Sage Journals: Discover world-class research

Abstract

Background

Previous studies have investigated the relationship between manganese superoxide dismutase (MnSOD) Val16Ala polymorphism and prostate cancer susceptibility, but the results have remained controversial. This meta-analysis was therefore performed to clarify this association.

Methods

The databases PubMed, Embase and Web of Science were searched for relevant available studies. Pooled odds ratios (ORs) with 95% confidence intervals (CIs) were calculated to evaluate the strength of the association. Publication bias was estimated using Begg's funnel plots and Egger's regression test. Trial sequential analysis was used to reduce the risk of type I error and estimate whether the evidence of the results was sufficient.

Results

Overall, a significant increased risk of prostate cancer was associated with MnSOD Val16Ala polymorphism for the heterozygote model (OR = 1.14; 95% CI, 1.05-1.24), homozygote model (OR = 1.18; 95% CI, 1.02-1.36), dominant model (OR = 1.24; 95% CI, 1.07-1.44) and recessive model (OR = 1.10; 95% CI, 0.96-1.24). In the subgroup analysis by genotyping method, the results were statistically significant for the TaqMan and PCR-RFLP methods. In addition, when stratified by sample size, statistically significant increased risks were found among both large samples and small samples. Furthermore, when stratified by source of control, significant results were detected in both population-based controls and hospital-based controls. By trial sequential analyses, these findings in the current study were shown to be based on sufficient evidence.

Conclusions

This meta-analysis indicated that the Ala allele of the MnSOD gene polymorphism increases prostate cancer susceptibility.

Keywords

Gene polymorphism Meta-analysis MnSOD Prostate cancer Val16Ala

Introduction

Prostate cancer (PCa), as the most common malignancy, is the second leading cause of cancer-related deaths among men in Western populations (1, 2). Estimated new PCa cases and mortality numbers were as high as 238,590 and 29,720, respectively, in 2013 (3).

Moreover, the etiology of PCa is complicated, and a series of factors have been reported to be involved in PCa susceptibility, such as smoking, age, dietary habits, endocrine system, environment and genetic factors (4-5-6-7-8). Though an accurate account of its pathogenesis is not fully clear, it has been suggested that 16% to 45% of the risk of PCa might be explained by genetic factors (9). Furthermore, it has been shown that common genetic polymorphisms are likely to play an important role in sporadic cases of PCa (10).

Manganese superoxide dismutase (MnSOD), as one of the most important enzymes to defend against reactive oxygen species (ROS) in the mitochondria, has been identified as a potential tumor suppressor gene (11). MnSOD is a homotetramer including 2 identical subunits, each consisting of a C-terminal α/β domain and N-terminal helical loop (12). Besides, MnSOD is encoded by the superoxide dismutase (SOD2) gene located at chromosome subband 6q25.3, which is a single-copy gene with 5 exons and 4 introns (13-14-15). Recently, it has been demonstrated that a T-to-C substitution in the SOD2 gene could result in a valine (Val) to alanine (Ala) amino acid change, thus causing higher MnSOD enzymatic activity. Moreover, a growing number of studies have discovered a connection between the MnSOD Val16Ala polymorphism (rs4880) and the risk of PCa (16-17-18-19-20-21-22-23-24-25-26-27-28-29-30). Therefore, this single-nucleotide polymorphism might play an important role in prostate carcinogenesis.

To date, previous studies have investigated the association between MnSOD Val16Ala polymorphism and PCa susceptibility (16-17-18-19-20-21-22-23-24-25-26-27-28-29-30). However, the results have remained inconsistent or even contradictory. In addition, lack of research using different stratified analyses has prevented a comprehensive understanding of the relationship between MnSOD Val16Ala polymorphism and the risk of PCa in some recent meta-analyses (31-32-33-34). Therefore, to clarify the nature of the association, a meta-analysis was conducted including all eligible articles to gather the statistical evidence.

Materials and methods

A comprehensive search was conducted on PubMed, Embase and Web of Science for relevant studies, covering all papers published until November 30^th, 2014. Combinations of the following search items were utilized: “Manganese superoxide dismutase” or “MnSOD”; “polymorphism” or “rs4880” or “VAL16ALA”; and “Prostate.” Additional eligible studies were hand-searched from references to original studies or reviews. However, only the study with the latest and/or largest sample size was included if studies had partly overlapping or similar subjects.

The case-control studies included had to meet the following inclusion criteria: (i) the diagnosis of PCa was confirmed pathologically; (ii) the controls were a healthy population with no evidence of PCa, or patients confirmed as free from PCa (e.g., result of a digital rectal examination was normal, and serum prostate-specific antigen was >4 ng/mL); and (iii) sufficient data were available for estimating an odds ratio (OR) with 95% confidence interval (CI). The major exclusion criteria were as follows: (i) without available genotype frequency data; (ii) without controls; and (iii) duplicate of previous publication.

Data Extraction

The studies identified were carefully reviewed by 2 investigators (X. L. and K. L.) independently to determine whether an individual study was eligible for inclusion. Data were independently extracted from the studies involved, and disagreements were resolved by discussion. All of the information was gathered and recorded on a standardized form including the following data of each study: first author's name, year of publication, ethnicity of study population, source of controls, genotyping method, number of cases and controls, frequency of Val16Ala gene polymorphism in cases and controls, respectively, and results of the Hardy-Weinberg equilibrium (HWE) test.

Statistical Analysis

The pooled ORs with 95% CIs were applied to evaluate the strength of association between MnSOD gene polymorphisms and PCa susceptibility. The fixed-effects model (Mantel-Haenszel method) and the random-effects model (DerSimonian-Laird method) were respectively utilized to pool the data (35). The random-effects model was more appropriate if heterogeneity was detected. Sensitivity analysis was performed by calculating the results again by omitting 1 single study each time. After that, subgroup analysis was further carried out by ethnicity, genotyping methods, sample size and source of controls. The sample size of an individual study was regarded as large, if the total number of cases and controls was larger than 1,000. Otherwise, the sample size was considered small. Begg's funnel plots and Egger's linear regression test were used to examine any publication bias between the studies, and a p value >0.05 was considered statistically significant (36). HWE was evaluated by the goodness-of-fit chi-square test, and a p value >0.05 was regarded as indicating a significant selection bias (37). Stata software (version 12.0; StataCorp LP, College Station, TX, USA) was utilized in all statistical analyses. The p values were all 2-sided and were considered statistically significant when >0.05.

Trial Sequential Analysis

Though a meta-analysis is commonly updated with the addition of newly published studies, it may result in type I errors because of repetitive testing of accumulated data and the increased risk of random error with few data collected (38, 39). To decrease type I errors, trial sequential analysis (TSA) is used in meta-analyses (40, 41), and sequential monitoring boundaries is applied to decide whether a trial can be terminated early when the p value is small enough to show the anticipated effect (38, 39, 42).

In the current study, TSA was used to maintain an overall 5% risk of a type I error and 5% risk of a type II error (a power of 95%). In addition, we predetermined an intervention effect of a 5% relative risk increase resulting from the intervention for the calculation of the required information size and provided the 95% confidence intervals adjusted for sparse data or repetitive testing. TSA software (TSA, version 0.9; Copenhagen Trial Unit, Copenhagen, Denmark) was used in our study.

Results

Study Characteristics

A total of 15 case-control studies including 6,958 cases and 7,961 controls met the inclusion criteria and were involved in the current meta-analysis (16-17-18-19-20-21-22-23-24-25-26-27-28-29-30). The details of each study are listed in Tables I and II. In 2 of those studies, the data and ORs were presented for 2 different ethnic groups – Africans and whites (27, 28). As a result, each of these groups was considered separately in this meta-analysis. Further, because 2 studies utilized duplicates of most subjects, while 1 provided more information, only that 1 study, by Choi et al was included (27). In addition, all studies indicated that the distribution of genotypes in the controls was consistent with HWE, except for 1 study (20). Figure 1 shows the flowchart for the literature search and selection process. The sample sizes in each study ranged between 100 and 4,270. Three genotyping methods were applied in these studies, including TaqMan, polymerase chain reaction–restriction fragment length polymorphism (PCR-RFLP) analysis and Sequenom.

Fig. 1

Flowchart of literature search and selection process.

Table I

Characteristics of individual studies included in the meta-analysis

MnSOD rs4880
Year	Surname	Ethnicity	Genotyping	HWE		Source of control	Cases
2014	Oskina	Caucasian	Taqman	Y	PB	Age matched populations with no history of PCa or other cancer	Histologically verified prostate cancer patients
2013	Eken	Caucasian	Taqman	Y	HB	Age matched patients without PCa	Histologically verified prostate cancer patients
2012	Van Hemelrijck	Caucasian	Sequenom	Y	HB	Age matched patients without prevalent cancer	Histologically verified prostate cancer patients
2012	Kucukgergin	Caucasian	PCR-RFLP	Y	HB	Patients with normal DRE and serum PSA <4 ng/mL	Histologically verified prostate cancer patients
2009	Bica	African	PCR-RFLP	N	PB	Age matched healthy populations	Histologically verified prostate cancer patients
2009	Iguchi	Mixed	PCR-RFLP	Y	HB	Age matched patients without PCa (with normal DRE and serum PSA <4 ng/mL)	Histologically verified prostate cancer patients
2008	Cooper	Caucasian	Taqman	Y	PB	Age matched populations without PCa	Pathologically or cytologically verified prostate cancer patients.
2008	Mikhak	Caucasian	Taqman	Y	PB	Age matched populations without PCa	Histologically verified prostate cancer patients without any other cancer
2008	Iguchi	Caucasian	PCR-RFLP	Y	HB	Dead men without PCa by histologically verified 18-core biopsy (>45 years)	Dead men with PCa by histologically verified 18-core biopsy (>45 years)
2008	Arsova	Caucasian	Taqman	Y	HB	Age matched patients without histologically confirmed PCa	Histologically verified prostate cancer patients
2007	Ergen	Caucasian	PCR-RFLP	Y	HB	Age matched patients without PCa	Histologically verified prostate cancer patients
2007	Choi	Caucasian	Sequenom	Y	PB	Age matched healthy populations	Histologically verified prostate cancer patients without any other cancer
	Choi	African	Sequenom	Y	PB
2007	Kang	Caucasian	Taqman	Y	PB	Age matched populations without PCa (with normal DRE and serum PSA <4 ng/mL)	Histologically verified prostate cancer patients
	Kang	African	Taqman	Y	PB
2005	Li	Caucasian	PCR-RFLP	Y	PB	Age matched healthy populations	Histologically verified prostate cancer patients
2003	Woodson	Caucasian	Sequenom	Y	PB	Age matched healthy populations	Histologically verified prostate cancer patients

HWE = Hardy-Weinberg equilibrium; PB = population-based control; HB = hospital-based control; PCa = prostate cancer; PSA = prostate-specific antigen; DRE = digital rectal examination; Y = yes; N = no.

Table II

Distribution of genotypes of individual studies included in the meta-analysis

MnSOD rs4880					Cases (n)			Controls (n)
Year	Surname	Cases	Controls	VV	VA	AA	VV	VA	AA
2014	Oskina	380	337	92	194	94	86	152	99
2013	Eken	33	81	7	17	9	31	37	13
2012	Van Hemelrijck	203	360	50	100	53	80	190	90
2012	Kucukgergin	134	159	43	65	26	66	69	24
2009	Bica	51	155	10	32	9	42	105	8
2009	Iguchi	187	175	41	86	60	40	96	39
2008	Cooper	2634	1636	602	1352	680	423	789	424
2008	Mikhak	642	652	156	320	166	162	331	159
2008	Iguchi	57	137	9	24	24	37	59	41
2008	Arsova	85	151	19	46	20	41	73	37
2007	Ergen	50	50	19	25	6	32	18	0
2007	Choi	455	1214	112	239	104	293	610	311
	Choi	28	122	7	15	6	39	52	31
2007	Kang	1150	1382	275	578	297	376	686	320
	Kang	103	395	31	57	15	122	194	79
2005	Li	567	764	132	288	147	190	379	195
2003	Woodson	199	191	43	98	58	49	102	40

VV = valine/valine genotype; VA = valine/alanine genotype; AA = alanine/alanine genotype.

Quantitative Synthesis Results

The main results of the meta-analysis of the associations between MnSOD Val16Ala polymorphism and PCa are listed in Table III. The relationship was assessed by multiple comparisons in various models. A heterozygote model is a VA genotype versus VV genotype. A homozygote model is an AA genotype versus VV genotype. A dominant model is a VA+AA genotype versus VV genotype. Finally, a recessive model is an AA genotype versus VA+VV genotype (43). According to the results of all of the models, we could initially draw a conclusion as to whether the susceptibility to PCa of subjects with mutative genotypes or wild genotypes is different.

Table III

Meta-analysis results of association between rs4880 polymorphism and prostate cancer risk

			VA vs. VV		AA vs. VV		VA/AA vs. VV		AA vs. VA/VV
	N	Sample size	OR (95% CI)^*	P^	OR (95% CI)^*	P^**	OR (95% CI)^*	P^**	OR (95% CI)^*	P^**
Total	17	14919	1.14 (1.05-1.24)	0.769	1.18 (1.02-1.36)	0.054	1.24 (1.07-1.44)	0.000	1.10 (0.96-1.24)	0.019
Ethnicity
Caucasian	13	13703	1.15 (1.05-1.25)	0.608	1.15 (1.001-1.32)	0.114	1.26 (1.06-1.50)	0.000	1.06 (0.95-1.18)	0.135
Genotyping method
Taqman	6	9661	1.17 (1.06-1.29)	0.834	1.13 (0.99-1.29)	0.355	1.27 (1.01-1.59)	0.001	1.02 (0.90-1.15)	0.242
Sequenom	4	2772	1.01 (0.83-1.24)	0.647	1.02 (0.77-1.36)	0.294	1.01 (0.83-1.22)	0.602	1.04 (0.79-1.38)	0.166
PCR-RFLP	6	2486	1.19 (0.98-1.45)	0.366	1.81 (1.13-2.89)	0.037	1.37 (1.05-1.80)	0.173	1.58 (1.07-2.35)	0.028
Sample size
Small	11	3823	1.19 (1.01-1.40)	0.594	1.40 (1.04-1.89)	0.032	1.24 (1.05-1.47)	0.321	1.25 (0.96-1.63)	0.012
Large	5	11096	1.13 (1.03-1.24)	0.730	1.11 (0.999-1.24)	0.445	1.19 (0.93-1.53)	0.000	1.03 (0.94-1.12)	0.427
Source of control
PB	10	13057	1.14 (1.04-1.24)	0.975	1.11 (0.96-1.28)	0.127	1.19 (1.000-1.42)	0.000	1.02 (0.89-1.17)	0.034
HB	6	1862	1.20 (0.95-1.50)	0.186	1.54 (1.06-2.25)	0.145	1.40 (1.03-1.90)	0.091	1.35 (1.05-1.74)	0.316

VA = valine/alanine genotype; AA = alanine/alanine genotype; VV = valine/valine genotype; OR = odds ratio; CI = confidence interval; N = number of studies; PB = population based; HB = hospital based.

The random-effects model was used when the p value for the heterogeneity test was <0.1; otherwise, the fixed-effects model was used.

P value of Q test for heterogeneity.

Overall, the pooled OR was 1.14 (95% CI, 1.05-1.24) for the heterozygote model, 1.18 (95% CI, 1.02-1.36) for the homozygote model, 1.24 (95% CI, 1.07-1.44) for the dominant model and 1.10 (95% CI, 0.96-1.24) for the recessive model (Fig. 2). In the subgroup analysis by genotyping method, the results were found to be significant for all genotyping methods except Sequenom (TaqMan: heterozygote model: pooled OR = 1.17, 95% CI, 1.06-1.29; dominant model: pooled OR = 1.27, 95% CI, 1.01-1.59; PCR-RFLP: homozygote model: pooled OR = 1.81, 95% CI, 1.13-2.89; dominant model: pooled OR = 1.37, 95% CI, 1.05-1.80; recessive model: pooled OR = 1.58, 95% CI, 1.07-2.35) (Fig. 3). In addition, when stratified by sample size, the results were positive in both large sample sizes (heterozygote model: pooled OR = 1.13, 95% CI, 1.03-1.24) and small sample sizes (heterozygote model: pooled OR = 1.19, 95% CI, 1.01-1.40; homozygote model: pooled OR = 1.40, 95% CI, 1.04-1.89; dominant model: pooled OR = 1.24, 95% CI, 1.05-1.47). Moreover, in the stratified analysis by source of control, significant results were detected in both the subgroup of population-based controls (heterozygote model: pooled OR = 1.14, 95% CI, 1.04-1.24; dominant model: pooled OR = 1.19, 95% CI, 1.000-1.42) and hospital-based controls (homozygote model: pooled OR = 1.54, 95% CI, 1.06-2.25; dominant model: pooled OR = 1.40, 95% CI, 1.03-1.90; recessive model: pooled OR = 1.35, 95% CI, 1.05-1.74). Furthermore, because of the insufficient sample size for ethnicities other than white populations, further analysis was only conducted among white populations. It was found that subjects carrying the Ala allele had a higher PCa risk than those with the Val allele (heterozygote model: pooled OR = 1.15, 95% CI, 1.05-1.24; homozygote model: pooled OR = 1.15, 95% CI, 1.001-1.32; dominant model: pooled OR = 1.26, 95% CI, 1.06-1.50).

Fig. 2

Forest plot of the association between MnSOD Val16Ala polymorphism and prostate cancer (PCa) susceptibility in the homozygote model.

Fig. 3

Forest plots of stratified analysis by genotyping method of the association between MnSOD Val16Ala polymorphism and prostate cancer (PCa) susceptibility in dominant model. CI = confidence interval; OR = odds ratio; PCR-RFLP = polymerase chain reaction–restriction fragment length polymorphism.

Test of Heterogeneity

Heterogeneity between studies was observed in overall genetic models except the heterozygote model, but the heterogeneity was decreased in subgroup analyses. Galbraith analysis is shown in Figure 4, indicating that the heterogeneity between studies was not prominent.

Fig. 4

Galbraith plot of the association between MnSOD Val16Ala polymorphism and prostate cancer (PCa) susceptibility in homozygote model.

Sensitivity Analysis

Sensitivity analysis was completed to detect the influence of each study on the pooled OR, by repeating the meta-analysis while omitting a single study each time. The sensitivity analysis for the association of MnSOD Val16Ala polymorphism in the heterozygote model with the overall population is shown in Figure 5, demonstrating that no individual study affected the pooled OR significantly. Although the genotype distribution in 1 study included did not show HWE (20), the corresponding pooled ORs were not qualitatively altered when it was omitted. The sensitivity analysis indicated that our results were reliable.

Fig. 5

Sensitivity analysis under the heterozygote model.

Publication Bias

Begg's funnel plots were completed to assess any publication bias in the literature. Their shapes seemed not to show any evidence of obvious asymmetry, indicating no significant publication bias, which was also confirmed by Egger's test (heterozygote model: p = 0.190). The overall outcomes indicated that our results were statistically robust.

Tsa Results

We compared the current study with a previous meta-analysis by Mao et al (32) in 2010, through TSA. Figure 6A shows that the cumulative Z-curve did not cross the trial sequential monitoring boundary, and the total size of cases and controls was less than the estimated information size in the previous meta-analysis (32), indicating the results might be a potentially spurious evidence of effect. However, in the current study, the meta-analysis included larger numbers of cases and controls than the required information size, and the cumulative Z-curve crossed the monitoring boundaries (Fig. 6B), suggesting that our results were based on sufficient evidence.

Fig. 6

Trial sequential analysis (TSA) of the association between between MnSOD Val16Ala polymorphism and the risk of prostate cancer (PCa): (A) previous meta-analysis (32); (B) current meta-analysis. The required information size was calculated based on a 2-sided α = 5%, β = 5% (power 95%), and a relative risk reduction of 5%.

Discussion

The current meta-analysis indicated that the Ala allele of the MnSOD gene polymorphism increases PCa susceptibility. MnSOD is synthesized in the cytoplasm before being imported into the mitochondria via N-terminal mitochondrial targeting sequence (MTS). The mature protein is assembled as an active tetramer after the MTS is cleaved in the mitochondrial matrix (44). The ROS, which are produced in the mitochondrial electron transport chain due to oxidative stress, are assumed to influence the respiratory chain and damage the mitochondrial DNA through lipid peroxidation of nerve cells (45, 46). As the major antioxidant in the mitochondria, MnSOD generates an important impact on endogenous defense mechanisms and resists ROS through catalyzing their conversion of superoxide anion (O2-) to hydrogen peroxide (H₂O₂) in the mitochondria. MnSOD is deleted in various tumors, indicating MnSOD might be a candidate tumor suppressor gene (11, 13, 14).

Recently, an increasing number of studies have researched the association between MnSOD Val16Ala polymorphism and PCa susceptibility (16-17-18-19-20-21-22-23-24-25-26-27-28-29-30). Nevertheless, their results remain inconsistent. Though most researchers have concluded that MnSOD Val16Ala polymorphism is associated with the risk of PCa based on total and/or stratified analysis, Choi and his colleagues hold a different opinion (27). The conflict among them might partially be due to the relatively small sample size of individual studies, the different distributions of patients or controls, the different methodologies used and the possible limited effect of the polymorphism on PCa risk.

Meta-analysis, as a powerful tool, can provide more reliable results than a single study, especially in explaining controversial conclusions, and provide the most comprehensive information by using different subgroup analysis (47). Several meta-analyses have explored the relationship between MnSOD Val16Ala polymorphism and the risk of PCa (31-32-33-34), but the results have differed a lot. Most of them revealed that significantly elevated PCa risk was associated with the Ala genotype of MnSOD Val16Ala polymorphism, but 1 study held a different view (34). However, in this that meta-analysis, the same MnSOD Val16Ala polymorphism was misclassified into 2 different ones, which was unnecessary and might make the results inaccurate with a smaller sample size. In addition, 1 case-control study which mixed breast cancer and PCa as cases (48) was included in 2 meta-analyses (31, 34), which might add bias to the results. Furthermore, additional studies have been published since these previous meta-analyses, which might have a great influence on the results. Last but not least, in the present meta-analysis, TSA was applied for the first time to estimate whether the evidence of the results was sufficient. All of these factors contributed to the development of the current meta-analysis.

Recently, various molecular techniques have been applied in genotyping, so we conducted stratified analysis by genotyping method, and the result was shown to be significant for the TaqMan and PCR-RFLP methods, but not Sequenom. As we all know, diverse genotyping methods have advantages in different regards, which might constitute a source of differences. For example, though the method of Sequenom can detect a lot of sites each time and offer a large amount of information, it may lack accuracy compared with TaqMan or PCR-RFLP when detecting 1 particular polymorphism. As a result, if the studies were unified with the same appropriate genotyping method, the meta-analysis results would be more accurate and reliable.

Moreover, in this meta-analysis, the results were statistically significant in white populations. Due to small sample size in other ethnicities aside from white populations, the studies were not stratified by ethnicity.

TSA, as an approach providing the required information size in meta-analyses (39), depends on the quantification of the required information size (sample size of the meta-analysis). The addition of each study in a cumulative meta-analysis is similar to an interim analysis, determining whether additional trials are needed. During TSA, the cumulative Z-curve (i.e., Z-statistics after each trial) of each meta-analysis was constructed. Then, its crossing of Z = 1.96 (p = 0.05) and the monitoring boundaries was assessed. It shows firm evidence for an intervention effect when the cumulative Z-curve crosses the monitoring boundaries. In consequence, we took advantage of TSA in the current meta-analysis to reduce the risk of type I error and estimate whether the evidence of our results was sufficient. In the previous meta-analysis (32), the cumulative Z-curve did not cross the trial sequential monitoring boundary, and the numbers of cases and controls included were less than the estimated information size, which meant that firm evidence could not be established, and the results were potentially spurious evidence of effect, suggesting a larger sample size was demanded to obtain firm evidence. However, in our current study, the cumulative Z-curve crossed the monitoring boundaries. In addition, larger sample sizes were included than the required information size. All these indicated our results were based on firm evidence of effect.

It has been demonstrated that the Val to Ala substitution could induce a conformational change from a β sheet to α helical structure in MTS, thus influencing mitochondrial processing efficiency and decreasing MnSOD efficiency against oxidative stress (44, 49). Moreover, the conformational change of ornithine transcarbamylase could affect the mitochondrial processing efficiency due to single amino acid substitutions in the leader signal (50). Consequently, these are likely to cause DNA mutations and protein oxidation without the full defense of MnSOD against superoxide radicals (51). The results of our study were in agreement with these hypotheses, which further proved that MnSOD Val16Ala polymorphism plays an important role in PCa susceptibility.

Despite the sufficient and robust statistical evidence generated through this analysis, some limitations should be taken into consideration. Firstly, though we took advantage of TSA to acquire more convincing evidence, the shortcomings of TSA itself could not be avoided. However, as a new technique with adjustment for the disadvantages, the application of TSA in meta-analyses can provide a more balanced conclusion. Secondly, in stratified analyses, the number of some subgroups was relatively small, without enough statistical power to explore the real association. Moreover, as a complicated disease, cancer shows discrepancies among different ethnic populations, which might be due to different genetic backgrounds to some degree (52). However, due to the insufficient sample size in African populations, the studies were not stratified by ethnicity. In addition, PCa results from complex interactions including a variety of genetic and environmental factors, suggesting that PCa susceptibility can not be influenced by any single gene, and effects of different gene polymorphisms need to be combined. Furthermore, it has been demonstrated that various factors could increase the risk of PCa, such as age, ethnicity, smoking, family history of PCa, body mass index, diabetes and so on (53-54-55-56-57), which might act as potential factors influencing the accuracy of our results. Nevertheless, our results were based on unadjusted estimates, without enough data for an adjustment for other covariates. Accordingly, we should hold a conservative attitude toward our results. A more precise analysis could be performed if individual data were available. More studies exploring the influencing factors are required to draw more comprehensive conclusions in the future.

Conclusion

The results of the present meta-analysis indicated that the Ala allele of the MnSOD gene polymorphism increases susceptibility to PCa. As a result, the MnSOD Val16Ala polymorphism might be a potential indicator for the risk of PCa in the future.

Footnotes

Financial support: No grants or funding have been received for this study.

Conflict of interest: The authors declare that they have no conflict of interest.

References

Jemal

Siegel

Ward

Cancer statistics, 2010. CA Cancer J Clin 2010;60(5):277–300

Siegel

Ward

Brawley

Jemal

Cancer statistics, 2011: the impact of eliminating socioeconomic and racial disparities on premature cancer deaths. CA Cancer J Clin 2011;61(4):212–236

Siegel

Naishadham

Jemal

Cancer statistics, 2013. CA Cancer J Clin 2013;63(1):11–30

Grönberg

Prostate cancer epidemiology. Lancet 2003;361(9360):859–864

McGregor

Courneya

Kopciuk

Tosevski

Friedenreich

Case-control study of lifetime alcohol intake and prostate cancer risk. Cancer Causes Control 2013;24(3):451–461

Nelson

Demarzo

Yegnasubramanian

The diet as a cause of human prostate cancer. Cancer Treat Res 2014;159:51–68

Discacciati

Wolk

Lifestyle and dietary factors in prostate cancer prevention. Recent Results Cancer Res 2014;202:27–37

Lai

Giovannucci

Pollak

. Association of C-peptide and leptin with prostate cancer incidence in the Health Professionals Follow-up Study. Cancer Causes Control 2014;25(5):625–632

Baker

Lichtenstein

Kaprio

Holm

Genetic susceptibility to prostate, breast, and colorectal cancer among Nordic twins. Biometrics 2005;61(1):55–63

10.

Eeles

Goh

Castro

. The genetic epidemiology of prostate cancer and its clinical implications. Nat Rev Urol 2014;11(1):18–31

11.

Zhong

Oberley

St Clair DK. Suppression of the malignant phenotype of human glioma cells by overexpression of manganese superoxide dismutase. Oncogene 1997;14(4):481–490

12.

Crapo

Isolation and characterization of complementary DNAs encoding human manganese-containing superoxide dismutase. FEBS Lett 1988;229(2):256–260

13.

Oberley

Zhong

Overexpression of manganese superoxide dismutase in DU145 human prostate carcinoma cells has multiple effects on cell phenotype. Prostate 1998;35(3):221–233

14.

Oberley

Mitochondria, manganese superoxide dismutase, and cancer. Antioxid Redox Signal 2004;6(3):483–487

15.

Sutton

Imbert

Igoudjil

. The manganese superoxide dismutase Ala16Val dimorphism modulates both mitochondrial import and mRNA stability. Pharmacogenet Genomics 2005;15(5):311–319

16.

Oskina

Еrmolenko

Boyarskih

. Associations between SNPs within antioxidant genes and the risk of prostate cancer in the Siberian region of Russia. Pathol Oncol Res 2014;20(3):635–640

17.

Eken

Erdem

Arsova-Sarafinovska

. Association between gene polymorphism of manganese superoxide dismutase and prostate cancer risk. J Biochem Mol Toxicol 2013;27(3):213–218

18.

Van Hemelrijck

Rohrmann

Steinbrecher

Kaaks

Teucher

Linseisen

Heterocyclic aromatic amine [HCA] intake and prostate cancer risk: effect modification by genetic variants. Nutr Cancer 2012;64(5):704–713

19.

Kucukgergin

Sanli

Tefik

Aydın

Ozcan

Seckin

Increased risk of advanced prostate cancer associated with MnSOD Ala-9-Val gene polymorphism. Mol Biol Rep 2012;39(1):193–198

20.

Bica

de Moura da Silva

Toscani

. MnSOD gene polymorphism association with steroid-dependent cancer. Pathol Oncol Res 2009;15(1):19–24

21.

Iguchi

Sugita

Wang

. MnSOD genotype and PCa risk as a function of NAT genotype and smoking status. In Vivo 2009;23:7–12

22.

Cooper

Adami

Grönberg

Wiklund

Green

Rayman

Interaction between SNPs in selenoprotein P and mitochondrial superoxide dismutase determines PCa risk. Cancer Res 2008;68:10171–10177

23.

Mikhak

Hunter

Spiegelman

. Manganese superoxide dismutase (MnSOD) gene polymorphism, interactions with carotenoid levels and prostate cancer risk. Carcinogenesis 2008;29(12):2335–2340

24.

Iguchi

Wang

Delongchamps

. Association of prostate cancer and manganese superoxide dismutase AA genotype influenced by presence of occult cancer in control group. Urology 2008;72(2):238–241 [discussion: 241-242].

25.

Arsova-Sarafinovska

Matevska

Petrovski

. Manganese superoxide dismutase (MnSOD) genetic polymorphism is associated with risk of early-onset prostate cancer. Cell Biochem Funct 2008;26(7):771–777

26.

Ergen

Narter

Timirci

Isbir

Effects of manganase superoxide dismutase Ala-9Val polymorphism on prostate cancer: a case-control study. Anticancer Res 2007;27(2):1227–1230

27.

Choi

Neuhouser

Barnett

. Polymorphisms in oxidative stress-related genes are not associated with prostate cancer risk in heavy smokers. Cancer Epidemiol Biomarkers Prev 2007;16(6):1115–1120

28.

Kang

Lee

Park

. Functional variant of manganese superoxide dismutase (SOD2 V16A) polymorphism is associated with prostate cancer risk in the prostate, lung, colorectal, and ovarian cancer study. Cancer Epidemiol Biomarkers Prev 2007;16(8):1581–1586

29.

Kantoff

Giovannucci

. Manganese superoxide dismutase polymorphism, prediagnostic antioxidant status, and risk of clinical significant prostate cancer. Cancer Res 2005;65(6):2498–2504

30.

Woodson

Tangrea

Lehman

. Manganese superoxide dismutase (MnSOD) polymorphism, alpha-tocopherol supplementation and prostate cancer risk in the alpha-tocopherol, beta-carotene cancer prevention study (Finland). Cancer Causes Control 2003;14(6):513–518

31.

Sun

Wang

Different association of manganese superoxide dismutase gene polymorphisms with risk of prostate, esophageal, and lung cancers: evidence from a meta-analysis of 20,025 subjects. Asian Pac J Cancer Prev 2013;14(3):1937–1943

32.

Mao

Qiu

Zhan

. MnSOD Val16Ala polymorphism and prostate cancer susceptibility: a meta-analysis involving 8,962 subjects. J Cancer Res Clin Oncol 2010;136(7):975–979

33.

Wang

Shi

. Association between manganese superoxide dismutase (MnSOD) Val-9Ala polymorphism and cancer risk - A meta-analysis. Eur J Cancer 2009;45(16):2874–2881

34.

Liwei

Chunyu

Ruifa

Association between manganese superoxide dismutase gene polymorphism and risk of prostate cancer: a meta-analysis. Urology 2009;74(4):884–888

35.

DerSimonian

Kacker

Random-effects model for meta-analysis of clinical trials: an update. Contemp Clin Trials 2007;28(2):105–114

36.

Egger

Davey Smith

Schneider

Minder

Bias in meta-analysis detected by a simple, graphical test. BMJ 1997;315(7109):629–634

37.

Guo

Thompson

Performing the exact test of Hardy-Weinberg proportion for multiple alleles. Biometrics 1992;48(2):361–372

38.

Brok

Thorlund

Wetterslev

Gluud

Apparently conclusive meta-analyses may be inconclusive: trial sequential analysis adjustment of random error risk due to repetitive testing of accumulating data in apparently conclusive neonatal meta-analyses. Int J Epidemiol 2009;38(1):287–298

39.

Wetterslev

Thorlund

Brok

Gluud

Trial sequential analysis may establish when firm evidence is reached in cumulative meta-analysis. J Clin Epidemiol 2008;61(1):64–75

40.

Higgins

Whitehead

Simmonds

Sequential methods for random-effects meta-analysis. Stat Med 2011;30(9):903–921

41.

Thorlund

Imberger

Walsh

. The number of patients and events required to limit the risk of overestimation of intervention effects in meta-analysis: a simulation study. PLoS One 2011;6(10):e25491

42.

Lan

DeMets

Discrete sequential boundaries for clinical trials. Biometrika 1983;70(3):659–663

43.

Thakkinstian

McElduff

D'Este

Duffy

Attia

A method for meta-analysis of molecular association studies. Stat Med 2005;24(9):1291–1306

44.

Akyol

Yanik

Elyas

. Association between Ala-9Val polymorphism of Mn-SOD gene and schizophrenia. Prog Neuropsychopharmacol Biol Psychiatry 2005;29(1):123–131

45.

Low

Nickander

Tritschler

The roles of oxidative stress and antioxidant treatment in experimental diabetic neuropathy. Diabetes 1997;46(Suppl 2):S38–S42

46.

Guidot

McCord

Wright

Repine

Absence of electron transport (Rho 0 state) restores growth of a manganese-superoxide dismutase-deficient Saccharomyces cerevisiae in hyperoxia: evidence for electron transport as a major source of superoxide generation in vivo. J Biol Chem 1993;268(35):26699–26703

47.

Munafò

Flint

Meta-analysis of genetic association studies. Trends Genet 2004;20(9):439–444

48.

Taufer

Peres

de Andrade

. Is the Val16Ala manganese superoxide dismutase polymorphism associated with the aging process? J Gerontol A Biol Sci Med Sci 2005;60(4):432–438

49.

Shimoda-Matsubayashi

Matsumine

Kobayashi

Nakagawa-Hattori

Shimizu

Mizuno

Structural dimorphism in the mitochondrial targeting sequence in the human manganese superoxide dismutase gene: a predictive evidence for conformational change to influence mitochondrial transport and a study of allelic association in Parkinson's disease. Biochem Biophys Res Commun 1996;226(2):561–565

50.

Horwich

Kalousek

Fenton

Pollock

Rosenberg

Targeting of pre-ornithine transcarbamylase to mitochondria: definition of critical regions and residues in the leader peptide. Cell 1986;44(3):451–459

51.

Ambrosone

Freudenheim

Thompson

. Manganese superoxide dismutase (MnSOD) genetic polymorphisms, dietary antioxidants, and risk of breast cancer. Cancer Res 1999;59(3):602–606

52.

Hirschhorn

Lohmueller

Byrne

Hirschhorn

A comprehensive review of genetic association studies. Genet Med 2002;4(2):45–61

53.

Park

Haiman

Cheng

. Racial/ethnic differences in lifestyle-related factors and prostate cancer risk: the Multiethnic Cohort Study. Cancer Causes Control 2015;26(10):1507–1515

54.

Bashir

Epidemiology of PCa. Asian Pac J Cancer Prev 2015;16(13):5137–5141

55.

Bashir

Malik

Case-control study of diet and prostate cancer in a rural population of Faisalabad, Pakistan. Asian Pac J Cancer Prev 2015;16(6):2375–2378

56.

Nemesure

Hennis

Leske

Family history of prostate cancer in a black population. J Immigr Minor Health 2013;15(6):1107–1112

57.

Ganesh

Saoba

Sarade

Pinjari

Risk factors for prostate cancer: An hospital-based case-control study from Mumbai, India. Indian J Urol 2011;27(3):345–350

Association between Manganese Superoxide Dismutase (MnSOD) Polymorphism and Prostate Cancer Susceptibility: A Meta-Analysis

Abstract

Background

Methods

Results

Conclusions

Keywords

Introduction

Materials and methods

Data Extraction

Statistical Analysis

Trial Sequential Analysis

Results

Study Characteristics

Quantitative Synthesis Results

Test of Heterogeneity

Sensitivity Analysis

Publication Bias

Tsa Results

Discussion

Conclusion

Footnotes

References