HOGG1 C1245G Gene Polymorphism Associated with Prostate Cancer: A Meta-Analysis

Abstract

Background

Prostate cancer (Pca) is one of the most frequently encountered multifactorial malignant diseases worldwide. The human oxoguanine glycosylase 1 (hOGG1) C1245G polymorphism (rs1052133) has been found to be associated with Pca. However, the conclusions have been controversial.

Methods

Based on the PubMed, Embase, HuGENet and Chinese National Knowledge Infrastructure (CNKI) databases, this meta-analysis was conducted with 4 models. Eleven qualified studies were included.

Results

Although no positive relation was discovered in the pooled analysis, significant associations between rs1052133 and Pca were found in the Asian population (recessive: odds ratio [OR] = 1.580, 95% confidence interval [95% CI], 1.189-2.098; GG vs. GC: OR = 1.504, 95% CI, 1.114-2.030; GG vs. CC: OR = 1.677, 95% CI, 1.201-2.342; allele analysis: OR = 1.249, 95% CI, 1.077-1.449), whites (dominant: OR = 2.138, 95% CI, 1.483-3.083; recessive: OR = 3.143, 95% CI, 1.171-8.437; GG vs. CC: OR = 3.992, 95% CI, 1.891-8.431; allele analysis: OR = 1.947, 95% CI, 1.467-2.586) and mixed populations (recessive: OR = 0.636, 95% CI, 0.484-0.834; GG vs. GC: OR = 0.654, 95% CI, 0.492-0.871; GG vs. CC: OR = 0.624, 95% CI, 0.473-0.823; allele analysis: OR = 0.857, 95% CI, 0.771-0.954). After excluding studies deviating from the Hardy-Weinberg equilibrium, a significant association was also found in the same ethnic groups. In addition, a new positive relation was identified in the “other country” subgroup (with China, South Korea and Australia included) (dominant: OR = 1.622, 95% CI, 1.163-2.261; recessive: OR = 1.773, 95% CI, 1.308-2.404; GG vs. GC: OR = 1.614, 95% CI, 1.169-2.230; GG vs. CC: OR = 2.108, 95% CI, 1.456-3.051; allele analysis: OR = 1.494, 95% CI, 1.235-1.808) and among the Chinese-Korean population.

Conclusions

In conclusion, we suggest that the hOGG1 C1245G polymorphism might be potentially associated with Pca risk in different ethnicities and countries, especially among Asians. Further studies are needed to confirm these relations.

Keywords

hOGG1 Human oxoguanine glycosylase 1 Polymorphism Prostate cancer

Introduction

Prostate cancer (Pca) is one of the most frequently encountered malignant diseases worldwide. It is the second leading cause of cancer mortality in Western countries (1). Among Asian countries, the incidence of Pca has also increased rapidly in recent years (2). It was estimated that 219,000 new cases of Pca and 27,000 deaths occurred in 2007 in the United States (3). The morbidity and death associated with this disease have also increased in recent years (4, 5). As a multifactorial disease, many factors are known to play a key role in Pca development, such as age, ethnicity, diet and geographic factors (6). Recently, hereditary factors have also been identified as being significantly associated with Pca (7), including the human oxoguanine glycosylase 1 (hOGG1) C1245G polymorphism (rs1052133) (8).

hOGG1 is located on 3p25 from 9749944 bp to 9787790 bp. It encodes for a glycosylase/AP lyase belonging to the DNA repair enzymes, and is hypothesized to play an important role in preventing carcinogenesis by repairing oxidative damage to DNA (9, 10). It is mainly responsible for recognizing and repairing 8-oxoguanine (8-oxoG) damage, which is a mutagenic base byproduct that occurs as a result of exposure to reactive oxygen species (ROS) (11, 12). There are several variants in this gene (http://www.ncbi.nlm.nih.gov/SNP/snp_ref.cgi?locusId=4968), among which the hOGG1 C1245G (rs1052133) gene polymorphism has been studied frequently. The rs1052133 polymorphism is situated in exon 7 of the hOGG1 gene; this polymorphism induces an amino acid substitution in codon 326 from serine to cysteine (13). It has been estimated that a nearly sevenfold greater DNA repair activity is carried out by the protein encoded by the 326Ser allele compared with 326Cys (13). Recently, this variant was found to be associated with many diseases, such as breast cancer (14), lung cancer (15) and urothelial carcinoma (16). Therefore, the status of this polymorphism might be important in oxidative DNA damage and the development and progression of many diseases. In addition, studies have suggested that rs1052133 might also play a key role in Pca risk. In 2002, Xu et al conducted a study of about 18 variants of the hOGG1 gene and Pca risk, in which 2 polymorphisms, including rs1052133, were shown to be significantly related to Pca risk (17). As a follow-up, Chen et al (18) showed that hOGG1 might have a role in the repair of 8-OH-dG adducts in prostate tissue and that the hOGG1 C1245G polymorphism is associated with Pca risk. However, the converse opinion has also been presented (19). Starting from these controversial conclusions, this meta-analysis was conducted. The present study was focused on this vital variant of the hOGG1 gene to provide new information on the relationship between the hOGG1 C1245G polymorphism and Pca risk, by analyzing different subgroups.

Methods and Materials

Search and Collection

Our eligible studies were mainly retrieved from the PubMed, Embase, HuGENet and Chinese National Knowledge Infrastructure (CNKI) databases. The following key words were used: “human 8-oxoguanine glycosylase 1”, “hOGG12”, “OGG12”, “OGG”, “hOGG”, “8-oxoguanine DNA glycosylase”, “8-oxoguanine glycosylase”, “Pca”, “prostate cancer”, “prostatic carcinoma” and “prostate carcinoma.” All the studies were published in English or Chinese, were primary research articles and were not replicated in other studies. Additional searches were collected from the references lists of eligible studies. It was estimated that we required about 2-3 weeks ensuring the veracity and comprehensiveness of this analysis. The latest search was performed in July 2014.

When selecting eligible studies, the inclusion criteria were as follows: (i) the essential contents were about the hOGG1 C1245G polymorphism and Pca risk; (ii) the study was designed as a case-control or cohort study, which distinguished the case group from controls; and (iii) the number of cases and controls for separate genotypes was provided. As for the exclusion criteria, the following studies were removed: (i) studies not focused on the association of Pca risk and the hOGG1 C1245G polymorphism but which were rather gene expression and functional studies; (ii) studies that were mainly conducted on Pca cells; (iii) studies that did not distinguish between cases and controls; (iv) studies with subjects who all had Pca; and (v) studies not providing raw data on the frequency of every genotype used in our analysis. A flowchart of the selection process is shown in Figure 1.

Fig. 1

Flowchart for study selection in this analysis.

When considering the studies included, we recoded the following characteristics: names of the authors; year of publication; country; ethnicity of the samples (categorized as white, Asian, African and mixed); methods of detection; the definition of cases and controls; the number of cases and controls for each studies. All of this information is shown in Table I.

TABLE I

Characteristics of studies included in analysis

Author (ref.)	Date	Country	Race	Method	Case	Control	Case G	Case C	Control G	Control C	No. of cases	No. of controls	HWE
Zhou et al (8)	2013	China	Asian	PCR-RFLP	Pathologically confirmed Pca	Cancer-free	104	96	85	115	100	100	Y
Mandal et al (20)	2012	India	Asian	PCR	Histologically confirmed Pca	Cancer-free, unrelated, age-matched healthy	110	274	121	327	192	224	Y
Mittal et al (21)	2012	India	Asian	PCR	Histologically confirmed Pca	Cancer-free, unrelated, age-matched healthy	111	279	128	372	195	250	Y
Yun et al (22)	2012	South Korea	Asian	PCR	Histologically confirmed primary adenocarcinoma of the prostate	BPH patients	305	227	265	267	266	266	Y
Zhang et al (23)	2010	USA	Mix	MS	Primary, incident, histologically confirmed Pca	Health gave rise to cases	69	313	85	307	191	196	Y
Lavender et al (19)	2010	USA	African	TaqMan-PCR	Pca	Health	66	322	215	1,077	194	646	Y
Dhillon et al (24)	2011	Australia	White	PCR	Histologically confirmed Pca	Age-matched controls	99	133	74	188	116	131	Y
Nock et al (25)	2006	USA	Mix	PCR	Pca	Unaffected brothers	183	695	204	752	439	478	N
Nam et al (26)	2005	Canada	Mix	MS	Adenocarcinoma of the prostate at biopsy	No evidence of invasive cancer	456	1,536	564	1,620	996	1,092	N
Chen et al (18)	2003	USA	White	RT-PCR	Primary adenocarcinoma of the prostate	No previous diagnosis of cancer	41	127	69	433	84	251	Y
Xu et al (17)	2002	USA	Mix	PCR	Hereditary Pca families and unrelated prostate cancer	Non-prostate cancer	126	470	93	255	298	174	Y

BPH = Benign Prostate Hyperplasia; HWE = Hardy-Weinberg equilibrium; MS = mass spectrometry-based genotyping analysis; N = no, does not satisfy conditions of the HWE; PCR = polymerase chain reaction; RFLP =; RT-PCR = RT-PCR = reverse transcription-polymerase chain reaction; Y = yes, satisfies conditions of the HWE.

Statistical Analysis

The frequencies of every genotype were included in the subsequent analysis. To describe the association between the hOGG1 C1245G polymorphism and Pca risk, we selected 4 common genotype models: the dominant (GG + GC vs. CC), recessive (GG vs. GC + CC) and codominant models (GG vs. GC; GG vs. CC) as well as per allele analysis (G vs. C). As for the effect models, I² was treated as standard (when I²<50%, the fixed effect was applied; otherwise, the random effect was selected). The effects incorporated an estimate of the interstudy variance and provided wider 95% confidence intervals (95% CI) if the results of the constituent studies differed among themselves. In all analyses, the odds ratio (OR) was used to evaluate the effect of the association with the corresponding 95% CI. Pooled ORs were estimated according to the individual ORs. Significant results were defined by the 95% CI of the OR values. Meanwhile, the chi-square-based Q statistic (p<0.10 as the standard) was applied to estimate heterogeneity, which represented the weighted sum of the squared difference in the overall effect sizes from each study (27).

In this study, subgroup analysis was conducted by collecting similar characteristics from eligible studies, such as country and ethnicity. Among the eligible studies, 6 countries were included: China, India, South Korea, United States, Australia and Canada, which included subjects of different ethnicities including Asian, white, African and mixed. On the basis of the countries, 2 different categories were applied: (i) India, United States and other; (ii) China-Korea, Indian Subcontinent, North America and Oceania. As for ethnicities, Asian, white and mixed were taken into consideration. In addition, to confirm our results and discover new positive signals, the Hardy-Weinberg equilibrium (HWE; p>0.05) in the control group was tested using Stata software (28). We excluded the divergent studies which did not achieve HWE. Then the meta-analysis was conducted with the remaining data.

Publication bias was evaluated using funnel plots and Begg's unweighted regression tests for every genotype model. All p values were 2-tailed, and the data were analyzed using Stata 9.0 (Stata Corporation, College Station, TX, USA).

Results

Characteristics of the Retrieval and Eligible Studies

In the retrieval, 4 databases (PubMed, HuGENet, Embase and CNKI) were searched after combining the key words. Finally, 163 studies (29 in PubMed, 61 in Embase, 15 in HuGENet and 58 in CNKI) were obtained. In the next stage, all of the abstracts were scanned. First, repeated studies in the 4 databases were removed. Then, according to our inclusion and exclusion criteria, 18 articles were left, which were found to investigate the association between polymorphisms in the hOGG1 gene and Pca risk. After a careful reading of all of the texts, 8 studies were excluded for the following reasons: (i) 1 study mainly discussed the relationship between variations in the hOGG1 gene and Pca recurrence, in which the samples were obtained from patients with Pca (29); (ii) although 1 study was focused on the hOGG1 gene and Pca, the specific frequencies of every genotype in the cases and controls were undefined (30); (iii) 1 study concentrated on rs3218997 and Pca risk (31); (iv) 4 studies were meta-analyses regarding hOGG1 gene polymorphisms and Pca risk (32–35); and (v) 1 study did not provide raw data on the numbers of each genotype (36). In addition, to ensure that as many eligible studies were included as possible, the reference lists in the articles were checked; 1 study with accessible data was discovered (19). Finally, 11 qualifying studies were included in this meta-analysis (8, 17–26) (Tab. I). The funnel plot for allele analysis (G vs. C) as an example is presented in Figure 2. All of the p values for publication bias were also evaluated with Begg's test.

Fig. 2

Begg's funnel plot for the allele analysis (G vs. C).

Among these eligible studies, the samples all came from local medical institutions or the general population. Cases were mainly defined as Pca patients who had been confirmed by histopathological examination. Patients who were cancer-free or who had other diseases (such as benign prostate hyperplasia) were identified as the controls, and were matched to the cases by age or other features. The participants in 4 of the eligible studies were Asian, mainly coming from China and India (8, 20–22). The other 4 studies from the United States and Canada included mixed ethnicities from Europe, Africa and other regions (17, 23, 25, 26). Two studies included whites from Australia and the United States (18, 24). The last study included subjects of African descent in the United States (19).

Results Assessment

In this meta-analysis, to ensure our that conclusions were more believable and valuable, we adopted many methods to avoid study bias and improve the power of effect sizes as much as possible: (i) The meta-analysis was mainly based on 4 authoritative databases (PubMed, HuGENet, Embase and CNKI) with comprehensive key words, which should help to include as many eligible studies as possible. The strict retrieval process should also have helped to ensure the veracity of the whole statistical analysis. (ii) Samples were mainly humans. And cases were defined as Pca patients with a diagnosis confirmed by histopathological examination. Cancer-free patients, those with other diseases (such as benign prostate hyperplasia) and healthy participants were identified as controls, who were matched with the cases by age or other features. The explicit definitions reduce the bias to some extent. (iii) In the processes of analysis, 4 common genotype models – the dominant (GG + GC vs. CC), recessive (GG vs. GC + CC) and codominant model (GG vs. GC; GG vs. CC) and per allele analysis (G vs. C) – and different subgroup analyses were applied in this study. This method may also reduce the study bias and improve the power of effect sizes. And (iv) last but not least, the HWE was applied. Studies not meeting the HWE were excluded in order to ensure the stability of the samples, which reduces the study and sample biases vastly. Thus, using these criteria and methods, we can propose that the results of our meta-analysis should be significant for further research.

hOGG1 C1245G Polymorphism Associated with Pca Risk in Different Ethnicities

In the first analysis, all 11 studies were included. The subgroups were mainly classified by country and ethnicity. For the ethnicities, Asian, white and mixed populations were identified. Although we did not find a positive relation in the pooled analysis, significant associations between the hOGG1 C1245G polymorphism and Pca risk were found in the Asian population (dominant model: OR = 1.216, 95% CI, 0.984-1.503, I² = 0%; recessive model: OR = 1.580, 95% CI, 1.189-2.098, I² = 0%; codominant model [GG vs. GC]: OR = 1.504, 95% CI, 1.114-2.030, I² = 0%; codominant model [GG vs. CC]: OR = 1.677, 95% CI, 1.201-2.342, I² = 0%; allele analysis: OR = 1.249, 95% CI, 1.077-1.449, I² = 0%), the white population (dominant model: OR = 2.138, 95% CI, 1.483-3.083, I² = 0%; recessive model: OR = 3.143, 95% CI, 1.171-8.437, I² = 41.2%; codominant model [GG vs. GC]: OR = 2.150, 95% CI, 0.828-5.581, I² = 33.6%; codominant model [GG vs. CC]: OR = 3.992, 95% CI, 1.891-8.431, I² = 6.9%; allele analysis: OR = 1.947, 95% CI, 1.467-2.586, I² = 0%) and mixed populations (dominant model: OR = 0.886, 95% CI, 0.778-1.008, I² = 0%; recessive model: OR = 0.636, 95% CI, 0.484-0.834, I² = 0%; codominant model [GG vs. GC]: OR = 0.654, 95% CI, 0.492-0.871, I² = 0%; codominant model [GG vs. CC]: OR = 0.624, 95% CI, 0.473-0.823, I² = 0%; allele analysis: OR = 0.857, 95% CI, 0.771-0.954, I² = 0%) (Tab. II). The forest plot for the recessive model is shown in Figure 3. For the country subgroups, there were 2 different analytical methods. As the first one, countries were divided into three groups (India, United States and other countries), in which we could not find any potentially significant signal for any country (Suppl Tab. I. Available online at www.biological-markers.com). Then, when the countries were split into China-Korea, Indian Subcontinent, North America and Oceania, it was found that, in the Chinese and Korean population, rs1052133 might play a key role in Pca risk (dominant model: OR = 1.376, 95% CI, 0.976-1.940, I² = 0%; recessive model: OR = 1.702, 95% CI, 1.221-2.373, I² = 0%; codominant model [GG vs. GC]: OR = 1.628, 95% CI, 1.144-2.318, I² = 0%; codominant model [GG vs. CC]: OR = 1.893, 95% CI, 1.249-2.868, I² = 0%; allele analysis: OR = 1.383, 95% CI, 1.126-1.700, I² = 0%) (Suppl Tab. II. Available online at www.biological-markers.com).

TABLE II

Results of meta-analysis with dominant model (GG+GC vs. CC), recessive model (GG vs. GC+CC), codominant model (GG vs. GC; GG vs. CC) and allele analysis (G vs. C)

Model	Subgroups	OR	95% CI	I²	p Value	p Value for bias
Dominant model (GG+GC vs. CC)	Asian	1.216	0.984-1.503	0.00%	0.812	–
Recessive model (GG vs. GC+CC)	Asian	1.58	1.189-2.098	0.00%	0.741	–
Codominant model (GG vs. GC)	Asian	1.504	1.114-2.030	0.00%	0.769	–
Codominant model (GG vs. CC)	Asian	1.677	1.201-2.342	0.00%	0.708	–
Allele analysis (G vs. C)	Asian	1.249	1.077-1.449	0.00%	0.544	–
Dominant model (GG+GC vs. CC)	White	2.138	1.483-3.083	0.00%	0.724	–
Recessive model (GG vs. GC+CC)	White	3.143	1.171-8.437	41.20%	0.192	–
Codominant model (GG vs. GC)	White	2.15	0.828-5.581	33.60%	0.22	–
Codominant model (GG vs. CC)	White	3.992	1.891-8.431	6.90%	0.3	–
Allele analysis (G vs. C)	White	1.947	1.467-2.586	0.00%	0.814	–
Dominant model (GG+GC vs. CC)	Mixed	0.886	0.778-1.008	0.00%	0.671	–
Recessive model (GG vs. GC+CC)	Mixed	0.636	0.484-0.834	0.00%	0.448	–
Codominant model (GG vs. GC)	Mixed	0.654	0.492-0.871	0.00%	0.603	–
Codominant model (GG vs. CC)	Mixed	0.624	0.473-0.823	0.00%	0.394	–
Allele analysis (G vs. C)	Mixed	0.857	0.771-0.954	0.00%	0.512	–
Dominant model (GG+GC vs. CC)	All	1.126	0.943-1.344	61.00%	0.004	0.025
Recessive model (GG vs. GC+CC)	All	1.122	0.759-1.657	72.00%	<0.001	0.573
Codominant model (GG vs. GC)	All	1.047	0.744-1.473	59.30%	0.006	0.657
Codominant model (GG vs. CC)	All	1.192	0.765-1.859	75.50%	<0.001	0.283
Allele analysis (G vs. C)	All	1.123	0.943-1.337	76.10%	<0.001	0.05

vs. = verses three race subgroups were divided: Asian, white and mixed races.

95% CI = 95% confidence interval; OR = odds ratio.

Fig. 3

Meta-analysis with random effects and recessive model (GG vs. GC+CC) for the association between hOGG1 C1245G polymorphism and prostate cancer. The first author and year of publication for each study are shown. In this analysis, 3 race subgroups are shown: Asian, white and mixed races. Odds ratio (OR) and accompanying 95% confidence interval (95% CI) are also presented for the association of the hOGG1 C1245G polymorphism and prostate cancer.

hOGG1 C1245G Polymorphism Associated with Pca Risk in the Country Subgroups

To ensure the correctness of our results, we filtered our included studies again by testing the HWE in every article. After excluding studies that did not satisfy the HWE, all analyses were conducted. In the ethnicities group, the same association between the hOGG1 C1245G polymorphism and Pca risk was also discovered for the same genotype models (Suppl Tab. III. Available online at www.biological-markers.com). Meanwhile, to our surprise, in the “other country” subgroup (with China, South Korea and Australia included), a latent relationship between this polymorphism and Pca risk was also identified for the 4 different genotype models (dominant model: OR = 1.622, 95% CI, 1.163-2.261, I² = 22.8%; recessive model: OR = 1.773, 95% CI, 1.308-2.404, I² = 0%; codominant model [GG vs. GC]: OR = 1.614, 95% CI, 1.169-2.230, I² = 0%; codominant model [GG vs. CC]: OR = 2.108, 95% CI, 1.456-3.051, I² = 0%; allele analysis: OR = 1.494, 95% CI, 1.235-1.808, I² = 7.8%). The forest plot for the recessive model is shown in Figure 4. Even so, a positive association was not detected in India and the United States (Suppl Tab. IV. Available online at www.biological-markers.com). In addition, there were no new distinct associations discovered in the “other country” subgroup analysis (Suppl Tab. V. Available online at www.biological-markers.com).

Fig. 4

Meta-analysis with random effects and recessive model (GG vs. GC+CC) for the association between hOGG1 C1245G polymorphism and prostate cancer after removing the studies that did not satisfy the criteria for the Hardy-Weinberg equilibrium. The first author and year of publication for each study are shown. In this analysis, 3 country subgroups are shown: India, United States and other countries. Odds ratio (OR) and accompanying 95% confidence interval (95% CI) are also presented for the association of the hOGG1 C1245G polymorphism and prostate cancer.

Discussion

Pca is a serious cancer and influences a large number of populations. As one aspect of a multifactorial disease, genetic mutation is known to play a key role in Pca development and progression. Recently, the hOGG1 C1245G polymorphism was found to be associated with Pca risk (8). However, the conclusion was inconclusive, so this meta-analysis was conducted. Our results suggested that the hOGG1 C1245G polymorphism might be a hereditary factor regulating Pca.

The hOGG1 gene encodes for a DNA glycosylase/AP lyase, which is mainly responsible for repairing oxidative damage to DNA (9). The mutation of this gene could lead to an alteration in heredity. Correspondingly, the transcription and translation of DNA could be influenced, which could result in many diseases, especially cancers. In 2012, Ma et al (37) suggested that the hOGG1 C1245G polymorphism may contribute to susceptibility to bladder cancer in a Chinese population. Meanwhile, in the same population, this polymorphism was found to be associated with susceptibility to childhood acute lymphoblastic leukemia (38). In 2009, Okasaka et al (39) also showed that the hOGG1 C1245G polymorphism might contribute to the risk of lung adenocarcinoma. In addition, other cancers such as urothelial carcinoma (16) were also presented. On the basis of these studies and the function of the hOGG1 gene, we believe that the hOGG1 C1245G polymorphism is important in regulating the development and progression of cancer.

As a common cancer, Pca is thought to be influenced by the hOGG1 C1245G polymorphism. In 2002, Xu et al (17) investigated hOGG1 sequence variants and Pca susceptibility, and suggested that hOGG1 C1245G is important in determining Pca risk. Then, in 2011, Dhillon et al (24) showed that Pca risk is significantly increased only in carriers of the G allele of the C1245G polymorphism of the hOGG1 gene. Moreover, Yun et al (22) confirmed a significant relationship in the Korean population. On the basis of these studies, this meta-analysis was conducted. In the first stage of analysis, 4 genotype models were applied in the ethnicity and country subgroups. Although there was no significant association discovered in the pooled analysis, the hOGG1 C1245G polymorphism seemed to be positively related to Pca risk in the Asian, white and mixed populations. In 2013, Zhou et al (8) investigated hOGG1 genetic variants and Pca risk, and found that the hOGG1 C1245G gene polymorphism might increase susceptibility to Pca in the Chinese population. This conclusion was also confirmed in the Korean population (23). As for the white population, Dhillon et al (24) and Chen et al (18) discovered a significant relationship between the hOGG1 C1245G polymorphism and Pca risk in Australia and United States, respectively. In addition, the association was also found in a mixed population, identified by Xu et al (17) in the white and African American populations. Thus, considering these studies, we can understand the potential function of the hOGG1 C1245G polymorphism in regulating Pca risk. To confirm our results further, the second stage of analysis was conducted, in which studies divergent from the HWE were excluded. Fortunately, our conclusion was confirmed.

In addition, we also conducted a comprehensive analysis in country subgroups with 2 different analytical methods: (i) the countries were divided into India, United States and other country; and (ii) countries were divided into China-Korea, Indian Subcontient, North America and Oceania. In the first category, we could not identify a possible association between the hOGG1 C1245G polymorphism and Pca risk in any of the country groups (India, United States and other). However, when 1 study was removed according to the HWE analysis (26), a positive relationship in the “other country” group was discovered, which suggests that, in different countries, the hOGG1 C1245G polymorphism might also contribute to the development and progression of Pca. We then tried to describe the features of this study. Firstly, this study diverged from the HWE, which suggests that its results cant not represent the general population. Secondly, the control samples in this study were with no evidence of invasive cancer. The definition seemed to be ambiguous which might result in bias. Meanwhile, the population in this study was mixed, which also reduced the function of representativeness by increasing deviation in the results. As for the second category (countries were divided into China-Korea, Indian Subcontinent, North America and Oceania), the results confirmed that, among the Asian population, rs1052133 might play a key role in the development and progression of Pca.

Combining the first- and second-stage analyses, we suggest that the potential association between the hOGG1 C1245G polymorphism and Pca risk cannot be ignored. More studies are needed to confirm our conclusions and the real associations in different countries and ethnicities. However, this study provides a new direction for genetically targeted therapy in the future.

Limitations

After a comprehensive meta-analysis, a relationship between Pca and the hOGG1 C1245G polymorphism was found. However, there were still some limitations to this study. Firstly, the number of studies on the association between the hOGG1 C1245G polymorphism and Pca is limited, especially for different ethnicities and countries, and may not illustrate the real function of this polymorphism in the development and progression of Pca. Secondly, the populations in the studies included were from different areas, especially for the mixed ethnicities subgroup, which resulted in some heterogeneity in the results. Finally, the definitions of cases and controls had some differences, which could also influence the results.

Conclusion

Pca is a global cancer. The hOGG1 C1245G polymorphism has been found to be associated with Pca risk. To make the relationship clear, this study was conducted. Our results suggest that the hOGG1 C1245G polymorphism is significantly associated with Pca, especially in the Asian population. However, the function of this locus for the risk of Pca in India and the United States remains uncertain. Further studies are needed to confirm these correlations.

Footnotes

Financial support: This study was supported by grants from the Guangxi science and technology project (no. 1355005-3-17).

Conflict of interest: None of the authors has any financial interest related to this study to disclose.

References

Jemal

Siegel

Ward

Cancer statistics, 2008.

CA Cancer J Clin 2008;58(2):71–96 PubMed: 18287387.

McCracken

Olsen

Chen

Cancer incidence, mortality, and associated risk factors among Asian Americans of Chinese, Filipino, Vietnamese, Korean, and Japanese ethnicities.

CA Cancer J Clin 2007;57(4):190–205 PubMed: 17626117.

Jemal

Siegel

Ward

Murray

Thun

Cancer statistics, 2007.

CA Cancer J Clin 2007;57(1):43–66 PubMed: 17237035.

Jemal

Siegel

Ward

Hao

Thun

Cancer statistics, 2009.

CA Cancer J Clin 2009;59(4):225–249 PubMed: 19474385.

Siegel

DeSantis

Virgo

Cancer treatment and survivorship statistics, 2012.

CA Cancer J Clin 2012;62(4):220–241 PubMed: 22700443.

Bostwick

Burke

Djakiew

Human prostate cancer risk factors.

Cancer 2004;10110 Suppl:2371–2490 PubMed: 15495199.

Chen

The RTK/ERK pathway is associated with prostate cancer risk on the SNP level: a pooled analysis of 41 sets of data from case-control studies.

Gene 2014;534(2):286–297 PubMed: 24177231.

Zhou

Xie

Lin

Yang

Mao

Cheng

Susceptibility of XPD and hOGG1 genetic variants to prostate cancer.

Biomed Rep 2013;1(4):679–683 PubMed: 24649009.

Rosenquist

Zharkov

Grollman

Cloning and characterization of a mammalian 8-oxoguanine DNA glycosylase.

Proc Natl Acad Sci USA 1997;94(14):7429–7434 PubMed: 9207108.

10.

Boiteux

Radicella

The human OGG1 gene: structure, functions, and its implication in the process of carcinogenesis.

Arch Biochem Biophys 2000;377(1):1–8 PubMed: 10775435.

11.

Kamiya

Miura

Ishikawa

Inoue

Nishimura

Ohtsuka

c-Ha-ras containing 8-hydroxyguanine at codon 12 induces point mutations at the modified and adjacent positions.

Cancer Res 1992;52(12):3483–3485 PubMed: 1596906.

12.

Boiteux

Radicella

The human OGG1 gene: structure, functions, and its implication in the process of carcinogenesis.

Arch Biochem Biophys 2000;377(1):1–8 PubMed: 10775435.

13.

Kohno

Shinmura

Tosaka

Genetic polymorphisms and alternative splicing of the hOGG1 gene that is involved in the repair of 8-hydroxyguanine in damaged DNA.

Oncogene 1998;16(25):3219–3225 PubMed: 9681819.

14.

Xie

Xia

Rong

Chen

Genetic polymorphism in hOGG1 is associated with triple-negative breast cancer risk in Chinese Han women.

Breast 2013;22(5):707–712 PubMed: 23369609.

15.

Cheng

Wang

Song

Kang

Xia

hOGG1, p53 genes, and smoking interactions are associated with the development of lung cancer.

Asian Pac J Cancer Prev 2012;13(5):1803–1808 PubMed: 22901126.

16.

Wang

Yeh

Shen

Association of hOGG1 and XPD polymorphisms with urothelial carcinoma in Taiwan.

Anticancer Res 2011;31(11):3939–3944 PubMed: 22110223.

17.

Zheng

Turner

Associations between hOGG1 sequence variants and prostate cancer susceptibility.

Cancer Res 2002;62(8):2253–2257 PubMed: 11956079.

18.

Chen

Elahi

Pow-Sang

Lazarus

Park

Association between polymorphism of human oxoguanine glycosylase 1 and risk of prostate cancer.

J Urol 2003;1706 Pt 1:2471–2474 PubMed: 14634453.

19.

Lavender

Komolafe

Benford

No association between variant DNA repair genes and prostate cancer risk among men of African descent.

Prostate 2010;70(2):113–119 PubMed: 19760636.

20.

Mandal

Gangwar

Kapoor

Mittal

Polymorphisms in base-excision & nucleotide-excision repair genes & prostate cancer risk in north Indian population.

Indian J Med Res 2012;135(1):64–71 PubMed: 22382185.

21.

Mittal

Mandal

Gangwar

Base excision repair pathway genes polymorphism in prostate and bladder cancer risk in North Indian population.

Mech Ageing Dev 2012;133(4):127–132 PubMed: 22019847.

22.

Yun

Chae

Kim

The hOGG1 mutant genotype is associated with prostate cancer susceptibility and aggressive clinicopathological characteristics in the Korean population.

Ann Oncol 2012;23(2):401–405 PubMed: 21515665.

23.

Zhang

Dhakal

Greene

Lang

Kadlubar

Polymorphisms in hOGG1 and XRCC1 and risk of prostate cancer: effects modified by plasma antioxidants.

Urology 2010;75(4):779–785 PubMed: 19914697.

24.

Dhillon

Yeoh

Fenech

DNA repair gene polymorphisms and prostate cancer risk in South Australia: results of a pilot study.

Urol Oncol 2011;29(6):641–646 PubMed: 19914098.

25.

Nock

Cicek

Polymorphisms in estrogen bioactivation, detoxification and oxidative DNA base excision repair genes and prostate cancer risk.

Carcinogenesis 2006;27(9):1842–1848 PubMed: 16569655.

26.

Nam

Zhang

Jewett

The use of genetic markers to determine risk for prostate cancer at prostate biopsy.

Clin Cancer Res 2005;11(23):8391–8397 PubMed: 16322300.

27.

Lau

Ioannidis

Schmid

Quantitative synthesis in systematic reviews.

Ann Intern Med 1997;127(9):820–826 PubMed: 9382404.

28.

Egger

Davey Smith

Schneider

Minder

Bias in meta-analysis detected by a simple, graphical test.

BMJ 1997;315(7109):629–634 PubMed: 9310563.

29.

Dluzniewski

Wang

Zheng

Variation in IL10 and other genes involved in the immune response and in oxidation and prostate cancer recurrence.

Cancer Epidemiol Biomarkers Prev 2012;21(10):1774–1782 PubMed: 22859398.

30.

Grundmark

Zethelius

Garmo

Holmberg

Serum levels of selenium and smoking habits at age 50 influence long term prostate cancer risk: a 34 year ULSAM follow-up.

BMC Cancer 2011;11(1):431 PubMed: 21982398.

31.

Agalliu

Kwon

Salinas

Koopmeiners

Ostrander

Stanford

Genetic variation in DNA repair genes and prostate cancer risk: results from a population-based study.

Cancer Causes Control 2010;21(2):289–300 PubMed: 19902366.

32.

Zhang

The hOGG1 Ser326Cys polymorphism and prostate cancer risk: a meta-analysis of 2584 cases and 3234 controls.

BMC Cancer 2011;11(1):391 PubMed: 21914193.

33.

Tong

Chen

Contribution of HOGG1 Ser³²⁶Cys polymorphism to the development of prostate cancer in smokers: meta-analysis of 2779 cases and 3484 controls.

PLoS ONE 2012;7(1):e30309 PubMed: 22279581.

34.

Zhu

Zhang

Tang

Wang

Polymorphisms in XPD and hOGG1 and prostate cancer risk: a meta-analysis.

Urol Int 2012;89(2):233–240 PubMed: 22890093.

35.

Wang

Chen

The hOGG1 Ser326Cys polymorphism contributes to cancer susceptibility: evidence from 83 case-control studies.

Mutagenesis 2012;27(3):329–336 PubMed: 22121210.

36.

Walsh

Associations between hOGG1 sequence variants and prostate cancer susceptibility.

J Urol 2002;1684 Pt 1:1636 PubMed: 12356046.

37.

Chu

Wang

hOGG1 Ser326Cys polymorphism is associated with risk of bladder cancer in a Chinese population: a case-control study.

Cancer Sci 2012;103(7):1215–1220 PubMed: 22463382.

38.

Huang

Rong

hOGG1 Ser326Cys polymorphism and risk of childhood acute lymphoblastic leukemia in a Chinese population.

Cancer Sci 2011;102(6):1123–1127 PubMed: 21401806.

39.

Okasaka

Matsuo

Suzuki

hOGG1 Ser326Cys polymorphism and risk of lung cancer by histological type.

J Hum Genet 2009;54(12):739–745 PubMed: 19881468.