RAD51 135G>C Polymorphism and Cancer Risk: An Updated Meta–Analysis Involving 54,239 Subjects

Gui-li Sun¹†*, Bei-Bei Zhang²†, Chao Xuan³, Kai- Feng Deng⁴, Ge Gao⁵,Li-Min Lun³

¹Department of Endocrinology, The Second Hospital of Nanning City, The Third Affiliated Hospital of Guangxi MedicalUniversity, Nanning 530000, PR China

²Graduate School of Medicine, Mie University, Mie, Japan

³Department of Clinical Laboratory, The Affiliated Hospital of Medical College, Qingdao University, Qingdao, PR China

⁴Department of Clinical Laboratory, The Fourth Affiliated Hospital of Guangxi Medical University, Liuzhou, PR China

⁵Center for Reproductive Medicine,Tianjin Central Hospital of Obstetrics and Gynecology,Tianjin,PR China † Contribute equally to this work;

*Corresponding author: : Gui-li Sun, MD, Department of Endocrinology, The Second Hospital of Nanning City,The Third Affiliated Hospital of Guangxi MedicalUniversity, Nanning, PR China

Received: February 05, 2014; Accepted: March 28, 2014; Published: April 07, 2014

Abstract

The RAD51 plays a pivotal role in homologous recombination repair of DNA double-strand breaks inducing chromosomal breaks and genomic instability. Previous studies yielded conflicting results for the association between RAD51 135G>C polymorphism and risk of cancer. The present study aimed at investigating the pooled association using a meta-analysis on the published studies, involving 27,895 cases and 26,344 controls to assess the effect of RAD51 135G>C on cancer susceptibility. Across all populations, our results indicated that significant associations were found between RAD51 135G>C polymorphism and risk of cancer under genotypic C allele vs. G allele (OR =1.36 95% CI: 1.31–1.41), CC vs. GG (OR =2.37 95% CI: 2.12–2.65), CC vs. CG (OR =4.02 95% CI: 3.62–4.46), recessive model (OR =3.74, 95% CI: 3.40–4.11), and dominant model (OR =1.08, 95% CI: 1.03–1.13). In subgroup analyses, similar associations were found among Caucasians but not Asians. Moreover, the significant associations were found in subgroups of breast cancer, hematologic malignances, colorectal cancer, endometrial cancer, and ovarian cancer. This meta–analysis suggests that the RAD51 135G>C polymorphism was associated with susceptibility of cancer. The effect of the variants on the expression levels and the possible functional role of the variants in cancer should be addressed in further studies.

Key words: RAD51, Polymorphism, Cancer risk, Meta–analysis

Introduction

Epidemiologic studies reveal a significant environmental contribution to the pathogenesis of cancer [1,2]. Familial aggregation and twin studies indicate that the presence of genetic factors are for susceptibility to this condition [3–5]. A number of genomic screens have been performed to find genetic linkage to cancer [6–8]. The faithful repair of DNA damage such as chromosomal double–strand breaks (DSBs) is crucial for genomic integrity [9]. DSBs may cause chromosomal breaks and genomic instability, thus increasing the probability of developing cancer [10]. Homologous recombination (HR), single–strand annealing and non–homologous end–joining are considered to be the main pathways for repairing the DSBs [11]. Among them, the central HR protein is RAD51 which ensures high fidelity DNA repair by facilitating strand exchange between amaged and undamaged homologous DNA segments [10]. Thus far, two SNPs (135G⁄C [rs1801320] and 172G⁄T [rs1801321]) were discovered in the 5 UTR of RAD51 [12]. The effect of 135G→C variant on the RAD51 was alternative splicing within the 5 UTR, while the latter SNP was found to have weak effect [13].

The genetic variations of RAD51 gene may contribute to the development and progression of cancers [14]. Many original studies have reported the role of RAD51 135G>C polymorphism and cancer risk, but the findings are inconclusive [15,16]. Partially, it may due to the fact that the RAD51 gene was a minor gene for risk of cancers and⁄or the relatively small sample–size in each published studies.Therefore, we performed this updated meta–analysis to derive a more precise estimation of the association between RAD51 135G>C polymorphism and cancer.

Materials and methods

Selection of published studies

Case–control studies reporting the association between the RAD51 135G>Cpolymorphisms and risk of cancer published in English before February 2013 were identified by comprehensive computer–based searches of Medline, EBSCO, and BIOSIS databases. The references of reviews and retrieved articles were also searched simultaneously to find additional eligible studies. The following keywords were used for searching: “RAD51” AND (“genetic variant*” or “genetic variation” or “polymorphism*”) AND (“cancer” or “carcinoma” or “tumor” or “leukemia” or “laukaemia”). The most complete and recent results were used when there were multiple publications from the same study group.

Two investigators reviewed all identified studies independently to determine whether an individual study was eligible for inclusion. The selection criteria for studies to be considered for this metaanalysis were as follows: 1) case–control or case–cohort study; 2) the RAD51135G>Cpolymorphism in cancer; 3) proper cancer diagnosis criteria; 4) original data; 5) not animal studies. The study would be excluded if the information could not be obtained.

Ethical consideration

The study has been approved by the Ethics committee of our Institutions.

Data extraction

The characteristics of selected studies were independently extracted through a standardized protocol by two authors, and the result was reviewed by a third investigator. The following information was sought from each study: first author, year of publication, study population (country, ethnicity), cancer types, the number of patients and controls for a study, genotype frequency for cases and controls, allele frequency in controls, and Hardy– Weinberg equilibrium (HWE).

Statistical analysis

Allele frequencies (–135C) at the RAD51 polymorphism from each respective study were determined by the allele counting method. Genotype distributions of controls were used to estimate the frequencies of the putative risk allele (–135C) using the inverse variance method [17,18]. The deviation from the Hardy–Weinberg Equilibrium (HWE) for distribution of the allele frequencies was analyzed by Fisher’s exact test in control groups, P < 0.05 was considered as representative of statistically significant. We examined the contrast of the C allele vs. G allele, CC vs. GG, CC vs. CG, and also examined the recessive genetic model (CC vs. CG+GG) and the dominant genetic model (CC+CG vs. GG). The associations between RAD51 (G135C) polymorphisms and cancer susceptibility were estimated by odds ratios (ORs) with 95% confidence intervals (CIs). The significance of the pooled OR was determined by the Z–test; P < 0.05 was considered statistically significant. Furthermore, to evaluate the ethnicity and cancer type–specific effects, subgroup analyses were performed.

Heterogeneity assumption was checked by a Chi–square based Q test, and it was considered statistically significant when P <0.1 [19]. Heterogeneity was also quantified with I² metric (I² = (Q–df)⁄Q×100%; I² < 25%, no heterogeneity; I² = 25–50%, moderate heterogeneity; I²=50–75%, large heterogeneity, I²>75%, extreme heterogeneity). When the effects were assumed to be homogenous (P > 0.1, I² < 50%), the fixed–effects model was used; otherwise, the random–effects model was more appropriate. Sensitivity analysis was performed to evaluate the stability of the results. If more than seven studies were included, Begg’s test was used to measure the publication bias which was shown as a funnel plot [20]. P < 0.05 was considered as representative of statistically significant publication bias. All analyses were performed using the software STATA software, version 12.0 (Stata Corporation, College Station, TX,USA) and R statistical software, version 2.15.2 (http:⁄⁄www.r–project.org).

Results

Characteristics of studies

A total of fifty studies that met the inclusion concerning the association between RAD51135G>C polymorphism and risk of cancer were considered in the meta–analysis [12,15,16,21–67]. These studies involved 27,895 patients and 26,344 controls, containing thirty–eight Caucasian, five Asian, and seven mixed studies. In subgroup analysis, thirty–eight Caucasian studies (14,180⁄12,726) and five Asian studies (1,946⁄2,945) were included in ethnic–specific group. Additionally, twenty–six (19,716⁄19,735) studies focusing on breast cancer, seven (2,169⁄3,629) studies focusing on hematologic malignances, four (753⁄720) studies focusing on colorectal cancer, three (500⁄506) studies focusing on endometrial cancer, three (1,085⁄1,160) studies focusing on head and neck cancer, and two (2,925⁄1,749) studies focusing on ovarian cancer were also respectively evaluated. 84% (42⁄50) of these studies included used polymerase chain reactionrestriction fragment length polymorphism (PCR–RFLP) analysis for genotyping. Main characteristics of included studies were listed in Table 1.

Table 1: Characteristics of the studies included in the meta-analysis.



  
    Target 
    Drug 
    Major mechanism of action 
    Antidepressant effects 
      in human 
    Antidepressant effects 
      in rodents 
    BDNF levels 
      in human 
    BDNF levels 
      in rodents 
    Reference 
  
  
    NMDAR 
    Ketamine
    Non-competitive    antagonist
    Yes
    Yes
    ↑ 
    ↑ 
    [67-72]
  
  
    NMDAR 
    MK-0657
    Selective NR2B antagonist
    Yes
     
    ↑ 
     
     
      [73]
  
  
    NMDAR 
    Acamprosate
    NMDA and mGluR5 antagonist
    Inconclusive
    Yes
    ↑ 
     
    [73-75]  
  
  
    NMDAR 
    Memantine
    Non-competitive    low-affinity antagonist
    No
     
     
    ↑ 
     
      [76-78]
  
  
    AMPAR 
    Ampakines
    Positive allosteric    modulator
    Yes
    Yes
     
    ↑ 
    [79-85]
  
  
    mGluR2/3 
    LY379268
    Agonist
    Clinical trials    undergoing
    Yes ↑ (enhancement with antidepressants)
     
    ↑ (with antidepressants)
    [86-90]
  
  
    mGluR2/3 
    LY341495
    Antagonist
    Clinical trials    undergoing
    Yes
     
    ↑ (enhancement with DOI)
    [91-92]
  
  
    mGluR5 
    MPEP
    Selective antagonist
     
    Yes
     
    ↑ (hippocampus) ↓ (cortex)
    [93-96]
  
  
    Other 
    Riluzole
    Reduces extra-synaptic    glutamate by inhibiting presynaptic release and enhancing glial uptake
    Yes (preliminary)
     
     
    ↑ 
    [97-99]



Table 1:  Characteristics of the studies included in the meta-analysis.

Frequency of the C allele in different groups

The pooled RAD51–135C frequencies were 17.77 % (95 % CI: 17.29 – 18.25 %), and 32.49 % (95 % CI: 30.66 % – 34.32 %) in the controls of Caucasian, and Asian population. Genotype distributions in the controls of all studies were in agreement with HWE, except ten studies [10, 23–25, 27, 39, 48, 49, 58].

Results of meta–analysis

For each study, we investigated the association between the 135G>C polymorphism and risk of cancer. Overall, RAD51 135 C allele was associated with a statistically increased risk of cancer, compared with the G allele (OR =1.36 95% CI: 1.31–1.41) under random–effect model. Significant associations were also observed in the genetic models for CC vs. GG (OR =2.37 95% CI: 2.12–2.65), CC vs. CG (OR =4.02 95% CI: 3.62–4.46), recessive model (OR =3.74, 95% CI: 3.40–4.11), and dominant model (OR =1.08, 95% CI: 1.03– 1.13, Figure 1.). Z–test indicated that the pooled ORs were statistically significant.