Introduction

Lung cancer incidence has increased rapidly in China in the past years [1, 2]. Mortality from lung cancer has increased by 465 % in China during the past 30 years, and it has become the main cause of death [1, 2]. Although cigarette smoking remains the predominant cause of lung cancer, it cannot fully explain epidemiologic characteristics of lung cancer in nonsmokers [3]. Many lung cancers occur in nonsmoker which suggests that genetic factors also play important roles in the development of lung cancer [4]. Currently, the genetic polymorphisms in DNA repair genes are increasingly studied as possible risk factors of lung cancer because of their roles in maintaining the genome integrity [5, 6]. Genomic instability has been considered to play key roles in the multistage carcinogenesis and a hallmark of the carcinogenesis of many cancers including lung cancer [7, 8]. Genetic variations in DNA repair genes have been reported to be associated with the genomic instability and increasing risk of genomic damages [9]. X-ray repair cross-complementing group 1 (XRCC1) is one of the major DNA repair proteins involved in the base excision repair (BER) and plays an important role in the maintenance of genomic integrity [5, 6]. Polymorphisms in XRCC1 may alter the function and repair capacity of XRCC1 protein which further results in the genetic instability and lung carcinogenesis [5, 6]. Previous studies investigating the association between XRCC1 Arg399Gln polymorphism and lung cancer risk in Chinese reported contradictory results [10–19]. A meta-analysis was performed to clarify the effect of XRCC1 Arg399Gln polymorphism on lung cancer.

Methods

Search strategy and inclusion criteria

A systemic literature search of PubMed and China National Knowledge Infrastructure (CNKI) databases was conducted from their inception to 06 January 2013. A search strategy combining both the Medical Subject Heading and text words was used. There was no language restriction. The search words included XRCC1, Arg399Gln, polymorphisms, polymorphism, lung cancer, and lung carcinoma. Reference lists of all the included articles, the related articles, and relevant reviews were also screened to avoid omitting any potentially relevant studies. Studies were included for the meta-analysis if they satisfied the following criteria: (1) Investigating the association between XRCC1 Arg399Gln polymorphism and lung cancer risk; (2) Participants were from China; (3) Baseline characteristics were comparable between the cases and controls; and (4) Distributions of XRCC1 Arg399Gln genotype were all reported.

Data extraction and quality assessment

Two investigators independently extracted data from each study with a predefined review form, and discrepancies were resolved by consensus of all investigators. Information extracted included author, year of publication, study design, age of participants, selection of control (population or hospital based), sample size, distributions of XRCC1 Arg399Gln genotype, and genotyping method. The quality of each study was assessed by the judgment of the deviation from Hardy–Weinberg equilibrium (HWE) in the control group. Studies without the deviation from HWE in the control group were defined as high-quality studies.

Statistical analysis

Deviation from HWE in the control group was examined using a two-tailed χ 2 test. The strength of association between XRCC1 Arg399Gln polymorphism and lung cancer risk was assessed using the pooled odds ratio (OR) with 95 % confidence intervals (95 %CI). We evaluated the risk using the allele model (Gln vs. Arg), the additive model (GlnGln vs. ArArg), the dominant model (GlnGln/ArgGln vs. ArArg), and the recessive model (GlnGln vs. ArArg/ArgGln). The statistical heterogeneity among studies was assessed with the I 2 test was used to quantify inconsistency [20]. A I 2 value ≥ 50 % was considered to represent significant heterogeneity, and the random (DerSimonian–Laird method) [21] effect model was used to calculate the pooled OR. A I 2 value < 50 % was considered to represent less heterogeneity, and the fixed (Mantel–Haenszel method) [22] effect model was used to calculate the pooled OR. To explore the potential effect of study quality on the overall effect estimates, sensitivity analysis was performed by excluding studies with the deviation from HWE in the control group. Funnel plot was generated as a visual aid to detect bias. Statistical analyses were undertaken using Stata version 11.0 (StataCorp, College Station, TX, USA).

Results

Description of included studies

A total of 85 potentially abstracts were identified from those two databases. After screening the abstracts and reviewing the full-texts, 66 studies were excluded because of case reports, or reviews, or non-relevant studies. Finally, 19 studies with a total of 12,835 participants were included into this meta-analysis [10–19, 23–31]. All studies included in the meta-analysis used a hospital-based case–control design, with a total of 6,288 cancer cases and 6,547 controls [10–19, 23–31]. Fifteen studies (78.9 %) used polymerase chain reaction–restriction fragment length polymorphisms (RFLP) to test the genotype of XRCC1 Arg399Gln polymorphism. All studies reported data enabling formal testing of whether genotype frequencies in the control group deviated from HWE, and the genotype distribution of XRCC1 Arg399Gln polymorphism in the control group was not consistent with HWE only in one study [19].

Meta-analysis

Table 1 shows the main results from the meta-analysis on the association between XRCC1 Arg399Gln polymorphism and lung cancer risk in Chinese. Overall, there was an obvious association between XRCC1 Arg399Gln polymorphism and increased risk of lung cancer under three genetic models (Gln vs. Arg: OR = 1.13, 95 %CI 1.01–1.25, P = 0.029; GlnGln vs. ArArg: OR = 1.41, 95 %CI 1.07–1.84, P = 0.013; GlnGln vs. ArArg/ArgGln: OR = 1.37, 95 %CI 1.07–1.76, P = 0.013) (Fig. 1, Table 1).

Table 1 Meta-analysis of the association between XRCC1 Arg399Gln polymorphism and lung cancer risk in Chinese
Full size table
Fig. 1
figure 1

Results of the association between XRCC1 Arg399Gln polymorphism and lung cancer risk in Chinese. a The allele model (Gln vs. Arg); b the recessive model (GlnGln vs. ArArg/ArgGln)

Full size image

Meta-analysis of 18 studies with high quality also found that there was an obvious association between XRCC1 Arg399Gln polymorphism and increased risk of lung cancer under three genetic models (Gln vs. Arg: OR = 1.12, 95 %CI 1.00–1.26, P = 0.045; GlnGln vs. ArArg: OR = 1.35, 95 %CI 1.02–1.77, P = 0.033; GlnGln vs. ArArg/ArgGln: OR = 1.31, 95 %CI 1.02–1.67, P = 0.034) (Table 1).

Publication bias

The shape of funnel plots in all four genetic models was symmetrical, which suggested no risk of publication bias in the meta-analysis. For example, the shape of funnel plot was symmetrical in the allele genetic model suggesting no risk of publication bias in the allele model of the meta-analysis (Fig. 2). Thus, there was no obvious risk of bias in the meta-analysis.

Fig. 2
figure 2

Funnel plot of publication bias in the meta-analysis of the association between XRCC1 Arg399Gln polymorphism and lung cancer risk in Chinese under the allele genetic model

Full size image

Discussion

Although cigarette smoking remains the predominant cause of lung cancer, it cannot fully explain epidemiologic characteristics of lung cancer in nonsmokers [3]. Many lung cancers occur in nonsmokers, which suggest that genetic factors also play important roles in the development of lung cancer [4, 32]. Some polymorphisms have been identified as risk factors of lung cancer, such as microsomal epoxide hydrolase 1 T113C polymorphism [33].

XRCC1 is one of the major DNA repair proteins involved in the base excision repair and plays an important role in the maintenance of genomic integrity. Polymorphisms in XRCC1 may alter the function and repair capacity of XRCC1 protein which further results in the genetic instability and lung carcinogenesis. Previous studies investigating the relationship between XRCC1 Arg399Gln polymorphism and lung cancer risk in Chinese reported contradictory results [10–19, 23–31]. A meta-analysis was performed to clarify the effect of XRCC1 Arg399Gln polymorphism on lung cancer. Nineteen studies with a total of 12,835 participants were included into this meta-analysis. Overall, there was an obvious association between XRCC1 Arg399Gln polymorphism and increased risk of lung cancer under three genetic models (Table 1). Meta-analysis of 18 studies with high quality also found that there was an obvious association between XRCC1 Arg399Gln polymorphism and increased risk of lung cancer under three genetic models. There was no obvious risk of bias in the meta-analysis. Data from the current meta-analysis support the obvious association between XRCC1 Arg399Gln polymorphism and increased risk of lung cancer in Chinese.

Currently, the genetic polymorphisms in DNA repair genes are increasingly studied as possible risk factors of lung cancer because of their roles in maintaining the genome integrity [5, 6]. XRCC1 Arg399Gln polymorphism is the most studied polymorphism in XRCC1 [5, 6]. However, previous meta-analyses of all possible studies from total populations showed that there was no association between XRCC1 Arg399Gln polymorphism and lung cancer [34, 35]. All those two meta-analyses showed obvious heterogeneity among the included studies [34, 35]. Though there were many possible sources of heterogeneity, the race-special effect usually was the main source of heterogeneity. Currently, there were many studies on the relationship between XRCC1 Arg399Gln polymorphism and lung cancer risk in Chinese, so we performed a meta-analysis only in Chinese. The findings from this meta-analysis showed that there was an obvious association between XRCC1 Arg399Gln polymorphism and increased risk of lung cancer in Chinese. Thus, our meta-analysis provided new findings in the relationships between XRCC1 Arg399Gln polymorphism and cancers.

Several limitations should be acknowledged in the meta-analysis. Firstly, lung cancer is a complex disease, and there are complex interactions between genetic background and environmental factors especially tobacco smoking. However, our meta-analysis did not analyze the interactions between XRCC1 Arg399Gln polymorphism and tobacco smoking in the development of lung cancer because there was not enough data from the included studies. Secondly, gene–gene interactions were also possible in the association between XRCC1 Arg399Gln polymorphism and lung cancer risk. Therefore, further studies are needed to assess the possible gene–gene interactions. Finally, our meta-analysis was pooled at the study’s level because of the lack of individual patient data from the original studies. A meta-analysis of individual patient data may provide a more precise estimation on the association between XRCC1 Arg399Gln polymorphism and lung cancer risk in Chinese. Therefore, a meta-analysis of individual patient data is needed in the future.

In conclusion, data from the current meta-analysis support the obvious association between XRCC1 Arg399Gln polymorphism and increased risk of lung cancer in Chinese. Besides, further studies are needed to assess the possible gene–gene or gene–environment interactions in the association above.