A study of incidence of EGFR mutations in non-smoker adenocarcinoma of the lung: disparity between north and south indian patients

Introduction

Lung cancer is the leading cause of cancer deaths worldwide for both men and women [1]. It is the number three cancer among Indian men [2]. Chemotherapy for non–small cell lung carcinoma (NSCLC), which accounts for approximately 85% of lung cancer cases, remains marginally effective [3]. Genomic research has unveiled tumour cell growth pathways that can be targeted with newer anticancer therapies having better efficacy & safety. One of such signalling growth pathway is human epidermal growth factor receptors (HER & EGFR members). EGFR is a transmembrane tyrosine kinase receptor that plays a crucial role in regulating cell division and death. EGFR is the cell surface receptor of extracellular protein ligands [4]. Genetic mutations in the EGFR gene may result in cancer [5]. Mutations in the tyrosine kinase domain of the EGFR gene increase the activity of the intracellular signalling cascades. This results in the increased proliferation activity of the tumour cells, finally leading to an inhibition in apoptosis and then metastasis [6-11,12].

Anti-EGFR therapies have been developed to intervene in the EGFR growth cascade [13-15]. EGFR belongs to the HER family of receptors which comprise EGFR(HER1/ErbB1), ERBB2(HER2), ERBB3(HER3) and ERBB4(HER4)). The HER receptors are known to be activated by binding to different ligands, including EGF, TGFA, heparin-binding EGF-like growth factor, amphiregulin, betacellulin, and epiregulin. After ligand binding to the extracellular domain of the receptor, the receptor forms functionally active dimmers. Dimerization induces the activation of the tyrosine kinase domain, which leads to autophosphorylation of the receptor on multiple tyrosine residues. This leads to activation of a series of intracellular signaling cascades to affect gene transcription, which in turn results in cancer cell proliferation, reduced apoptosis, invasion and metastasis.

The gene is located on the chromosome 7p12 & has many conserved domains. It has about 28 exons consisting of 1211 amino acids. Exons 18 through 21 corresponds to the tyrosine kinase domain which is very critical for the response to anti-EGFR therapies. It has been seen that presence of activating or driver mutations in this domain of EGFR gene has been associated with response to the anti-EGFR therapies [16-20]. Mutation incidence in the EGFR gene has been shown to be different in different ethnicities [21-23]. Lung adenocarcinoma patients with Asian descent have higher incidence of EGFR mutations than that of European descent [23]. Presence of mutation was also correlated with smoking status & gender [24,25]. It has been seen that non-smoker NSCLC patients have much higher incidence of mutations in EGFR gene than smokers [24,25]. It has been observed that the mutations are profound in non-smoking individuals (about 6.5 times higher) than smokers, 4.4 times higher in adenocarcinoma than other sub class of NSCLC, 1.7 times higher in females than males and 27–60% East Asians (compared to Europeans 8–13.3%) [26,44]. Therefore, it is important to understand the mutation status before starting the treatment with anti-EGFR therapies.

There have been very few studies involving Indian lung cancer patients response to tyrosine kinase inhibitors [27,28] & these studies showed some benefit of using TKIs. These studies didn't include the genetic evaluations of the tumour samples before initiation of the anti-EGFR treatment & therefore those results were based on patient population that has not been selected using genetic evaluations of the tumor samples.

However, a recent survey of EGFR mutations in Asian population (PIONEER study) [44] showed only 22% of EGFR mutation positive rate in a subgroup of Indian ethnic background patients which is lower than the rate observed by Sahoo et al., [26]. In our study we sought for the presence of EGFR mutations in non-smokers NSCLC patients in Indian population.

Methods

Patients
Tissue blocks of tumor specimen were obtained from 55 non-smokers and ex-smokers suffering with lung adenocarcinoma from two centres. One centre was in North India (Sir Ganga Ram Hospital, New Delhi) and second centre was in South India (HCG Oncology, Bengaluru). This study was approved by the Institutional Ethics Committee of Sir Gangaram Hospital, New Delhi & HCG Oncology, Bengaluru. Written informed consent of all participants was taken. This consent was approved by the ethics committees of the respective institutions mentioned above & it was conducted by these institutions as per the ethical principles for medical research (Declaration of Helsinki). Non smokers were defined as those who had smoked fewer than 100 cigarettes in their lifetime. Ex-smokers were defined as those who had smoked more than 100 cigarettes in their lifetime but had stopped smoking for at least 1 year before recruitment. Light ex-smokers were defined as those who had stopped smoking more than 15 years ago and had smoked fewer than 10 packs per year. Written informed consent for the use of tissues for molecular analysis was taken from patients at the time of procurement of tumor specimens.

DNA extraction and mutation detection
Lung biopsies/surgical tissue specimens of non-smoker or ex-smoker adenocarcinomas were analyzed for somatic EGFR mutation status. DNA was extracted from the paraffin embedded tissue specimens using QIAamp DNA FFPE tissue kit (Qiagen, GmBH, Germany) according to manufacturer's protocol. The extracted DNA was eluted in 50ul of buffer ATE.

The 29 reported somatic mutations were detected using DxS EGFR mutation detection kit (Qiagen). This method detects presence of 29 known mutations in exons 18 through 21 of EGFR gene. In exon 18 it checks for presence of G719X mutation (detects G719S, G719A or G719C but does not distinguish them). In exon 19 it checks for presence of 19 deletions but does not distinguish between them. In exon 20 it checks for presence of T790M, S768I & 3 insertions (does not distinguish the insertions). In exon 21 it checks for presence of L858R & L861Q. The method is based on ARMS PCR technology in which primers for 29 mutations are designed based on Scorpions technology. A 25ul reaction containing 5ul of not more than 20ng of DNA was set up for real-time PCR reaction according to the manufacturer's protocol. For quality assessment we sent two EGFR mutation positive samples to the external laboratory for confirming our results & both of them matched with our findings.

Data analyses
Samples were called positive or negative qualitatively depending on presence of mutations. The statistical significance was calculated using Chi-square test. The main variables in our analyses are gender & ethnic background (no smoking status since all are non-smokers). We have done the analyses separately within each variable & therefore we have not done statistical correction for multiple testing in our analyses.

Results

In this study, we enrolled 55 cases of non-smoker, advanced, inoperable adenocarcinoma of lung (grade 3 & higher) after June 2008 in a retrospective fashion. 26 patients were enrolled from Sir Ganga Ram Hospital, New Delhi (North India) and 29 from Healthcare Global Hospital, Bangalore (South India) (Table 1). Mutation analysis was possible in 50 of the 55 samples. The five samples did not have sufficient amplifiable DNA to perform the mutation analyses. There were 28 females & 22 males. There was no difference in male to female ratio among the samples from North & South India.

Of the 50 samples 28 (56%±14%) were positive for EGFR mutations (Table 2). Of the 28 females 17 (61%) and of 22 males 11 (50%) were positive for EGFR mutations. There was no statistically significant difference in mutation rate between males & females (p=0.45).

Of the 22 samples from North India 9 were positive for mutations (41%±21%). Of the 28 samples from South India 19 were positive for mutations (68%±17%). There is a trend of higher mutation positives in South Indians compared to North Indians (p=0.06). Among the females 5 of the 13 (38%±26%) North Indian NSCLC samples were mutation positive compared to 12 of the 15 (80%±20%) South Indian NSCLC samples. There was a statistically significant difference in the mutation status among north & south Indian females (p=0.02). No such difference was seen in the male NSCLC samples.

The most common mutations observed in this study were deletions in exon 19 (17/28,61%) followed by L858R in exon 21 (10/28–36%), G719X in exon 18 (2/28-7%) and insertions in exon 20 (1/28–3.6%). Two samples (4%) were found to have two different mutations (L858R in exon 21 and deletion in exon 19). Interestingly both samples were from South India.

Discussion

Our study in non-smoker Indian lung adenocarcinoma patients showed 56% EGFR mutation positive rate. This rate is higher than that of the European or other western populations in which it is in the range of 17 to 25% [29]. Such a high rate of mutation positives in non-smoker lung adenocarcinoma is remarkable & has important implications in case of newer targeted therapies like tyrosine kinase inhibitors. This mutation rate was similar to previous study in Indian lung cancer patients [26] and consistent with results from the studies involving Asian populations [25,30,31,44].

We chose ARMS PCR technology for detection of mutations because of its higher sensitivity & specificity than sequencing based methods resulting in high false negative rates [32,33]. This method has also been used previously in many studies involving response to TKIs [34-36].

Interestingly, we observed that in our sample south Indians have higher rate of mutation positive than north Indians (68% vs 41%). This is surprising considering we were looking at somatic mutations & not germline mutations or polymorphisms in which case the genetic differentiation may exist due to difference in ancestral background. However, the difference in the incidence of somatic mutation rates in EGFR gene between caucasians & east-asians is known & widely reported. This can be one of the explanations for having such difference in our study population considering that there is a genetic differentiation between North (Indo-european speaking) & South Indians [37] (Dravidian speaking). North Indians are closer to European genetically & have caucasoid admixture while South Indians are ethnically considered as australoid [38].

Liu et al., [39] previously showed that presence of high frequency of EGFR somatic mutations in Asian patients was also associated with higher incidence of germline polymorphisms in the EGFR gene. It will be interesting to study whether South Indians have higher rates of EGFR germline mutations and polymorphisms than that of North Indians. If so this could also explain why South Indians are tend to have higher somatic mutations in EGFR gene than North Indians.

We also need to confirm the observed higher difference in the mutation rates between North & South Indians in larger sample size. It will be interesting to see if the difference remains same if we increase the sample size. We also want to point out that trend is driven by primarily by higher mutation rate in female samples from South India compared to that of North India (p=0.02). The same trend was not seen in male patient samples. However, this needs to be confirmed in larger sample size and also by looking at the presence of germline polymorphisms between males & females which may also drive the somatic mutation rates. This is because a trend of decrease in lung adenocarcinoma risk was found in female subjects with hormone replacement therapy (HRT) and harbouring a higher number of protective EGFR SNPs (single nucleotide polymorphisms) near the tyrosine kinase domain of EGFR gene [40]. The EGFR SNPs have a cumulative effect on decreasing lung adenocarcinoma risk in never-smoking women with HRT use.

There was another study from India by Sahoo et al., [26] which screened lung cancer samples from all of India. In this study the overall incidence of mutation was around 52% which is similar to what we have seen in this study (56%). In the same study it was shown that there was a higher mutation rate in females than males which was not significant in our case probably due to the smaller sample size in our study. They also report differences in mutation rates specifically for mutations in exon 18&19& gender which was also not significant in our study. They didn't report the regional difference in the mutation rates as seen in our study. However, a recent epidemiological study of EGFR mutations in Asian patients [44] showed much lower EGFR mutation positive rate among Indian subgroup patients than what has been seen in our study & Sahoo et al., It will be interesting to see the ethnic background of the Indian patients included in the PIONEER study as well as the smoking status of these patients. If the predominant Indian patient population in their study group include North Indians & smokers then this may justify for the lower mutation rate (22%) observed in their study. This falls within the confidence interval for the North Indians mutation positive rate in our study (Table 3).

We also observed similar mutation rates with respect to the type of mutations as seen in the East Asian [30] and Indian [26] study. In our study the deletions in exon 19 were most common (61%) followed by L858R mutation (36%) followed by other mutations (10.6%). In the Mok et al., [30], the percentages were 54%, 43% & 8% respectively. The explanation for having high mutation rate in exon 19 could be due to the local haplotype structure favouring the occurrence of EGFR mutations especially deletions in exon 19 as was shown in a study from Taiwan [41]. We had two samples (4%) with double mutations (L858R & Deletions). It is interesting to have samples with multiple activating mutations & it has also been observed previously in Chinese [31,42] & other populations [43,45].

Our study indicates a higher EGFR mutation positive rate in the lung adenocarcinoma patients compared to that of populations of European descent. It is also comparable to the findings from the studies in other Asian populations. There is also a disparity in the mutation rate between north & south India especially in women. It is therefore important to evaluate the EGFR mutation status before prescribing the tyrosine kinase inhibitors in the case of non-smoker lung adenocarcinoma patients from North & South India to have effective treatment outcome.

Competing interests

The authors declare that they have no competing interests.

Authors' contributions

Acknowledgement

We thank Dr Reddy's Laboratories for their support.

Publication history

EIC: G. J. Peters, VU University Medical Center, Netherlands.
Received: 30-Mar-2014 Final Revised: 02-May-2014
Accepted: 15-May-2014 Published: 03-Jun-2014