OPN gene polymorphisms, rs17524488 GG/G, rs11730582 T/C, and rs9138 C/A, and cancer risk in a Chinese population

Previous studies have investigated the association between osteopontin (OPN) gene polymorphisms, rs17524488 (−156 GG/G), rs11730582 (−443 T/C), and rs9138 (C/A) and cancer risk in the Chinese population. However, the results are controversial and indefinite. We therefore carried out a meta-analysis to derive a more precise estimation of these associations. The PubMed database was systematically searched to identify potentially eligible reports. Crude odds ratios (OR) and 95% confidence intervals (CI) were used to assess the strength of associations between 3 OPN gene polymorphisms and cancer risk in a Chinese population. A total of 10 articles involving 2,391 cases and 3,007 controls were evaluated. The pooled OR indicated that OPN rs17524488 (−156 GG/G) polymorphism was significantly associated with cancer risk in Chinese population. In a stratified analysis by source of control, significant associations were also observed among rs17524488 (−156 GG/G) and rs11730582 (−443 T/C) polymorphisms and cancer. In addition, a stronger association was observed between rs9138 (C/A) polymorphism and cancer risk. In conclusion, this meta-analysis suggests that OPN rs17524488 (−156 GG/G), rs11730582 (−443 T/C), and rs9138 (C/A) polymorphisms may be associated with cancer susceptibility in the Chinese population. Nevertheless, further investigation on a larger population covering different ethnicities are warranted.

Scientific RepoRts | 5:14164 | DOi: 10.1038/srep14164 This is because OPN is a metastasis-related gene that contributes to the progression of over 30 types of cancer 14,15 .
The gene encoding OPN, also known as secreted phosphoprotein 1 (SPP1), is mapped on human chromosome 4q21-q25, together with other members of the SIBLING family of proteins, bone sialoprotein, and dentin matrix protein-1, which share some structural characteristics 16 .
More than 10 SNPs have been identified in the OPN promoter. These polymorphisms may affect the transcriptional activity of OPN and some of them are thought to be genetic risk factors for disease susceptibility [17][18][19] . Several polymorphisms in the human gene encoding OPN have been identified in the Chinese population, of which the rs17524488 (−156 GG/G), rs11730582 (−443 T/C), and rs9138 (C/A) polymorphisms are most frequently studied.  Considering the impact of the cancer risk potentially resulting from OPN gene, a number of studies have explored the association between these polymorphisms and cancer. However, individual studies have yielded inconsistent or conflicting findings, possibly caused by limitations associated with an individual study. To shed light on these contradictory results and to more precisely evaluate the relationship

Methods
Search strategy and inclusion criteria. We did our best to include all case-control studies published until date, regarding the association between OPN rs17524488 (−156 GG/G), rs11730582 (−443 T/C), and rs9138 (C/A) polymorphisms and cancer risk. Eligible studies were found by searching the PubMed database for relevant reports published between 2010 and 2014. The search terms were "Osteopontin", "polymorphism", and "cancer". In addition, the references of all retrieved articles were also manually searched for additionally relevant publications. The inclusion criteria were as follows: (1)      Heterogeneity was evaluated using a chi-square-based Q-test 30 , and the summary OR was determined with the Z-test. If P > 0.10 for the Q-test, a lack of heterogeneity among studies was found, meanwhile the fixed effects model should be used, otherwise, the random effects model should be used 31,32 . The HWE was assessed by a chi-square test in controls; P < 0.05 was considered significant. Sensitivity analysis was performed on excluded individual studies to assess the stability of the results. Publication bias was assessed by both Egger's test and Begg's test 33 . All statistical tests were used by Stata software (version 11.0; StataCorp LP, College Station, TX). Figure 1 and Tables 1, 2 show the study selection process and main characteristics of included studies, respectively. A total of 24 articles were retrieved based on the search criteria. Among them, 10 articles were excluded because they did not provide information about OPN gene polymorphism. An additional 4 articles, without control group data, were excluded. Thus, a total of 10 articles with 2,391 cases and 3,007 controls were included in the meta-analysis [20][21][22][23][24][25][26][27][28][29] . For the rs17524488 (−156 GG/G) polymorphism, 7 studies were available, including a total of 2,130 cases and 2,644 controls. For the rs11730582 (−443 T/C) polymorphism, 8 studies involved a total of 2,241 cases and 2,857 controls. For the rs9138 (C/A) polymorphism, 2 studies involved a total of 418 cases and 424 controls.     20,21 were not satisfied with the HWE, to make our analysis more powerful, we excluded these two studies and re-analysis. To our regret, no association was found, which indicated that the heterogeneity may exist in this SNP. Sensitivity analysis and publication bias. Sensitivity analysis was used to determine whether modification of the inclusion criteria affected the final results. The sensitivity analysis did not influence the results excessively by omitting any single study for rs17524488 (−156 GG/G) (Fig. 6). However, for rs11730582 (−443 T/C), a single study named Mu et al. 23 may influence the whole results (Fig. 7). Because only two studied of rs9138 (C/A), the sensitivity analysis was not examed. Egger's and Begg's tests were performed to assess publication bias and the funnel plot symmetry was examined. Finally, no proof of publication bias was obtained (Table 4, Figs 8,9).

Discussion
The overall goal of a meta-analysis is to combine the results of previous studies to arrive at a summary conclusion about a body of research. It is most useful in summarizing prior research when individual studies are too small to yield a valid conclusion. In this study, we analyzed the associations between OPN rs17524488 (−156 GG/G), rs11730582 (−443 T/C), and rs9138 (C/A) polymorphisms and cancer risk using a meta-analysis to obtain a powerful conclusion. To the best of our knowledge, this is the first meta-analysis providing comprehensive insights into the effects of the OPN rs17524488 (−156 GG/G), rs11730582 (−443 T/C), and rs9138 (C/A) polymorphisms and risk associated with all types of cancer in a Chinese population. Our meta-analysis included 2,391 cases and 3,007 controls [20][21][22][23][24][25][26][27][28][29] .
For the rs17524488 (−156 GG/G) polymorphism in the OPN promoter region, the overall results suggested that the subjects with G allele showed decreased susceptibility to cancer in a Chinese population. Moreover, individuals carrying either −156 GG or −443 CC genotype may have lower cancer susceptibility. However, people with AA genotype may have decreased cancer risk than CC-, CT, or CC + CT carriers. Considering that the previous single-institution study for cancer had a small sample size and may not justify the significance of current work, further studies are needed to clarify the effect of the 3 polymorphisms on the risk of cancer. A possible explanation for this phenomenon is that different polymorphisms may exert different effects on gene function, subsequently resulting in varying cancer susceptibility. Furthermore, a single gene or a single environmental factor is not likely to have a large effect on cancer susceptibility. Complex interactions between several genetic and environmental factors may be involved in cancer development.
Meta-analysis has been recognized as an effective method to answer a wide variety of clinical questions by summarizing and reviewing previously published, quantitative research. However, some limitations in our meta-analysis should be mentioned. First, our results were based on unadjusted estimates; more accurate outcomes would result from adjustments for other confounders such as gender, age, body mass index, lifestyle, and so on. Second, the studies included in this analysis were insufficient, especially in terms of a subgroup analysis. Thus, potential publication bias is very likely to exist, in spite of no evidence obtained from our statistical tests. Third, language of studies was limited to English, which may result in potential language bias. Fourth, a comparison of mRNA expression levels of the OPN gene between cancer and normal tissue should have been reported and included, which would better explain genetic function. Next, inter-gene and gene-environment interactions had not been evaluated owing to the absence of original data. Fifth, our study focused on Chinese people, other ethnicities should be reported and included. Sixth, a single study 23 may influence the whole results in the sensitivity analysis for the rs11730582 (−443 T/C) polymorphism, which suggested our study may be poorly powerful and stable. Finally, we also could not integrate different studies to look at the association between these three OPN SNPs and one specific cancer type, owning to insufficient publications. If there have a number of studies related different cancers in the future, this work may be carried out. In contrast, some advantages should also be highlighted. Our analysis comprehensively and systematically sheds light on the relationship between OPN rs17524488 (−156 GG/G), rs11730582 (−443 T/C), and rs9138 (C/A) polymorphisms and the susceptibility to cancer in the Chinese population. Additionally, due to the larger sample size, our meta-analysis increases the power and plausibility of our conclusion when compared with previous, individual studies. Finally, the studies included in this analysis were published between 2010 and 2014; thus, these studies are quite recent.
In summary, our meta-analysis suggests that OPN rs17524488 (−156 GG/G), rs11730582 (−443 T/C), and rs9138 (C/A) polymorphisms are associated with cancer risk in the Chinese population. Larger sample sizes of different ethnic populations are required to confirm our findings.