PTGS2 (COX-2) -765G > C Promoter Variant Reduces Risk of Colorectal
Total Page:16
File Type:pdf, Size:1020Kb
616 Cancer Epidemiology, Biomarkers & Prevention PTGS2 (COX-2) -765G > C Promoter Variant Reduces Risk of Colorectal Adenoma among Nonusers of Nonsteroidal Anti-inflammatory Drugs Cornelia M. Ulrich,1,2 John Whitton,1 Joon-Ho Yu,1,2 Justin Sibert,1 Rachel Sparks,1 John D. Potter,1,2 and Jeannette Bigler1 1Fred Hutchinson Cancer Research Center and 2Department of Epidemiology, Seattle, Washington Abstract Prostaglandin H synthase 2 (PTGS2) or cyclooxygenase-2 were GC 0.97 (95% CI, 0.65-1.46) and CC 0.24 (95% CI, 0.05- (COX-2) has been shown to play a key role in the regulation 1.11). Risk associated with the À765G > C variant differed of inflammation, and its inhibition is associated with a by aspirin or other nonsteroidal anti-inflammatory drug reduced risk of colon cancer. The PTGS2 (COX-2) À765G > (NSAID) use. Among nonusers of aspirin or other NSAIDs, C promoter variant is located in a putative SP1 binding site the CC genotype conferred a significant decrease in risk of and reduces PTGS2 expression. In a Minnesota-based case- adenoma (OR, 0.26; 95% CI, 0.07-0.89). Use of aspirin or control study of cases with adenomatous (n = 494) or other NSAIDs reduced risk of adenoma only among those hyperplastic polyps (n = 186) versus polyp-free controls with the À765GG (wild type) and possibly À765CG (n = 584), we investigated the role of the PTGS2 À765G > C genotypes (OR, 0.66; 95% CI, 0.48-0.92 and OR, 0.64; 95% promoter polymorphism. Multiple logistic regression ana- CI, 0.40-1.02, respectively). These data suggest that COX-2 lysis was used, adjusting for age, body mass index, caloric expression or activity may be beneficially suppressed, and intake, alcohol, fiber, sex, hormone use, and smoking. For risk of colorectal polyps reduced, by aspirin or other colorectal adenoma, odds ratios (OR) compared with PTGS2 NSAIDs in PTGS2 À765GG (wild type) individuals and À765GG as reference were GC 1.00 [95% confidence interval by the À765 CC variant genotype in nonusers of (95% CI), 0.74-1.35] and CC 0.53 (95% CI, 0.22-1.28). For NSAIDs. (Cancer Epidemiol Biomarkers Prev 2005; hyperplastic polyps, the comparable adjusted odds ratios 14(3):616–9) Introduction Prostaglandin H synthase (PTGS) or cyclooxygenase (COX) is a No nonsynonymous PTGS2 polymorphisms have yet been key enzyme in prostaglandin synthesis. This enzyme is reported among Caucasians (22, 23). A V511A polymorphism, bifunctional; the initial COX reaction converts arachidonic acid which occurs at about 5% allele frequency among African to prostaglandin G2; subsequently, the peroxidase reaction Americans, has been found to possibly decrease risk of both converts prostaglandin G2 to prostaglandin H2. Whereas colorectal adenoma and colorectal cancer (24). A common PTGS1 has been generally found to be constitutively expressed polymorphism in the regulatory region of PTGS2 has been and involved in cell-cell signaling and maintenance of tissue recently described (À765G > C; dbSNP rs20417); it results in homeostasis, PTGS2 expression occurs in a more limited reduced gene expression and is also associated with decreased number of cell types and is regulated or induced by specific serum concentrations of C-reactive protein in patients following stimulatory events (1-10). Thus, PTGS2 plays a role in the coronary bypass surgery (25) and reduced risk of myocardial prostanoid synthesis involved in inflammation and mitogene- infarction and stroke (26). We investigated associations between sis. A possible PTGS3 mRNA has been reported by several the PTGS2 promoter variant and risk of colorectal adenomatous groups, although expression of a functional PTGS3 protein and hyperplastic polyps. Furthermore, we investigated inter- in vivo remains controversial (11-14). actions between the PTGS2 polymorphism and use of aspirin or Whereas PTGS2 expression is low in normal colonic epithe- other nonsteroidal anti-inflammatory drugs (NSAID), poly- lium, its expression is elevated in up to 90% of colon carcinomas morphisms in TGFb1 (a PTGS2 inducer; ref. 27) and PTGS1. and 40% of colon adenoma (15-17). Furthermore, evidence from mouse experiments implicates PTGS2 in colorectal carcinogen- esis (18, 19), potentially through effects on prostaglandin E2 levels (20, 21). Oshima et al. (18) crossed Min mice (carrying an Materials and Methods APC-truncating mutation) with a PTGS2-deficient strain and showed a substantial reduction in polyps at 10 weeks among the Study Subjects. Participant recruitment for the Minnesota PTGS2À/À compared with PTGS2+/+ Min mice. These results case-control study has been described previously (28). Briefly, provide direct genetic evidence that PTGS2 plays a key role in cases with colorectal adenomatous and/or hyperplastic polyps tumorigenesis at the stage of adenoma formation. and polyp-free control subjects were recruited through a large multiclinical private gastroenterology practice in metropolitan Minneapolis. Patients ages 30 to 74 years who were scheduled for a colonoscopy between April 1991 and April 1994 were Received 7/13/04; revised 9/23/04; accepted 10/1/04. recruited before colonoscopy so as to blind patients and Grant support: Grants R01CA89445 and R01CA59045. recruiters to the final diagnosis. The study was approved by The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. the internal review boards of the University of Minnesota and Section 1734 solely to indicate this fact. each endoscopy site. Written informed consent was obtained. Requests for reprints: Cornelia M. Ulrich, Cancer Prevention Program, Fred Hutchinson Cases were identified as meeting eligibility criteria: resident Cancer Research Center, 1100 Fairview Avenue N, M4-B402, Seattle, WA 98109-1024. Phone: 206-667-7617; Fax: 206-667-7850. E-mail: [email protected] of Twin Cities metropolitan area, English speaking, no known Copyright D 2005 American Association for Cancer Research. genetic syndrome associated with predisposition to colonic Cancer Epidemiol Biomarkers Prev 2005;14(3). March 2005 Downloaded from cebp.aacrjournals.org on September 27, 2021. © 2005 American Association for Cancer Research. Cancer Epidemiology, Biomarkers & Prevention 617 Table 1. Risk of colorectal adenomatous and hyperplastic Genotyping methods for polymorphisms in PTGS1 (R8W, polyps associated with PTGS2 À765G > C genotype L15-L16del, P17L, and L237M) or TGFb1 (L10P) by our group (multivariate-adjusted ORs) have been published previously (30, 31). Case versus PTGS2 À No. cases/ ORs* Statistical Data Analysis. Logistic regression analysis was polyp-free control 765G > C controls (95% CI) used to estimate odds ratios (OR) and corresponding 95% confidence intervals (95% CI) comparing cases (with adeno- Adenoma GG (wt) 344/405 1.00 (reference) GC (het) 140/159 1.00 (0.74-1.35) matous or hyperplastic polyps) with polyp-free controls in CC (hzv) 10/20 0.53 (0.22-1.28) association with PTGS2 genotypes. Multivariate adjustment by Hyperplastic GG (wt) 136/405 1.00 (reference) previously identified risk factors (age, sex, body mass index, polyps dietary intakes of alcohol, fiber, kilocalories, hormone replace- GC (het) 47/159 0.97 (0.65-1.46) ment therapy, or smoking) was used, because there was some CC (hzv) 3/20 0.24 (0.05-1.11) confounding in the stratified analyses. To evaluate possible NOTE: Abbreviations: wt, wild type; het, heterozygote; hzv, homozygote variant. interactions between the PTGS2 polymorphism and aspirin/ * Sex, age, body mass index, caloric intake, alcohol, fiber, hormone use, and other NSAID use or polymorphisms in TGFb1 or PTGS1, the smoking. respective multiplicative interaction terms were included in the logistic regression models. All statistical tests were two sided and analyses were undertaken with SAS 8.02 (SAS neoplasia, no individual history of cancer (except nonmela- Institute, Cary, NC). noma skin cancer), no history of inflammatory bowel disease and having a first diagnosis of colon or rectal adenomatous (n = 521) or hyperplastic polyp (n = 194) at the time of the Results colonoscopy. Control subjects were free of polyps during colonoscopy (n = 621). Patients for whom the colonoscopy did Characteristics of the study population and risk factors for not reach the cecum were ineligible; removed polyps were colorectal polyps have been described previously (28, 32-34). examined histologically using standard diagnostic criteria (29). Briefly, adenoma cases were older than individuals with Information on use of aspirin and NSAID, lifestyle factors and hyperplastic polyps or polyp-free controls and more likely to diet, anthropometry, demographics, and medical information, be male. including family history of cancer and polyps, were obtained We investigated the risk of colorectal polyps associated with by questionnaire. Regular, current use of NSAIDS was defined the À765G > C polymorphism in the promoter region of as NSAID use >1/wk versus V1/wk. The participation rate for PTGS2 (variant allele frequency among controls = 0.17). ORs all colonoscoped patients was 68%. associated with the À765C variant are shown in Table 1. Risk of adenoma or hyperplastic polyps did not significantly differ Genotyping. Genomic DNA was extracted from peripheral by genotype, although there was an indication that the WBCs using the Puregene kit (Gentra Systems, Minneapolis, homozygote variant CC genotype may be associated with a MN). The À765G > C polymorphism in the COX-2 promoter reduced risk (adenoma OR, 0.53; 95% CI, 0.22-1.28); hyper- region was genotyped on a 7900HT Sequence Detection System plastic polyps OR 0.24 (95% CI, 0.05-1.11). These associations (Applied Biosystems, Foster City, CA) using a fluorescent allelic did not differ by polyp location (proximal, distal, or rectal) or discrimination assay. Amplification was done in 20-AL reactions polyp size (data not shown). using the Taqman core reagent kit (Applied Biosystems) and Regular aspirin and NSAID use (>1/wk for at least 1 year) minor groove binding probes with nonfluorescent quenchers has been previously identified in this population as associated (Applied Biosystems).