The role of genetics in progression of cancer is an established fact, and susceptibility risk and difference in outcome to chemotherapy may be caused by the variation in low-penetrance alleles of risk genes. We selected seven genes (CRP, GPC5, ACTA2, AGPHD1, SEC14L5, RBMS3, and GKN1) that previously reported link to lung cancer (LC) and genotyped single nucleotide polymorphisms (SNPs) of these genes in a case-control study. A protective allele “C” was found in rs2808630 of the C-reactive protein (CRP). Model association analysis found genotypes “T/C” and “C/C” in the dominant model and genotype “T/C” in the overdominant model of rs2808630 associated with reduced LC risk. Gender-specific analysis in each model showed that genotypes “T/T” and “C/C” in rs2352028 of the Glypican 5 (GPC5) were associated with increased LC risk in males. Logistic regression analysis showed “C/T” genotype carriers of rs4254535 in the Gastrokine 1 (GKN1) had less likelihood to have chemotherapy response. Our results suggest a potential association between CRP and GPC5 variants with LC risk; variation in GKN1 is associated with chemotherapy response in the Chinese Han population.

1. Introduction

Lung cancer is the most common malignancy in the world and is reported to have an increasing incidence in developing countries [1, 2]. According to the global cancer statistics, in 2008 approximately 1.6 million people were diagnosed with lung cancer, and there were 1.4 million deaths [3]. Tobacco smoke, environmental pollution, occupational exposures, and preexisting lung disease increase the risk of lung cancer. However, patients have been diagnosed with lung cancer in the absence of these risk factors [46]. Genetic susceptibility to lung cancer independent of established risk factors has not yet been clearly defined.

Despite considerable advances in the field of tumor biology, the majority of patients with lung cancer are diagnosed at an already advanced stage and thus surgical resection is not a feasible treatment option. Platinum-based doublet chemotherapy is the current standard of therapy in this situation. However, the response to chemotherapy among lung cancer patients has significant variation. We hypothesize that the susceptibility risk and variation in outcome to chemotherapy may be caused by the variation in low-penetrance alleles.

In this study, we selected single nucleotide polymorphisms (SNPs) from seven different genes (CRP (C-reactive protein), GPC5 (Glypican 5), ACTA2 (actin, alpha 2, smooth muscle, aorta), AGPHD1 (aminoglycoside phosphotransferase domain containing 1), SEC14L5 (SEC14-like 5), RBMS3 (RNA binding motif, single stranded interacting protein 3), and GKN1 (Gastrokine 1)) that have been linked to lung cancer [713]. We analyzed each tag single nucleotide polymorphism (tSNP) for lung cancer risk in a case-control study involving Chinese population. Multivariate logistic regression analysis was used to test the association between gene polymorphisms and chemotherapy response.

2. Materials and Methods

2.1. Study Participants

A case-control study involving the Chinese study population of 309 lung cancer patients and 310 controls was conducted at the First Affiliated Hospital of Xi’an Jiaotong University. All included patients had recently diagnosed and histopathologically confirmed primary lung cancer. The control subjects were recruited from the health check-up center of the First Affiliated Hospital of Xi’an Jiaotong University, which they had visited for an annual health examination. Patients were ascertained to be free from any acute or chronic pathology. Their cancer-free status was reconfirmed by testing for plasma levels of carcinoembryonic antigen and alpha-fetoprotein. Blood samples from the patients were collected before initiation of chemotherapy or radiotherapy. Demographic and related clinical data of the study population was collected by a face-to-face questionnaire and medical case record. Patients were categorized as smokers or nonsmokers. The smokers were defined as those who smoked one cigarette/pipe per day for twelve months or longer at any time in their life. All of the participants were genetically unrelated ethnic Han Chinese from Shaanxi Province and provided written informed consent for their participation in the present study. The protocols for this study were conducted according to the Declaration of Helsinki and were approved by the Institutional Review Boards of both the First Affiliated Hospital of Xi’an Jiaotong University and Northwest University.

Five milliliters of whole blood were collected from each subject into tubes containing ethylenediaminetetraacetic acid (EDTA) at the time of initial diagnosis. After centrifugation, the samples were stored at −80°C until further use.

2.2. Evaluation of Cisplatin-Based Chemotherapeutic Response

There are all together 113 lung cancer patients who received cisplatin based first-line chemotherapy and satisfied the following criteria: Eastern Cooperative Oncology Group (ECOG) performance status (PS) ≤ 1, age > 18 years, and adequate bone marrow reserve, as well as satisfactory liver and renal function. These patients were in clinical stage III or IV and had a measurable lesion on computed tomography scan at the beginning of treatment. The patients received chemotherapy every 3 weeks, for a maximum of six cycles or until disease progression or unacceptable toxicity occurred. Response to treatment was determined according to the Response Evaluation Criteria in Solid Tumor Group (RECIST) guidelines after two cycles of chemotherapy and every two cycles thereafter [14]. For data analysis, patients achieving complete response (CR) or partial response (PR) were considered “responders,” and patients with stable disease (SD) or progressive disease (PD) were defined as “nonresponders” [15]. Multivariate logistic regression analysis was used to test the association between gene polymorphisms and chemotherapy response.

2.3. tSNP Selection and Genotyping

All seven tSNPs in the selected genes were associated with lung cancer and with minor allele frequencies (MAF) greater than 5% in the HapMap CHB (Chinese Han Beijing) population. DNA was extracted from whole blood by GoldMag-Mini Whole Blood Genomic DNA Purification Kit (GoldMag Co., Ltd., Xi’an City, China). The concentration was measured by NanoDrop 2000 (Thermo Scientific, Waltham, Massachusetts, USA). The design of primers, SNP genotyping, and data processing were performed by Sequenom MassARRAY platform Software (Sequenom Co., Ltd., San Diego, California, USA) [16, 17].

2.4. Statistical Analysis

Statistical analysis was undertaken using statistical software (SPSS 16.0; Chicago, IL) and Microsoft Excel. A two-sided value < 0.05 was considered the threshold for statistical significance. Hardy-Weinberg equilibrium (HWE) of each tSNP in control group was tested by Fisher’s exact test. The differences in allelic frequencies between case and control groups were compared via the Chi-squared test [18]. Associations between genotypes and lung cancer risk were tested in different genetic models (codominant, dominant, recessive, overdominant, and log-additive) by SNPStats website software http://bioinfo.iconcologia.net/snpstats/start.htm [19]. Testing of odds ratios (ORs) with 95% confidence intervals (CIs) was performed by unconditional logistic regression analysis with adjustment for gender and age [20]. Akaike’s Information Criterion and Bayesian Information Criterion were applied to estimate the best-fit model for each SNP. Association between genotypes and lung cancer risk was determined by SNPStats for gender-specific populations under each model [19].

3. Results

We recruited 309 patients (74 females and 235 males, mean age at diagnosis 58 years, range 25–85, SD ± 10) and 310 healthy (113 females and 197 males, mean age at diagnosis 50 years, range 29–75, SD ± 8) individuals into our study (Table 1). The genotype profiles of our study patients are shown in Supplementary Table S1 in the Supplementary Material available online at http://dx.doi.org/10.1155/2015/824304. The SNPs and primers used in the multiplexed SNP MassEXTENED assay are presented in Table 2. None of the tSNPs that we evaluated among the control group deviated from HWE (Table 3). We hypothesized that the minor allele of each SNP was a risk factor compared with the wild-type allele.

A significant protective allele “C” was found in rs2808630 of the CRP gene based on the crude value of 0.05 (OR = 0.66; 95% CI, 0.48–0.91; ) by Chi-square test (Table 3). Various genetic models were applied to calculate genetic risk. Reduced risk for lung cancer was associated with the genotypes “T/C” and “C/C” in rs2808630 (OR = 0.66, 95% CI, 0.44–0.98; ) in the dominant model and the genotype “T/C” (OR = 0.65, 95% CI, 0.43–0.98; ) in the overdominant model (Table 4). Each tSNP was analyzed in a gender-specific population under each model. We found that the genotypes “T/T” and “C/C” in rs2352028 of the GPC5 gene were associated with increased lung cancer risk in males in the overdominant model (Table 5). rs2808630 in CRP and rs2352028 in GPC5 were both associated with lung cancer risk.

“C/T” genotype distribution in the rs4254535 of the GKN1 gene was significantly higher in nonresponders than in responders (34.62% versus 14.29%, ) (Table 6). Logistic regression analysis showed that “C/T” genotype carriers had poor response for chemotherapy as compared to “T/T” genotype carriers (OR 3.287, 95% CI, 1.135–9.522; ) after adjustment for age, gender, smoking status, histology, stage, and chemotherapy regimens.

However, as shown in Tables 3, 4, and 6, the significance levels were attenuated after applying a strict Bonferroni correction, indicating a likely association between positive tSNPs and risk of lung cancer and chemotherapy response.

4. Discussion

In this case-control study, we selected tSNPs with MAF greater than 5% in the HapMap CHB population to ensure that the statistical power was sufficient for data analysis. Our results firstly suggest that polymorphisms in CRP and GPC5 genes have an association with susceptibility risk of lung cancer in the Chinese Han population. The multivariate logistic regression analysis shows that polymorphism in GKN1 influences chemotherapy response.

The CRP gene, located in 1q23.2, encodes CRP protein which has several host defense-related functions, including recognition and elimination of foreign pathogens and damaged host cell. CRP is an acute-phase protein that increases during the host response to tissue injuries, including infection, trauma, surgery, myocardial infarct, and cancer [8, 21]. There are three potential mechanisms linking CRP to cancers. One is that tumor growth promotes tissue inflammation and increases the level of CRP. Another possibility is that cancer cells increase production of inflammatory proteins, which leads to high CRP levels in cancer patients. Besides, CRP may promote tumor growth in chronic inflammation [22]. Elevated CRP levels are associated with poor prognosis of lung, hepatic, renal, colorectal, and ovarian cancers [2329].

Our study found that rs2808630, an intronic SNP within the CRP gene, was significantly linked with lung cancer risk in both allelic and genotypic association analysis of a Chinese population. We also ascertained a significant allele “C” and genotypes “T/C” and “C/C” in rs2808630 in the dominant model and genotype “T/C” in the overdominant model that is protective against lung cancer development. We hypothesize that rs2808630 variant of the CRP gene could have decreased the level of CRP or reduced the activity of CRP in the presence of allele “C”. A recent study by Xu et al. [30] found that 5 SNPs in the CRP gene (including rs2808630) were uncorrelated with lung cancer risk. They recruited 96 lung cancer patients and 124 controls of different races. This disparity in findings could be attributed to the small sample size and racial or regional differences in study populations. To our knowledge, our study is the first genotype/allele-based study that describes the association between SNPs within the CRP locus and lung cancer risk in a Chinese population.

The GPC5 gene is a member of the glypican gene family and has eight exons encoding 572 amino acids in a large genomic region (1.47 Mb) of chromosome 13q31.3. Reduction of GPC5 protein is linked to lung cancer [7]. A previous study involving American population reported an association (OR = 1.46, 95% CI 1.26–1.70, ) between the single nucleotide polymorphism rs2352028 and lung cancer risk in never smokers [31] but failed to replicate in Caucasian [32] and Chinese [33] populations, indicating that the sensitivity and specificity of rs2352028 in terms of smoking status may not be similar in between races. Our study observed the variation between gender and found that genotypes “T/T” and “C/C” in rs2352028 of the GPC5 gene are associated with increased lung cancer risk in males (under the overdominant model, after adjusting for age).

The GKN1 gene is located in 2p13.3 and has a protective function on gastric antral mucosa by facilitating restoration and proliferation after injury. As it is expressed in normal gastric tissue but absent in gastric cancer tissues, GKN1 protein is treated as a potential biomarker for gastric cancer [34]. It is also found downregulated in placental tissue and cell [35]. Although current research focuses on the potential clinical use of GKN1 in the treatment of tumor, little is known about its expression and function in other organ systems or the significance of GKN1 polymorphisms in cancer. Our study firstly reports that polymorphismin GKN1 is influence cisplatin based chemotherapy response in lung cancer patients. The SNPs from the other four genes (ACTA2, AGPHD1, SEC14L5, and RBMS3) included in this study did not reach any statistically significant association with lung cancer risks or cisplatin based chemotherapy response in our study population.

There are certain intrinsic limitations in our study and must be noted. The sample size was not as large as some other lung cancer association studies. We performed Bonferroni correction in our statistical analysis and found no statistical significant associations between CRP and GPC5 SNPs and lung cancer risk, neither in GKN1 polymorphisms nor in response to cisplatin-based chemotherapy, which could be attributed to the relatively small sample size that may not satisfy all the seven independent hypotheses at the same time. Adjustments for multiple tests, like Bonferroni correction, are needed for medical association studies but may create more problems. The main weakness of Bonferroni correction is that the results depend on the number of other tests performed. True important differences may be deemed nonsignificant since the likelihood of type II errors is also increased [36]. Cumulatively, our findings provide evidence that polymorphisms in C-reactive protein and Glypican 5 genes are associated with lung cancer risk, and GKN1 determines chemotherapy response in Chinese population. We believe our results will encourage further studies to understand the function of these genes.


tSNP:Tag single nucleotide polymorphism
CRP: C-reactive protein gene
GPC5: Glypican 5 gene
ACTA2: Actin, alpha 2, smooth muscle, aorta gene
AGPHD1: Aminoglycoside phosphotransferase domain containing 1 gene
SEC14L5: SEC14-like 5 gene
RBMS3: RNA binding motif, single stranded interacting protein 3 gene
GKN1: Gastrokine 1 gene
LC:Lung cancer
MAF:Minor allele frequency
HWE:Hardy-Weinberg equilibrium
OR:Odds ratio
CI:Confidence intervals.

Conflict of Interests

The authors have no conflict of interests regarding the publication of this paper.


This work was supported by the National Science and Technology Major Project of the Ministry of Science and Technology of China (no. 2012ZX09506001) and Collaboration and Communication Project of International Science and Technology of Shaanxi Province (Grant no. 2014KW24-03).

Supplementary Materials

Genotype frequencies of the seven tSNPs in lung cancer patients and controls.

  1. Supplementary Material