Cyclooxygenase‐2 gene polymorphisms reduce the risk of oral premalignant lesions

Abstract

BACKGROUND: Oral premalignant lesions (OPLs) have the potential to transform into malignant oral cancers. Overexpression of the cyclooxygenase‐2 gene (COX‐2) is observed frequently in OPLs and oral cancers, suggesting that this gene may play an important role in the progression of oral cancer. Single‐nucleotide polymorphisms of COX‐2 have been associated with the risk of multiple cancers; however, to date, their effects on OPL susceptibility have not been evaluated sufficiently. METHODS: The authors conducted a case‐control study that included 147 patients with OPL and a group of 147 healthy, matched controls. The effects of 3 potentially functional COX‐2 polymorphisms on the risk of OPL were evaluated: the −765 G→C polymorphism (rs20417), the exon 10 +837 T→C polymorphism (rs5275), and the exon 10 −90 C→T polymorphism (rs689470). RESULTS: The variant‐containing genotypes of COX‐2 exon 10 +837T→C variant were associated with a significantly reduced risk of OPL (odds ratio [OR], 0.48; 95% confidence interval [95% CI], 0.28‐0.80). This protective effect also was significant in men, younger individuals, ever smokers, and ever drinkers. Consistently, a common halotype WMW (in the following order: −765G→C, exon 10 +837T→C, and exon 10 −90C→T; w, widetype; M, variable allele) and a common diplotype (WWW/WMW) that contained the variant allele of exon 10 +837T→C, both were associated with a reduced risk of OPL (WMW: OR, 0.55; 95% CI, 0.33‐0.93; WWW/WMW: OR, 0.44; 95% CI, 0.22‐0.89). In addition, using never smokers with the variant‐containing genotypes as the reference group, interaction effects were observed between specific COX‐2 variants and tobacco smoking in the modulation of OPL risk. CONCLUSIONS: Overall, the current results provided the first epidemiologic evidence indicating that potentially functional polymorphisms of the COX‐2 gene may have an impact on individual susceptibility to OPLs. Cancer 2009. © 2009 American Cancer Society.

Role of COX-2 activity and CRP levels in patients with non-melanoma skin cancer. −765G>C PTGS2 polymorphism and NMSC risk

“…We observed higher COX-2 activity in SCC patients with ¡765GG genotype (2.17 § 1.6 U/mg protein) versus BCC patients with the same genotype (1.91 § 1.12 U/mg protein) that sustains no link between genotype and tumor type. Intra-group comparison showed an intra-group variation in BCC patients with signiWcant lower COX-2 activity in C-allele carriers compared to non-carriers (t test, p < 0.001), and also in SCC patients (allele C carriers vs. non-carriers, t test, p < 0.0001) which is similar to other Wndings regarding signiWcantly lower COX-2 activity in C-allele carriers [35,36]. On this basis, our results suggest that COX-2 activity might be a contributing factor to nonmelanoma skin cancer, but these Wndings reXect diVerences in unidentiWed characteristics other than PTGS2 G>C polymorphism.…”

“…These Wndings are in line with prior evidence regarding signiWcant lower COX-2 activity in C-allele carriers [35,36]. We found a signiWcant variation in COX-2 activity between BCC and SCC patients (p = 0.0019).…”

TNF superfamily gene polymorphism as prognostic factor in early breast cancer

“…CC genotype of this SNP was associated with an unfavorable distant DFS (DDFS) of EBC. A few previous studies also demonstrated that this genotype was significantly associated with the risk of Korean and Chinese lung adenocarcinomas (Hu et al 2005;Park et al 2006), bladder cancer (Yang et al 2008), and oral premalignant lesions adjusted for ethnicity (Pu et al 2009). However, further study is warranted to define the significance of this polymorphism in breast cancer due to the lack of statistical correlations with DFS and OS in the current study.…”

An association between the PTGS2 rs5275 polymorphism and colorectal cancer risk in families with inherited non-syndromic predisposition

“…There are reported significant associations with basal cell, bladder, breast, colorectal, gastric, lung, prostate and ovarian cancers. [14][15][16][17][18][19][20][21][22][23][24][25][26][27][28] Across cancer tissues of different origin, each allele paradoxically has been associated with increased risk of cancer. However, if studies are stratified by cancer type, there is greater consistency in the assignment of the risk allele.…”

“…29 Other biological factors such as gender and age and environmental factors such as non-steroidal antiinflammatory drug use and smoking status also influence the strength of association with cancer incidence. 15,25,27,28,30,31 Considered together, the prevailing evidence suggests that rs5275 is associated with cancer incidence, but the associations are complex.…”