Abstract

Manganese superoxide dismutase (MnSOD) plays a critical role in the defense against reactive oxygen species. The association between MnSOD Val16Ala polymorphism and cancer risk has been widely studied, but the results are contradictory. To obtain more precision on the association, we performed the current meta-analysis with 33,098 cases and 37,831 controls from 88 studies retrieved from PubMed, Embase, Chinese National Knowledge Infrastructure (CNKI), and Wanfang databases. Pooled odds ratios (ORs) and 95% confidence intervals (CIs) were used to assess the strength of association. We found that the polymorphism was associated with an increased overall cancer risk (homozygous: , 95% CI = 1.00–1.19; heterozygous: , 95% CI = 1.02–1.12; dominant: , 95% CI = 1.02–1.14; and allele comparison: , 95% CI = 1.02–1.11). Stratification analysis further showed an increased risk for prostate cancer, Asians, Caucasians, population-based studies, hospital-based studies, low quality and high quality studies. However, the increased risk for MnSOD Val16Ala polymorphism among Asians needs further validation based on the false-positive report probability (FPRP) test. To summarize, this meta-analysis suggests that the MnSOD Val16Ala polymorphism is associated with significantly increased cancer risk, which needs further validation in single large studies.

1. Introduction

Cancer is one of the leading causes of death across the world, with an estimate of over 20 million new cancer cases that will occur per year as early as 2025 [1]. Although great efforts have been devoted to cancer treatment, cancer still poses a huge threat to human health. Carcinogenesis is rather complex, and mounting evidence suggests that reactive oxygen species- (ROS-) related oxidative damage is involved in this process [24].

Among the endogenous antioxidants, manganese superoxide dismutase (MnSOD) is one of the critical enzymes which defends against ROS in the mitochondria. The MnSOD gene, located on chromosome 6q25.3, is composed of four introns and five extrons. Currently, several single-nucleotide polymorphisms (SNPs) in the MnSOD gene have been reported, of which the most extensively studied one is Val16Ala. Since this residue is 9 amino acids upstream of the cleavage site, it has also been called Val9Ala (rs4880) polymorphism [5]. A previous study has shown that Ala-MnSOD allowed more efficient MnSOD localized to the mitochondria than the Val-variant form [6]. In view of this, it is speculated that the Val form of MnSOD may be associated with higher levels of ROS and increased susceptibility to cancer.

Several studies have found the associations between the Val form of the MnSOD gene and increased cancer risk [79], but a majority of studies showed the Ala form to be associated with higher cancer risk, such as breast cancer [10, 11], esophageal cancer [12], colorectal cancer [13], and cervical cancer [14], and some other studies find no significant association between this polymorphism and cancer risk [1518]. To draw a more comprehensive estimation of this possible association, we conducted the present meta-analysis to evaluate the relevance of this variant with susceptibility of cancer.

2. Materials and Methods

2.1. Search Strategy

We systematically searched the PubMed, Embase, Chinese National Knowledge Infrastructure (CNKI), and Wanfang databases for all related publications using the following keywords: “MnSOD or manganese superoxide dismutase,” “polymorphism or variant or variation,” and “cancer or carcinoma or tumor or neoplasm” (the last search was updated on February 22, 2018). Additional relevant studies were searched manually from the references or review articles about this topic. If studies had overlapped data, only the one with the most participants was included in this analysis.

2.2. Inclusion and Exclusion Criteria

The inclusion criteria were as follows: (1) case-control studies, (2) studies assessing the association between MnSOD Val16Ala polymorphism and cancer risk, (3) and provision of detailed data about genotype and allele distribution of the studied polymorphism. Studies were excluded if any of the following aspects existed: (1) duplicate publications, (2) review articles or meta-analyses, (3) not a case-control study, and (4) genotype frequencies in the control departure from Hardy-Weinberg equilibrium (HWE).

2.3. Data Extraction

Two authors (Ping Wang and Yanfeng Zhu) independently extracted the data from included studies according to the criteria mentioned above. Disagreement was resolved by discussion until a consensus was reached. The following information was collected from each study: first author’s surname, year of publication, country of origin, ethnicity, cancer type, control source (hospital-based or population-based), genotyping methods, and numbers of cases and controls with the Val/Val, Val/Ala, and Ala/Ala genotypes.

2.4. Quality Assessment

The quality of each included study was assessed independently by two authors using the criteria from a previous study [19]. Quality scores were rated from 0 to 15, and the studies were classified as high-quality studies () and low-quality studies ().

2.5. Statistical Analysis

The strength of association between the MnSOD Val16Ala polymorphism and cancer risk was assessed by calculating the odd ratios (ORs) with the corresponding 95% confidence intervals (CIs). The pooled ORs of five comparison models were calculated: homozygous model (Ala/Ala versus Val/Val), heterozygous model (Val/Ala versus Val/Val), recessive model [Ala/Ala versus (Val/Val + Val/Ala)], dominant model [(Ala/Ala + Val/Ala) versus Val/Val], and an allele comparison (Ala versus Val). We used the chi-square-based test to check the between-study heterogeneity, and the fixed-effects model (the Mantel-Haenszel method) [20] was used when no significant heterogeneity was found (). Otherwise, the random-effects model (the Dersimonian and Laird method) [21] was applied. The stratification analysis was performed by cancer type (cancer types with less than three studies would be merged into the “others” group), ethnicity (Asians, Caucasians, Africans, or mixed which contained more than one ethnic group), control source (hospital-based studies and population-based studies), and quality scores (≤9 and >9). Publication bias was examined using Begg’s funnel plot [22] and Egger’s linear regression test [23]. Sensitivity analysis was carried out to assess the results stability by excluding one study each time and revaluating the pooled ORs and 95% CIs.

The false-positive report probability (FPRP) was calculated for all the significant findings in the present study. We set 0.2 as a FPRP threshold and assign a prior probability of 0.1 to detect an OR of 0.67/1.50 (protective/risk effects) for an association with the genotypes under investigation [24, 25]. FPRP values less than 0.2 were considered as noteworthy associations. All the statistical tests were performed with STATA software (version 12.0; Stata Corporation, College Station, TX). Two-sided values <0.05 were considered statistically significant.

3. Results

3.1. Study Characteristics

As shown in Figure 1, a total of 348 articles were identified from PubMed, Embase, CNKI, and Wanfang databases, and 34 more articles were identified by reading the references of retrieved publications. After reading the titles and abstracts, 266 articles were excluded, leaving 116 articles for further assessment. Among them, six were excluded as case-only studies [2631], five [3236] were covered by other included publications [7, 37, 38], three were without detailed data for further analysis [3941], and 18 deviated from HWE [4259]. Finally, a total of 84 case-control publications [718, 37, 38, 60129] were included in this meta-analysis. Of the 84 publications, three publications [37, 69, 82] with two ethnic groups were considered as two independent studies and one publication [119] with two cancer types were also considered as two independent studies.

For the two studies in the publication [119] with the same control group, the number of control was only calculated once in the total number. Overall, 88 studies with 33,098 cases and 37,831 controls were included in this meta-analysis. Of the 88 studies, 24 studies focused on breast cancer [911, 16, 38, 60, 61, 68, 69, 71, 72, 77, 88, 93, 96, 97, 100, 105, 109, 114, 119, 122, 127]; 17 on prostate cancer [37, 66, 74, 79, 82, 85, 86, 89, 95, 106, 111, 113, 120, 125, 128]; six for each of the following cancer types, such as lung cancer [7, 17, 18, 65, 92, 118], bladder cancer [8, 15, 67, 75, 112, 117], and pancreatic cancer [64, 91, 102, 107, 108, 121]; five on colorectal cancer [13, 63, 73, 94, 101]; three for each of the following cancer types, such as ovarian cancer [70, 81, 87], hepatocellular carcinoma [98, 99, 129], and non-Hodgkin’s lymphoma [76, 78, 110]; and the other with fewer than three studies for each cancer type. Of all the studies, 56 studies were performed on Caucasians, 18 studies on Asians, and seven studies on Africans and mixed ethnicity, respectively. When classified by source of control, 48 were population-based and 40 were hospital-based. In addition, according to the quality score, 49 studies were considered as high-quality and 39 studies were considered as low-quality. The characteristics of the included studies are shown in Table 1.

3.2. Meta-Analysis Results

The overall results suggested there was a significant association between MnSOD Val16Ala polymorphism and cancer risk (homozygous: , 95% CI = 1.00–1.19, ; heterozygous: , 95% CI = 1.02–1.12, ; dominant: , 95% CI = 1.02–1.14, ; and allele comparison: , 95% CI = 1.02–1.11, ) (Table 2, Figure 2). In the subgroup analysis, a statistically significant association was found for prostate cancer (heterozygous: , 95% CI = 1.05–1.24, ; dominant: , 95% CI = 1.05–1.23, ; and allele comparison: , 95% CI = 1.00–1.15, ), Asians (homozygous: , 95% CI = 1.15–2.88, , and recessive: , 95% CI = 1.16–2.68, ), Caucasians (heterozygous: , 95% CI = 1.03–1.13, ; dominant: , 95% CI = 1.02–1.14, ; and allele comparison: , 95% CI = 1.00–1.09, ), population-based studies (homozygous: , 95% CI = 1.01–1.19, ; heterozygous: , 95% CI = 1.02–1.12, ; dominant: , 95% CI = 1.02–1.13, ; and allele comparison: , 95% CI = 1.00–1.08, ), hospital-based studies (recessive: , 95% CI = 1.01–1.34, , and allele comparison: , 95% CI = 1.03–1.24, ), low-quality studies (allele comparison: , 95% CI = 1.02–1.23, ) and high-quality studies (homozygous: , 95% CI = 1.00–1.17, ; heterozygous: , 95% CI = 1.02–1.13, ; dominant: , 95% CI = 1.02–1.14, ; and allele comparison: , 95% CI = 1.00–1.09, ).

3.3. Heterogeneity and Sensitivity Analysis

As shown in Table 2, substantial heterogeneities were found among all studies for the MnSOD Val16Ala polymorphism and overall cancer risk (homozygous: ; heterozygous: ; recessive: ; dominant: ; and allele comparison: ). Therefore, the random-effects model was used to generate wider CIs. The leave-one-out sensitivity analysis indicated that no single study could change the pooled ORs obviously (data not shown).

3.4. Publication Bias

Begg’s funnel plot and Egger’s test were performed to evaluate the publication bias of 88 studies, and we found significant publication bias for the homozygous model (), recessive model (), dominant model (), and allele comparison (), but not for the heterozygous model (). Therefore, the Duval and Tweedie nonparametric “trim and fill” method was used to adjust for publication bias. The “trim and fill” method did not draw different conclusions (data not shown), indicating that our findings were statistically robust.

3.5. False-Positive Report Probability (FPRP) Analysis

The FPRP values were calculated for all the significant findings (Table 3). With the assumption of a prior probability of 0.1, the FPRP results revealed that three genetic models [Val/Ala versus Val/Val, (Ala/Ala + Val/Ala) versus Val/Val, and Ala versus Val] of the MnSOD Val16Ala polymorphism were truly associated with increased cancer risk (, 0.045, and 0.106, resp.). In addition, according to the FPRP results, we confirmed that the MnSOD Val16Ala polymorphism was associated with cancer risk for prostate cancer (heterozygous: FPRP = 0.020 and dominant: FPRP = 0.006), Caucasians (heterozygous: FPRP = 0.008 and dominant: FPRP = 0.045), population-based studies (homozygous: FPRP = 0.136, heterozygous: FPRP = 0.032 and dominant: FPRP = 0.119), hospital-based studies (allele comparison: FPRP = 0.082), low-quality studies (allele comparison: FPRP = 0.138), and high-quality studies (heterozygous: FPRP = 0.119).

4. Discussion

In this meta-analysis, we comprehensively assessed the association between MnSOD Val16Ala polymorphism and cancer risk through 88 studies, and we found that this gene polymorphism was significantly associated with overall cancer risk. Further, stratification analysis revealed that the association was more obvious for risk of prostate cancer, Asians, Caucasians, population-based studies, hospital-based studies, low-quality studies, and high-quality studies. To avoid the false-positive results of the meta-analysis, we performed the FPRP analysis for the significant findings by setting as the prior probability of 0.1, and the results suggested that the association between MnSOD Val16Ala polymorphism and cancer risk for Asians was false positive, which may due to limited sample size.

MnSOD is a mitochondrial enzyme that converts superoxide radical O2 into H2O2, and it plays a critical role in human cells. Studies have revealed that the aberrant expression of MnSOD is involved in many types of cancers. Our current study indicated that the MnSOD Val16Ala polymorphism was significantly associated with an increased overall cancer risk. Previous meta-analyses have also assessed the association of MnSOD Val16Ala polymorphism with cancer susceptibility. The study carried out by Kang [130] analyzed MnSOD Val16Ala polymorphism and cancer risk, consisting 52 studies with 26,865 cases and 32,464 controls, in which no significant association was found between this polymorphism and overall cancer risk. In the subgroup analysis, statistically significant associations were found between this polymorphism and non-Hodgkin lymphoma, lung cancer, and colorectal cancer. Another meta-analysis [131] including 7366 cases and 9102 controls found no overall association of MnSOD Val16Ala polymorphism for cancer risk. Some of the significant associations detected in the previous meta-analyses were not found in the present study; for example, MnSOD Val16Ala polymorphism was associated with the risk of hepatocellular carcinoma [132, 133], esophageal cancer [134], and lung cancer [134]. The discrepancy that occurred may be because our current study was based on a much larger sample size, allowing the more precise detection of the association. In the subgroup analysis by cancer type, we found a significant association between MnSOD Val16Ala polymorphism and elevated prostate cancer risk, and no significant association between this polymorphism and breast cancer, which were consistent with previous meta-analyses [131, 134137].

In spite of genetic importance, environment factors such as dietary pattern and exercise play important roles in the development of cancer. Recently, several studies have investigated the association between dietary intake of antioxidant-rich foods and MnSOD Val16Ala polymorphism in breast cancer [60], prostate cancer [89], and cervical cancer [14]. Despite the lack of consistent data, the results suggested that the MnSOD Val16Ala polymorphism and cancer risk could be modulated by dietary factors. Besides, a previous study had shown that moderate exercise training is beneficial for prostate cancer [138], and evidence showed that exercise training may result in positive MnSOD modulation through redox sensitive pathways [139].

The current meta-analysis has several advantages. First, we included the latest publications in the present study and also the publications written in Chinese. Second, the quality of included studies was assessed by the quality score criteria. Third, the FPRP test was performed to make the results more trustworthy and robust. Although the study is the largest and most comprehensive one regarding the association between MnSOD Val16Ala polymorphism and all cancer types, there were still some limitations that should be addressed. First, the number of cases in each study was small (<1000) in all but seven studies [11, 38, 69, 78, 82, 86, 119], which may have an effect on the investigation of the real association. Second, the results were based on unadjusted estimates, which might make the results imprecise. Third, only publications in English and Chinese were included, which could lead to selection bias. Fourth, in the subgroup analysis by cancer type, less than three studies were included for some types of cancer, which may affect the detection of the real association. Finally, the potential gene-gene, and gene-environment interactions were not investigated due to the lack of original information.

Despite of these limitations, this meta-analysis indicated there was a significant association between MnSOD Val16Ala polymorphism and cancer risk, which should be further validated by single large studies.

Conflicts of Interest

The authors declare that they have no conflicts of interest.

Acknowledgments

This work was supported by the Key Research Programs for Institutions of Higher Education in Henan Province (Grant no. 18A180012).