Polymorphisms of EGFR predict clinical outcome in advanced non-small-cell lung cancer patients treated with Gefitinib

Lung Cancer 66 (2009) 114–119

Contents lists available at ScienceDirect

Lung Cancer journal homepage: www.elsevier.com/locate/lungcan

Polymorphisms of EGFR predict clinical outcome in advanced non-small-cell lung cancer patients treated with Gefitinib Fei Ma a,1 , Tong Sun b,1 , Yuankai Shi a , Dianke Yu b , Wen Tan b , Ming Yang b , Chen Wu b , Datong Chu a , Yan Sun a , Binghe Xu a,∗ , Dongxin Lin b,∗∗ a b

Department of Medical Oncology, Cancer Hospital and Institute, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China Department of Etiology and Carcinogenesis, Cancer Hospital and Institute, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China

a r t i c l e

i n f o

Article history: Received 30 September 2008 Received in revised form 2 December 2008 Accepted 25 December 2008 Keywords: EGFR Polymorphism Lung cancer Gefitinib therapy Clinical outcome

a b s t r a c t Purpose: Genetic variations in EGFR may alter protein function and therefore the therapeutic efficacy of epidermal growth factor receptor inhibitors. This study investigated the association between polymorphisms in EGFR and clinical outcome in patients with advanced non-small-cell lung cancer (NSCLC) treated with Gefitinib. Methods: A whole gene-based tag-SNP approach was used to determine the candidate SNPs in EGFR. Four tag SNPs, one CA simple sequence repeat (CA-SSR) in intron 1, one coding region SNP (R521K), and SNPs identified by resequencing in the tyrosine kinase domain of EGFR were selected to analyze their association with therapeutic outcome and survival in 84 advanced NSCLC patients treated with Gefitinib. Progression-free and overall survivals were computed by Cox model adjusted for clinical factors. Results: We identified two EGFR polymorphisms, rs2293347 (D994D) and CA-SSR in intron 1, associated with clinical outcome of Gefitinib therapy. The response rate for the rs2293347GG or shorter CA repeat genotype was significantly higher than that for the rs2293347GA or AA or longer CA repeat genotype (71.2% versus 37.5%, P = 0.0043 and 88.5% versus 48.3%, P = 0.0005). The rs2293347GG genotype was also associated with longer progression-free survival compared with the rs2293347GA or AA genotype (11 months versus 3 months, P = 0.0018). A combination of rs2293347GG and shorter CA repeat genotypes had more pronounced clinical benefit. Conclusion: The D994D and CA-SSR polymorphisms in EGFR are potential predictors for clinical outcome in advanced NSCLC patients treated with Gefitinib. © 2009 Elsevier Ireland Ltd. All rights reserved.

1. Introduction Lung cancer is the leading cause of cancer-related death in the world [1]. Non-small-cell lung cancer (NSCLC) accounts for ∼80% of all types of lung cancer and has only ∼14% of 5-year survival rate [2]. Although majority of NSCLC patients may benefit from chemotherapy, the platinum-based first-line therapy produces median survival times of 8–11 months and 1- and 2-year survival rates of 35–40% and 10–20%, respectively [3,4]. Gefitinib (IressaTM , ZD1839) is a first molecularly targeting agent for the treatment of patients with advanced NSCLC refractory to chemotherapy. It is an orally active, low molecular weight, synthetic quinazoline, which selectively targets the epidermal growth factor receptor (EGFR) and

∗ Corresponding author. Tel.: +86 10 87788826; fax: +86 10 87715711. ∗∗ Corresponding author. Tel.: +86 10 87788491; fax: +86 10 67722460. E-mail addresses: [email protected] (B. Xu), [email protected] (D. Lin). 1 These authors contributed equally to this work. 0169-5002/$ – see front matter © 2009 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.lungcan.2008.12.025

blocks the tyrosine kinase activity [5]. The response rates to Gefitinib among advanced NSCLC patients vary from 8.9 to 69% [6–8]. Studies have shown ethnic differences in response to Gefitinib and Asian patients appeared to have significantly higher probability of response. Although other factors, such as patient’s age, sex, diet, and performance status, might contribute to the interindividual variability [9], it is widely accepted that a significant portion of interindividual variability is due to genetic variations [10–12]. A significant difference in response to Gefitinib has been shown to depend on mutation status of EGFR, with the mutation-positive patients showing a much greater response than mutation-negative patients [13–15]. Compared with the wild-type EGFR, mutant EGFR exhibits increased activation following ligand binding and thus enhanced inhibitory effect of Gefitinib [13,16]. Although direct examination of EGFR mutations is possible, the clinical practice is somewhat difficult because somatic mutations can only be detected in target tissues [13]. It is necessary to obtain cancer specimens to get reliable results. If blood samples can be used for prediction to Gefitinib response, it would be more helpful in clinic. In addition, about 20% of patients showing response to Gefitinib do not have

F. Ma et al. / Lung Cancer 66 (2009) 114–119

somatic EGFR mutation, indicating that mutations are not the sole determinant for response to the drug [17]. Other genetic factors contributing to sensitivity of Gefitinib in NSCLC patients are largely unknown yet and, thus, efforts have been made to identify genetic polymorphisms in EGFR that have impact in response to Gefitinib treatment. A few studies evaluating the association between genetic polymorphisms in EGFR and efficacy of Gefitinib treatment have been published. The most studied polymorphism is a CA simple sequence repeat (CA-SSR) in intron 1 of EGFR [18,19]. It was shown that transcription activity of EGFR declines with the numbers of CA dinucleotides increasing and this polymorphism has been associated with the efficacy of EGFR inhibitors [18,20–23]. A EGFR promoter polymorphism (−216G > T) was also reported to be associated with altered EGFR expression and differential sensitivity and toxicity to EGFR inhibitors [24–28]. However, these studies investigated only one or two but not all polymorphisms in the EGFR gene. Many polymorphisms have been identified in the EGFR gene according to the International HapMap Project (http://www.hapmap.org/) and Environmental Genome Project (http://egp.gs.washington.edu/). It is highly warranted to identify all variants that influence EGFR expression or function and examine their effects on sensitivity to EGFR inhibitors. Here we reported a comprehensive analysis of EGFR polymorphisms and their effect on outcome of Gefitinib therapy.

115

cal Sciences (Beijing, China) between March 2004 and March 2006. The blood samples were collected at the time of enrollment. The primary outcome of interest in this study was the best response achieved during the study until disease progression. The clinical response to the therapy was evaluated every 8 weeks and at the end of Gefitinib treatment according to the Response Evaluation Criteria in Solid Tumors [29]. Complete response (CR) and partial response (PR) had a confirmation at a minimum of 4 weeks between the assessments, and stable disease (SD) should be present for a minimum of 24 weeks from the first dose. Progression within 24 weeks was defined as progressive disease (PD). In this study, patients achieving CR, PR, or SD for at least 24 weeks were classified as responders whereas patients having PD within 24 weeks were classified as non-responders. Progression-free survival (PFS) was defined as the time from first dose of Gefitinib to the date of disease progression or death due to any cause. Overall survival (OS) was defined as the time between first dose of Gefitinib and death or the last follow-up. Dates of death were obtained from (a) inpatient and outpatient records, (b) patient’s family, or (c) local Public Security Census Register Office. Patients who were not departed were censored at the last date they were known to be alive based on the date of last contact. This study was approved by the Institutional Review Board of the Chinese Academy of Medical Sciences Cancer Hospital. 2.2. Polymorphism analysis

2. Materials and methods 2.1. Patients and treatment This study recruited 84 patients with advanced NSCLC. All patients were Han Chinese. Eligible patients had at least one measurable lesion with a minimum size in at least one diameter of ≥10 mm for liver, lung, brain or lymph node metastases, WHO performance status of 0–1, and life expectancy ≥3 months. Each patient received orally Gefitinib at daily dose of 250 mg as 2nd or 3rd line monotherapy at Cancer Hospital, Chinese Academy of Medi-

Tag-SNPs were selected according to the genotyped SNPs in Asian populations in the International HapMap Project (http://www.hapmap.org/) and the Environmental Genome Project (http://egp.gs.washington.edu/). Genotypes of 235 SNPs across a 195.3-kb region spanning EGFR gene locus (NM 005228) on chromosome 7p11 were surveyed. Tag-SNP selection process was performed in a pairwise mode, by selecting a set of markers in which all correlated alleles are captured at r2 > 0.8 as described [30]. Four tag SNPs were thus chosen as candidate SNPs and each of them represents ≥5 SNPs (Fig. 1). If one or more tag SNPs had impact on

Fig. 1. The genomic organization and candidate polymorphic loci or region of EGFR gene. (A) EGFR gene located in chromosome 7p12. (B) EGFR gene structure and genotyped polymorphic loci or region. The candidate polymorphic loci included 4 representative tag SNPs and one well-known CA simple sequence repeat (SSR) polymorphism. The region encoding EGFR tyrosine kinase domain and intron/exon border were chosen for sequencing. (C) The linkage disequilibrium plot of EGFR for Han Chinese population using HapMap Phase II data. LD plots were identified by strong LD.

116

F. Ma et al. / Lung Cancer 66 (2009) 114–119

Table 1 Information and genotyping approach of some polymorphisms in the EGFR gene. No. 1 2 3 4 5 6

Identity rs4947492 rs11977388 rs2075102 rs7809028 rs2293347 rs11543848

a b

Base change A>G T>C C>A G>A G>A A>G

Locationa 101022 145276 166335 – 181946 (D994D) 142285 (K521R)

Position Intron 1 Intron 14 Intron 20 Downstream Exon 25 Exon 13

PCR primersb

Tm









5 -GAAGTGGAGATAGTCACATATTAGGC-3 ; 5 -TGACCAAGAGCACTCTACACAG-3 5 -TCAATATCGCCAATTATTTTAATGCC-3 ; 5 -TGTGTTTTGAGGGATAGGCTGC-3 5 -AGACTGTAATCTGCATTAGGTGCTGA-3 ; 5 -ATGCCCAGGACTGCAGAGTC-3 5 -CATCAGAGAGTGGACCAGCA-3 ; 5 -AGCATTAGGGGTGGTGTGAG-3 5 -ATGAGGTACTCGTCGGCATC-3 ; 5 -GAACCAAGGGGGATTTCATT-3 5 -TGCTGTGACCCACTCTGTCT-3 ; 5 -CAACGCAAGGGGATTAAAGA-3

Endonuclease ◦

63 C 58 ◦ C 57 ◦ C 68 ◦ C 60 ◦ C 63 ◦ C

Bgl I Msp I Dde I BsmA I Tfi I Sty I

Nucleotide location counting from the ATG codon of the EGFR gene. The underlined-nucleotides were mismatch nucleotides introduced to create a restriction enzyme recognizing site.

response to Gefitinib, functional SNPs was addressed among those SNPs the tag SNPs represent. SNPs in the most important tyrosine kinase domain of EGFR (exons 18–21 and their exon/intron boundaries) and two well-studied polymorphisms, CA-SSR in intron 1 and rs11543848 (R521K, also assigned as R497K in literature) were also included (Fig. 1). The genotypes of SNPs in the kinase domain and CA-SSR were determined by DNA-sequencing in an ABI 3730 system. Polymerase chain reaction-restriction fragment length polymorphism (PCR-RFLP) methods were established to analyze other candidate SNPs (Table 1; Supplementary Table 1). 2.3. Statistical analysis 2 -Test was used to compare clinical characteristics and response according to EGFR genotypes. A Fisher’s exact test was used when necessary. Progression-free and overall survival curves were drawn with the Kaplan–Meier product limit method for each of the different genotypes. Comparisons were made with the log-rank test. Hazard ratios of both progression and death with 95% confidence intervals were estimated by using the Cox model, with genotype, sex, smoking status and histological subtype as covariates. All statistical tests are two-sided tests and P < 0.05 was considered significance. The analyses were performed with the SPSS software package (Version 12.0, SPSS Inc., Chicago, IL). 3. Results 3.1. Patient characteristics and response to Gefitinib treatment Patients’ baseline characteristics and response to Gefitinib are shown in Table 2. Median age of the patients was 56 years (range, 34–76 years). Among patients, female, non-smoker, and adenocarcinoma accounted for 57.1%, 82.1%, and 79.8%, respectively. Median follow-up period was 21 months (range, 14–51 months). As of December 2007, 73 patients (86.9%) had disease progression, 46 patients (54.8%) were confirmed as dead, and the survival status of 5 patients (6.0%) was unknown. Among all patients treated with Gefitinib, there was no complete response (CR) whereas 31 (36.9%) had partial response (PR), 20 (23.8%) had stable disease more than 24 weeks (SD), and 33 (39.3%) had progressive disease (PD). Overall, 51 (60.7%) patients were responder whereas 33 (39.3%) patients were non-responder. 3.2. EGFR tag-SNPs and response to Gefitinib therapy Four tag-SNPs, rs4947492, rs11977388, rs2075102, and rs7809028, which represent 6, 17, 16, and 5 common SNPs, respectively, were genotyped. The genotype frequencies of these tag-SNPs and their distributions among responders and non-responders of Gefitinib treatment were summarized in Table 3. No significant difference in response to Gefitinib therapy was observed among different genotypes of 3 tag-SNPs, i.e., rs4947492, rs11977388, and rs2075102. However, patients carrying the rs7809028G allele tended to have higher response rate than those carrying the

Table 2 Clinical characteristics of patients with advanced NSCLC and clinical response to Gefitinib therapy. Patients (n = 84) No.

%

Age (years) <60 ≥60

49 35

58.3 41.7

Sex Male Female

36 48

42.9 57.1

WHO performance status 0 1

22 62

26.2 73.8

Smoking status Non-smoker Smoker

69 15

82.1 17.9

Histological subtype Adenocarcinoma Othera

67 17

79.8 20.2

Gefitinib therapy 2nd line 3rd line

44 40

52.4 47.6

Clinical response Partial response Stable disease Progressive disease

31 20 33

36.9 23.8 39.3

a

Includes squamous cell carcinoma (n = 12) and large cell carcinoma (n = 5).

Table 3 Genotype distribution of 4 EGFR tag-SNPs among responders and non-responders of advanced NSCLC patients treated with Gefitinib. Respondera

Non-responderb

No.

%

No.

rs4947492 A > G AA AG GG

22 22 7

56.4 66.7 70.0

17 11 3

43.6 33.3 30.0

rs11977388 T > C TT TC CC

17 29 5

58.6 65.9 45.5

12 15 6

41.4 34.1 54.5

rs2075102 C > A CC CA AA

33 18 0

60.0 62.1 0.0

22 11 1

40.0 37.9 100.0

rs7809028 G > A GG GA AA

24 18 7

66.7 69.2 38.9

12 8 11

33.3 30.8 61.1

Genotype

a

P

% 0.578

0.444

0.854

0.085c

Includes partial response and stable disease. Includes progressive disease. c P = 0.027 when rs7809028 AA genotype was compared with combined GG and GA genotype. b

F. Ma et al. / Lung Cancer 66 (2009) 114–119

A allele (67.3% versus 51.6%) although the difference was only marginally significant (P = 0.067). Specifically, the response rate was 67.7% among 62 patients with the GG and GA genotype, and 38.9% among 18 patients with the AA (P = 0.027). These results imply that one or more SNPs represented by this tag SNP might have impact on response to Gefitinib. Among the 5 SNPs that rs7809028 represents, rs2293347 is the only SNP that is located in coding region (exon 25) and may has potential function. The correlation between rs2293347 genotype and response to Gefitinib was further analyzed as described below.

117

CA-SSR alleles on attenuating response to the therapy. Genotype frequencies of the 2293347G > A and CA-SSR were not significantly different in terms of sex, age, histological subtype of the patients (Supplementary Table 2). 3.4. Genotypes of rs2293347 and CA-SSR and outcome of Gefitinib therapy We further investigated whether there was significant difference in progression-free survival (PFS) and overall survival (OS) among patients carrying different EGFR rs2293347 and/or CA-SSR genotypes after treatment with Gefitinib. It was found that patients carrying the rs2293347GG genotype had a longer median PFS than patients carrying the GA or AA genotypes [10 months (range, 1–33) versus 3 months (range, 1–20), P = 0.002; Fig. 2a]. Patients carrying the rs2293347GG genotype also had longer median OS than patients carrying the AG or AA genotypes although the difference was not statistically significant probably due to limited sample size in analysis [21 months (range, 3–48) versus 13 months (range, 1–51), P = 0.176; Fig. 2b). However, no significant difference in either PFS [10 months (range, 2–32) for shorter CA carriers versus 6 months (range, 1–33) for longer CA carriers, P = 0.580; Fig. 2c] or OS [22 months (range, 4–36) for shorter CA carriers versus 16 months (range, 1–51) for longer CA carriers, P = 0.662; Fig. 2d] in patients with different CA-SSR genotypes was observed. Multivariate analysis was taken to evaluate whether the apparent association of rs2293347 or CA-SSR with outcome could be accounted for by other clinical factors which may affect the outcome of Gefitinib, such as sex, smoking status, performance status and histological subtype. For both functional polymorphisms, the rs2293347G > A polymorphism (adjusted hazard ratio = 2.29, 95% CI = 1.30–4.03, P = 0.004) was a predictor of poor PFS for patients treated with Gefitinib (Supplementary Table 3).

3.3. Other EGFR polymorphisms and response to Gefitinib therapy Additional 10 common SNPs with heterozygosity >0.01 were identified in EGFR exons 18–21 and their exon/intron boundaries by DNA sequencing (Supplementary Table 1). Although none of these SNPs showed significant association with response to Gefitinib therapy, several interesting tendency were observed. Three responder patients carried the variant GA genotype at 155077 polymorphic site in intron 18, which has never been reported before, but none of non-responder patients carried this genotype. Patients carried the variant allele (AG or AA genotype) of rs1050171, a synonymous cSNP in exon 20, tend to have higher response rate than patients carried the wild-type GG genotype (78.9% versus 55.4%, P = 0.064). Similar result was seen for rs10241326, a T > C SNP located in intron19, probably due to its strong linkage to the cSNP rs1050171 (Supplementary Table 1). Because rs11543848 (K521R) and rs2293347 (D994D) SNPs are located in the coding region and might have functional significance, we further analyzed these SNPs and outcome of patients treated with Gefitinib. It was found that although rs11543848 SNP was not significantly associated with response to Gefitinib therapy, patients carrying the rs2293347A allele had significantly lower response rate than patients carrying the rs2293347G allele (37.5% versus 71.2%, P = 0.004). Besides the rs2293347, another functional polymorphism in intron 1, CA-SSR, was also associated with Gefitinib response. Patients carrying at least one shorter CA repeat allele [≤16 (CA)n ] had significantly higher response rate than patients carrying two longer CA repeat alleles (88.5% versus 48.3%, P < 0.001). We further analyzed whether there is a joint effect between rs2293347 and CA-SSR genotypes on response to Gefitinib (Table 4). Interestingly, 90.5% of patients carrying both rs2293347GG and shorter CA repeat genotypes response to Gefitinib therapy; however, the response rate decreased to 26.3% in patients carried both rs2293347GA or AA and longer CA repeat genotypes (P < 0.001). These results suggest a significant joint effect between the rs2293347A and longer

4. Discussion Because response to Gefitinib has been shown to be dependent on mutation status of EGFR, we examined whether germline polymorphisms of EGFR had effect on clinical response and outcome in advanced NSCLC patients treated with Gefitinib. Using whole gene-based tag-SNP approach, we found that the rs2293347G > A polymorphism in exon 25 of EGFR was associated with poor response to Gefitinib. We also found that the CA-SSR polymorphism in intron 1 of the gene had effect on response to Gefitinib and patients carrying the longer CA-SSR allele had significantly lower response rate than patients carrying the shorter CA-SSR allele. In

Table 4 Analysis of response to Gefitinib in advanced NSCLC: association with EGFR genetic polymorphisms. Genotype

Respondera

Non-responderb

P

No.

%

No.

%

rs2293347 G > A (D994D) GG GA + AAc

42 9

71.2 37.5

17 15

28.8 62.5

CA-SSR in intron 1d Shorter CA-SSR Longer CA-SSR

23 28

88.5 48.3

3 30

11.5 51.7

Combination of rs2293347 G > A and CA-SSR rs2293347 GG and shorter CA-SSR rs2293347 AG + AA and shorter CA-SSR rs2293347 GG and longer CA-SSR rs2293347 AG + AA and longer CA-SSR

19 4 23 5

90.5 80.0 60.5 26.3

2 1 15 14

9.5 20.0 39.5 73.7

a b c d

0.004

<0.001

<0.001

Includes partial response and stable disease. Includes progressive disease. Only one AA genotype that was non-responseder. Shorter CA-SSR, with any allele of ≤16 (CA)n ; longer CA-SSR, with both alleles of >16 (CA)n .

118

F. Ma et al. / Lung Cancer 66 (2009) 114–119

Fig. 2. Kaplan–Meier plots and log-rank tests for progression-free survival and overall survival in advanced NSCLC patients treated with Gefitinib: association with EGFR polymorphisms. Progression-free survival curves for rs2293347 genotypes (A) and CA-SSR genotypes (C); Overall survival curves for rs2293347 genotypes (B) and CA-SSR genotypes (D).

addition to the response rate, the rs2293347G > A polymorphism was significantly associated with poor PFS in patients treated with Gefitinib. These results indicate that, besides somatic mutations, naturally occurring sequence variations in EGFR may also influence the efficacy of Gefitinib treatment and thus should be considered as important modifiers for advanced NSCLC treated with EGFR inhibitors. The rs2293347G > A polymorphism located in exon 25 is a synonymous SNP (D994D) and a synonymous SNP does not change amino acid sequence of the protein, which might not influence biological function of the protein itself. However, synonymous SNPs may affect mRNA stability, translational kinetics, and splicing, resulting in alteration of protein amount, structure or function [31]. For example, it has been reported that synonymous SNPs alter the interaction of ABC transporter ABCB1, with its substrates and inhibitors plausibly mediated by translational pauses induced by rare codons, because the synonymous SNPs that produce the phenotype all use relatively rare codons [32]. Synonymous SNPs were also shown to influence protein expression (and thus function) by either alteration in the stability of the mRNA or mRNA splicing [33–35]. Our result showing an association between the rs2293347G > A polymorphism and poor response to Gefitinib therapy among patients with NSCLC suggests that this synonymous SNP might be of functional relevance. Further studies are under way to examine whether this SNP affects EGFR expression or its biological function. Several studies addressing the effects of naturally occurring EGFR polymorphisms on efficacy of EGFR inhibitors have been published. Polymorphisms in the essential promoter of EGFR (−216G > T and −191C > A), which might affect transcription factor Sp1 binding, have been shown to be associated with the clinical outcomes and toxicity in patients treated with EGFR inhibitor [24,26–28]. However, the frequencies of the two SNPs are very low in Asian populations [28], and were therefore not analyzed in our study.

Another cSNP, rs11543848 (R521K, also assigned as R497K in literature), which seems to decrease the activity of EGFR [36], was reported to be weekly associated with relative resistance to some EGFR inhibitors in NCI60 cell lines in vitro only in combination with the −216G > T SNP [37]. In the present study, we did not observe any significant association between the R521K polymorphism and response to Gefitinib in advanced NSCLC patients. The CA-SSR polymorphism in intron 1 of EGFR has been studied more intensively. It has been shown in an in vitro study that head and neck carcinoma cell lines with shorter CA repeat displayed higher level of EGFR mRNA and protein expression and were significantly more sensitive than those with longer CA repeat to inhibitory effect of Erlotinib, another EGFR inhibitor [18]. In this report, an association between shorter CA repeat genotype and skin toxicity of Gefitinib, which may be related to the antitumor activity of this agent, was also observed in colorectal cancer patients [18]. Recently, Nie et al. reported that Chinese NSCLC patients with shorter CA repeat genotype were more sensitive to Gefitinib than patients with the longer repeat genotype [38], which is consistent with our results. In Nie’s study, a correlation was found between the shorter CA repeat genotype and longer patients’ survival time [38]; in our study, however, no such effect was observed in patients treated with Gefitinib. Several other studies showed that the shorter CA repeat genotype was a poor prognostic indicator for NSCLC and other cancers [23,38–40]. The conflict results in these studies might be due to Gefitinib therapy, subsequent treatment and/or other unknown factors such as the presence of other SNPs or somatic mutations in EGFR. Because the CA-SSR polymorphism is associated with the expression of EFGR [18], Gefitinib therapy may subvert the intrinsic clinical outcome in patients with different CA-SSR genotypes. However, the sample size in Nie’s study (70 patients) and ours (84 patients) was too limited to address this hypothesis. The exact effect of the CA-SSR genotype on survival time of patients treated with or without EGFR inhibitors needs further clinical investigations with larger sample size.

F. Ma et al. / Lung Cancer 66 (2009) 114–119

The present study might have some limitations. With the number of variables we analyzed, in a sample size of 84, there is a likelihood of identifying spurious associations by random chance. The confirmation of apparent correlation is in progress in some independent data set. In addition, although we captured most functional SNPs in EGFR, we did not analyze somatic mutations of EGFR in target tissues because of lacking lung cancer tissue specimens. It has been shown that some polymorphisms in EGFR might be associated with somatic mutations [28]. In addition, we observed only certain variants but not all polymorphisms that were associated with efficacy of Gefitinib, suggesting that these effects may not be due to EGFR somatic mutations. Furthermore, genetic polymorphisms of EGFR have been reported to affect the therapeutic efficacy of EGFR inhibitors [18,24,37–38]. A study has being undertaken to compare the effect between the EGFR CA-SSR and D994D polymorphisms and EGFR mutations in terms of outcome of Gefitinib therapy. 5. Conclusion This comprehensive analysis of EGFR polymorphisms showed that rs2293347 (D994D) and CA-SSR polymorphisms are associated with outcome of Gefitinib therapy in advanced NSCLC patients. These results suggest that rs2293347 and CA-SSR genotypes may have potential to serve as predictors for clinical outcome of NSCLC patients treated with Gefitinib. Conflict of interest None declared. Acknowledgement This work was supported in part by the Chinese Society of Clinical Oncology Science Foundation Y-2005-0018 and National High Technology Project Grant 2006AA02A401. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.lungcan.2008.12.025. References [1] Walker S. Updates in non-small cell lung cancer. Clin J Oncol Nurs 2008;12:587–96. [2] Spira A, Ettinger DS. Multidisciplinary management of lung cancer. N Engl J Med 2004;350:379–92. [3] Schiller JH, Harrington D, Belani CP, et al. Comparison of four chemotherapy regimens for advanced non-small cell lung cancer. N Engl J Med 2002;346:92–8. [4] Fossella F, Pereira JR, von Pawel J, et al. Randomized, multinational, phase III study of docetaxel plus platinum combinations versus vinorelbine plus cisplatin for advanced non-small-cell lung cancer: the TAX 326 study group. J Clin Oncol 2003;21:3016–24. [5] Fukuoka M, Yano S, Giaccone G, et al. Multi-institutional randomized phase II trial of Gefitinib for previously treated patients with advanced non-small-cell lung cancer. J Clin Oncol 2003;21:2237–46. [6] Kris MG, Natale RB, Herbst RS, et al. Efficacy of Gefitinib, an inhibitor of the epidermal growth factor receptor tyrosine kinase, in symptomatic patients with non-small cell lung cancer: a randomized trial. JAMA 2003;290:2149–58. [7] Lee DH, Han JY, Lee HG, et al. Gefitinib as a first-line therapy of advanced or metastatic adenocarcinoma of the lung in never-smokers. Clin Cancer Res 2005;11:3032–7. [8] van Puijenbroek R, Bosquee L, Meert AP, et al. Gefitinib monotherapy in advanced nonsmall cell lung cancer: a large Western community implementation study. Eur Respir J 2007;29:128–33. [9] Ando M, Okamoto I, Yamamoto N, et al. Predictive factors for interstitial lung disease, antitumor response, and survival in non-small-cell lung cancer patients treated with Gefitinib. J Clin Oncol 2006;24:2549–56. [10] Bonjour JP, Chevalley T, Rizzoli R, et al. Gene-environment interactions in the skeletal response to nutrition and exercise during growth. Med Sport Sci 2007;51:64–80.

119

[11] Saposnik B, Reny JL, Gaussem P, et al. A haplotype of the EPCR gene is associated with increased plasma levels of sEPCR and is a candidate risk factor for thrombosis. Blood 2004;103:1311–8. [12] Walgren RA, Meucci MA, McLeod HL. Pharmacogenomic discovery approaches: will the real genes please stand up? J Clin Oncol 2005;23:7342–9. [13] Lynch TJ, Bell DW, Sordella R, et al. Activating mutations in the epidermal growth factor receptor underlying responsiveness of non-small-cell lung cancer to Gefitinib. N Engl J Med 2004;350:2129–39. [14] Paez JG, Jänne PA, Lee JC, et al. EGFR mutations in lung cancer: correlation with clinical response to Gefitinib therapy. Science 2004;304:1497–500. [15] Han SW, Kim TY, Hwang PG, et al. Predictive and prognostic impact of epidermal growth factor receptor mutation in non-small-cell lung cancer patients treated with Gefitinib. J Clin Oncol 2005;23:2493–501. [16] Bell DW, Lynch TJ, Haserlat SM, et al. Epidermal growth factor receptor mutations and gene amplification in non-small-cell lung cancer: molecular analysis of the IDEAL/INTACT Gefitinib trials. J Clin Oncol 2005;23:8081–92. [17] Sharma SV, Bell DW, Settleman J, et al. Epidermal growth factor receptor mutations in lung cancer. Nat Rev Cancer 2007;7:169–81. [18] Amador ML, Oppenheimer D, Perea S, et al. An epidermal growth factor receptor intron 1 polymorphism mediates response to epidermal growth factor receptor inhibitors. Cancer Res 2004;64:9139–43. [19] Perea S, Oppenheimer D, Amador M. Genotypic bases of EGFR inhibitors pharmacological actions. Proc Am Soc Clin Oncol 2004;23(196):A3005. [20] Brandt B, Meyer-Staeckling S, Schmidt H, et al. Mechanisms of egfr gene transcription modulation: relationship to cancer risk and therapy response. Clin Cancer Res 2006;12:7252–60. [21] Chrysogelos SA. Chromatin structure of the EGFR gene suggests a role for intron 1 sequences in its regulation in breast cancer cells. Nucleic Acids Res 1993;21:5736–41. [22] Etienne-Grimaldi MC, Pereira S, Magne N, et al. Analysis of the dinucleotide repeat polymorphism in the epidermal growth factor receptor (EGFR) gene in head and neck cancer patients. Ann Oncol 2005;16:934–41. [23] Han SW, Kim TY, Lee KW, et al. EGFR mutation and intron 1 CA repeat polymorphism as predictive markers of Gefitinib responsiveness in non-small cell lung cancer (NSCLC). J Clin Oncol ASCO Annual Meeting Proceedings (Post-Meeting Edition) 2006;24(18S):A7173. [24] Liu W, Innocenti F, Wu MH, et al. A functional common polymorphism in a Sp1 recognition site of the epidermal growth factor receptor gene promoter. Cancer Res 2005;65:46–53. [25] Rudin CM, Liu W, Desai A, et al. Pharmacogenomic and pharmacokinetic determinants of erlotinib toxicity. J Clin Oncol 2008;26:1119–27. [26] Liu G, Gurubhagavatula S, Zhou W, et al. Epidermal growth factor polymorphisms and clinical outcomes in non-small-cell lung cancer patients treated with gefitinib. Pharmacogenomics J 2008;8:129–38. [27] Gregorc V, Hidalgo M, Spreafico A, et al. Germline polymorphisms in EGFR and survival in patients with lung cancer receiving gefitinib. Clin Pharmacol Ther 2008;83:477–84. [28] Nomura M, Shigematsu H, Li L, et al. Polymorphisms, mutations, and amplification of the EGFR gene in non-small cell lung cancers. PLoS Med 2007;4: e125. [29] Therasse P, Arbuck SG, Eisenhauer EA, et al. New guidelines to evaluate the response to treatment in solid tumors. J Natl Cancer Inst 2000;92:205–16. [30] Stram DO. Software for tag single nucleotide polymorphism selection. Hum Genomics 2005;2:144–51. [31] Sauna ZE, Kimchi-Sarfaty C, Ambudkar SV, et al. Silent polymorphisms speak: how they affect pharmacogenomics and the treatment of cancer. Cancer Res 2007;67:9609–12. [32] Kimchi-Sarfaty C, Oh JM, Kim I-W, et al. A “Silent” polymorphism in the MDR1 gene changes substrate specificity. Science 2007;315:525–8. [33] Nackley AG, Shabalina SA, Tchivileva IE, et al. Human catechol-Omethyltransferase haplotypes modulate protein expression by altering mRNA secondary structure. Science 2006;314:1930–3. [34] Capon F, Allen MH, Ameen M, et al. A synonymous SNP of the corneodesmosin gene leads to increased mRNA stability and demonstrates association with psoriasis across diverse ethnic groups. Hum Mol Genet 2004;13:2361–8. [35] Nielsen KB, Sorensen S, Cartegni L, et al. Seemingly neutral polymorphic variants may confer immunity to splicing-inactivating mutations: a synonymous SNP in exon 5 of MCAD protects from deleterious mutations in a flanking exonic splicing enhancer. Am J Hum Genet 2007;80:416–32. [36] Moriai T, Kobrin MS, Hope C, et al. A variant epidermal growth factor receptor exhibits altered type transforming growth factor binding and transmembrane signalling. Proc Natl Acad Sci USA 1994;91:10217–21. [37] Liu W, Wu X, Zhang W, et al. Relationship of EGFR mutations, expression, amplification, and polymorphisms to epidermal growth factor receptor inhibitors in the NCI60 cell lines. Clin Cancer Res 2007;13:6788–95. [38] Nie Q, Wang Z, Zhang GC, et al. The epidermal growth factor receptor intron 1 (CA)n microsatellite polymorphism is a potential predictor of treatment outcome in patients with advanced lung cancer treated with Gefitinib. Eur J Pharmacol 2007;570:175–81. [39] Liu W, Montenegro RC, Zhang W, et al. Response to EGFR inhibitors in fibroblast cell lines and its association with germline polymorphisms. J Clin Oncol ASCO Annual Meeting Proceedings (Post-Meeting Edition) 2006;24(18S): A13167. [40] Dubey S, Stephenson P, Levy DE, et al. EGFR dinucleotide repeat polymorphism as a prognostic indicator in non-small cell lung cancer. J Thorac Oncol 2006;1:406–12.