Lung cancer remains the leading cause of cancer morbidity and mortality in males, comprising 17% of the total new cancer cases and 23% of the total cancer deaths. On the other hand, lung cancer is the fourth most commonly diagnosed cancer and the second leading cause of cancer death. In developing countries, the mortality burden for lung cancer accounts for 11% of the total female cancer deaths, as high as the burden for cervical cancer [1
]. Surgery is the predominant approach for treatment of lung cancer, in combination with other approaches depending on the disease status. In recent years, targeted drug treatment has become a highlight for lung cancer, especially with the use of epidermal growth factor receptor-tyrosine kinase inhibitors (EGFR-TKIs).
Targeting EGFR is a promising strategy for the treatment of non-small cell lung cancer (NSCLC). Previous studies have demonstrated various EGFR
mutations in NSCLC, including adenocarcinomas and nonadenocarcinomas [2
]. Unlike drug resistant mutations (Exon 20 T790M point mutation (T790M) and exon 20 insertion mutation (20-ins)), NSCLC harboring EGFRTKI sensitive mutations such as exon 19 deletion mutation (19-del) and exon 21 L858R point mutation (L858R) respond to EGFR-TKIs [14
]. Other drug sensitive mutations consist of exon 21 L861Q point mutation (L861Q) and exon 18 G719X point mutations (G719X, including G719C, G719S, G719A). Large-scale studies have also demonstrated that TKIs could apparently improve the therapeutic outcome of patients with EGFR
-mutant NSCLC [12
]. Thus, screening for EGFR
mutations in NSCLC is significant in the decision-making on the treatment of NSCLC.
Surgically resected tumor specimens are the optimal DNA source for EGFR
mutation detection. Complete and sufficient DNA can be extracted from surgically resected fresh samples, while specimens obtained from transbronchial lung biopsy or percutaneus aspiration lung biopsy could not demonstrate the whole tumor genomics because of the existence of intratumor genetic heterogeneity [17
]. Surgical specimens can show relatively complete tumor genomics, thus avoiding or reducing false-negative results of EGFR
The amplification refractory mutation system (ARMS) is a method for point mutation in DNA based on allele specific polymerase chain reaction [18
]. It is quick, relatively easy and more sensitive than DNA direct sequencing, and EGFR
mutation detection kits according to the principle of ARMS can detect 29 common EGFR
]. As some mutations may be present in a minor population of tumor cells, and normal cells can be mixed in tumor tissue, this highly sensitive assay can be an appropriate method for EGFR
The frequency of EGFR
mutations in NSCLC is known to be associated with many factors, including race, gender, smoking status, and tumor histology. About 8%–15% of European patients with NSCLC harbor EGFR
mutations. In Asian patients, the frequency is up to 31% and the mutations are associated with Asian ethnicity, female gender, never-smoking or light-smoking, and adenocarcinoma [7
]. As most previous large-scale studies on EGFR
mutations primarily focused on adenocarcinoma, few studies have evaluated the frequency of EGFR
mutations in non-adenocarcinoma NSCLC, such as squamous-cell carcinoma, adenosquamous carcinoma, and large-cell carcinoma. In this study, we used ARMS to demonstrate the status of EGFR
mutations in Chinese patients with NSCLC after lung resection and clarify correlations between EGFR
mutations and clinical features.
The present study was intended to improve our understanding about the EGFR mutation status in NSCLC, especially in non-adenocarcinoma.
Our results showed that the overall frequency of TKIs sensitive EGFR
mutations was 33.7%, and the EGFR
mutation rate in squamous-cell carcinoma (14.5%), adenocarcinoma (52.9%) and adenosquamous carcinoma (39.5%) was higher than the data from previous Asian population-based studies [2
]. We also identified that one of the 11 large-cell carcinomas, 2 of the 18 sarcomatoid carcinomas and 2 of the 11 mucoepidermoid carcinomas harbored TKIs sensitive EGFR mutations. Meanwhile, we found that TKIs sensitive EGFR
mutations were associated with the female gender (p
< 0.001), non-smoking history (p
= 0.045), and adenocarcinoma subtype (p
There are several underlying causes for the higher frequency of TKIs sensitive EGFR mutations in our study. Fresh and sufficient tumor samples after lung resection could be one of them. Owing to tumor heterogeneity and limitations of needle biopsy, the small amount of tissue from needle biopsies may not be emblematic of the complete pictures of tumors, leading to false negative results; while adequate surgical samples can provide enough DNA for EGFR mutation detection. On the other hand, as formaldehyde may cause crosslinking and degradation of DNA, surgically resected fresh samples can offer less damaged DNA than the formalin-fixed ones, thus making the results of EGFR mutation detection more accurate. Another reason for the higher rate of TKIs sensitive EGFR mutations could be the use of the more sensitive method ARMS. EGFR mutation is a somatic mutation, the detection of which requires a highly specific and sensitive method, for EGFR mutant cells are mixed with wild type cells in the tumor sample. ARMS is based on allele specific polymerase chain reaction with a sensitivity at 1% (this means that at least 1% mutant DNAs can be detected within a “normal” DNA background via ARMS), which also makes the results more accurate.
In a subgroup analysis where the subtypes of EGFR
mutations were studied, we found that exon 19 deletion and exon 21 L858R point mutation were the two dominant subtypes of TKIs sensitive EGFR
mutations in Chinese patients with NSCLC, comprising 43.4% and 48.1% of all the TKIs sensitive EGFR
mutations respectively. It is interesting that the frequency was not consistent with previous Asian population-based studies [13
], which demonstrated that exon 19 in-frame deletion was more frequent. It could be explained by unselected tumor subtypes and stages in our study, ethnic variations between Chinese and other populations, or sampling errors.
Two major TKIs resistant EGFR
mutations (exon 20 insertion and T790M point mutation) were found in Chinese patients with NSCLC prior to any treatment. Exon 20 insertion mutations are associated with primary TKIs resistance. It can promote the activation of EGFR kinase domain, leading to carcinogenesis. It can also affect ATP and the affinity of EGFRs to gefitinib or erlotinib, causing the resistance against gefitinib or its sister drug. Patients harboring exon 20 insertion mutations should receive irreversible inhibitors rather than gefitinib or erlotinib. T790M point mutation in exon 20 is responsible for approximately 50% patients with acquired resistance against TKIs [24
]. This mutation does not reduce the affinity of gefitinib or erlotinib to the receptors but it enhances the affinity to ATP and thereby causes resistance. However, de novo
T790M mutations were found in Chinese patients with NSCLC before administration of TKIs in our study, indicating that T790M could also lead to primary resistance against TKIs; this might confirm that a low frequency T790M mutation may have been present in the primary cancer, but under the selective pressure imposed by targeted therapies it may expand and lead to TKIs resistance [25
]. As reported by Tetsuya Mitsudomi et al.
]; ASCO 2012, abstract 7521), compared with patients harboring T790M wild type, the T790M mutant counterparts could enjoy longer progress free survival (PFS) and overall survival (OS), no matter what type of treatments they received (TKIs or chemotherapy). Thus, the identification of these resistant EGFR
mutations is as important as the identification of TKIs sensitive ones for the treatment patterns of NSCLC.
In addition, our study failed to find a significant association between the subtypes of EGFR
mutations and gender, smoking status, and tumor histology, which was different from the study of Tanaka et al.
], who reported that there was a gender difference in EGFR
mutation subtypes. Tumor stages (unselected vs.
advanced stage), mutation detection methods (ARMS vs.
PNA-LNA PCR clamp), specimen types (all fresh cases vs.
fresh cases/archival tissue), racial differences (Chinese vs.
Japanese) or sampling errors are likely to be the underlying causes of the discordance between our study and Tanaka’s. More studies are required to elucidate such discord.
Even though targeting therapy is a brand new strategy for the treatment of non-small cell lung cancer, surgical resection is still regarded as the predominant method of controlling the tumor. Apart from the early stage cases, patients at late stages, provided the risks of the procedure are low, could also undergo surgeries for more sufficient tumor samples for pathologic and molecular diagnoses rather than needle biopsies and benefit the most from treatments. The explanations for this are mentioned above. In addition to that, drug resistance and heterogeneity of the EGFR mutations that results from tumor heterogeneity, contribute to the rationale for resections when the disease recurs or metastasizes, for significantly more adequate tissue could make the reevaluations of the pathologic and molecular status of the diseases more accurate.
Using ARMS to detect EGFR mutations in NSCLC, led to three limitations in our study. First, the primers in the ADx EGFR Mutations Detection Kit we used are designed for the 29 already known EGFR mutations, in some rare instances polymorphisms may be present that would not be recognized by this assay. Second, although ARMS is sensitive, routinely being able to detect at least 1% mutant in a normal DNA background, when the DNA concentration is below that level, the results would be false-negative; also, if the samples are contaminated, the results would be less accurate. Third, compared with DNA sequencing, ARMS is not readily available and less economic, although it is superior to sequencing in both sensitivity and robustness on a large and diverse set of clinical tumor samples.