Literature DB >> 30408066

Associations between three XRCC1 polymorphisms and hepatocellular carcinoma risk: A meta-analysis of case-control studies.

Yao Xiong1, Qian Zhang2, Jiaxiang Ye1, Shan Pan1, Lianying Ge1.   

Abstract

Conflicting results have been obtained regarding the association between X-ray repair cross complementation group 1 (XRCC1) and susceptibility to hepatocellular carcinoma (HCC). In this study, associations between HCC and three polymorphisms (Arg194Trp, Arg280His, and Arg399Gln) were evaluated using a meta-analysis approach. PubMed, Web of Science, Cochrane Library, the Chinese National Knowledge Infrastructure, and the Wanfang standard database were systematically searched to identify all relevant case-control studies published through March 2018. A total of 32 case-control studies, including 13 that evaluated Arg194Trp, 14 that evaluated Arg280His, and 26 that evaluated Arg399Gln, were analyzed. In the entire study population, XRCC1 Arg399Gln was significantly associated not only with overall risk of HCC (homozygous model, OR = 1.61, 95% CI: 1.40-1.85, P < 0.05; recessive model, OR = 1.40, 95% CI: 1.23-1.59, P < 0.05) but also with the risk of HCC in Chinese patients (homozygous model, OR = 1.78, 95% CI: 1.53-2.08, P < 0.05; recessive model, OR = 1.47, 95% CI: 1.27-1.70, P < 0.05). Limiting the analysis to studies demonstrating Hardy-Weinberg equilibrium (HWE), the results were consistent and robust. Similarly, a significant association between XRCC1 Arg399Gln and HCC risk was found in healthy controls in the general population but not in hospital controls. Trial sequential analysis (TSA), false-positive report probabilities (FPRP), and combined genotype analysis revealed that XRCC1 Arg399Gln is mainly associated with susceptibility to liver cancer. However, there was no association between Arg194Trp or Arg280His and the risk of HCC. These results, indicating that the Arg399Gln polymorphism of XRCC1 is associated with the risk of HCC in the Chinese population, provide a basis for the development of improved detection and treatment approaches.

Entities:  

Mesh：

Substances：

Year:  2018        PMID： 30408066      PMCID： PMC6226104          DOI： 10.1371/journal.pone.0206853

Source DB:  PubMed          Journal:  PLoS One        ISSN： 1932-6203            Impact factor:   3.240


Introduction

Hepatocellular carcinoma (HCC) is a primary malignant tumor of the liver that ranks second in cancer deaths in developing countries, sixth in cancer deaths in developed countries [1], and third in the incidence of malignant tumors in China [2]. The onset of HCC is occult, and early symptoms and signs are not easy to detect. Most patients are diagnosed with advanced-stage disease; therefore, treatment is not effective. According to recent epidemiological data, the 5-year survival rate of patients with HCC is only 18% [1]. To improve the prevention and treatment of HCC, it is necessary to clarify its pathogenesis. The formation of HCC is a multistep process of multiple pathogenies. The causes include chronic hepatitis virus infection, Aspergillus flavus toxin damage, long-term drinking, and extensive smoking. However, not everyone exposed to these factors develops HCC. Increasing evidence suggests that HCC is triggered not only by external factors but also by genetic factors. The base excision repair pathway repairs damaged DNA, thereby maintaining genomic integrity. However, this pathway is prone to errors, resulting in DNA damage and cancer [3]. XRCC1 is a key molecule in the DNA repair process, with a key role in the integrity and stability of the genome and in the pathogenesis and carcinogenesis of various types of tumors. It has been reported that XRCC1 gene polymorphisms are associated with lung, esophageal, breast, bladder, and gastrointestinal cancer [3-7]. Additionally, a clinical study has shown that XRCC1 280 is significantly associated with the number of tumors, tumor size, and tumor location and is an independent risk factor for poor prognosis in patients with HCC [8]. Similarly, XRCC1 399 is significantly associated with clinical prognosis. After transcatheter arterial chemoembolization, the risk of death in patients with the A/A+G/A genotype is lower and the median survival time is longer (11.2 months) than those in patients with other genotypes [9]. Many studies have explored the relationship between gene polymorphisms and HCC susceptibility, but a unified conclusion is lacking. In this study, a meta-analysis of studies of XRCC1 Arg194Trp, Arg280His, and Arg399Gln was used to determine the relationship between these polymorphisms and susceptibility to HCC.

Materials and methods

Literature search

A comprehensive search was performed against various databases, i.e., PubMed, Web of Science, Cochrane Library, the Chinese National Knowledge Infrastructure, and the Wanfang standard database, to identify case–control studies published through March 1, 2018 that examined the association between XRCC1 polymorphisms and HCC risk. Searches were performed using various combinations of customized terms and the MeSH-indexed terms “X-ray repair cross complementation group 1” OR “XRCC1” AND “variation” OR “variability” AND “hepatocellular carcinoma” OR “liver cancer”, without restrictions on publication language. The following sequential search strategy was applied for each database: (#1) ‘DNA repair pathway’: ab, ti OR ‘repair gene’/exp ‘OR ‘repair reaction’/exp OR ‘repair response’/exp OR ‘Base Excision Repair/BER’/exp; (#2) ‘X-ray repair cross-complementation group 1’: ab, ti OR ‘XRCC1’: ab, ti OR’ X-ray repair complementing defective repair in Chinese hamster cells 1’/exp; (#3) ‘variation’: ab, ti OR ‘polymorphism’: ab, ti OR ‘SNP’: ab, ti OR ‘Single Nucleotide Polymorphism’/exp OR ‘genetic polymorphism’/exp OR ‘genetic variability’/exp; (#4) ‘liver cancer’: ab, ti OR ‘hepatocellular carcinoma’: ab, ti OR ‘primary hepatic carcinoma’/exp OR ‘primary liver cancer’/exp; (#5) #1 AND #2 AND #3 AND #4.

Study inclusion and exclusion criteria

Inclusion criteria

The inclusion criteria were as follows: (1) studies that examined the association between XRCC1 Arg194Trp, Arg280His, and Arg399Gln and susceptibility to HCC in the Chinese population; (2) studies of humans; (3) case–control studies; (4) studies reporting genotype distributions in the case group and the control group.

Exclusion criteria

The exclusion criteria were as follows: (1) simple case reports, reviews, or commentaries; (2) subjects were single HCC families, animals, or other organisms; (3) the association between XRCC1 Arg194Trp, Arg280His, and Arg399Gln and susceptibility to HCC was not evaluated; (4) data were incomplete; (5) repeated publication (only the most recent or most complete studies were included). The process of literature screening is shown in Fig 1.
Fig 1

Flow diagram of study selection for the meta-analysis.

CNKI, Chinese National Knowledge Infrastructure Database. WFSD, the Wanfang standard database.

Flow diagram of study selection for the meta-analysis.

CNKI, Chinese National Knowledge Infrastructure Database. WFSD, the Wanfang standard database. The following data were carefully extracted and examined by two assessors: author, publication year, country, case/control number, source, method, Hardy–Weinberg equilibrium (HWE) test, and quality score. The basic features of these studies are shown in Table 1.
Table 1

The general data of the observation group and the control group were included in the meta-analysis.

VariableYearsCountryCases/ControlsCaseControlSourceMethodHWEScore
c1/c1c1/c2c2/c2c1/c1c1/c2c1/c2
Arg194Trp
Su [10]2008China100/11146504574311PBPCR-RFLP0.5010
Kiran [11]2009India63/14384312276452PBPCR-RFLP0.367
Zeng [12]2010China500/5072801833727019938HBTaqman0.8711
Bo [13]2011China130/13094315116122PBPCR-RFLP0.029
Tang [14]2011China150/150944115815811PBPCR-RFLP0.8910
Bo [15]2012aChina60/60411365352PBPCR-RFLP0.008
Han [16]2012China150/158724731844628PBPCR-RFLP0.008
Yuan [17]2012aChina252/2501191151812810121HBPCR-RFLP0.869
Zeng [18]2012China46/4623232620HBPCR-RFLP—
Wu [19]2014China218/27715155121986811PBPCR-RFLP0.1010
Yang [20]2015China118/120555310584517HBPCR-RFLP0.105
Krupa [21]2017Polish65/5057534181HBTaqman0.435
Guo [22]2012China410/410264109372929623HBPCR-RFLP0.0011
Arg280His
Su [10]2008China100/1117920187213PBTaqman0.2310
Wu [23]2009China100/607722147130PBPCR-RFLP0.347
Kiran [11]2009India63/155193014912935PBPCR-RFLP0.006
Zeng [12]2010China500/507414797417873HBTaqman0.5011
Tang [14]2011China150/150138111123261PBPCR-RFLP0.7710
Han [16]2012China150/158813534823640PBPCR-RFLP0.008
Yuan [17]2012aChina252/250193536206395HBPCR-RFLP0.069
Yuan [24]2012bChina350/400272735329647HBPCR-RFLP0.0710
Bo [15]2012aChina60/60421265163PBPCR-RFLP0.008
Bo [25]2012bChina90/90641887893PBPCR-RFLP0.008
Zeng [18]2012China46/463973511HBPCR-RFLP—
Gulnaz [26]2013Pakistan50/742417944273HBPCR-RFLP0.656
He [27]2015China77/40611603640PBPCR-RFLP0.747
Krupa [21]2017Polish65/50577136113HBTaqman0.125
Arg399Gln
Yao [28]2014China1486/1996777608101143752039PBPCR-RFLP0.3113
Yu [29]2003China577/3893012235321814328PBPCR-RFLP0.5011
Yang [30]2004China69/13634728581563HBPCR-RFLP0.007
Long [31]2004China140/5367263536215915HBPCR-RFLP0.6210
Kirk [32]2005Gambia149/294120263248433HBPCR-RFLP0.4611
Borentain [33]2007France56/8927218274319PBTaqman0.818
Ren [34]2008China50/923214446415PBPCR-RFLP0.287
Su [10]2008China100/11140537693111PBTaqman0.019
Kiran [11]2009India63/14225335457027PBPCR-RFLP0.987
Jia [35]2010China136/136536617784513HBPCR-RFLP0.1010
Zeng [12]2010China500/5072861803430416736HBTaqman0.0511
Pan [36]2011China202/23645105526811256PBPCR-RFLP0.469
Tang [14]2011China150/150419415845412PBPCR-RFLP0.4310
Guo [22]2012China410/4102031367122712855PBPCR-RFLP0.0011
He [37]2012China113/11380231097124PBPCR-RFLP0.0010
Han [16]2012China150/158327840467339PBPCR-RFLP0.359
Bo [15]2012aChina60/60381485253PBPCR-RFLP0.008
Zeng [18]2012China46/4633132521HBPCR-RFLP—
Mohana [38]2013India93/93364512325110HBPCR-RFLP0.125
Bose [39]2013India55/20922294758846HBPCR-RFLP0.048
Gulnaz [24]2013Pakistan50/74191417273215HBPCR-RFLP0.346
Wu [19]2014China218/27710874361618729PBPCR-RFLP0.009
He [27]2015China77/404726427121PBPCR-RFLP0.807
Krupa [21]2017Polish65/504215832126HBTaqman0.024
Santonocito [40]2017Italia89/993745759382HBPCR0.145
Bazgir [41]2017Irania50/101121820315614HBPCR-RFLP0.1610

Notes: PB, population-based; HB, hospital-based; HWE, Hardy-Weinberg equilibrium; c1:Arg; c2: For Arg194Trp, Trp; ForArg280His, His; ForArg399Gln, Gln.

Notes: PB, population-based; HB, hospital-based; HWE, Hardy-Weinberg equilibrium; c1:Arg; c2: For Arg194Trp, Trp; ForArg280His, His; ForArg399Gln, Gln.

Quality assessment

The Newcastle–Ottawa Scale (NOS) was used to evaluate the quality of all eligible studies. The NOS provides a quality rating, ranging from 0 to 10, based on criteria covering three study aspects: study group selection, comparability of cases and controls, and exposure of cases and controls. Results of the quality assessment are shown in Table 2. We also used the quality assessment criteria (S1 Table), derived from a previously published meta-analysis of non-Hodgkin lymphoma [42], for further assessment. Quality scores of studies ranged from 0 to 15. Studies with scores ≤ 9 were considered of low quality, while those with scores > 9 were considered of high quality.
Table 2

Results of quality assessment using the Newcastle-Ottawa Scale for cohort studies.

Study (au, year)A1A2A3A4BC1C2C3Score
Su 2008★★★★★★★★8
Kiran 2009★★★★★★★一7
Zeng 2010★★一★★★★★7
Bo 2011★★★★★★★一7
Tang 2011★★★★★★★一7
Bo 2012a★★★★★★★一7
Han 2012★★★★★ ★★★★9
Yuan 2012a★★一★★★★★7
Zeng 2012★★一★★★★一6
Yao 2014★★★★★ ★★★一8
Wu 2014★★★★★★★一7
Yang 2015★★一★★★★一6
Krupa 2017★★一★★★★一6
Wu 2009★★★★★★★一7
Yuan 2012b★★一★★★★★7
Gnlnaz 2013★★一★★★★一6
He 2015★★★★★★★★8
Yu 2003★★★★★★★一7
Yang 2004★★一★★ ★★★一7
Long 2004★★一★★★★一6
Bo 2012b★★★★★★★一7
Kirk 2005★★一★★ ★★★一7
Borentain 2007★★★★★★★一7
Ren 2008★★★★★★★一7
Jia 2010★★一★★★★一6
Pan 2011★★★★★ ★★★一8
Guo 2012★★★★★ ★★★一8
He 2012★★★★★★★一7
Mohana 2013★★一★★★★一6
Bose 2013★★一★★ ★★★一7
Santonocito 2017★★一★★ ★★★一7
Bazgir 2017★★★★★★★★8

Notes: A1, Representativeness of the exposed cohort; A2, Selection of the non-exposed cohort; A3, Ascertainment of exposure; A4, Demonstration that outcome of interest was not present at start of study; B, Comparability of cohorts on the basis of the design or analysis; C1, Assessment of outcome; C2, Was follow-up long enough for outcomes to occur; C3, Adequacy of follow up of cohorts; A, B, C represent Selection, Comparability, Outcome, respectively;★and★★indicate compliance with the requirements of the definition, for which specific meaning see S1 Text.

Notes: A1, Representativeness of the exposed cohort; A2, Selection of the non-exposed cohort; A3, Ascertainment of exposure; A4, Demonstration that outcome of interest was not present at start of study; B, Comparability of cohorts on the basis of the design or analysis; C1, Assessment of outcome; C2, Was follow-up long enough for outcomes to occur; C3, Adequacy of follow up of cohorts; A, B, C represent Selection, Comparability, Outcome, respectively;★and★★indicate compliance with the requirements of the definition, for which specific meaning see S1 Text.

Data analysis

RevMan 5.3 was used for meta-analysis, and Q and I2 values were used to evaluate heterogeneity. If the heterogeneity test showed P > 0.1 or I2 < 50%, a fixed effect model was used, and heterogeneity was considered relatively low; if P < 0.1 or I2 > 50%, heterogeneity was considered high, and a subgroup analysis or sensitivity analysis was used to study the source of heterogeneity. The odds ratio (OR) and 95% confidence interval (CI) were used as indicators of the effect for each result. HWE was calculated using HWSIM (http://krunch.med.yale.edu/hwsim/website). P > 0.05 indicated equilibrium; P < 0.05 indicated a departure from HWE. A sensitivity analysis was used to evaluate the stability of the results. False-positive report probabilities (FPRP) were calculated using the FPRP calculation spreadsheet (see http://jncicancerspectrum.oupjournals.org/jnci-/content/vol96/issue6) to assess positive results. The FPRP threshold was set to 0.2, and the prior probability was set to 0.1 to detect the OR. A significant result with an FPRP value of less than 0.2 indicated a notable finding. All statistical tests were two-sided, and P < 0.05 was considered statistically significant. Trial sequential analysis (TSA) was used to reduce random errors and increase the robustness of the conclusions, using a 5% significance level for type I errors and a 20% significance level for type II errors, and the amount of information and a TSA monitoring boundary were determined.

Results

Study selection and characteristics

The meta-analysis was conducted according to the recommendations of the “Preferred Reporting Items for Systematic Reviews and Meta-Analyses” (PRISMA) statement (S1 Checklist) and “Meta-analysis on Genetic Association Studies” statement (S2 Checklist). Systemic literature searches identified 32 articles [10-41], eight of which discussed all three loci, four of which discussed two loci, and twenty of which discussed a single locus. Of the 32 studies, 13 analyzed the XRCC1 Arg194Trp polymorphism, 14 analyzed Arg280His, and 26 analyzed Arg399Gln. The XRCC1 Arg194Trp, Arg280His, and Arg399Gln polymorphisms were evaluated by calculating ORs and 95% CIs under homozygous, heterozygous, dominant, and recessive models. The results are summarized in Table 3.
Table 3

Overall and subgroup analysis of the XRCC1 polymorphisms and cancer risk.

VaribleNHomozygous genetic modelHeterozygous genetic modelDominant genetic modelRecessive genetic model
OR(95%CI)PhetI2OR(95%CI)PhetI2OR(95%CI)PhetI2OR(95%CI)PhetI2
Arg194TrpTrp/Trp vs Arg/ArgTrp/Trp vs Arg/TrpTrp/Trp + Arg/Trp vs Arg/ArgTrp/Trp vs Arg/Arg + Arg/Trp
All131.13(0.90,1.41)0.34101.42(1.24,1.62)0.01571.14(1.01,1.29)0.01531.02(0.82,1.26)0.0544
All-China111.14(0.91,1.44)0.25211.41(1.23,1.61)0.03521.15(1.01,1.30)0.01581.11(0.89,1.39)0.2027
All-HWE80.92(0.69,1.21)0.7801.39(1.18,1.64)0.01601.00(0.87,1.16)0.5200.82(0.63,1.06)0.1535
All-HWE-China60.91(0.68,1.23)0.6401.38(1.16,1.62)0.06530.99(0.85,1.15)0.4400.91(0.68,1.21)0.3313
All-PB71.18(0.84,1.67)0.3781.72(1.37,2.14)0.01681.26(1.03,1.54)0.01680.97(0.71,1.33)0.0455
All-HB61.09(0.82,1.47)0.23291.26(1.07,1.50)0.5001.08(0.94,1.26)0.37141.06(0.79,1.40)0.2034
Arg280HisHis/His vs Arg/ArgHis/His vs Arg/HisHis/His + Arg/His vs Arg/ArgHis/His vs Arg/Arg + Arg/His
All131.43(0.91,2.25)0.15311.20(1.02,1.41)0.00691.19(1.02,1.38)0.00671.15(0.84,1.56)0.2223
All-China101.14(0.77,1.69)0.4701.14(0.96,1.35)0.03531.13(0.96,1.33)0.01561.10(0.76,1.59)0.510
All-HWE91.28(0.61,2.67)0.18331.05(0.88,1.26)0.03531.08(0.91,1.28)0.02571.31(0.77,2.23)0.1734
All-HWE-China71.15(0.59,2.22)0.6801.08(0.90,1.31)0.03561.10(0.91,1.32)0.05521.09(0.58,2.05)0.590
All-PB71.37(0.80,2.35)0.24261.34(1.03,1.73)0.00771.30(1.03,1.64)0.00751.03(0.71,1.48)0.450
All-HB61.47(0.62,3.47)0.11461.12(0.91,1.37)0.11471.12(0.92,1.36)0.06531.49(0.84,2.63)0.1245
Arg399GlnGln/Gln vs Arg/ArgGln/Gln vs Arg/GlnGln/Gln + Arg/Gln vs Arg/ArgGln/Gln vs Arg/Arg + Arg/Gln
All251.61(1.40,1.85)0.00691.55(1.42,1.68)0.00741.56(1.45,1.69)0.00791.40(1.23,1.59)0.0064
All-China171.78(1.53,2.08)0.00671.66(1.52,1.82)0.00771.68(1.54,1.82)0.00811.47(1.27,1.70)0.0060
All-HWE171.80(1.51,2.13)0.00721.58(1.44,1.73)0.00791.64(1.50,1.79)0.00821.53(1.30,1.79)0.0067
All-HWE-China102.00(1.65,2.42)0.00731.71(1.55,1.89)0.00821.77(1.61,1.95)0.00841.57(1.31,1.87)0.0069
All-PB141.83(1.55,2.17)0.00741.66(1.51,1.83)0.00811.73(1.57,1.90)0.00841.51(1.29,1.77)0.0069
All-HB111.19(0.93,1.53)0.00741.29(1.10,1.50)0.04471.24(1.07,1.43)0.00571.17(0.92,1.48)0.0158

Quantitative synthesis

The XRCC1 Arg399Gln polymorphism was related to the risk of HCC in the Arg399Gln homozygous genetic model (OR = 1.61, 95% CI: 1.40–1.85, Pheterogeneity < 0.05; Fig 2), recessive genetic model (OR = 1.40, 95% CI: 1.23–1.59, Pheterogeneity < 0.05; Fig 3), dominant genetic model (OR = 1.56, 95% CI: 1.45–1.69, Pheterogeneity < 0.05; Fig 4), and heterozygous genetic model (OR = 1.55, 95% CI: 1.42–1.68, Pheterogeneity < 0.05). Arg399Gln was also associated with susceptibility to HCC in the Chinese population based on the homozygous genetic model (OR = 1.78, 95% CI: 1.53–2.08, Pheterogeneity < 0.05) and recessive genetic model (OR = 1.47, 95% CI: 1.27–1.70, Pheterogeneity < 0.05), suggesting that Gln/Gln is a risk factor for HCC. Limiting the analysis to studies demonstrating HWE, inconsistent results were obtained (Table 3). In the Indian population, the Arg399Gln homozygous genetic model (OR = 0.49, 95% CI: 0.27–0.87, Pheterogeneity = 0.15) and recessive genetic model (OR = 0.51, 95% CI: 0.30–0.87, Pheterogeneity = 0.07) indicated that Gln/Gln is a protective factor for liver cancer. Similarly, the funnel plot for Arg399Gln was asymmetric, implying a slight publication bias (Fig 5, funnel plot for the Arg399Gln homozygous model; Fig 6, funnel plot for the Arg399Gln recessive model; Fig 7, funnel plot for the Arg399Gln dominant model).
Fig 2

Forest plot of the association between XRCC1 Arg399Gln polymorphism and HCC risk under a homozygous model.

Fig 3

Forest plot of the association between XRCC1 Arg399Gln polymorphism and HCC risk under a recessive model.

Fig 4

Forest plot of the association between XRCC1 Arg399Gln polymorphism and HCC risk under a dominant model.

Fig 5

Funnel plot to detect publication bias in data on XRCC1 Arg399Gln polymorphism according to a homozygous model.

Fig 6

Funnel plot to detect publication bias in data on XRCC1 Arg399Gln polymorphism according to a recessive model.

Fig 7

Funnel plot to detect publication bias in data on XRCC1 Arg399Gln polymorphism according to a dominant model.

The XRCC1 Arg280His polymorphism was not associated with the risk of HCC under the Arg280His homozygous genetic model (OR = 1.43, 95% CI: 0.91–2.25, Pheterogeneity = 0.15), recessive genetic model (OR = 1.15, 95% CI: 0.84–1.56, Pheterogeneity = 0.22), dominant genetic model (OR = 1.19, 95% CI: 1.02–1.38, Pheterogeneity < 0.01), or heterozygous genetic model (OR = 1.20, 95% CI: 1.02–1.41, Pheterogeneity < 0.01). In addition, no association was observed for any subgroups. Similarly, the XRCC1 Arg194Trp was not related to susceptibility to HCC under the Arg194Trp homozygous genetic model (OR = 1.13, 95% CI: 0.90–1.41, Pheterogeneity = 0.34), recessive genetic model (OR = 1.02, 95% CI: 0.82–1.26, Pheterogeneity = 0.05), dominant genetic model (OR = 1.14, 95% CI: 1.01–1.29, Pheterogeneity < 0.05), or heterozygous genetic model (OR = 1.42, 95% CI: 1.24–1.62, Pheterogeneity < 0.05). No significant association was found in any of the subgroups.

Sensitivity analysis

Owing to the slight heterogeneity of the results for Arg194Trp, a sensitivity analysis was performed, which indicated that Tang et al. [14] was the source of the heterogeneity. After eliminating this study, Arg194Trp was unrelated to susceptibility to HCC based on the homozygous genetic model (OR = 1.13, 95% CI: 0.78–1.64, Pheterogeneity = 0.55), heterozygous genetic model (OR = 1.17, 95% CI: 0.97–1.39, Pheterogeneity < 0.01), dominant genetic model (OR = 1.14, 95% CI: 0.96–1.35, Pheterogeneity < 0.01), and recessive genetic model (OR = 0.96, 95% CI: 0.67–1.36, P = 0.79). The full analysis suggested that Arg280His was not associated with susceptibility to HCC under the homozygous genetic model (OR = 1.56, 95% CI: 1.11–2.18, Pheterogeneity = 0.19) or recessive genetic model (OR = 1.45, 95% CI: 1.05–1.99, Pheterogeneity = 0.07). However, many studies had small sample sizes. After those with N < 200 were eliminated, the Arg280His homozygous genetic model (OR = 1.13, 95% CI: 0.78–1.64, Pheterogeneity = 0.55), heterozygous genetic model (OR = 1.17, 95% CI: 0.97–1.39, Pheterogeneity < 0.01), dominant genetic model (OR = 1.14, 95% CI: 0.96–1.35, Pheterogeneity < 0.01), and recessive genetic model (OR = 0.96, 95% CI: 0.67–1.36, Pheterogeneity = 0.79), still indicated a lack of evidence for an association with susceptibility to HCC.

TSA, combined genotype analysis, and FPRP analysis

We performed a TSA to reduce random errors and increase the robustness of the conclusions. The TSA of the Arg194Trp polymorphism model showed that the cumulative z-curve did not cross the traditional cut-off value, nor did it cross the TSA threshold. Moreover, the expected amount of information was not obtained, indicating that the difference in the XRCC1 Arg194Trp polymorphism between the HCC group and the control group was not statistically significant and that additional experiments are needed (Fig 8). The TSA of the allele models for the Arg280His polymorphism showed that the cumulative z-curve crossed the traditional cut-off value but did not cross the TSA threshold, and the cumulative amount of information was insufficient (Fig 9). The TSA of the allele models for the Arg399Gln polymorphism showed that the cumulative z-curve crossed both the traditional threshold and the TSA threshold, and the accumulated information was sufficient, indicating that no further evidence was needed to verify the conclusion (Fig 10).
Fig 8

Trial sequential analysis for XRCC1 Arg194Trp gene polymorphism under the allele contrast model.

Fig 9

Trial sequential analysis for XRCC1 Arg280His gene polymorphism under the allele contrast model.

Fig 10

Trial sequential analysis for XRCC1 Arg399Gln gene polymorphism under the allele contrast model.

HWE-based studies were performed using the combined genotype analysis. XRCC1 Arg194Trp + Arg280His was not associated with HCC susceptibility under the homozygous genetic model (OR = 1.00, 95% CI: 0.78–1.29) or recessive genetic model (OR = 0.90, 95% CI: 0.71–1.14). When Arg399Gln was combined with either or both of the other polymorphisms, correlations with HCC susceptibility were detected, indicating that the main SNP related to HCC risk is XRCC1 Arg399Gln (Table 4).
Table 4

Combined genotype analysis for three XRCC1 single nucleotide polymorphisms.

All-HWEHomozygous genetic modelHeterozygous genetic modelDominant genetic modelRecessive genetic model
OR(95%CI)OR(95%CI)OR(95%CI)OR(95%CI)
Arg194Trp + Arg280His0.96(0.73,1.26)1.22(1.08,1.38)1.03(0.93,1.15)0.90(0.71, 1.14)
Arg194Trp + Arg399Gln1.50(1.29,1.74)1.53(1.41,1.66)1.42(1.32,1.53)1.29(1.12,1.48)
Arg280His + Arg399Gln1.77(1.49,2.10)1.45(1.33,1.57)1.50(1.39,1.62)1.51(1.29,1.76)
Arg194Trp + Arg280His + Arg399Gln1.49(1.29,1.73)1.43(1.33,1.54)1.36(1.27,1.46)1.29(1.13,1.47)
Table 5 shows the FPRP values for our positive results using different prior probability levels. Assuming a prior probability of 0.1 and a specific genotype with an OR of 1.5, the statistical power was 0.856, and the FPRP value was < 0.001 for the recessive model of the XRCC1 Arg399Gln polymorphism. Thus, the risk of liver cancer was elevated for all individuals. In addition, the FPRP values for the Chinese population, the all-HWE-compliant population, the China-HWE-compliant subgroup, and the population-based (PB) group were all less than 0.2, indicating reliable results.
Table 5

False-positive report probability values for associations between the risk of hepatocellular carcinoma and the frequency of genotypes of XRCC1 gene.

Arg399Gln HomozygousCrude OR(95%CI)Statistical powerP-valuePrior probability
0.250.10.010.0010.0001
All1.61(1.40,1.85)0.1590.0000.0000.0000.0000.0000.000
All-China1.78(1.53,2.08)0.0160.0000.0000.0000.0000.0000.000
All-HWE1.80(1.51,2.13)0.1740.0000.0410.1140.5850.9340.993
All-HWE-China2.00(1.65,2.42)0.0020.0000.0000.0000.0000.0000.000
All-PB1.83(1.55,2.17)0.0110.0000.0000.0000.0000.0000.000
Arg399Gln Recessive
All1.40(1.23,1.59)0.8560.0000.0000.0000.0040.0030.025
All-China1.47(1.27,1.70)0.6070.0000.0000.0000.0000.0000.003
All-HWE1.53(1.30,1.79)0.4020.0000.0000.0000.0000.0000.003
All-HWE-China1.57(1.31,1.87)0.3050.0000.0000.0000.0000.0010.014
All-PB1.51(1.29,1.77)0.4670.0000.0000.0000.0000.0010.008

Discussion

HCC is a serious digestive system tumor that is typically detected at an advanced stage, when treatment approaches are limited and prognosis is poor. Studies have shown that bad eating habits, alcohol consumption, environment, work stress, and emotional changes are risk factors for HCC in high-incidence areas. However, not all individuals exposed to these risk factors develop HCC, indicating that genetic susceptibility may be important. Defects or inadequate DNA repair caused by polymorphisms in DNA repair genes increase the risk of cancer. Previous studies have reported that XRCC1 expression is elevated in colorectal [43], esophageal [44], and lung cancer tissues [45]. Similarly, Krupa et al. [21] showed that the mRNA expression of XRCC1 in HCC tissues was significantly lower than that in adjacent non-lesional tissues. The expression of XRCC1 in cancer tissues is closely related to the intrinsic genetic phenotype. Thus, DNA repair gene polymorphisms may play an important role in susceptibility to liver cancer. Many studies have shown that the XRCC1 Arg399Gln polymorphism is associated with HCC risk, while susceptibility is lower for carriers of Arg194Trp and Arg280His; however, the results of previous studies have been inconsistent. Guo et al. [22] found that, compared with Arg/Arg, XRCC1 194Trp/Trp was significantly associated with risk of HCC, and individuals carrying XRCC1 399Gln/Gln showed an increased risk of HCC. However, Yang et al. [30], found that this polymorphism was not related to HCC risk. Su et al. [10] suggested that the Arg194Trp and Arg280His polymorphisms are not related to susceptibility to HCC but that the Arg399Gln polymorphism is a susceptibility factor for HCC, with Gln/Gln as a risk factor, consistent with the results of this study. Jia et al. [35] found that the XRCC1 399 Arg/Gln genotype conferred increased HCC risk. Han et al. [16] found that the median survival rate of individuals carrying the XRCC1 Gln/Gln genotype was significantly lower than that of individuals carrying the XRCC1 Arg/Arg genotype. However, Zeng et al. [12] suggested that XRCC1 Arg194Trp, Arg280His, and Arg399Gln are not predisposing factors for HCC but found that there was an additive interaction between XRCC1 polymorphisms and smoking, drinking, and chronic hepatitis B virus (HBV) infection. Similarly, Yuan et al. [17] found that XRCC1 Arg194Trp and Arg280His are not associated with the risk of HCC but that Arg399Gln is associated with a significantly increased risk of HCC in the background of HBV infection and family history. In a previous meta-analysis, Xu et al. [46] found that Arg280His was associated with the risk of HCC and that His/His increases the risk of disease. Li et al. [47] found that Arg194Trp and Arg280His were not related to the risk of HCC, but that 399 Arg/Gln was significantly associated with the risk of HCC, and the results were still significant in studies demonstrating HWE. Similarly, Liu et al. [48] indicated that there was a significant correlation between Arg399Gln and susceptibility to HCC in the Chinese population. Shi et al. [49] found that 399 Arg/Gln was unrelated to HCC but that it was significantly correlated with the incidence of HCC in southern China, suggesting that there was genetic heterogeneity. In our comprehensive meta-analysis of Arg194Trp, Arg280His, and Arg399Gln case–control studies, we grouped the populations that satisfied HWE, calculated FPRPs, and performed a TSA to increase the robustness of the conclusions. Our findings showed that the Arg399Gln polymorphism increased susceptibility to liver cancer, while Arg194Trp and Arg280His were not associated with susceptibility to liver cancer. However, additional samples are needed to further evaluate these findings. Zhu et al. [50] studied the relationship between the XPC genotype and DNA repair ability in an alkaline comet assay challenged by benzo[a]pyrene diol epoxide (BPDE) and γ radiation. Healthy subjects with the XPC Lys939Gln variant allele (AC and CC) were found to have significantly increased rates of DNA damage induced by BPDE and γ irradiation compared to homozygous wild-type (AA) subjects. In contrast, subjects with the Ala499Val variant allele (CT and TT) showed reduced BPDE- and γ radiation-induced DNA damage. Reinardy et al. [51] evaluated echinoderms after 24 h of exposure to genotoxic agents (UV-C, hydrogen peroxide, and bleomycin) and found that adult sea urchin coelomocytes and larvae with XRCC1 polymorphisms showed more mutations in the body cavity after recovery, indicating the heterogeneous response of echinoderms to DNA damage and revealing that DNA repair ability within host cells may be regulated by specific gene polymorphisms. Therefore, XRCC1 plays a crucial role in maintaining genomic stability and preventing cancer. We hypothesize that people exposed to risk factors for liver cancer are more likely to develop XRCC1 mutations, resulting in an altered DNA repair capacity and increased susceptibility to liver cancer. In addition, Kuptsova et al. [52] found that after a standard chemotherapy induction regimen in elderly patients with acute myeloid leukemia, different DNA repair gene variants repaired chemotherapy-induced DNA damage, which may affect drug toxicity and treatment response to varying degrees. Xuan et al. [53] found that XRCC1 can increase the base repair ability and promote tumor resistance via the tumor drug resistance pathway, suggesting that variants in the DNA repair pathway may impact patient outcomes and treatment-related responses. Wang et al. [54] found that XRCC1 protein levels are significantly down-regulated in gastric cancer lesions compared with levels in adjacent non-cancerous tissues in a study of the prognosis and predictive role of XRCC1 in patients treated with surgery alone or in combination with adjuvant chemotherapy. Low expression of XRCC1 was significantly associated with shorter overall survival and clinicopathological features of unassisted patients. The prognosis of patients treated with adjuvant fluorouracil-leucovorin-oxaliplatin was significantly improved compared with that for surgery alone. However, this effect was only significant in the low expression group; therefore, XRCC1 protein expression in tumors is a novel candidate prognostic marker and response predictor. Li et al. [55] performed a prognostic analysis of 150 patients with HCC and found that patients carrying the Gln/Gln genotype showed a significantly lower median survival than individuals with the Arg/Arg genotype. Compared with Arg/Arg carriers, the median survival rate of Arg/Gln + Gln/Gln carriers was reduced. Therefore, we hypothesize that the XRCC1 Gln/Gln genotype can be used as a negative indicator in liver cancer and that XRCC1 can serve as a potential indicator for clinical diagnosis and prognosis, as well as a new potential target for clinical treatment in HCC cases. Our study had some limitations. We observed high heterogeneity among studies, which may be related to the choice of the control population, differences in living environments, and differences in family genetic background. Second, many studies included in the analysis had small sample sizes. To ensure the stability of the results, we evaluated the FPRP and performed a TSA. In addition, we detected a slight publication bias, suggesting that additional well-designed studies are needed. Our results showed that the XRCC1 Arg399Gln Gln/Gln genotype is a risk factor for liver cancer in the Chinese population.

Conclusions

XRCC1 polymorphisms are still a major topic in cancer research. Previous meta-analyses of these polymorphisms have yielded inconsistent results. In this study, relevant literature was obtained to resolve this controversy. Our results indicated that XRCC1 Arg399Gln is significantly associated with the risk of HCC, especially in the Chinese population. In addition, there was a slight publication bias, suggesting the need for further research.

PRISMA 2009 checklist.

(DOC) Click here for additional data file.

Meta-analysis of Genetic Association Studies checklist.

(DOCX) Click here for additional data file.

The Newcastle-Ottawa Scale (NOS) as a quality assessment method in this systematic review.

(PDF) Click here for additional data file.

Score of quality assessment.

(DOC) Click here for additional data file.

Gene polymorphisms distribution of XRCC1 gene polymorphic sites in HCC group and control group.

(DOC) Click here for additional data file.

The detailed reasons for excluded articles.

(DOC) Click here for additional data file.

The differences from the previous meta-analyses.

(DOCX) Click here for additional data file.

The detailed search criteria and the full data range for the search in each database.

(DOC) Click here for additional data file.
  40 in total

1.  DNA repair capacity, DNA-strand break repair gene polymorphisms, and the incidence of hepatocellular carcinoma in southwestern Guangxi of China.

Authors:  Xiaoyun Zeng; Shun Liu; Hongping Yu; Long Ji; Longman Li; Jinmei Huang; Hua Bai; Xiaoqiang Qiu
Journal:  DNA Cell Biol       Date:  2012-06-12       Impact factor: 3.311

2.  Cancer statistics, 2018.

Authors:  Rebecca L Siegel; Kimberly D Miller; Ahmedin Jemal
Journal:  CA Cancer J Clin       Date:  2018-01-04       Impact factor: 508.702

3.  Roles of XRCC1/XPD/ERCC1 Polymorphisms in Predicting Prognosis of Hepatocellular Carcinoma in Patients Receiving Transcatheter Arterial Chemoembolization.

Authors:  Xu-Cai Wang; Fan Wang; Qing-Qing Quan
Journal:  Genet Test Mol Biomarkers       Date:  2016-02-26

4.  Analysis of genetic damage and gene polymorphism in hepatocellular carcinoma (HCC) patients in a South Indian population.

Authors:  Subramaniam Mohana Devi; Vellingiri Balachandar; Meyyazhagan Arun; Shanmugam Suresh Kumar; Balasubramanian Balamurali Krishnan; Keshavarao Sasikala
Journal:  Dig Dis Sci       Date:  2012-10-02       Impact factor: 3.199

5.  Relationship between XRCC1 and XPD polymorphisms and the risk of the development of hepatocellular carcinoma: A case-control study.

Authors:  Tao Yuan; Shaoli Deng; Hongming Liu; Menggang Liu; Ping Chen
Journal:  Exp Ther Med       Date:  2012-05-17       Impact factor: 2.447

6.  Germline Mutations in FAN1 Cause Hereditary Colorectal Cancer by Impairing DNA Repair.

Authors:  Nuria Seguí; Leonardo B Mina; Conxi Lázaro; Rebeca Sanz-Pamplona; Tirso Pons; Matilde Navarro; Fernando Bellido; Adriana López-Doriga; Rafael Valdés-Mas; Marta Pineda; Elisabet Guinó; August Vidal; José Luís Soto; Trinidad Caldés; Mercedes Durán; Miguel Urioste; Daniel Rueda; Joan Brunet; Milagros Balbín; Pilar Blay; Silvia Iglesias; Pilar Garré; Enrique Lastra; Ana Beatriz Sánchez-Heras; Alfonso Valencia; Victor Moreno; Miguel Ángel Pujana; Alberto Villanueva; Ignacio Blanco; Gabriel Capellá; Jordi Surrallés; Xose S Puente; Laura Valle
Journal:  Gastroenterology       Date:  2015-06-05       Impact factor: 22.682

7.  Polymorphisms in XRCC1 and glutathione S-transferase genes and hepatitis B-related hepatocellular carcinoma.

Authors:  Ming-Whei Yu; Shi-Yi Yang; I-Jen Pan; Chih-Lin Lin; Chun-Jen Liu; Yun-Fan Liaw; Shi-Ming Lin; Pei-Jer Chen; Shou-Dong Lee; Chien-Jen Chen
Journal:  J Natl Cancer Inst       Date:  2003-10-01       Impact factor: 13.506

8.  Association of XRCC1, XRCC3, and XPD genetic polymorphism with an increased risk of hepatocellular carcinoma because of the hepatitis B and C virus.

Authors:  Asma Gulnaz; Ali H Sayyed; Farah Amin; Abrar ul Haq Khan; Muhammad A Aslam; Rehan S Shaikh; Muhammad Ali
Journal:  Eur J Gastroenterol Hepatol       Date:  2013-02       Impact factor: 2.566

9.  [Importance of expression of DNA repair proteins in non-small-cell lung cancer].

Authors:  S Rybárová; J Muri; I Hodorová; J Vecanová; M Benický; D Piovarči; P Janík; J Mihalik; L Mirossay
Journal:  Klin Onkol       Date:  2012

10.  XRCC1 and XPD polymorphisms and their relation to the clinical course in hepatocarcinoma patients.

Authors:  Qinghai Guan; Zhiqiang Chen; Qiangpu Chen; Xuting Zhi
Journal:  Oncol Lett       Date:  2017-07-05       Impact factor: 2.967
View more
  2 in total

1.  Comprehensive analysis of the correlation between base-excision repair gene SNPs and esophageal squamous cell carcinoma risk in a Chinese Han population.

Authors:  Yu Pu; Liang Zhao; Nan Dai; Mingfang Xu
Journal:  Mol Clin Oncol       Date:  2020-06-09

2.  Identification of Ten Core Hub Genes as Potential Biomarkers and Treatment Target for Hepatoblastoma.

Authors:  Rui Sun; Simin Li; Ke Zhao; Mei Diao; Long Li
Journal:  Front Oncol       Date:  2021-04-01       Impact factor: 6.244
  2 in total

北京卡尤迪生物科技股份有限公司 © 2022-2023.