The association between −1195G>A and digestive system cancers was further stratified by ethnicity, and we only found significantly increased risk in Asians compared to Caucasians, although the between-groups heterogeneity test was not significant, except for the recessive model (P < 0.001 for the heterogeneity test between groups;
race can explain nearly 100% of the heterogeneity between groups by meta regression; the P-value of the dummy variable was 0.004 for race) (Table 3). We used the Funnel plot and Egger’s test to address potential publication bias in the available literature. As shown in Figure 2, for −1195G>A, the shape of the funnel plot seemed symmetrical in the dominant model comparison in digestive system cancers, suggesting the absence of publication Selleckchem ABC294640 bias. Egger’s test was then used to provide statistical evidence for funnel plot symmetry, which is more pronounced when the larger of the intercept deviated from zero in the linear regression analysis. We also did not find significant publication bias (P = 0.147 for −1195G>A in the dominant model GA/AA vs GG). In late 1980s, COX-2 was discovered and postulated to be distinct from the constitutive COX (COX-1), because its activity was not regulated by glucocorticoids but induced at sites of inflammation.66,67 Subsequently, a large body
of studies investigated the role of COX-2 in cancer development. Up to now, it has been well known that COX-2 plays a key role in the carcinogenic process, especially for digestive system cancers.10,11 LGK-974 cell line In our meta-analysis, the COX-2−1195 variant A allele was associated with significantly increased risk of digestive system
cancers, but not for other cancers. The −1195G>A polymorphism is within the promoter region, which contains several key cis-acting regulatory elements, and has decisive roles in the regulation of COX-2 transcription.68 Zhang et al. reported that −1195 G>A change created a transcriptional factor c-myeloblastosis oncogene-binding site, and the −1195 A allele displayed higher transcription activity and mRNA expression compared with the −1195 G allele.69 Because COX-2 overexpression can increase proliferation, inhibit apoptosis, and enhance the invasiveness of cancer cells, the increased risk for the variant A alleles is biologically 上海皓元 plausible. COX-2−765G>C, also located in the promoter region, appears to disrupt a stimulatory protein 1 binding site, and leads to a 30% reduction of COX-2 promoter activity in vitro.70 In the current study, no significant association between COX-2−765G>C and the risk of both digestive system cancers and other cancers was observed. However, when we remove the three studies deviated from Hardy–Weinberg equilibrium, −765G>C was significantly associated with an increased risk of both total cancer and digestive system cancer. Further larger studies are needed to validate its real association with cancer risk.