-sequencing kit (Applied Biosystems) according to the manufacturer's instruction. Sequence reactions were then subjected to electrophoresis on an ABI Prism 3100 (Applied Biosystems) instrument. For the SMAP assay, ∼5 mg of tumor sample were minced and mixed with 1 mL of 30 mmol/L NaOH, stirred, and incubated at 95°C for 5 min. After chilling on ice, the tumor lysis extract was diluted 1:10 in water, and 1 μL of the diluted solution was used as the template for SMAP assays carried out in the same way as described in the previous sections.

PCR-RFLP analysis of clinical samples. For confirmation of the point mutation in exon 21, a tumor sample was amplified by PCR using the same primers as used in the sequencing protocol. The wild-type EGFR gene on exon 21 encoded an MscI restriction site that was not present in the L858R mutant. PCR products were digested with MscI overnight at 37°C. The digested products were analyzed on a 3% of NuSieve GTG agarose gel, which was stained with ethidium bromide. Genotype assessments were based on the gel pattern of the restricted PCR fragments.

Identification of EGFR exon 18 and exon 21 point mutations by SMAP. Before testing clinical samples with our mutant primer sets and SMAP assay, controlled tests on known genomic templates were done. The SMAP reaction was done in the presence of an intercalating dye (SYBR Green I) and was monitored with the Mx3000P System. Genomic DNA isolated from the NC1-H1975 cell line known to be carrying the L858R substitution mutation (exon 21) was used as a template to test the reliability of the SMAP mutation detection assay. SMAP amplification primer sets with a FP specific for detecting the L858R point mutation (2573T > G) as described in Fig. 2 were shown to rapidly amplify (in 20 min) the NC1-H1975 cell line DNA, whereas the same primer set was incapable of amplifying wild-type genomic control DNA even after 60 min (Fig. 4A). A “no-template” control reaction was also negative after 60 min. The graphs in Fig. 4A are composite graphs, each line representing a different SMAP assay using a wild-type primer set (left) and a mutant primer set (right) and a different genomic template. Each assay was done in duplicate. Note that NCl-H1975 shows amplification with both primer sets, confirming that the cell line is heterozygous for the L858R mutation.

Electrophoretic analysis of the amplified DNA showed ladder patterns expected and typical of SMAP amplification (Fig. 4B). The DNA bands of ∼50 bp in size (indicated by asterisk) are likely to correspond to the monomeric self-primed amplification product bounded by the 5′ ends of the FP and TP primers. Larger sized bands are alternative forms (derived from intermediate forms, IM1 and IM2), self-hybridized species, and multiple-unit length amplicons of each species. To confirm the accuracy of genotype-specific products by SMAP, restriction analysis followed by electrophoresis was done. The MscI enzyme recognition sequence exists in the wild-type sequence of EGFR exon 21, but not the L858R mutant. As expected, the amplification products derived from the H1975 cell line were digested by MscI. In similar experiments on genomic DNA templates, we could also detect the G719S mutation by SMAP within 30 min (data not shown). The G719S mutation assay also showed similar high specificity, allowing reliable discrimination of mutants from wild-type specimens.

Identification of EGFR exon 19 deletions by SMAP. Again, before testing clinical samples with our mutant primer sets and SMAP assay, controlled tests on known templates were done. For deletions detection, engineered plasmid templates were used because no cell lines were available that have these specific EGFR mutations. The SMAP reaction was done in the presence of an intercalating dye (SYBR Green I) and was monitored with the Mx3000P real-time PCR instrument. Seven different allele-specific primer sets for deletion detection were developed and examined for specificity in the SMAP assay. The primer sets described previously and illustrated in Fig. 3 were each used in SMAP assays with exact match plasmid templates and against all other deletion mutants. In all cases, only the template completely identical with each allele-specific primer set was amplified within 30 min. No amplification was observed from any mismatched combination of primers and templates even when monitored for 60 min (Fig. 4C, Supplementary Fig. S2). Each experiment was done in duplicate; hence, each graph shows two positive amplification profiles for each of the deletion primer sets.

Sensitivity of SMAP-based mutation detection in mixed-cell populations. Tumor samples frequently consist of numerous subpopulations of cancer cells. A useful diagnostic for mutation detection must be able to detect mutations in heterogenous genomic DNA samples. To test SMAP for this capability, we conducted serial dilution experiments to examine detection sensitivity and genomic DNA mixing experiments to examine the sensitivity of detecting mutants as a subpopulation in a background of wild-type DNA. In the serial dilution experiments, using the NC1-H1975 and NSC-H1650 cell line genomic DNAs, the allele-specific primers for amplification of the exon 21 L858R point mutation, and the exon 19 E746-A750 deletion could each detect 30 copies in SMAP amplification (25 μL reaction) in 30 min when using the respective full-match cell line DNA (Fig. 4D).

To determine the minimal detection limits for mutant sequences in a background of wild-type DNA, mutant cell line DNA was mixed with increasing amounts of wild-type DNA, and SMAP assays were done with full-match mutant primer sets. Using the NC1-H1975 cell line genomic DNA and the allele-specific primers for amplification of the exon 21 L858R point mutation, the mutant sequences could be detected even when present at only 0.5% in 60 min. Likewise, using the NC1-H1650 cell line genomic DNA and the allele-specific primers for amplification of the exon 19 E746-A750 deletion, the mutant sequences could be detected even when present at only 0.1% in 60 min (Fig. 4E).

Genotyping of clinical samples. We purposefully set out to design a mutation detection system that was both accurate and fast. We therefore minimized the genomic DNA sample preparation to a simple lysis in NaOH as described in Materials and Methods and usually within several minutes had material ready for the SMAP assays. With this crude extraction procedure and SMAP analysis, we were able to diagnose specific mutations from clinical samples in about 30 min.

The genomic status of the EGFR gene was evaluated in a series of 45 primary NSCLC specimens by both SMAP and conventional sequencing. Nine mutations were found by sequencing, of which four cases were substitution in exon 21, and five cases were a deletion in exon 19. All of these mutations were believed to be heterozygous, having also one wild-type allele. The nine cases that were proven to have mutations by sequencing were also verified independently by SMAP (Fig. 5, Table 1). The four clinical samples that had exon 21 substitutions all displayed amplification profiles similar to case 16 (Fig. 5A). Both wild-type and mutant amplification curves were evident in all four cases, indicating that the samples were most likely heterozygous for both wild-type and mutant alleles. Based on these data, we cannot dismiss the possibility that a homozygous mutant subpopulation existed in a nearly equivalent ratio to a normal (homozygous wild-type) subpopulation and accounted for the seemingly heterozygous results. However, this scenario is not likely due to the equivalent ratio of mutant and wild-type alleles in all four independent cases, which is a statistically improbable event. In one case, case 3, the mutation was not detected by direct sequencing but was identified as the L858R mutation by SMAP. This mutation was not detected by sequencing because the abundance of the mutant-containing cells within the tumor sample was very low. Although the amplification curve for detecting this mutant displays exponential kinetics and is easily detectable by SMAP, its long delay in appearance (relative to the wild-type kinetic profile) also suggests that the population of mutant-containing cells is very low in the sample. Visual inspection of the sequencing data shows hardly identifiable mutation peaks of CT for AG at the nucleotide position 2572 to 2573 (Fig. 5B). In the initial calls, these low sequencing peaks were dismissed for background and not considered indicative of a mutation. To prove the validity of the SMAP assay results that indicate a mutation in case 3, we did PCR-RFLP analysis as a confirmatory test. In PCR-RFLP analysis, homozygous wild-type results would result in only two bands when examined by MscI digestion and agarose gel electrophoresis. Whereas samples that contain mutant-type DNA will show amplicon products that are resistant to cleavage by MscI, and hence, three bands would be evident on the gel under conditions favoring complete digestion. The results of case 3 shows three bands, indicating the existence of a mutation and confirming the previous results of the SMAP assay. In total, the PCR-RFLP results found the L858R point mutation in five samples, corresponding completely with the findings of SMAP.

The SMAP amplification profiles for the five cases having deletions in exon 19 are reported (Fig. 5D). Only one mutant primer set and the wild-type primer set amplified for each specimen. No amplification was observed from mismatched combination of primers and templates within 60 min. This result is consistent with the interpretation that the samples were heterozygous for mutant and wild type. Again, we acknowledge the possibility of two homozygous subpopulations, but that likelihood is remote for the reasons described previously.

Among the 10 cases where we could detect mutations by SMAP, four cases were treated with gefitinib. Two cases including case 3 showed a complete response to gefitinib. Case 22 showed a progressive disease in spite of the gefitinib. We were not able to judge the drug response to gefitinib in case 28 because of the short dosage period.

The SMAP method has several advantages for EGFR mutation detection over genotyping technologies besides conventional sequencing. More newly developed methods including PCR-related technologies (PNA-LNA PCR clamp, TaqMan PCR assay, SURVEYOR analysis, mutant-enriched PCR, PCR-SSCP, ARMS, TaqMan-MGB probes, and others), other sequencing platforms (microreactor-based pyrosequencing), and other methods such as high-resolution melting amplicon analysis and nanoscale engineered biomagnetite, all have serious limitations on comparison to SMAP. These new methods are more accurate, sensitive, and convenient than sequencing, but they are still time consuming (shortest time, 3.5 h), require careful DNA extraction, and involve several steps (DNA extraction, PCR, electrophoresis, etc.). They are therefore unsuitable for routine pretherapeutic examination and difficult to implement at most medical institution. The SMAP method is based on strand-displacing DNA polymerase activity and can amplify and detect mutations directly from blood samples requiring only a simple heat lysis and denaturation step. We have also shown that the simple sample preparation is effective on lysed tissue samples from human lung cancer. It is not known why some strand-displacing polymerases can effectively amplify DNA from such crude samples, but this feature is one crucial merit of the SMAP method.

The application of SMAP to molecular diagnostics has significant potential as evident from this study. Sample preparation was extremely minimal, setup was very simple, detection time was <30 min, and the specificity displayed single-nucleotide precision with no miscalls. The unique primer design, background suppression technology, and isothermal nature of the assay provide a powerful combination of attributes for mutation detection that stands alone, making SMAP perhaps the first practical tool for bedside usage or POCT in molecular diagnostics.

In this study, we employed the SMAP method for EGFR mutation detection to explore its utility as a tool to rapidly diagnose NSCLC tumors that are potentially gefitinib sensitive. We have shown that the SMAP method is capable of detecting as little as 30 copies per reaction of the EGFR mutation. This high sensitivity of SMAP is an advantage, particularly when analyzing small biopsies, or in cases where cytodiagnosis of the sputum and bronchoalveolar lavage are used such as for inoperable patients from which it is not possible to obtain enough clinical material for analysis using existing mutation detection methods. In addition, SMAP allows the detection of as little as 0.1% to 0.5% of targeted mutant alleles in DNA samples without any background amplification, which is similar or better than other newly developed methods. In our study, a gefitinib-sensitive case with a two bases-substitution mutation could be detected by SMAP, where sequencing was unable to identify it at all. This observation suggests that some clinical cases may be inappropriately treated if the diagnosis was dependent exclusively on DNA sequencing for EGFR mutation detection.

One limitation of SMAP technology is that it requires pre-engineered primer sets and, therefore, is effective only on already known mutations. Among cases which have mutations in the EGFR gene, 387 (45.6%) mutations consisted of small in-frame deletion between codons E746 and P753, and 325 (38.3%) were at codon 858 (L858R), and 21 (2.5%) were at G719S. Taken together, these three alterations account for 86.4% of the total number of mutations thus far detected in the EGFR gene. We decided to evaluate these three hotspots because they were mainly related to gefitinib sensitivity. It is possible that similar, but nonidentical mutations, might exist in these same codons and also render the cancer sensitive to gefitinib. These new mutations would be missed by the SMAP method because of its precision and, thus, present a risk of misdiagnosis. Sequencing, which has sensitivity limits and its own risks (already discussed), may be able to detect these new mutations. However, at the pace of current research in this area, it is likely that most occurring EGFR mutations will be identified and mapped for sensitivity to gefitinib therapy. SMAP assays can then be created for detection of these new mutations, as well.

The SMAP method can be applied to known point mutations, in-frame deletions, insertions, and duplications at fixed positions. To facilitate the design of primer sets for SMAP applications, a software specific for SMAP primer design can be freely accessed online.⁸

The software generates primer sequences and suggested arrangements suitable for any target sequence of interest, and optimization and assay design recommendations are supported as well. Cancer-related gene polymorphisms are said to participate in every cancer phase such as carcinogenesis, cell proliferation, metastasis, and drug sensitivity. Ongoing global research have already found several mutations, such as those in p-53 (22), k-ras (23), c-kit (24, 25), BRCA (26), etc. As more and more clinically important cancer-related mutations and genes are described, a quick and simple diagnostic method will become indispensable for prescribing effective cancer therapies. Rapid and sensitive quantification methods for detecting mutated DNA has great potential for providing new intraoperating (real-time) diagnosis of micrometastases in lymph nodes, peritoneal disseminations, resection margins, etc. The SMAP genotyping method for cancer mutation detection (SMAP), pioneered here first for EGFR mutation in NSCLC, is likely to become widely employed in the near future as the age of tailor-made medicine emerges in the operating room and clinic.