- Email: [email protected]
SNPs in the transforming growth factor-β pathway as predictors of outcome in advanced lung adenocarcinoma with EGFR mutations treated with gefitinib
Annals of Oncology Advance Access published June 13, 2014 1
SNPs in the transforming growth factor beta pathway as predictors of outcome in advanced lung adenocarcinoma with EGFR mutations treated with Gefitinib
L Zhang1, Q.X. Li1, H.L. Wu1, X Lu2, M Yang3, S.Y. Yu1, X.L. Yuan1,
*
Department of Oncology, Tongji Hospital, Huazhong University of Science and
Technology, Wuhan, Hubei Province, China 2
Department of Transplantation, Tongji Hospital, Huazhong University of Science and
Technology, Wuhan, Hubei Province, China 3
College of Life Science and Technology, Beijing University of Chemical Technology,
Beijing, China *
Corresponding author: Dr. X.L. Yuan, Department of Oncology, Tongji Hospital,
Huazhong University of Science and Technology, Wuhan 430030, China. Tel: 86-027-83663406; Fax:86-027-83662834; E-mail: [email protected]
Key Message: "The manuscript describes the original research investigating the use of genetic variations to predict response of lung adenocarcinoma to tyrosine kinase inhibitor (TKI) treatment. We found that the genetic variations in the TGF‐β signaling pathway are associated with survival in NSCLC patients treated with EGFR‐TKI, especially in the patients with EGFR mutation. Our study also showed that other genetic factors are associated with acquired resistance to EGFR‐TKI using systematic genetic analysis. Our study was base on a complete clinical dataset with EGFR mutation status determined in all the patients and the genetic variations determined by analyzing SNPs of candidate genes involved in TGF‐β pathways. We used dual luciferase reporter assay and Western Blot to analyze the function of SMAD3 rs11632964, which was significantly associated with survival time of patients.Our findings will help to select patients who can benefit from TKI therapy."
© The Author 2014. Published by Oxford University Press on behalf of the European Society for Medical Oncology. All rights reserved. For permissions, please email: [email protected].
Downloaded from http://annonc.oxfordjournals.org/ at Memorial University of Newfoundland, Health Sciences Library on July 18, 2014
1
2
Abstract Background: The aim of this study was to evaluate whether genetic variations in the transforming growth factor beta (TGF-β) pathway influenced clinical outcome of advanced lung ademocarcinoma with EGFR mutations treated with gefitinib.
were enrolled in this study. EGFR mutation in these tumors was detected. Among them 106 patients with EGFR mutation and 37 of 100 patients with wild type were treated with gefitinib. Genotype of 33 SNPs from 13 genes involved in the TGF-β signaling pathway was determined and their association with survival time was analyzed. Univariate and multivariate analyses were performed to assess the role of biological/clinical parameters in progression-free survival (PFS) and overall survival (OS) using Pearson’s χ2 test, log-rank test and Cox proportional hazards model.
Results: Among SNPs analyzed, multivariate analysis showed the CT genotype of SMAD3: rs11632964 was associated with a longer OS and PFS when the entire cohort of 143 patients were included; the association was significant in the patients with EGFR mutant tumors (30.8 months vs. 17.5 months; Log-rank P=0.020; and 20.8 months vs. 9.4 months; Log-rank P=0.001), as compared with patients with wild-type EGFR tumors. In patients with mutant EGFR, the CT genotype of SMAD3: rs11071938 and the CC genotype of SMAD3: rs6494633 were also found to be associated with better PFS. Dual luciferase reporter assays showed gefitinib-resistant PC9/G cells transfeted with SMAD3: rs11632964T allelic reporter construct showed significantly lower luciferase activities compared to cells expression C allelic reporter construct. There was significantly decreased expression of SMAD3 and pi-SMAD3 in the PC-9/G cells compared with PC-9.
Conclusions: Among the candidate genes involved in the TGF-β pathway, the polymorphisms of SMAD3 appear to be highly predictive of outcome of patients with lung ademocarcinoma after gefitinib treatment, especially in those with EGFR mutations.
Key words: Polymorphisms; TGF-β; EGFR-TKI; SMAD3; Survival
Downloaded from http://annonc.oxfordjournals.org/ at Memorial University of Newfoundland, Health Sciences Library on July 18, 2014
Patients and methods: Two hundred six patients with advanced lung adenocarcinomas
3
Introduction The overall prognosis of patients with non-molecularly selected advanced non-small cell lung cancer (NSCLC) is poor. In East Asia, the epidermal growth factor receptor (EGFR) tyrosine kinase inhibitors (TKI) gefitinib is widely used [1]. In clinical trials, dramatic responses have been observed soon after initiation of treatment, particularly in Asians,
activating EGFR mutations, gefitinib could significantly prolong progression-free survival (PFS) compared to cytotoxic chemotherapy [3]. However, drug resistance eventually develops in most patients despite the initial good response, with a median PFS of about 1 year [4].
Resistance to EGFR-TKI has been associated with the acquisition of EGFR T790M mutation) [5] and overexpression and/or activation of c-Met [6, 7]. There are also other potential genetic factors associated with EGFR gene mutations that may be useful to predict the likelihood of response or resistance to EGFR-TKI.
In contrast with EGFR signaling pathway, the transforming growth factor-beta (TGF-β) signaling pathway is strongly anti-proliferative, which plays an important role in tumorigenesis，cancer progression and metastasis via crosstalk with Smad and EGFR signaling pathways, like P13K/AKT and RPTP-к [8,9]. To our knowledge, TGF-β signaling pathway has two major mechanisms to affect the outcome of EGFR-TKI. First, epithelial-mesenchymal transition (EMT) is one of the important mechanisms of TKI resistant, and changes in EMT status can explain resistance to EGFR-TKI [10,11]. Receptor Smads (Smad2 and particularly Smad3), the direct targets of the activated TGF-β receptor have been implicated as critical mediators in EMT [11]. Second, Bcl-2L11 (BIM) associated with TKI resistance [12], and one target of Smad3 in inducing apoptosis is Bcl-2, a key antiapoptotic inhibitor. TGF-β/Smad3 signaling facilitates a permissive cellular context that is tending to promote apoptosis by attenuating the level of the major apoptosis inhibitor Bcl-2 [13].
Downloaded from http://annonc.oxfordjournals.org/ at Memorial University of Newfoundland, Health Sciences Library on July 18, 2014
women, nonsmokers, and adenocarcinoma cases [2]. For patients with tumors harboring
4
The purpose of our study is to determine whether genetic variations in the TGF-β signaling pathway are associated with survival in NSCLC patients treated with EGFR-TKI, especially in the patients with EGFR mutation.
Materials and Methods This study was approved by the Review Board of Tongji Hospital. Written informed consent from each patient for the use of their DNA and clinical information was obtained. . This study included 206 patients with advanced lung adenocarcinoma. Patients were recruited between January 2009 and December 2011 at Tongji Hospital, Huazhong University of Science and Technology (Wuhan, Hubei Province). Eligible and exclusion criteria described in Supplement S1. The EGFR mutations were detected by DNA sequencing (Supplement S1), Amplification Refractory Mutation System (ARMS) or Mutant-Enriched-Liquidchip (MEL) (Table 1) [14]. This study was approved by Ethics Committee of Tongji Hospital Medical, and registered with National Natural Science Foundation of China (Grant No. 81001035).
Treatment and Follow-up Among 206 patients, 143 patients were treated with oral gefitinib between January 2009 and December 2011. Survival information was collected every 2 months from the specific doctor’s follow-up records; and all the patients were asked to return to the hospital for examination [15]. Free gefitinib program (Iressa Means-Tested Drug Donation) is supported by China Charity Federation (www. Iressaccf.org), The patients needed pay for the treatment in the first six month, than they received the free drugs until disease progression or death. PFS was defined as the time from first dose of gefitinib to the date of obvious disease progression or death due to any causes. 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 reports. Living Patients were censored at the last date they were known to be alive based on the date of the last contact.
Downloaded from http://annonc.oxfordjournals.org/ at Memorial University of Newfoundland, Health Sciences Library on July 18, 2014
Patients
5
Polymorphism selection and genotyping In this study, a total of 33 SNPs in 13 candidate genes of the TGF-β signaling pathway were selected based on the following criteria (Supplement S2). All the SNPs were either located in the promoter untranslated region or coding region of the gene or those that had been previously reported as being associated with survival, lung cancer, or metastasis
a pairwise mode and all correlated alleles were captured at r2 > 0.8 [16]. In this study, the genes (number of SNPs) were including TGF-β1(3), TGFβR1(3), BMP1(2), BMP2(1), BMP4(2), SMAD1(1) , SMAD3(7), SMAD4(6), SMAD6(3), SMAD7(1), SMAD8(2), VEGF(4), CDH(1), INHBC(1), ACVR2A(1) [16,17] (for details see Supplement S2).
Dual luciferase reporter assay and Western blot We used dual luciferase reporter assay and Western Blot to analyze the function of SMAD3 rs11632964, which was significantly associated with survival time of patients. Detailed dual luciferase reporter assay and western blot analysis were described in Supplement S1.
Statistical analysis The primary endpoint of the study was OS and the second was PFS. Pearson’s χ2 test was used to calculate the difference of patient clinical characteristics. The genotype affect on patient’s survival was estimated with the Kaplan-Meier method and compared with the log-rank tests. Multivariate Cox proportional hazards models were applied to estimate the effect of prognostic factors on OS and PFS, using proverbial clinical factors, including age, sex, smoking status, EGFR mutation status and clinical stages. Hazard ratios (HRs) and their 95% confidence intervals (CIs) of were computed by Cox model [18]. Statistical significance was set at a level of 0.05 and checked by Bonferroni correction. All the analyses were performed using the SPSS software package (version 16.0, SPSS Inc., Chicago, IL).
Downloaded from http://annonc.oxfordjournals.org/ at Memorial University of Newfoundland, Health Sciences Library on July 18, 2014
(Table 3). All of them were selected using HaploView software with HapMap HCB data in
6
Results Patient characteristics and clinical outcomes The distribution of demographic and clinical characteristics of patients is showed in Table 1. All the patients were diagnosed with lung adenocarcinoma. EGFR mutations were detected in 106 patients (51.5%). Del 19 (47 tumors) and L858R mutations (43 tumors)
(16 tumors). All these patients as well as 37 patients with wild type EGFR received gefitinib treatment. The median follow-up time of the all patients was 18.3 months at the final analysis (November 2012). The median age of patients was 55.8 years (range, 32.0-80.0 years); 47.6% were male. There were no significant difference between those patients who were alive and those who had died with respect to age (P=0.53), gender (P=0.19), smoking status (P=0.92), ECOG (P=0.32), alcohol use (P=0.19), difference of
mutation (P=0.14) and weight loss at the time of diagnosis (P=0.23). Using UICC TNM staging manual (7th Edition)[19], twenty eight (19.6%) had stage IIIB disease and 115 (80.4%) had stage IV disease. In this cohort, ninety one patients (63.6%) died and the median OS time was 16.8 months (range, 1.2-45.7 months). The median PFS time was 15.9 months (range, 0.7-36.5 months).
Individual SNPs and survival The association of all SNPs with OS and PFS is summarized in supplementary Table 2-5. We identified 2 SNPs with a P<0.05 and 1 SNP with a P<0.01, and all of them belonged to SMAD3 (Table 2). The most significant association in this study was found to be with SMAD3: rs11632964.
Among the whole group patients, wild type SMAD3: rs11632964
was associated with an increased risk of death and disease progression. The patients with SMAD3: rs11632964 had a decreased median survival time (MST) of 15.9 months compared with 30.8 months for those with variant allele (Log-rank P=0.037, Figure 2A). Similar result was also found in PFS. Patients carrying variant allele had a longer PFS (13.2 months vs. 9.2 months) compared to those with the wild type (Figure 2B). In the patients with EGFR mutations, the efficacy of SMAD3: rs11632964 was found to be more obvious. The differences of OS and PFS in the patients with wild type versus variant
Downloaded from http://annonc.oxfordjournals.org/ at Memorial University of Newfoundland, Health Sciences Library on July 18, 2014
were most common; other mutations included missense mutations in exon 18, 19 and 21
7
genotypes of SMAD3: rs11632964 were more significant. Patients with variant allele also had much longer MST (30.8 months vs. 17.5 months; Log-rank P=0.020) and PFS (20.8 months vs. 9.4 months; Log-rank P=0.001) than those with wild type (Figure 2C, 2D).
SMAD3: rs11071938 variant genotype was also significantly associated with better PFS
patients with the EGFR mutations, SMAD3: rs11071938 variant genotype was associated better PFS (14.2months vs. 9.9 months; HR=1.75, 95%CI=1.06-2.89, P=0.029) (Figure 2E).
Different from the two SNPs described above, it was the wild type genotype of SMAD3: rs6494633 that was associated with improved PFS in EGFR mutated patients. These individuals had more than two months disease control advantage with PFS than those with variant genotype (11.3 months vs. 8.6 months, Log-rank P=0.048). (Figure 2F). The remaining 30 selected SNPs showed no associations between their genotype and the survival time (Table 3, Supplementary S3-S6).
Functional evaluation of the SMAD3 rs11632964 Polymorphism SMAD3 protein and pi-SMAD3 expression in PC-9 cells and PC9/G cells were detected firstly. We found that there was significantly decreased smad3 as well as pi-smad3 in the Gefitinib resistant PC-9/G cells compared with wild type PC-9 (Figure 3A,3B). We then examined the regulator role of SMAD3 rs11632964 SNP in these two cell lines. Relative luciferase expression assays indicated that the 651-bp intron 1 region which we cloned in either pGL3b-rs11632964T construct or pGL3b-rs11632964C construct showed promoter activity. This demonstrates that there might be an intronic promoter region around rs11632964 polymorphism. We next examined whether the survival-related SNP rs11632964 has an allele-specific effect on the intronic promoter activity. PC9/G cells transfected with SMAD3 rs11632964T allelic reporter construct showed significantly lower luciferase activities compared to cells expression C allelic reporter construct (both
Downloaded from http://annonc.oxfordjournals.org/ at Memorial University of Newfoundland, Health Sciences Library on July 18, 2014
(95%CI=1.01-2.31, P=0.045) in the whole cohort (11.8 months vs. 9.2 months); and the
8
P<0.05). However, no such difference was observed in PC9 cells (Figure 3C).
Discussion Currently, the detection of EGFR-TKI sensitive mutations are essential for selecting the treatments for the patients with advanced NSCLC since these mutations have been
our previous studies suggested that the polymorphisms of EGFR signaling pathway might be prognostic markers of OS in advanced lung adenocarcinoma patients who were treated with gefitinib [21]. But we did not find any significant correlation between EGFR SNPs and clinical outcome in the patients carrying EGFR mutations. In the current study, we systematically evaluated genetic variations in 33 SNPs from13 candidate genes and treatment outcome in the patients with advanced lung adenocarcinomas. Among the SNPs we have analyzed, the genetic variations in TGF-β pathway were found to be highly associated with outcome, especially in those with EGFR mutations. To our knowledge, our study is the first to link TGF-β signaling pathway SNPs with lung cancer outcome for in patients with EGFR mutant adenocarcinomas.
The variations of genes in TGF-β pathway are detected in several human cancers. TGF-β signaling pathway possesses a tumor suppressor function [22]. TGF-β and its signaling effectors act as the key determinants of carcinoma cell behavior.[23]. There is evidence to suggest that Smad3 plays a major role in TGF-β-mediated growth inhibition. First, Smad3 works in TGF-β–induced antiproliferative transcriptional responses. Second, Smad3 works in TGF-β–mediated proapoptotic transcriptional responses. [24]. On the other side, TGF-β promotes tumor progression during the late stages of cancer by suppressing immune surveillance, inducing EMT, and enhancing cell migration and transcription of factors favorable to metastasis. Smad3 is directly involved in EMT [25]. Changes of EMT status may underlie acquired resistance of gefitinib[26]. Sordella et al found erlotinib-resistant cells H1650-M3 were characterized by an up-regulation of TGF-β mediated signaling and increased levels of Smad3 phosphorylation in H1650-M3 cells compared with H1650 cells. Therefore, Smad3 functions as both a negative and positive
Downloaded from http://annonc.oxfordjournals.org/ at Memorial University of Newfoundland, Health Sciences Library on July 18, 2014
shown to be predictive of drug sensitivity. In addition to the well-known EGFR mutations,
9
regulator of carcinogenesis and cancer progression depending on cell type and clinical stage of the tumor [27].
In the main analysis, SMAD3: rs11632964, showed strong association with treatment outcome. It is possible that the variant allele of SMAD3: rs11632964 affect gene
suggested gefitinib-resistant PC9/G cells transfected with SMAD3 rs11632964T allelic reporter construct showed significantly lower luciferase activities compared to cells expression C allelic reporter construct and no such difference was observed in PC9 cells. We also found that there was significantly decreased smad3 as well as pi-smad3 in PC-9/G cells compared with wild type PC-9. The results of two functional tests were consistent and demonstrated that individuals with decreased expression of SMAD3 caused by the rs11632964T allele might contribute to shortened survival time after gefitinib treatment.
Our study has a complete clinical dataset with EGFR mutation status determined in all the patients. Although function is not known for all gene variations, this study suggests biological implications of these genetic variations in treatment selection and outcome. This study will help us choose the patients who can get more clinical benefit from TKI therapy except EGFR mutation. It has been reported that a haplotype of SMAD3 is associated with overall survival of NSCLC patients, which is the support of our results [16]. In conclusion, our data show that genetic variations in TGF-β pathway are the potential predictors of outcome in advanced lung adenocarcinoma with EGFR mutations. Analysis of these genetic variations, in addition to EGFR mutation detection, may refine the drug selection process for the treatment of advanced lung adenocarcinoms, improve the prediction of the treatment outcome. However, as a single institutional study, our patient number was relatively small. We need a larger patient population and better mapping and functional assays to validate the results.
Acknowledgements: The authors are thankful to Dr. Shiyu Song and Dr. Sheng Wei, for
Downloaded from http://annonc.oxfordjournals.org/ at Memorial University of Newfoundland, Health Sciences Library on July 18, 2014
transcription thus altering protein level [16]. Our functional test in lung cancer cell lines
10
reviewing the paper and analyzing the data. This work was supported in part by National Natural Science Foundation of China (no. 81001035; no. 81200498; no. 31271382; no. 81201586; no. 81071832). Disclosure: The authors have declared no conflicts of interest.
References Downloaded from http://annonc.oxfordjournals.org/ at Memorial University of Newfoundland, Health Sciences Library on July 18, 2014
1. Ciardiello F, De Vita F, Orditura M, Tortora G. The role of EGFR inhibitors in non small cell lung cancer. Curr Opin Oncol. 2004; 16: 130–135. 2. Fukuoka M, Wu YL, Thongprasert S, et al. Biomarker analyses and final overall survival results from a phase III, randomized, open-label, first-line study of gefitinib versus carboplatin/paclitaxel in clinically selected patients with advanced non-small-cell lung cancer in Asia (IPASS). J Clin Oncol. 2011; 29: 2866-2874. 3. Mitsudomi T, Morita S, Yatabe Y, et al. Gefitinib versus cisplatin plus docetaxel in patients with non-small-cell lung cancer harbouring mutations of the epidermal growth factor receptor (WJTOG3405): an open label, randomized phase 3 trial. Lancet Oncol. 2010; 11: 121-128. 4. Zhou C, Wu YL, Chen G, et al. Erlotinib versus chemotherapy as first-line treatment for patients with advanced EGFR mutation-positive non-small-cell lung cancer (OPTIMAL, CTONG-0802): a multicentre, open-label, randomised, phase 3 study. Lancet Oncol. 2011 ;12:735-742. 5. Kobayashi S, Boggon TJ, Dayaram T, et al. EGFR mutation and resistance of non-small-cell lung cancer to Gefitinib. N Engl J Med. 2005; 352: 786–792. 6. Engelman JA, Zejnullahu K, Mitsudomi T, et al., MET amplification leads to gefitinib resistance in lung cancer by activating ERBB3 signaling. Science. 2007; 316:1039-1043. 7. Yao Z, Fenoglio S, Gao DC, et al. TGF-beta IL-6 axis mediates selective and adaptive mechanisms of resistance to molecular targeted therapy in lung cancer. Proc Natl Acad Sci U S A. 2010; 107: 15535-15540. 8. Zhang Y.E. Non-Smad pathways in TGF-beta signaling. Cell Res, 2009; 19, 128-39. 9. Xu Y, Baker D, Quan T, et al. Receptor type protein tyrosine phosphatase-kappa mediates cross-talk between transforming growth factor-betaand epidermal growth factor receptor signaling pathways in human keratinocytes. Mol Biol Cell 2010; 21: 29-35. 10. Uramoto H, Yamada T, Yano S, et al. Prognostic value of acquired resistance-related molecules in Japanese patients with NSCLC treated with an EGFR-TKI. Anticancer Res 2012; 32: 3785-3790. 11. Chung JH, Jin KR, Xu XH, et al. Clinical and molecular evidences of epithelial to mesenchymal transition in acquired resistance to EGFR-TKIs. Lung Cancer 2011; 73: 176-182. 12. Ng KP, Hillmer AM, Chuah CT, et al. A common BIM deletion polymorphism mediates intrinsic resistance and inferior responses to tyrosine kinase inhibitors in cancer. Nat Med 2012; 18: 521-528. 13. Caroline M and Ying E. Zhang. Roles of Smad3 in TGF-β Signaling During
11
Carcinogenesis. Crit Rev Eukaryot Gene Expr. 2007; 17:281–293.
Downloaded from http://annonc.oxfordjournals.org/ at Memorial University of Newfoundland, Health Sciences Library on July 18, 2014
14. Takuya A, Hideaki Y, Kimihiro S, et al. Review of the Treatment of Non-Small Cell Lung Cancer with Gefitinib. Clin Med Insights Oncol. 2012; 6: 407-421. 15. Zhang L, Yang M, Bi N, et al. Association of TGF-β1 and XPD polymorphisms with severe acute radiation-induced esophageal toxicity in locally advanced lung cancer patients treated with radiotherapy. Radiother Oncol. 2010; 97: 19-25. 16. Lin M, Stewart DJ, Spitz MR, et al. Genetic variations in the transforming growth factor-beta pathway as predictors of survival in advanced non-small cell lung cancer. Carcinogenesis. 2011; 32: 1050-1056. 17. Li Q, Wu H, Chen B, et al. SNPs in the TGF-β signaling pathway are associated with increased risk of brain metastasis in patients with Non-Samll-Cell lung cancer. PloS One. 2012; 7: e51713. 18. Yang M, Zhang L, Bi N, et al. Association of P53 and ATM polymorphisms with risk of radiation-induced pneumonitis in lung cancer patients treated with radiotherapy. Int J Radiat Oncol Biol Phys. 2011; 79: 1402-1407. 19. TNM Classification of Malignant Tumours. Seventh Edition. Wilry-Blackwell: The Union for International Cancer Control (UICC), 2009; 138. 20. Zhang X, Wei J, Zhou L, et al. A functional BRCA1 coding sequence genetic variant contributes to risk of esophageal squamous cell carcinoma. Carcinogenesis. 2013;34:2309-2313. 21. Zhang L, Yuan X, Chen Y, et al. Role of EGFR SNPs in survival of advanced lung adenocarcinoma patients treated with Gefitinib. Gene. 2013; 517: 60-64. 22. Halder SK, Beauchamp RD, Datta PK. Smad7 induces tumorigenicity by blocking TGF-beta-induced growth inhibition and apoptosis. Exp Cell Res. 2005; 307: 231–246. 23. Siegel PM, Massague J. Cytostatic and apoptotic actions of TGF-β in homeostasis and cancer. Nat Rev Cancer. 2003; 3: 807–821. 24. Caroline M and Ying E. Zhang. Roles of Smad3 in TGF-β signaling during carcinogenesis. Crit Rev Eukaryot Gene Expr. 2007; 17:281–293. 25. Saika S, Kono-Saika S, Ohnishi Y, et al. Smad3 signaling is required for epithelial-mesenchymal transition of lens epithelium after injury. Am J Pathol. 2004; 164: 651–663. 26. Uramoto H, Shimokawa H, Hanagiri T, et al. Expression of selected gene for acquired drug resistance to EGFR-TKI in lung adenocarcinoma. Lung Cancer. 2011; 73: 361-365. 27. Masszi A, Kapus A. Smaddening complexity: the role of Smad3 in Epithelial-Myofibroblast transition. Cells Tissues Organs. 2011; 193: 41-52.
12
Downloaded from http://annonc.oxfordjournals.org/ at Memorial University of Newfoundland, Health Sciences Library on July 18, 2014
Fig 1. Patients flow diagram. Starting with advanced lung adenocarcinoma with EGFR mutation status treated in Tongji Hospital during 2009 to 2011.
13
Downloaded from http://annonc.oxfordjournals.org/ at Memorial University of Newfoundland, Health Sciences Library on July 18, 2014
Fig2. Kaplan-Meier survival curves for adenocarcinoma patients according to SMAD3: rs11632964 SNP. (A) Overall survival in the total population; (B) Progression-free survival in the total population; (C) Overall survival in patients with EGFR mutation; (D) Progression-free survival in patients with EGFR mutation. (E) SMAD3:rs11071938 Progression-free survival in patients with EGFR mutation; (F) SMAD3:rs6494633 Progression-free survival in patients with EGFR mutation.
14
A
B
Fig 3. T, cells transfected with pGL3b-rs11632964T construct; C, cells transfected with pGL3b-rs11632964C construct. Smad3 and pi-smad3 protein expression in PC9 cells and PC9/G cells(A,B). Upper panel, Smad3 expression in PC9 cells and PC9/G cells; lower panel, pi-smad3 expression in in PC9 cells and PC9/G cells. Transient luciferase reporter gene expression assays with constructs containing different alleles of SMAD3 intron 1 region in PC9 and gefitinib-resistant PC9/G lung cancer cells (C).
Downloaded from http://annonc.oxfordjournals.org/ at Memorial University of Newfoundland, Health Sciences Library on July 18, 2014
C
Table 1. Demographic and base line clinical characteristics of patients (N=143) Variable
Dead
P-value
57.7 (9.20)
54.7 (10.11)
0.53
21 (40.4) 31 (59.6)
47 (51.6) 44 (48.4)
0.19
17 (32.7) 35 (67.3)
29 (31.9) 62 (68.1)
0.92
12 (23.1) 40 (76.9)
16 (17.6) 75 (82.4)
0.43
38 (73.1) 14 (26.9)
73 (80.2) 18 (19.8)
0.32
16 (30.8) 36 (69.2)
38 (41.8) 53 (58.2)
0.19
33 (63.5) 7 (13.5) 10 (19.2) 2 (3.8)
72 (79.1) 8 (8.8) 9 (9.9) 2 (2.2)
0.23
42 (80.8) 10 (19.2)
64 (70.3) 27 (29.7)
<0.001
18 (46.2) 12 (30.8) 9 (23.0)
29 (43.3) 31 (46.3) 7 (10.4)
0.14
NOTE. Multivariate analyses were adjusted for all factors listed in Table.
Downloaded from http://annonc.oxfordjournals.org/ at Memorial University of Newfoundland, Health Sciences Library on July 18, 2014
Age, mean(SD), years Sex Male Female Smoking Status, n (%) smoker Non smoker Stage (TNM) IIIB IV ECOG, n (%) 0-1 ≥2 Alcohol, n (%) used Non Weight loss, n (%) Weight gain or stable 1-5% 5-10% >10% EGFR Status, n (%) Positive Negative EGFR mutations, n (%) Del 19 L858R others
Alive
Table 2. Association between Smad3 genotype and survival SNPs and genotypes Smad3: Rs11632964 CC CT Smad3: Rs11071938 CC CT Smad3: Rs6494633 CC CT
No of patients (all) (%) 143 120 (83.9) 23 (16.1) 143 98 (68.5) 45 (31.5) 143 113 (79.0) 30 (21.0)
No of patients (EGFR mutation) (%) 106 85 (80.2) 21 (19.8) 106 70 (66.0) 36 (34.0) 106 81 (76.4) 25 (23.6)
Median survival of all Patients (N=143) a OS P PFSa P
Median survival of patients with EGFR mutation (N=106) OSa P PFSa P
15.9 30.8
0.063
17.5 19.4
0.045
17.5 30.8
0.096
21.0 18.2
16.5 19.4 17.5 9.9
0.037
9.2 13.2
0.062
9.2 11.8
0.148
9.8 8.1
0.154
9.4 20.8
0.001
0.020
9.9 14.2
0.029
0.059
11.3 8.6
0.048
a
Adjusted for age, sex, Smoking status, Clinical stage, ECOG and EGFR mutation status. Results in bold are significant at P < 0.002 (Bonferroni correction). Abbreviations: EGFR, epidermal growth factor receptor; SNP, single nucleotide polymorphism; No, number of patients.
Downloaded from http://annonc.oxfordjournals.org/ at Memorial University of Newfoundland, Health Sciences Library on July 18, 2014
Table 3. Association between mutated-type genotype with OS and PFS for all patients and patients with EGFR mutation
TGF-β1 TGF-β1 TGF-β1 TGFβR1 TGFβR1 TGFβR1
BMP1 BMP1 BMP2 BMP4 BMP4 SMAD1 SMAD3 SMAD3 SMAD3 SMAD3 SMAD3 SMAD3 SMAD3 SMAD4 SMAD4
OS of all Patients (N=143) P HRa
SNP
Rs18004699 Rs4803455 Rs1800470 Rs64789974 Rs10733710 Rs10819638 Rs3857979 Rs783891 Rs235756 Rs8014071 Rs17563 Rs11939979 Rs11632964 Rs11071938 Rs6494633 Rs4776342 Rs4776343 Rs750766 Rs12102171 Rs7244227 Rs10502913
0.98 (0.63-1.53) 0.87 (0.59-1.35) 0.96 (0.61-1.49) 0.92 (0.59-1.44) 1.31 (0.84-2.07) 1.07 (0.68-1.67) 1.16 (0.75-1.80) 1.06 (0.69-1.63) 1.02 (0.62-1.71) 0.87 (0.56-1.36) 0.90 (0.59-1.37) 1.08 (0.69-1.70) 1.95 (1.04-3.67) 1.57 (0.98-2.52) 0.69 (0.42-1.14) 0.96 (0.62-1.50) 1.39 (0.65-2.97) 0.93 (0.61-1.44) 1.07 (0.70-1.66) 0.95 (0.58-1.54) 0.85 (0.54-1.33)
0.926 0.543 0.839 0.708 0.236 0.774 0.508 0.788 0.927 0.551 0.620 0.731 0.037 0.062 0.148 0.866 0.403 0.758 0.748 0.831 0.484
PFS of all patients (N=143) P HRa
0.96 (0.64-1.42) 0.96 (0.65-1.40) 1.03 (0.70-1.52) 1.04 (0.71-1.52) 0.96 (0.65-1.41) 1.30 (0.89-1.90) 0.91 (0.62-1.33) 0.89 (0.61-1.30) 0.93 (0.60-1.45) 0.93 (0.63-1.38) 1.22 (0.84-1.76) 1.08 (0.73-1.59) 1.67 (0.97-2.85) 1.53 (1.01-2.31) 0.68 (0.43-1.07) 0.90 (0.59-1.37) 1.30 (0.69-2.46) 1.03 (0.70-1.51) 1.20 (0.82-1.77) 1.11 (0.73-1.70) 0.94 (0.64-1.39)
0.819 0.815 0.881 0.855 0.826 0.182 0.626 0.540 0.757 0.731 0.290 0.706 0.063 0.045 0.096 0.583 0.424 0.893 0.348 0.625 0.750
OS for patients with EGFR mutation (N=106) P HRa
PFS for patients with EGFR mutation (N=106) P HRa
0.85 (0.50-1.45) 1.48 (0.67-3.27) 0.85 (0.49-1.47) 0.83 (0.48-1.41) 1.24 (0.74-2.09) 0.99 (0.58-1.67) 0.92 (0.55-1.54) 0.91 (0.55-1.51) 1.18 (0.62-2.26) 0.94 (0.55-1.59) 0.92 (0.56-1.51) 1.19 (0.70-2.01) 2.38 (1.15-4.94) 1.50 (0.86-2.63) 0.58 (0.33-1.02) 0.96 (0.57-1.62) 1.30 (0.50-3.36) 0.96 (0.57-1.60) 0.90 (0.54-1.51) 0.98 (0.55-1.74) 1.05 (0.62-1.76)
0.84 (0.53-1.35) 1.33 (0.70-2.53) 0.94 (0.59-1.50) 0.90 (0.57-1.41) 1.07 (0.68-1.69) 1.13 (0.71-1.80) 0.69 (0.44-1.08) 0.69 (0.44-1.08) 0.99 (0.56-1.75) 0.87 (0.54-1.38) 1.16 (0.75-1.78) 1.29 (0.82-2.03) 3.01 (1.54-5.86) 1.75 (1.06-2.89) 0.55 (0.37-1.00) 0.90 (0.56-1.44) 1.06 (0.49-2.29) 1.01 (0.64-1.60) 1.00 (0.62-1.59) 1.12 (0.68-1.86) 1.09 (0.69-1.72)
0.555 0.332 0.566 0.483 0.419 0.963 0.761 0.712 0.622 0.809 0.727 0.524 0.020 0.154 0.059 0.877 0.587 0.861 0.689 0.944 0.871
Downloaded from http://annonc.oxfordjournals.org/ at Memorial University of Newfoundland, Health Sciences Library on July 18, 2014
Gene
0.470 0.379 0.792 0.639 0.770 0.600 0.100 0.108 0.982 0.541 0.499 0.278 0.001 0.029 0.048 0.652 0.885 0.951 0.987 0.651 0.724
SMAD4 SMAD4 SMAD4 SMAD4 SMAD6 SMAD6 SMAD6 SMAD7 SMAD8 SMAD8 INHBC ACVR2A
Rs12455792 Rs12456284 Rs12958604 Rs948588 Rs12906898 Rs12913975 Rs4776318 Rs7227023 Rs511674 Rs7333607 Rs4760259 Rs1424954
1.27 (0.83-1.94) 1.27 (0.83-1.94) 1.10 (0.70-1.71) 1.02 (0.52-2.00) 1.02 (0.58-1.80) 0.85 (0.55-1.31) 1.03 (0.64-1.67) 2.98 (0.40-22.2) 0.88 (0.58-1.34) 1.16 (0.71-1.91) 0.87 (0.57-1.35) 1.17 (0.73-1.87)
0.281 0.281 0.690 0.958 0.941 0.451 0.907 0.287 0.553 0.556 0.543 0.512
1.22 (0.85-1.76) 1.22 (0.85-1.76) 1.05 (0.71-1.54) 1.53 (0.78-3.00) 1.14 (0.69-1.89) 0.93 (0.63-1.38) 1.17 (0.76-1.81) 1.30 (0.39-4.34) 1.00 (0.69-1.45) 1.27 (0.82-1.97) 0.95 (0.65-1.40) 1.17 (0.77-1.77)
0.284 0.284 0.825 0.212 0.613 0.733 0.468 0.668 0.995 0.279 0.804 0.460
1.13 (0.68-1.89) 1.13 (0.68-1.89) 1.07 (0.62-1.83) 0.77 (0.35-1.70) 0.80 (0.41-1.56) 0.76 (0.46-1.25) 0.88 (0.49-1.59) 4.50 (0.60-33.55) 0.69 (0.42-1.13) 1.46 (0.79-2.70) 1.01 (0.60-1.69) 1.37 (0.80-2.36)
0.641 0.641 0.821 0.514 0.515 0.281 0.672 0.143 0.142 0.233 0.968 0.256
1.23 (0.80-1.89) 1.23 (0.80-1.89) 1.11 (0.70-1.77) 1.43 (0.63-3.25) 0.91 (0.51-1.63) 0.86 (0.55-1.34) 1.11 (0.66-1.86) 2.28 (0.67-7.73) 0.83 (0.54-1.27) 1.58 (0.93-2.69) 1.09 (0.70-1.70) 1.19 (0.74-1.92)
a
0.353 0.353 0.663 0.397 0.748 0.497 0.703 0.185 0.386 0.093 0.715 0.480
Adjusted for age, sex, Smoking status, Clinical stage, ECOG and EGFR mutation status. Results in bold are significant at P < 0.002 (Bonferroni correction). Abbreviations: HR, hazard ratio; CI, confidence interval
Downloaded from http://annonc.oxfordjournals.org/ at Memorial University of Newfoundland, Health Sciences Library on July 18, 2014
SNPs in the transforming growth factor beta pathway as predictors of outcome in advanced lung adenocarcinoma with EGFR mutations treated with Gefitinib
L Zhang1, Q.X. Li1, H.L. Wu1, X Lu2, M Yang3, S.Y. Yu1, X.L. Yuan1,
*
Department of Oncology, Tongji Hospital, Huazhong University of Science and
Technology, Wuhan, Hubei Province, China 2
Department of Transplantation, Tongji Hospital, Huazhong University of Science and
Technology, Wuhan, Hubei Province, China 3
College of Life Science and Technology, Beijing University of Chemical Technology,
Beijing, China *
Corresponding author: Dr. X.L. Yuan, Department of Oncology, Tongji Hospital,
Huazhong University of Science and Technology, Wuhan 430030, China. Tel: 86-027-83663406; Fax:86-027-83662834; E-mail: [email protected]
Key Message: "The manuscript describes the original research investigating the use of genetic variations to predict response of lung adenocarcinoma to tyrosine kinase inhibitor (TKI) treatment. We found that the genetic variations in the TGF‐β signaling pathway are associated with survival in NSCLC patients treated with EGFR‐TKI, especially in the patients with EGFR mutation. Our study also showed that other genetic factors are associated with acquired resistance to EGFR‐TKI using systematic genetic analysis. Our study was base on a complete clinical dataset with EGFR mutation status determined in all the patients and the genetic variations determined by analyzing SNPs of candidate genes involved in TGF‐β pathways. We used dual luciferase reporter assay and Western Blot to analyze the function of SMAD3 rs11632964, which was significantly associated with survival time of patients.Our findings will help to select patients who can benefit from TKI therapy."
© The Author 2014. Published by Oxford University Press on behalf of the European Society for Medical Oncology. All rights reserved. For permissions, please email: [email protected].
Downloaded from http://annonc.oxfordjournals.org/ at Memorial University of Newfoundland, Health Sciences Library on July 18, 2014
1
2
Abstract Background: The aim of this study was to evaluate whether genetic variations in the transforming growth factor beta (TGF-β) pathway influenced clinical outcome of advanced lung ademocarcinoma with EGFR mutations treated with gefitinib.
were enrolled in this study. EGFR mutation in these tumors was detected. Among them 106 patients with EGFR mutation and 37 of 100 patients with wild type were treated with gefitinib. Genotype of 33 SNPs from 13 genes involved in the TGF-β signaling pathway was determined and their association with survival time was analyzed. Univariate and multivariate analyses were performed to assess the role of biological/clinical parameters in progression-free survival (PFS) and overall survival (OS) using Pearson’s χ2 test, log-rank test and Cox proportional hazards model.
Results: Among SNPs analyzed, multivariate analysis showed the CT genotype of SMAD3: rs11632964 was associated with a longer OS and PFS when the entire cohort of 143 patients were included; the association was significant in the patients with EGFR mutant tumors (30.8 months vs. 17.5 months; Log-rank P=0.020; and 20.8 months vs. 9.4 months; Log-rank P=0.001), as compared with patients with wild-type EGFR tumors. In patients with mutant EGFR, the CT genotype of SMAD3: rs11071938 and the CC genotype of SMAD3: rs6494633 were also found to be associated with better PFS. Dual luciferase reporter assays showed gefitinib-resistant PC9/G cells transfeted with SMAD3: rs11632964T allelic reporter construct showed significantly lower luciferase activities compared to cells expression C allelic reporter construct. There was significantly decreased expression of SMAD3 and pi-SMAD3 in the PC-9/G cells compared with PC-9.
Conclusions: Among the candidate genes involved in the TGF-β pathway, the polymorphisms of SMAD3 appear to be highly predictive of outcome of patients with lung ademocarcinoma after gefitinib treatment, especially in those with EGFR mutations.
Key words: Polymorphisms; TGF-β; EGFR-TKI; SMAD3; Survival
Downloaded from http://annonc.oxfordjournals.org/ at Memorial University of Newfoundland, Health Sciences Library on July 18, 2014
Patients and methods: Two hundred six patients with advanced lung adenocarcinomas
3
Introduction The overall prognosis of patients with non-molecularly selected advanced non-small cell lung cancer (NSCLC) is poor. In East Asia, the epidermal growth factor receptor (EGFR) tyrosine kinase inhibitors (TKI) gefitinib is widely used [1]. In clinical trials, dramatic responses have been observed soon after initiation of treatment, particularly in Asians,
activating EGFR mutations, gefitinib could significantly prolong progression-free survival (PFS) compared to cytotoxic chemotherapy [3]. However, drug resistance eventually develops in most patients despite the initial good response, with a median PFS of about 1 year [4].
Resistance to EGFR-TKI has been associated with the acquisition of EGFR T790M mutation) [5] and overexpression and/or activation of c-Met [6, 7]. There are also other potential genetic factors associated with EGFR gene mutations that may be useful to predict the likelihood of response or resistance to EGFR-TKI.
In contrast with EGFR signaling pathway, the transforming growth factor-beta (TGF-β) signaling pathway is strongly anti-proliferative, which plays an important role in tumorigenesis，cancer progression and metastasis via crosstalk with Smad and EGFR signaling pathways, like P13K/AKT and RPTP-к [8,9]. To our knowledge, TGF-β signaling pathway has two major mechanisms to affect the outcome of EGFR-TKI. First, epithelial-mesenchymal transition (EMT) is one of the important mechanisms of TKI resistant, and changes in EMT status can explain resistance to EGFR-TKI [10,11]. Receptor Smads (Smad2 and particularly Smad3), the direct targets of the activated TGF-β receptor have been implicated as critical mediators in EMT [11]. Second, Bcl-2L11 (BIM) associated with TKI resistance [12], and one target of Smad3 in inducing apoptosis is Bcl-2, a key antiapoptotic inhibitor. TGF-β/Smad3 signaling facilitates a permissive cellular context that is tending to promote apoptosis by attenuating the level of the major apoptosis inhibitor Bcl-2 [13].
Downloaded from http://annonc.oxfordjournals.org/ at Memorial University of Newfoundland, Health Sciences Library on July 18, 2014
women, nonsmokers, and adenocarcinoma cases [2]. For patients with tumors harboring
4
The purpose of our study is to determine whether genetic variations in the TGF-β signaling pathway are associated with survival in NSCLC patients treated with EGFR-TKI, especially in the patients with EGFR mutation.
Materials and Methods This study was approved by the Review Board of Tongji Hospital. Written informed consent from each patient for the use of their DNA and clinical information was obtained. . This study included 206 patients with advanced lung adenocarcinoma. Patients were recruited between January 2009 and December 2011 at Tongji Hospital, Huazhong University of Science and Technology (Wuhan, Hubei Province). Eligible and exclusion criteria described in Supplement S1. The EGFR mutations were detected by DNA sequencing (Supplement S1), Amplification Refractory Mutation System (ARMS) or Mutant-Enriched-Liquidchip (MEL) (Table 1) [14]. This study was approved by Ethics Committee of Tongji Hospital Medical, and registered with National Natural Science Foundation of China (Grant No. 81001035).
Treatment and Follow-up Among 206 patients, 143 patients were treated with oral gefitinib between January 2009 and December 2011. Survival information was collected every 2 months from the specific doctor’s follow-up records; and all the patients were asked to return to the hospital for examination [15]. Free gefitinib program (Iressa Means-Tested Drug Donation) is supported by China Charity Federation (www. Iressaccf.org), The patients needed pay for the treatment in the first six month, than they received the free drugs until disease progression or death. PFS was defined as the time from first dose of gefitinib to the date of obvious disease progression or death due to any causes. 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 reports. Living Patients were censored at the last date they were known to be alive based on the date of the last contact.
Downloaded from http://annonc.oxfordjournals.org/ at Memorial University of Newfoundland, Health Sciences Library on July 18, 2014
Patients
5
Polymorphism selection and genotyping In this study, a total of 33 SNPs in 13 candidate genes of the TGF-β signaling pathway were selected based on the following criteria (Supplement S2). All the SNPs were either located in the promoter untranslated region or coding region of the gene or those that had been previously reported as being associated with survival, lung cancer, or metastasis
a pairwise mode and all correlated alleles were captured at r2 > 0.8 [16]. In this study, the genes (number of SNPs) were including TGF-β1(3), TGFβR1(3), BMP1(2), BMP2(1), BMP4(2), SMAD1(1) , SMAD3(7), SMAD4(6), SMAD6(3), SMAD7(1), SMAD8(2), VEGF(4), CDH(1), INHBC(1), ACVR2A(1) [16,17] (for details see Supplement S2).
Dual luciferase reporter assay and Western blot We used dual luciferase reporter assay and Western Blot to analyze the function of SMAD3 rs11632964, which was significantly associated with survival time of patients. Detailed dual luciferase reporter assay and western blot analysis were described in Supplement S1.
Statistical analysis The primary endpoint of the study was OS and the second was PFS. Pearson’s χ2 test was used to calculate the difference of patient clinical characteristics. The genotype affect on patient’s survival was estimated with the Kaplan-Meier method and compared with the log-rank tests. Multivariate Cox proportional hazards models were applied to estimate the effect of prognostic factors on OS and PFS, using proverbial clinical factors, including age, sex, smoking status, EGFR mutation status and clinical stages. Hazard ratios (HRs) and their 95% confidence intervals (CIs) of were computed by Cox model [18]. Statistical significance was set at a level of 0.05 and checked by Bonferroni correction. All the analyses were performed using the SPSS software package (version 16.0, SPSS Inc., Chicago, IL).
Downloaded from http://annonc.oxfordjournals.org/ at Memorial University of Newfoundland, Health Sciences Library on July 18, 2014
(Table 3). All of them were selected using HaploView software with HapMap HCB data in
6
Results Patient characteristics and clinical outcomes The distribution of demographic and clinical characteristics of patients is showed in Table 1. All the patients were diagnosed with lung adenocarcinoma. EGFR mutations were detected in 106 patients (51.5%). Del 19 (47 tumors) and L858R mutations (43 tumors)
(16 tumors). All these patients as well as 37 patients with wild type EGFR received gefitinib treatment. The median follow-up time of the all patients was 18.3 months at the final analysis (November 2012). The median age of patients was 55.8 years (range, 32.0-80.0 years); 47.6% were male. There were no significant difference between those patients who were alive and those who had died with respect to age (P=0.53), gender (P=0.19), smoking status (P=0.92), ECOG (P=0.32), alcohol use (P=0.19), difference of
mutation (P=0.14) and weight loss at the time of diagnosis (P=0.23). Using UICC TNM staging manual (7th Edition)[19], twenty eight (19.6%) had stage IIIB disease and 115 (80.4%) had stage IV disease. In this cohort, ninety one patients (63.6%) died and the median OS time was 16.8 months (range, 1.2-45.7 months). The median PFS time was 15.9 months (range, 0.7-36.5 months).
Individual SNPs and survival The association of all SNPs with OS and PFS is summarized in supplementary Table 2-5. We identified 2 SNPs with a P<0.05 and 1 SNP with a P<0.01, and all of them belonged to SMAD3 (Table 2). The most significant association in this study was found to be with SMAD3: rs11632964.
Among the whole group patients, wild type SMAD3: rs11632964
was associated with an increased risk of death and disease progression. The patients with SMAD3: rs11632964 had a decreased median survival time (MST) of 15.9 months compared with 30.8 months for those with variant allele (Log-rank P=0.037, Figure 2A). Similar result was also found in PFS. Patients carrying variant allele had a longer PFS (13.2 months vs. 9.2 months) compared to those with the wild type (Figure 2B). In the patients with EGFR mutations, the efficacy of SMAD3: rs11632964 was found to be more obvious. The differences of OS and PFS in the patients with wild type versus variant
Downloaded from http://annonc.oxfordjournals.org/ at Memorial University of Newfoundland, Health Sciences Library on July 18, 2014
were most common; other mutations included missense mutations in exon 18, 19 and 21
7
genotypes of SMAD3: rs11632964 were more significant. Patients with variant allele also had much longer MST (30.8 months vs. 17.5 months; Log-rank P=0.020) and PFS (20.8 months vs. 9.4 months; Log-rank P=0.001) than those with wild type (Figure 2C, 2D).
SMAD3: rs11071938 variant genotype was also significantly associated with better PFS
patients with the EGFR mutations, SMAD3: rs11071938 variant genotype was associated better PFS (14.2months vs. 9.9 months; HR=1.75, 95%CI=1.06-2.89, P=0.029) (Figure 2E).
Different from the two SNPs described above, it was the wild type genotype of SMAD3: rs6494633 that was associated with improved PFS in EGFR mutated patients. These individuals had more than two months disease control advantage with PFS than those with variant genotype (11.3 months vs. 8.6 months, Log-rank P=0.048). (Figure 2F). The remaining 30 selected SNPs showed no associations between their genotype and the survival time (Table 3, Supplementary S3-S6).
Functional evaluation of the SMAD3 rs11632964 Polymorphism SMAD3 protein and pi-SMAD3 expression in PC-9 cells and PC9/G cells were detected firstly. We found that there was significantly decreased smad3 as well as pi-smad3 in the Gefitinib resistant PC-9/G cells compared with wild type PC-9 (Figure 3A,3B). We then examined the regulator role of SMAD3 rs11632964 SNP in these two cell lines. Relative luciferase expression assays indicated that the 651-bp intron 1 region which we cloned in either pGL3b-rs11632964T construct or pGL3b-rs11632964C construct showed promoter activity. This demonstrates that there might be an intronic promoter region around rs11632964 polymorphism. We next examined whether the survival-related SNP rs11632964 has an allele-specific effect on the intronic promoter activity. PC9/G cells transfected with SMAD3 rs11632964T allelic reporter construct showed significantly lower luciferase activities compared to cells expression C allelic reporter construct (both
Downloaded from http://annonc.oxfordjournals.org/ at Memorial University of Newfoundland, Health Sciences Library on July 18, 2014
(95%CI=1.01-2.31, P=0.045) in the whole cohort (11.8 months vs. 9.2 months); and the
8
P<0.05). However, no such difference was observed in PC9 cells (Figure 3C).
Discussion Currently, the detection of EGFR-TKI sensitive mutations are essential for selecting the treatments for the patients with advanced NSCLC since these mutations have been
our previous studies suggested that the polymorphisms of EGFR signaling pathway might be prognostic markers of OS in advanced lung adenocarcinoma patients who were treated with gefitinib [21]. But we did not find any significant correlation between EGFR SNPs and clinical outcome in the patients carrying EGFR mutations. In the current study, we systematically evaluated genetic variations in 33 SNPs from13 candidate genes and treatment outcome in the patients with advanced lung adenocarcinomas. Among the SNPs we have analyzed, the genetic variations in TGF-β pathway were found to be highly associated with outcome, especially in those with EGFR mutations. To our knowledge, our study is the first to link TGF-β signaling pathway SNPs with lung cancer outcome for in patients with EGFR mutant adenocarcinomas.
The variations of genes in TGF-β pathway are detected in several human cancers. TGF-β signaling pathway possesses a tumor suppressor function [22]. TGF-β and its signaling effectors act as the key determinants of carcinoma cell behavior.[23]. There is evidence to suggest that Smad3 plays a major role in TGF-β-mediated growth inhibition. First, Smad3 works in TGF-β–induced antiproliferative transcriptional responses. Second, Smad3 works in TGF-β–mediated proapoptotic transcriptional responses. [24]. On the other side, TGF-β promotes tumor progression during the late stages of cancer by suppressing immune surveillance, inducing EMT, and enhancing cell migration and transcription of factors favorable to metastasis. Smad3 is directly involved in EMT [25]. Changes of EMT status may underlie acquired resistance of gefitinib[26]. Sordella et al found erlotinib-resistant cells H1650-M3 were characterized by an up-regulation of TGF-β mediated signaling and increased levels of Smad3 phosphorylation in H1650-M3 cells compared with H1650 cells. Therefore, Smad3 functions as both a negative and positive
Downloaded from http://annonc.oxfordjournals.org/ at Memorial University of Newfoundland, Health Sciences Library on July 18, 2014
shown to be predictive of drug sensitivity. In addition to the well-known EGFR mutations,
9
regulator of carcinogenesis and cancer progression depending on cell type and clinical stage of the tumor [27].
In the main analysis, SMAD3: rs11632964, showed strong association with treatment outcome. It is possible that the variant allele of SMAD3: rs11632964 affect gene
suggested gefitinib-resistant PC9/G cells transfected with SMAD3 rs11632964T allelic reporter construct showed significantly lower luciferase activities compared to cells expression C allelic reporter construct and no such difference was observed in PC9 cells. We also found that there was significantly decreased smad3 as well as pi-smad3 in PC-9/G cells compared with wild type PC-9. The results of two functional tests were consistent and demonstrated that individuals with decreased expression of SMAD3 caused by the rs11632964T allele might contribute to shortened survival time after gefitinib treatment.
Our study has a complete clinical dataset with EGFR mutation status determined in all the patients. Although function is not known for all gene variations, this study suggests biological implications of these genetic variations in treatment selection and outcome. This study will help us choose the patients who can get more clinical benefit from TKI therapy except EGFR mutation. It has been reported that a haplotype of SMAD3 is associated with overall survival of NSCLC patients, which is the support of our results [16]. In conclusion, our data show that genetic variations in TGF-β pathway are the potential predictors of outcome in advanced lung adenocarcinoma with EGFR mutations. Analysis of these genetic variations, in addition to EGFR mutation detection, may refine the drug selection process for the treatment of advanced lung adenocarcinoms, improve the prediction of the treatment outcome. However, as a single institutional study, our patient number was relatively small. We need a larger patient population and better mapping and functional assays to validate the results.
Acknowledgements: The authors are thankful to Dr. Shiyu Song and Dr. Sheng Wei, for
Downloaded from http://annonc.oxfordjournals.org/ at Memorial University of Newfoundland, Health Sciences Library on July 18, 2014
transcription thus altering protein level [16]. Our functional test in lung cancer cell lines
10
reviewing the paper and analyzing the data. This work was supported in part by National Natural Science Foundation of China (no. 81001035; no. 81200498; no. 31271382; no. 81201586; no. 81071832). Disclosure: The authors have declared no conflicts of interest.
References Downloaded from http://annonc.oxfordjournals.org/ at Memorial University of Newfoundland, Health Sciences Library on July 18, 2014
1. Ciardiello F, De Vita F, Orditura M, Tortora G. The role of EGFR inhibitors in non small cell lung cancer. Curr Opin Oncol. 2004; 16: 130–135. 2. Fukuoka M, Wu YL, Thongprasert S, et al. Biomarker analyses and final overall survival results from a phase III, randomized, open-label, first-line study of gefitinib versus carboplatin/paclitaxel in clinically selected patients with advanced non-small-cell lung cancer in Asia (IPASS). J Clin Oncol. 2011; 29: 2866-2874. 3. Mitsudomi T, Morita S, Yatabe Y, et al. Gefitinib versus cisplatin plus docetaxel in patients with non-small-cell lung cancer harbouring mutations of the epidermal growth factor receptor (WJTOG3405): an open label, randomized phase 3 trial. Lancet Oncol. 2010; 11: 121-128. 4. Zhou C, Wu YL, Chen G, et al. Erlotinib versus chemotherapy as first-line treatment for patients with advanced EGFR mutation-positive non-small-cell lung cancer (OPTIMAL, CTONG-0802): a multicentre, open-label, randomised, phase 3 study. Lancet Oncol. 2011 ;12:735-742. 5. Kobayashi S, Boggon TJ, Dayaram T, et al. EGFR mutation and resistance of non-small-cell lung cancer to Gefitinib. N Engl J Med. 2005; 352: 786–792. 6. Engelman JA, Zejnullahu K, Mitsudomi T, et al., MET amplification leads to gefitinib resistance in lung cancer by activating ERBB3 signaling. Science. 2007; 316:1039-1043. 7. Yao Z, Fenoglio S, Gao DC, et al. TGF-beta IL-6 axis mediates selective and adaptive mechanisms of resistance to molecular targeted therapy in lung cancer. Proc Natl Acad Sci U S A. 2010; 107: 15535-15540. 8. Zhang Y.E. Non-Smad pathways in TGF-beta signaling. Cell Res, 2009; 19, 128-39. 9. Xu Y, Baker D, Quan T, et al. Receptor type protein tyrosine phosphatase-kappa mediates cross-talk between transforming growth factor-betaand epidermal growth factor receptor signaling pathways in human keratinocytes. Mol Biol Cell 2010; 21: 29-35. 10. Uramoto H, Yamada T, Yano S, et al. Prognostic value of acquired resistance-related molecules in Japanese patients with NSCLC treated with an EGFR-TKI. Anticancer Res 2012; 32: 3785-3790. 11. Chung JH, Jin KR, Xu XH, et al. Clinical and molecular evidences of epithelial to mesenchymal transition in acquired resistance to EGFR-TKIs. Lung Cancer 2011; 73: 176-182. 12. Ng KP, Hillmer AM, Chuah CT, et al. A common BIM deletion polymorphism mediates intrinsic resistance and inferior responses to tyrosine kinase inhibitors in cancer. Nat Med 2012; 18: 521-528. 13. Caroline M and Ying E. Zhang. Roles of Smad3 in TGF-β Signaling During
11
Carcinogenesis. Crit Rev Eukaryot Gene Expr. 2007; 17:281–293.
Downloaded from http://annonc.oxfordjournals.org/ at Memorial University of Newfoundland, Health Sciences Library on July 18, 2014
14. Takuya A, Hideaki Y, Kimihiro S, et al. Review of the Treatment of Non-Small Cell Lung Cancer with Gefitinib. Clin Med Insights Oncol. 2012; 6: 407-421. 15. Zhang L, Yang M, Bi N, et al. Association of TGF-β1 and XPD polymorphisms with severe acute radiation-induced esophageal toxicity in locally advanced lung cancer patients treated with radiotherapy. Radiother Oncol. 2010; 97: 19-25. 16. Lin M, Stewart DJ, Spitz MR, et al. Genetic variations in the transforming growth factor-beta pathway as predictors of survival in advanced non-small cell lung cancer. Carcinogenesis. 2011; 32: 1050-1056. 17. Li Q, Wu H, Chen B, et al. SNPs in the TGF-β signaling pathway are associated with increased risk of brain metastasis in patients with Non-Samll-Cell lung cancer. PloS One. 2012; 7: e51713. 18. Yang M, Zhang L, Bi N, et al. Association of P53 and ATM polymorphisms with risk of radiation-induced pneumonitis in lung cancer patients treated with radiotherapy. Int J Radiat Oncol Biol Phys. 2011; 79: 1402-1407. 19. TNM Classification of Malignant Tumours. Seventh Edition. Wilry-Blackwell: The Union for International Cancer Control (UICC), 2009; 138. 20. Zhang X, Wei J, Zhou L, et al. A functional BRCA1 coding sequence genetic variant contributes to risk of esophageal squamous cell carcinoma. Carcinogenesis. 2013;34:2309-2313. 21. Zhang L, Yuan X, Chen Y, et al. Role of EGFR SNPs in survival of advanced lung adenocarcinoma patients treated with Gefitinib. Gene. 2013; 517: 60-64. 22. Halder SK, Beauchamp RD, Datta PK. Smad7 induces tumorigenicity by blocking TGF-beta-induced growth inhibition and apoptosis. Exp Cell Res. 2005; 307: 231–246. 23. Siegel PM, Massague J. Cytostatic and apoptotic actions of TGF-β in homeostasis and cancer. Nat Rev Cancer. 2003; 3: 807–821. 24. Caroline M and Ying E. Zhang. Roles of Smad3 in TGF-β signaling during carcinogenesis. Crit Rev Eukaryot Gene Expr. 2007; 17:281–293. 25. Saika S, Kono-Saika S, Ohnishi Y, et al. Smad3 signaling is required for epithelial-mesenchymal transition of lens epithelium after injury. Am J Pathol. 2004; 164: 651–663. 26. Uramoto H, Shimokawa H, Hanagiri T, et al. Expression of selected gene for acquired drug resistance to EGFR-TKI in lung adenocarcinoma. Lung Cancer. 2011; 73: 361-365. 27. Masszi A, Kapus A. Smaddening complexity: the role of Smad3 in Epithelial-Myofibroblast transition. Cells Tissues Organs. 2011; 193: 41-52.
12
Downloaded from http://annonc.oxfordjournals.org/ at Memorial University of Newfoundland, Health Sciences Library on July 18, 2014
Fig 1. Patients flow diagram. Starting with advanced lung adenocarcinoma with EGFR mutation status treated in Tongji Hospital during 2009 to 2011.
13
Downloaded from http://annonc.oxfordjournals.org/ at Memorial University of Newfoundland, Health Sciences Library on July 18, 2014
Fig2. Kaplan-Meier survival curves for adenocarcinoma patients according to SMAD3: rs11632964 SNP. (A) Overall survival in the total population; (B) Progression-free survival in the total population; (C) Overall survival in patients with EGFR mutation; (D) Progression-free survival in patients with EGFR mutation. (E) SMAD3:rs11071938 Progression-free survival in patients with EGFR mutation; (F) SMAD3:rs6494633 Progression-free survival in patients with EGFR mutation.
14
A
B
Fig 3. T, cells transfected with pGL3b-rs11632964T construct; C, cells transfected with pGL3b-rs11632964C construct. Smad3 and pi-smad3 protein expression in PC9 cells and PC9/G cells(A,B). Upper panel, Smad3 expression in PC9 cells and PC9/G cells; lower panel, pi-smad3 expression in in PC9 cells and PC9/G cells. Transient luciferase reporter gene expression assays with constructs containing different alleles of SMAD3 intron 1 region in PC9 and gefitinib-resistant PC9/G lung cancer cells (C).
Downloaded from http://annonc.oxfordjournals.org/ at Memorial University of Newfoundland, Health Sciences Library on July 18, 2014
C
Table 1. Demographic and base line clinical characteristics of patients (N=143) Variable
Dead
P-value
57.7 (9.20)
54.7 (10.11)
0.53
21 (40.4) 31 (59.6)
47 (51.6) 44 (48.4)
0.19
17 (32.7) 35 (67.3)
29 (31.9) 62 (68.1)
0.92
12 (23.1) 40 (76.9)
16 (17.6) 75 (82.4)
0.43
38 (73.1) 14 (26.9)
73 (80.2) 18 (19.8)
0.32
16 (30.8) 36 (69.2)
38 (41.8) 53 (58.2)
0.19
33 (63.5) 7 (13.5) 10 (19.2) 2 (3.8)
72 (79.1) 8 (8.8) 9 (9.9) 2 (2.2)
0.23
42 (80.8) 10 (19.2)
64 (70.3) 27 (29.7)
<0.001
18 (46.2) 12 (30.8) 9 (23.0)
29 (43.3) 31 (46.3) 7 (10.4)
0.14
NOTE. Multivariate analyses were adjusted for all factors listed in Table.
Downloaded from http://annonc.oxfordjournals.org/ at Memorial University of Newfoundland, Health Sciences Library on July 18, 2014
Age, mean(SD), years Sex Male Female Smoking Status, n (%) smoker Non smoker Stage (TNM) IIIB IV ECOG, n (%) 0-1 ≥2 Alcohol, n (%) used Non Weight loss, n (%) Weight gain or stable 1-5% 5-10% >10% EGFR Status, n (%) Positive Negative EGFR mutations, n (%) Del 19 L858R others
Alive
Table 2. Association between Smad3 genotype and survival SNPs and genotypes Smad3: Rs11632964 CC CT Smad3: Rs11071938 CC CT Smad3: Rs6494633 CC CT
No of patients (all) (%) 143 120 (83.9) 23 (16.1) 143 98 (68.5) 45 (31.5) 143 113 (79.0) 30 (21.0)
No of patients (EGFR mutation) (%) 106 85 (80.2) 21 (19.8) 106 70 (66.0) 36 (34.0) 106 81 (76.4) 25 (23.6)
Median survival of all Patients (N=143) a OS P PFSa P
Median survival of patients with EGFR mutation (N=106) OSa P PFSa P
15.9 30.8
0.063
17.5 19.4
0.045
17.5 30.8
0.096
21.0 18.2
16.5 19.4 17.5 9.9
0.037
9.2 13.2
0.062
9.2 11.8
0.148
9.8 8.1
0.154
9.4 20.8
0.001
0.020
9.9 14.2
0.029
0.059
11.3 8.6
0.048
a
Adjusted for age, sex, Smoking status, Clinical stage, ECOG and EGFR mutation status. Results in bold are significant at P < 0.002 (Bonferroni correction). Abbreviations: EGFR, epidermal growth factor receptor; SNP, single nucleotide polymorphism; No, number of patients.
Downloaded from http://annonc.oxfordjournals.org/ at Memorial University of Newfoundland, Health Sciences Library on July 18, 2014
Table 3. Association between mutated-type genotype with OS and PFS for all patients and patients with EGFR mutation
TGF-β1 TGF-β1 TGF-β1 TGFβR1 TGFβR1 TGFβR1
BMP1 BMP1 BMP2 BMP4 BMP4 SMAD1 SMAD3 SMAD3 SMAD3 SMAD3 SMAD3 SMAD3 SMAD3 SMAD4 SMAD4
OS of all Patients (N=143) P HRa
SNP
Rs18004699 Rs4803455 Rs1800470 Rs64789974 Rs10733710 Rs10819638 Rs3857979 Rs783891 Rs235756 Rs8014071 Rs17563 Rs11939979 Rs11632964 Rs11071938 Rs6494633 Rs4776342 Rs4776343 Rs750766 Rs12102171 Rs7244227 Rs10502913
0.98 (0.63-1.53) 0.87 (0.59-1.35) 0.96 (0.61-1.49) 0.92 (0.59-1.44) 1.31 (0.84-2.07) 1.07 (0.68-1.67) 1.16 (0.75-1.80) 1.06 (0.69-1.63) 1.02 (0.62-1.71) 0.87 (0.56-1.36) 0.90 (0.59-1.37) 1.08 (0.69-1.70) 1.95 (1.04-3.67) 1.57 (0.98-2.52) 0.69 (0.42-1.14) 0.96 (0.62-1.50) 1.39 (0.65-2.97) 0.93 (0.61-1.44) 1.07 (0.70-1.66) 0.95 (0.58-1.54) 0.85 (0.54-1.33)
0.926 0.543 0.839 0.708 0.236 0.774 0.508 0.788 0.927 0.551 0.620 0.731 0.037 0.062 0.148 0.866 0.403 0.758 0.748 0.831 0.484
PFS of all patients (N=143) P HRa
0.96 (0.64-1.42) 0.96 (0.65-1.40) 1.03 (0.70-1.52) 1.04 (0.71-1.52) 0.96 (0.65-1.41) 1.30 (0.89-1.90) 0.91 (0.62-1.33) 0.89 (0.61-1.30) 0.93 (0.60-1.45) 0.93 (0.63-1.38) 1.22 (0.84-1.76) 1.08 (0.73-1.59) 1.67 (0.97-2.85) 1.53 (1.01-2.31) 0.68 (0.43-1.07) 0.90 (0.59-1.37) 1.30 (0.69-2.46) 1.03 (0.70-1.51) 1.20 (0.82-1.77) 1.11 (0.73-1.70) 0.94 (0.64-1.39)
0.819 0.815 0.881 0.855 0.826 0.182 0.626 0.540 0.757 0.731 0.290 0.706 0.063 0.045 0.096 0.583 0.424 0.893 0.348 0.625 0.750
OS for patients with EGFR mutation (N=106) P HRa
PFS for patients with EGFR mutation (N=106) P HRa
0.85 (0.50-1.45) 1.48 (0.67-3.27) 0.85 (0.49-1.47) 0.83 (0.48-1.41) 1.24 (0.74-2.09) 0.99 (0.58-1.67) 0.92 (0.55-1.54) 0.91 (0.55-1.51) 1.18 (0.62-2.26) 0.94 (0.55-1.59) 0.92 (0.56-1.51) 1.19 (0.70-2.01) 2.38 (1.15-4.94) 1.50 (0.86-2.63) 0.58 (0.33-1.02) 0.96 (0.57-1.62) 1.30 (0.50-3.36) 0.96 (0.57-1.60) 0.90 (0.54-1.51) 0.98 (0.55-1.74) 1.05 (0.62-1.76)
0.84 (0.53-1.35) 1.33 (0.70-2.53) 0.94 (0.59-1.50) 0.90 (0.57-1.41) 1.07 (0.68-1.69) 1.13 (0.71-1.80) 0.69 (0.44-1.08) 0.69 (0.44-1.08) 0.99 (0.56-1.75) 0.87 (0.54-1.38) 1.16 (0.75-1.78) 1.29 (0.82-2.03) 3.01 (1.54-5.86) 1.75 (1.06-2.89) 0.55 (0.37-1.00) 0.90 (0.56-1.44) 1.06 (0.49-2.29) 1.01 (0.64-1.60) 1.00 (0.62-1.59) 1.12 (0.68-1.86) 1.09 (0.69-1.72)
0.555 0.332 0.566 0.483 0.419 0.963 0.761 0.712 0.622 0.809 0.727 0.524 0.020 0.154 0.059 0.877 0.587 0.861 0.689 0.944 0.871
Downloaded from http://annonc.oxfordjournals.org/ at Memorial University of Newfoundland, Health Sciences Library on July 18, 2014
Gene
0.470 0.379 0.792 0.639 0.770 0.600 0.100 0.108 0.982 0.541 0.499 0.278 0.001 0.029 0.048 0.652 0.885 0.951 0.987 0.651 0.724
SMAD4 SMAD4 SMAD4 SMAD4 SMAD6 SMAD6 SMAD6 SMAD7 SMAD8 SMAD8 INHBC ACVR2A
Rs12455792 Rs12456284 Rs12958604 Rs948588 Rs12906898 Rs12913975 Rs4776318 Rs7227023 Rs511674 Rs7333607 Rs4760259 Rs1424954
1.27 (0.83-1.94) 1.27 (0.83-1.94) 1.10 (0.70-1.71) 1.02 (0.52-2.00) 1.02 (0.58-1.80) 0.85 (0.55-1.31) 1.03 (0.64-1.67) 2.98 (0.40-22.2) 0.88 (0.58-1.34) 1.16 (0.71-1.91) 0.87 (0.57-1.35) 1.17 (0.73-1.87)
0.281 0.281 0.690 0.958 0.941 0.451 0.907 0.287 0.553 0.556 0.543 0.512
1.22 (0.85-1.76) 1.22 (0.85-1.76) 1.05 (0.71-1.54) 1.53 (0.78-3.00) 1.14 (0.69-1.89) 0.93 (0.63-1.38) 1.17 (0.76-1.81) 1.30 (0.39-4.34) 1.00 (0.69-1.45) 1.27 (0.82-1.97) 0.95 (0.65-1.40) 1.17 (0.77-1.77)
0.284 0.284 0.825 0.212 0.613 0.733 0.468 0.668 0.995 0.279 0.804 0.460
1.13 (0.68-1.89) 1.13 (0.68-1.89) 1.07 (0.62-1.83) 0.77 (0.35-1.70) 0.80 (0.41-1.56) 0.76 (0.46-1.25) 0.88 (0.49-1.59) 4.50 (0.60-33.55) 0.69 (0.42-1.13) 1.46 (0.79-2.70) 1.01 (0.60-1.69) 1.37 (0.80-2.36)
0.641 0.641 0.821 0.514 0.515 0.281 0.672 0.143 0.142 0.233 0.968 0.256
1.23 (0.80-1.89) 1.23 (0.80-1.89) 1.11 (0.70-1.77) 1.43 (0.63-3.25) 0.91 (0.51-1.63) 0.86 (0.55-1.34) 1.11 (0.66-1.86) 2.28 (0.67-7.73) 0.83 (0.54-1.27) 1.58 (0.93-2.69) 1.09 (0.70-1.70) 1.19 (0.74-1.92)
a
0.353 0.353 0.663 0.397 0.748 0.497 0.703 0.185 0.386 0.093 0.715 0.480
Adjusted for age, sex, Smoking status, Clinical stage, ECOG and EGFR mutation status. Results in bold are significant at P < 0.002 (Bonferroni correction). Abbreviations: HR, hazard ratio; CI, confidence interval
Downloaded from http://annonc.oxfordjournals.org/ at Memorial University of Newfoundland, Health Sciences Library on July 18, 2014