Association of the MDM2 SNP285 Polymorphism with Cancer Susceptibility: A Meta-Analysis

The mouse double minute 2 (MDM2) gene encodes a negative regulator for p53, and the polymorphism SNP285 in the promoter region of MDM2 gene has been implicated in cancer risk, but individual published studies had inconclusive results. Therefore, we performed this meta-analysis to obtain a more precise estimation between MDM2 SNP285 polymorphism and risk of cancer. A systematic literature search was performed using the PubMed, Embase, and Chinese Biomedical (CBM) databases. Ultimately, 16 published studies comprising 14,573 cases and 9,115 controls were included. Pooled odds ratios (ORs) and 95% confidence intervals (CIs) were calculated to assess the strength of associations. Overall, MDM2 SNP285 polymorphism was significantly associated with a decreased overall cancer risk with the heterozygous model (OR = 0.89, 95% CI = 0.79–0.99), and reduced ORs were observed with other genetic models (dominant: OR = 0.90, 95% CI = 0.79–1.01 and allele comparison: OR = 0.91, 95% CI = 0.80–1.03) but not reaching statistical significance. Stratification analysis indicated a decreased risk for ovarian cancer, Caucasians, and studies with relatively large sample size. Despite some limitations, this meta-analysis indicated that the MDM2 SNP285 polymorphism was associated with a decreased cancer risk, which warrants further validation in large and well-designed studies.


Introduction
Cancer is a major public health problem and about 14.1 million new cancer cases and 8.2 million deaths occurred in 2012 worldwide according to the GLOBOCAN estimates [1]. The development and progression of cancer is a multistage process and alterations of oncogenes, tumor suppressor genes, and stability genes are responsible for tumorigenesis [2]. Tumor suppressor gene p53 is one of the most frequently mutated genes and numerous studies have reported that the p53 mutations play an important role in human cancers [3][4][5][6].
The mouse double minute 2 (MDM2) gene encodes a protein which could negatively regulate the activity of p53 tumor suppressor protein by binding to the latter and leading to its ubiquitination [7,8]. Elevated MDM2 levels have been detected in several human cancers even though with wildtype p53 due to the abnormal expression of MDM2 gene and/ or protein [9,10], which suggested as an alternative way for p53 inactivation in tumorigenesis [11]. Therefore, polymorphisms and mutations affecting MDM2 expression may contribute to the susceptibility to various cancers.
Two functional SNPs have been discovered which are located in the MDM2 intronic promoter (P2): SNP309 (rs2279744 T>G) [12] and SNP285 (rs117039649 G>C) [13] located 24 base pairs upstream of SNP309. Compared with the SNP309T allele, the G-variant of SNP309 increases MDM2 transcription through enhancing the binding of the transcription factor Sp1 [12]. In contrast, the presence of 2 Disease Markers SNP285C allele would reduce the binding strength between Sp1 and the MDM2 promoter in comparison with the SNP285G allele [13]. Although the presence of the SNP285C variants seems to antagonize the effect of SNP309G [13], there are some contradictory studies about MDM2 SNP285 in different types of cancers [13][14][15][16][17][18][19][20][21][22]. One of the possible reasons may be the relatively small sample size in individual published studies. Hence, we performed a meta-analysis to provide a more precise estimation of the relationship between MDM2 SNP285 polymorphism and cancer risk.

Search Strategy.
We systematically searched all relevant articles from PubMed, Embase, and Chinese Biomedical (CBM) databases using the search terms: "MDM2 or mouse double minute 2", "variant or polymorphism or variation", and "cancer or carcinoma or tumor or neoplasm" (last search was updated on May 19, 2016). We also checked all references of relevant reviews and eligible articles for additional studies. Only the latest or the largest study would be included in the current meta-analysis if studies were carried out with the same or overlapped subjects.

Inclusion and Exclusion Criteria.
Studies included in the meta-analysis had to meet the following criteria: (a) evaluating the association between MDM2 SNP285 polymorphism and cancer risk; (b) case-control design; (c) written in English or Chinese; (d) providing sufficient data to calculate odds ratios (ORs) with corresponding 95% confidence intervals (CIs).
The following exclusion criteria were used: (a) case-only studies or case reports; (b) conference abstracts, reviews, or meta-analysis; (c) duplicate publications; and (d) no available data reported.

Data Extraction. Two authors (Ping Wang and Meilin
Wang) independently extracted data from all the publications according to the inclusion and exclusion criteria. Disagreements were resolved through discussion until a consensus was reached. The following information was extracted from each study: first author's surname, year of publication, country of origin, ethnicity, cancer type, control source, total number of cases and controls, genotype methods, and numbers of cases and controls with the GG, GC, and CC genotypes. Ethnic backgrounds were categorized as Caucasians, Africans, or Mixed which contained more than one ethnic group. Chi-square-based test was used to assess the heterogeneity between studies, and < 0.10 was considered as significant heterogeneity exists. The fixed-effects model (the Mantel-Haenszel method) [23] was used when there was no significant heterogeneity; otherwise, the random-effects model (the DerSimonian and Laird method) [24] would be applied. Potential publication bias was assessed by the funnel plot as described previously [25]. The asymmetry of the funnel plot was evaluated using Egger's linear regression test [26]. Subgroup analysis was performed by cancer type (if one cancer type contained only one study, it would be merged into the "other" group), ethnicity, and sample size (<500 and ≥500). Sensitivity analysis was used to evaluate the effect of individual investigations on the overall cancer risk by excluding each investigation individually and recalculating the ORs and the 95% CIs [27]. All the statistical tests were performed with Stata (version 12.0; Stata Corporation, College Station, TX). All values were two-sided, and a < 0.05 was considered as statistically significant. Figure 1, a total of 1,282 published records were retrieved by using the search terms described above, consisting of 542 related studies from PubMed, 690 from Embase (494 overlapped studies were deducted), and 50 studies from CBM database. After rigorous assessment of abstracts and contents, only 14 publications met the crude inclusion criteria and were subjected for further evaluation. Of these 14 publications, four were excluded for being without detailed data for further evaluation [28][29][30][31], and a total of 10 publications met the inclusion criteria [13][14][15][16][17][18][19][20][21][22]. Of the 10 publications, one publication [18] with two ethnic groups was separated as two independent studies, two publications [13,16] with two cancer types were separated as two independent studies, and one publication [21] with four cancer types was also separated as four independent studies. A total of 10 publications including 16 studies were included in the final meta-analysis (Table 1).

Study Characteristics. As shown in
For those studies [13,15,16,21] that used the same control group, the control numbers were only calculated once in the total number of controls, and overlapped controls and cases were subtracted from the total number. Overall, 16 published studies of 14,573 cases and 9,115 controls were included in the final meta-analysis. Of the 16 studies, sample sizes of case ranged from 119 to 2501, in which four studies focused on breast cancer, three on lung cancer, two on ovarian cancer, prostate cancer, and cervical cancer, and others (colon cancer, hepatocellular carcinoma, and endometrial cancer) with only one study. There were 13 studies on Caucasians, two studies on mixed ethnicity, and only one study on Africans. Of all the studies, 14 were population-based, while two were hospitalbased, 7 studies with sample size less than 500 and 9 studies with sample size greater than 500. Most of the cases were histologically confirmed and controls were mainly matched for sex, age, and ethnicity.

Meta-Analysis
Results. The overall results for the MDM2 SNP285 polymorphism and cancer risk are shown in Table 2 and Figure 2. We found that there was a significant association between SNP285 polymorphism and overall cancer risk with the heterozygous model (OR = 0.89, 95% CI = 0.79-0.99), and reduced ORs were observed with other genetic models (dominant: OR = 0.90, 95% CI = 0.79-1.01 and allele comparison:   These suggested that publication bias might not have severe influence on the results of the current meta-analysis on the association between MDM2 SNP285 polymorphism and cancer risk.

Discussion
The association between MDM2 SNP285 polymorphism and cancer risk has been investigated by several research groups, but the conclusions were inconsistent. The most possible reason for the differences between studies is the small sample size in individual studies, which limits the statistical power to detect the real effects of polymorphism. We performed the current meta-analysis to combine the results of all available studies through a systematic search of relevant literatures, which may be useful for evaluating the contribution of SNP285 polymorphism to cancer. In this meta-analysis, which included 16 published studies of 14,573 cases and 9,115 controls, we found that MDM2 SNP285 polymorphism was significantly associated with a decreased overall cancer risk in the heterozygous model. Furthermore, the subgroup analysis showed that the association was more evident in the studies of ovarian cancer, Caucasians subjects, and relatively large sample size. MDM2 encodes a protein that binds to and facilitates the degradation of the p53 tumor suppressor protein via the ubiquitination pathway [32,33]. The cellular level of MDM2 controls both the p53 and pRb pathways and keeps the balance between growth arrest, cell death, senescence, and apoptosis; disturbances of these processes contribute to malignant transformation of cell [34]. MDM2 protein levels and function are precisely controlled at the transcription, translational, and posttranslation levels. Therefore, SNPs occurring in the MDM2 gene could potentially dysregulated both transcription and translation.     The others included endometrial cancer [16], hepatocellular carcinoma [19], colon cancer [21], and cervical cancer [22].   The MDM2 SNP285G>C is a newly discovered polymorphism, and this is the first report on the association of this polymorphism and cancer risk. It has been suggested that the MDM2 SNP285C allele might reduce the risk of female cancers (such as breast, ovarian, and endometrial cancer), due to the estrogen receptor (ESR) reported to be as a transcriptional activator with Sp1 [35] and one of the estrogen receptor binding elements (EREs) in the MDM2 P2 promoter overlapped with the Sp1 binding site harboring SNP285 [16]. In the present meta-analysis, a significantly decreased risk was found between MDM2 SNP285 polymorphism and ovarian cancer, and a much lower OR was found for breast cancer than that for male cancers (lung and prostate cancer), which was consistent with previous studies.
There are some limitations in our meta-analysis that remain to be addressed. Firstly, there is a limited number of studies that have actually analysed MDM2 SNP285 polymorphism with cancer susceptibility and the total sample size was Disease Markers 7 relatively small, which may lead to relatively weak power to detect the real association. Secondly, only published studies were included in this meta-analysis. It may be possible that some related unpublished studies as well as literatures published in languages other than English and Chinese were not included, which might lead to a bias to some extent. Thirdly, in the subgroup analysis, only one study was carried out in Africans. This subgroup did not have enough statistical power for us to investigate the real relationship. Fourthly, genotyping methods were different in each study, which will affect the bias. Additionally, our results were based on unadjusted estimates of ORs without adjustment for individual's age, sex, smoking status, drinking status, environmental factors, and other lifestyles.
In summary, this meta-analysis suggests that MDM2 SNP285 polymorphism was significantly associated with a decreased overall cancer risk with the heterozygous model. However, large and well-designed studies are warranted to validate our finding. Moreover, further studies estimating the effect of gene-gene and gene-environment interactions may provide a better, comprehensive understanding of the association between MDM2 SNP285 polymorphism and cancer risk.