Two isoforms of prostaglandin H synthase (PTGS = COX) are key enzymes in prostaglandin synthesis and primary targets for aspirin and other non-steroidal anti-inflammatory drugs (NSAIDs). Use of aspirin or other NSAIDs is associated with a lower risk and reduced recurrence of colorectal adenomas, established precursors of adenocarcinoma. This study investigated risk of colorectal adenomatous and hyperplastic polyps associated with several polymorphisms in the coding region of PTGS1. Within the Minnesota polyp case-control study, patients with colorectal adenomatous (n = 521) or hyperplastic (n = 194) polyps and n = 621 polyp-free controls were genotyped for four PTGS1 polymorphisms (R8W, L15-L16del, P17L, L237M); these had been predicted to affect protein function based on sequence-homology software. Age- and sex-adjusted odds ratios (ORs) and 95% confidence intervals (CIs) were computed. Whereas there was no appreciable difference in adenoma or hyperplastic polyp risk associated with R8W, P17L, and L237M, an increased risk was observed for individuals heterozygous for the L15-L16del polymorphism (OR = 3.6, 95% CI 1.2–11.2). The variant L15-L16del allele appeared to be associated with a stronger increase in adenoma risk among nonusers of aspirin/other NSAIDs. The reduced risk observed with aspirin/other NSAID use was limited to those wild type for P17L [PP users: OR = 0.6 (0.5–0.8) versus PP nonusers: 1.0 (referent) (P interaction = 0.03)]. To our knowledge, this study represents the first investigation of polymorphisms in PTGS1 and risk of colorectal polyps. The L15-L16del variant allele may result in an increased risk of colorectal adenomas, whereas P17L may be relevant to the pharmacogenetics of aspirin. These preliminary findings require confirmation in larger studies of colorectal neoplasia.

Prostaglandin H synthase (PTGS) catalyzes the formation of prostaglandin G2, and, through its peroxidase activity, prostaglandin H2 (1). The latter serves as the precursor for a number of important prostanoids, including prostaglandin E2. Two PTGS isoforms, PTGS1 and PTGS2, have been well characterized and share approximately 60% homology. PTGS1 is constitutively expressed and appears to be important in maintaining prostanoid levels for “housekeeping” functions; PTGS2 is inducible and plays a critical role in many inflammatory responses (2). There is strong evidence that use of PTGS inhibitors, such as aspirin or other non-steroidal anti-inflammatory drugs (NSAIDs), is associated with a reduced risk of colorectal polyps and cancer (3, 4), and decreases the recurrence of colorectal adenomas (5, 6). There is also some evidence connecting regular NSAID use to lower risks of cancers of the digestive tract, breast, and lung cancer (3, 4, 7–12). Evidence from mouse experiments implicates not only PTGS2, but also PTGS1 in colorectal carcinogenesis (13, 14), potentially through effects on prostaglandin E2 levels (15, 16). We recently identified several polymorphisms in the coding region of PTGS1 (17), predicted that several of these may affect protein function based on sequence homology software applications (R8W, P17L, L237M), and show here that some of these variants affect risk of colorectal adenomas, possibly in interaction with aspirin or other NSAID use.

Study Subjects

Participant recruitment for this case-control study has been described previously (18). Briefly, cases with colorectal adenomatous and/or hyperplastic polyps and polyp-free control subjects were recruited through a large multiclinic private gastroenterology practice in metropolitan Minneapolis. Patients aged 30–74 years who were scheduled for a colonoscopy between April 1991 and April 1994 were recruited before colonoscopy so as to blind patients and recruiters to the final diagnosis. The study was approved by the internal review boards of the University of Minnesota and each endoscopy site. Written informed consent was obtained.

Cases were identified as meeting eligibility criteria (see Ref. 18) and having a first diagnosis of colon or rectal adenomatous (n = 521) or hyperplastic polyp (n = 194) at the time of the colonoscopy. Control subjects were free of polyps during colonoscopy (n = 621). Patients for whom the colonoscopy did not reach the cecum were ineligible; removed polyps were examined histologically using standard diagnostic criteria (19). Information on use of aspirin and NSAIDs, lifestyle factors and diet, anthropometry, demographics, and medical information, including family history of cancer and polyps, were obtained by questionnaire. The participation rate for all colonoscoped patients was 68%.

Genotyping

Genomic DNA was extracted from peripheral white blood cells using the Puregene kit (Gentra Systems, Minneapolis, MN). PTGS1 genotyping was performed at the Core Laboratory of the Public Health Sciences Division of the Fred Hutchinson Cancer Research Center (J.B.). RFLP was used to genotype the PTGS1 R8W polymorphism as described (17). Genotyping of the P17L, and L237M polymorphisms was performed by sequencing, which resulted in the discovery of a new polymorphism, L15-L16del (17). For quality control purposes, 94 randomly selected samples were genotyped twice for each polymorphism. There were no discrepancies. The genotype frequencies did not deviate from Hardy-Weinberg equilibrium at the 0.05% significance level.

Statistical Data Analysis

Logistic regression analysis was used to estimate odds ratios (ORs) and corresponding 95% confidence intervals (CIs) comparing cases (with adenomatous or hyperplastic polyps), to polyp-free controls in association with PTGS1 genotypes, adjusting for age and sex. Multivariate adjustment by previously identified risk factors (body mass index, race, physical activity, dietary intakes of alcohol, fiber, or kilocalories, hormone replacement therapy, or smoking) did not alter the ORs appreciably (data not shown). For evaluating interaction between PTGS1 polymorphisms and NSAID use, the respective interaction terms were included in the logistic regression models. All statistical tests were two sided and analyses were undertaken with SAS 8.02 (SAS Institute, Cary, NC).

Haplotype frequencies were estimated using the expectation maximization algorithm as implemented in the program EH (20). Permutation testing (1000 replications) was used to evaluate the significance of (a) differences between observed and expected haplotype frequencies in controls, and (b) differences between observed haplotype frequencies in cases and controls (21).

Characteristics of the study population and risk factors for colorectal polyps have been described previously (18, 22–24). Briefly, adenoma cases were older than individuals with hyperplastic polyps or polyp-free controls and more likely to be male.

We investigated four polymorphisms within the coding region of PTGS1, R8W (variant allele frequency among controls = 0.07), L15-L16del (allele frequency = 0.01), P17L (allele frequency = 0.07), and L237M (allele frequency = 0.03). Genotype frequencies among case groups and controls are given in Table 1. The R8W, P17L, and L237M variants were selected because of their likely phenotypic impact based on predictions using sequence homology software (25, 26), and expected allele frequency (≥0.03) in the Minnesota Caucasian population (17). A new polymorphism, L15-L16del (a deletion of two leucines), was discovered as part of genotyping for this study. Whereas R8W, L15-L16del, and P17L are found in the PTGS1 signal peptide sequence which targets the protein for translocation into the lumen of the endoplasmic reticulum, L237M is located near the PTGS1 dimer interface connecting two identical monomeric subunits (27).

Table 1.

PTGS1 genotype frequencies among adenoma cases, hyperplastic polyp cases, and controls

PTGS1 polymorphismGenotypeAdenomas N (%)Hyperplastic polyps N (%)Controls N (%)
R8W     
 Wt/wt 445 (85.4) 164 (84.5) 539 (86.8) 
 Wt/var 73 (14.0) 30 (15.5) 79 (12.7) 
 Var/var 3 (0.6) 0 (0.0) 3 (0.5) 
L15-L16del     
 Wt/wt 510 (97.9) 191 (98.5) 616 (99.2) 
 Wt/var 11 (2.1) 3 (1.5) 5 (0.8) 
P17L     
 Wt/wt 451 (86.6) 173 (89.2) 527 (84.9) 
 Wt/var 63 (12.1) 20 (10.3) 90 (14.5) 
 Var/var 7 (1.3) 1 (0.5) 4 (0.6) 
L237M     
 Wt/wt 493 (94.6) 183 (94.3) 585 (94.2) 
 Wt/var 28 (5.4) 11 (5.7) 36 (5.8) 
PTGS1 polymorphismGenotypeAdenomas N (%)Hyperplastic polyps N (%)Controls N (%)
R8W     
 Wt/wt 445 (85.4) 164 (84.5) 539 (86.8) 
 Wt/var 73 (14.0) 30 (15.5) 79 (12.7) 
 Var/var 3 (0.6) 0 (0.0) 3 (0.5) 
L15-L16del     
 Wt/wt 510 (97.9) 191 (98.5) 616 (99.2) 
 Wt/var 11 (2.1) 3 (1.5) 5 (0.8) 
P17L     
 Wt/wt 451 (86.6) 173 (89.2) 527 (84.9) 
 Wt/var 63 (12.1) 20 (10.3) 90 (14.5) 
 Var/var 7 (1.3) 1 (0.5) 4 (0.6) 
L237M     
 Wt/wt 493 (94.6) 183 (94.3) 585 (94.2) 
 Wt/var 28 (5.4) 11 (5.7) 36 (5.8) 

In this population, no haplotypes were estimated to carry two of the variant PTGS1 alleles. We observed no evidence for linkage disequilibrium between the variant alleles, neither considering haplotypes based on all four polymorphisms (P = 0.10) nor considering two polymorphisms at a time (all six P values > 0.05). However, due to the rarity of the variant alleles, a much larger sample size would be needed to reliably test for linkage disequilibrium.

Haplotype frequencies did not differ between either one of the case groups and controls (P = 0.56 and P = 0.63, respectively). ORs estimating the relative risk associated with each PTGS1 variant are shown in Table 2. On the basis of the risk patterns observed, individuals with at least one variant allele of R8W and P17L were grouped together. There was no appreciable difference in adenoma or hyperplastic polyp risk associated with PTGS1 polymorphisms R8W, P17L, and L237M. A statistically significant increase in risk was seen for those heterozygous for the L15-L16del polymorphism (age and sex-adjusted OR = 3.6, 95% CI 1.2–11.2). This risk was perhaps further increased among those with synchronous adenomatous and hyperplastic polyps (OR = 5.8, 95% CI 1.2–28.3) (data not shown). However, due to the small number of individuals carrying the L15-L16del allele, all of these estimates were quite imprecise. For all variants, there were no appreciable differences based on stratification by age group (>50 years versus younger), or polyp location.

Table 2.

Risk of colorectal adenomatous and hyperplastic polyps associated with PTGS1 polymorphisms (age- and sex-adjusted ORs)

PTGS1 polymorphismGenotypeNumber of cases/controlsOR95% CI
Adenomas     
R8W Wt/wt 445/539 1.0 Ref 
 Wt/var or var/var 76/82 1.1 (0.8–1.6) 
L15-L16del Wt/wt 510/616 1.0 Ref 
 Wt/var 11/5 3.6 (1.2–11.2) 
P17L Wt/wt 451/527 1.0 Ref 
 Wt/var or var/var 70/94 0.9 (0.6–1.2) 
L237M Wt/wt 493/585 1.0 Ref 
 Wt/var 28/36 0.8  (0.5–1.4) 
Hyperplastic polyps     
R8W Wt/wt 164/539 1.0 Ref 
 Wt/var or var/var 30/82 1.2 (0.8–1.9) 
L15-L16del Wt/wt 191/616 1.0 Ref 
 Wt/var 3/5 2.1 (0.5–9.1) 
P17L Wt/wt 173/527 1.0 Ref 
 Wt/var or var/var 21/94 0.7 (0.4–1.1) 
L237M Wt/wt 183/585 1.0 Ref 
 Wt/var 11/36 1.0  (0.5–2.0) 
PTGS1 polymorphismGenotypeNumber of cases/controlsOR95% CI
Adenomas     
R8W Wt/wt 445/539 1.0 Ref 
 Wt/var or var/var 76/82 1.1 (0.8–1.6) 
L15-L16del Wt/wt 510/616 1.0 Ref 
 Wt/var 11/5 3.6 (1.2–11.2) 
P17L Wt/wt 451/527 1.0 Ref 
 Wt/var or var/var 70/94 0.9 (0.6–1.2) 
L237M Wt/wt 493/585 1.0 Ref 
 Wt/var 28/36 0.8  (0.5–1.4) 
Hyperplastic polyps     
R8W Wt/wt 164/539 1.0 Ref 
 Wt/var or var/var 30/82 1.2 (0.8–1.9) 
L15-L16del Wt/wt 191/616 1.0 Ref 
 Wt/var 3/5 2.1 (0.5–9.1) 
P17L Wt/wt 173/527 1.0 Ref 
 Wt/var or var/var 21/94 0.7 (0.4–1.1) 
L237M Wt/wt 183/585 1.0 Ref 
 Wt/var 11/36 1.0  (0.5–2.0) 

Regular aspirin and NSAID use (>1/week) have been previously identified in this population as associated with reduced adenoma risk (24) [age and sex-adjusted OR = 0.70 (0.53–0.92) and 0.65 (0.53–0.92), respectively]. We evaluated whether associations with current regular use of aspirin or NSAIDs differed depending on PTGS1 genotypes (Table 3). There was indication that the L15-L16del polymorphism was associated with a stronger increase in adenoma risk among non-users of aspirin/other NSAIDs. Furthermore, the P17L polymorphism modified risk of adenomas associated with regular current use of aspirin/other NSAIDs: among those with 17PP genotypes (wt/wt) regular use of aspirin/other NSAIDs was associated with a significantly decreased risk [OR = 0.6 (0.5–0.8)]; yet, among individuals carrying at least one variant 17 L allele, no risk reduction was observed [compared to the reference group of wt/wt non-users: wt/var or var/var non-users OR = 0.6 (0.5–1.0) versus wt/var or var/var users OR = 0.8 (0.5–1.4) (P interaction = 0.03)]. Similarly, the inverse association observed with aspirin/NSAID use was limited to those wild type for the L237M polymorphism [users: OR = 0.6 (0.5–0.9) versus nonusers: 1.0 (ref)], although this interaction was not statistically significant (P = 0.22). However, these preliminary findings may be spurious and clearly require confirmation in a larger population. Risk patterns for hyperplastic polyps were generally similar to those for adenomas (data not shown).

Table 3.

Risk of colorectal adenomas associated with PTGS1 polymorphisms stratified by use of aspirin or other NSAIDs (age- and sex-adjusted ORs)

PTGS1 polymorphismGenotypeRegular, current use of aspirin or other NSAIDs
No
Yes
OR (95% CI) (N cases/n controls)OR (95% CI) (N cases/n controls)
R8W Wt/wt 1.0 (ref) 0.7 (0.5–0.9) 
  278/302 167/237 
 Wt/var or var/var 1.3 (0.8–2.1) 0.6 (0.4–1.1) 
  50/42 26/40 
   P interaction = 0.31 
L15-L16del Wt/wt 1.0 (ref) 0.7 (0.5–0.9) 
  320/343 190/273 
 Wt/var 12.8 (1.4–115) 1.0 (0.2–4.9) 
  8/1 3/4 
   P interaction = 0.12 
P17L Wt/wt 1.0 (ref) 0.6 (0.5–0.8) 
  287/288 164/239 
 Wt/var or var/var 0.6 (0.4–0.97) 0.8 (0.5–1.4) 
  41/56 29/38 
   Pinteraction = 0.03 
L237M Wt/wt 1.0 (ref) 0.6 (0.5–0.9) 
  314/324 179/261 
 Wt/var 0.6 (0.3–1.3) 0.8 (0.4–1.6) 
  14/20 14/16 
   P interaction = 0.22 
PTGS1 polymorphismGenotypeRegular, current use of aspirin or other NSAIDs
No
Yes
OR (95% CI) (N cases/n controls)OR (95% CI) (N cases/n controls)
R8W Wt/wt 1.0 (ref) 0.7 (0.5–0.9) 
  278/302 167/237 
 Wt/var or var/var 1.3 (0.8–2.1) 0.6 (0.4–1.1) 
  50/42 26/40 
   P interaction = 0.31 
L15-L16del Wt/wt 1.0 (ref) 0.7 (0.5–0.9) 
  320/343 190/273 
 Wt/var 12.8 (1.4–115) 1.0 (0.2–4.9) 
  8/1 3/4 
   P interaction = 0.12 
P17L Wt/wt 1.0 (ref) 0.6 (0.5–0.8) 
  287/288 164/239 
 Wt/var or var/var 0.6 (0.4–0.97) 0.8 (0.5–1.4) 
  41/56 29/38 
   Pinteraction = 0.03 
L237M Wt/wt 1.0 (ref) 0.6 (0.5–0.9) 
  314/324 179/261 
 Wt/var 0.6 (0.3–1.3) 0.8 (0.4–1.6) 
  14/20 14/16 
   P interaction = 0.22 

In addition to investigating genotypes at each polymorphic site in direct comparison (e.g., wt/wt versus wt/het), we also undertook analyses restricting the reference group to individuals wt/wt for all four polymorphisms (wt/wt haplotypes). ORs were virtually identical, although with wider CIs, attributable to the smaller number of individuals in the reference category.

To our knowledge, this study represents the first investigation of polymorphisms in PTGS1 and risk of colorectal polyps. Although more attention has been given to PTGS2 as a key player in colorectal carcinogenesis, recent experiments among Min mice, an animal model for intestinal polyp formation, implicate PTGS1 as equally important for intestinal carcinogenesis (14). PTGS1 knock-out mice showed a similar reduction in polyp burden as PTGS2 knock-outs compared to the wild-type PTGS1 (+/+) or PTGS2 (+/+) Min/+ mouse (14). Epidemiological studies of aspirin intake show reduced risks of colorectal adenomas and cancer associated with regular use (3), and effective prevention of adenoma recurrence (6, 28). These reduced risks are seen with low-dose and low-frequency aspirin, most likely insufficient to inhibit COX-2, and suggest that mechanisms other than direct inhibition of COX-2 may be involved. One hypothesis is that COX-1 activity in platelets releases lipid and protein mediators that may induce COX-2 via a paracrine mechanism in the formation of neoplasia (3). Thus, both animal and epidemiological studies support the concept that both PTGS1 and PTGS2 are involved in colorectal carcinogenesis.

We here report that a variant in the signal peptide of PTGS1 (L15-L16del) may result in an elevated risk of colorectal adenomas. Risk associated with this genotype appeared stronger among individuals currently not using other NSAIDs or aspirin on a regular basis. The consistency of the findings for L15-L16del for both hyperplastic and adenomatous polyps provides support for a true association. Yet, given the overall rarity of the variant PTGS1 alleles, our findings could also be attributable to chance and should be interpreted with caution until confirmed.

Furthermore, the P17L polymorphism appeared to modify the effects of regular aspirin/other NSAID use on colorectal polyp risk. Halushka et al. recently presented data indicating that this variant (which exists in complete linkage disequilibrium with the promoter polymorphism A-842G) may slightly increase the inhibitory effect of aspirin on platelet cyclooxygenase activity (29). This in vitro finding is seemingly at odds with the observation in our study population of a benefit of aspirin/other NSAIDSs limited to those with the wild-type PP genotype. Yet, the effects of the P17L variant on cyclooxygenase inhibition by aspirin in vitro may not be conclusive, as they were based on changes in the levels of prostaglandin F2a, which is a minor cyclooxygenase metabolite in platelets (30). Clearly, evaluation of the pharmacogenetic relevance of this PTGS1 polymorphism will require further experimental and pharmacological data.

We targeted our study to PTGS1 polymorphisms located in regions encoding amino acids that are highly conserved among PTGS1 proteins of different species, and have been predicted to exert phenotypic effects by two sequence-based algorithms (17, 25, 26). There is now some limited experimental evidence suggesting that the P17L variant has functional consequences, whereas R8W was not found to alter the formation of prostaglandin F in human platelets in vitro (31). Unfortunately, L15-L16del and L237M were not evaluated for functional impact in these experimental studies. The L15-L16del polymorphism results in the deletion of two amino acids in the PTGS1 signal peptide, which targets the protein for translocation into the lumen of the endoplasmic reticulum (32). Evaluation of the PTGS1 signal peptide polymorphisms with the SignalP software program, which predicts the likelihood of an NH2-terminal segment being recognized as a signal peptide and the site of signal peptidase cleavage (33), suggested that the polymorphisms would have little impact.2

2

R. Kulmacz, personal communication.

These findings reduce but do not abolish the likelihood that the L15-L16del polymorphism could cause alterations in protein location as a result of miscompartmentalization. A signal peptide may have multiple functions and consequences of genotypic variation in signal peptides are not completely understood (34).

In conclusion, the L15-L16del polymorphism in the PTGS1 gene may result in an increased risk of colorectal adenomas, particularly among non-aspirin/other NSAID users. Our finding of effect modification between the PTGS1 P17L variant and aspirin/other NSAID use could potentially be of pharmacogenetic relevance. Yet in light of the rarity of the variant alleles of all PTGS1 polymorphisms, our findings should be considered preliminary. Larger studies of both colorectal adenomas and cancer should be undertaken investigating these PTGS1 variants and their interaction with NSAID use.

Grant support: Grants R01CA89445 and R01CA59045.

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. Section 1734 solely to indicate this fact.

We thank Dr. Richard Kulmacz for discussions on the potential functional consequences of the PTGS1 polymorphisms and Joon-Ho Yu for assistance with the manuscript resubmission.

1
Kulmacz RJ. Cellular regulation of prostaglandin H synthase catalysis.
FEBS Lett
1998
;
430
:
154
-7.
2
Funk CD. Prostaglandins and leukotrienes: advances in eicosanoid biology.
Science
2001
;
294
:
1871
-5.
3
Thun MJ, Henley SJ, Patrono C. Nonsteroidal anti-inflammatory drugs as anticancer agents: mechanistic, pharmacologic, and clinical issues.
J Natl Cancer Inst
2002
;
94
:
252
-66.
4
Schreinemachers DM, Everson RB. Aspirin use and lung, colon, and breast cancer incidence in a prospective study [see comments].
Epidemiology
1994
;
5
:
138
-46.
5
Sandler RS, Halabi S, Baron JA, et al. A randomized trial of aspirin to prevent colorectal adenomas in patients with previous colorectal cancer.
N Engl J Med
2003
;
348
:
883
-90.
6
Baron JA, Cole BF, Sandler RS, et al. A randomized trial of aspirin to prevent colorectal adenomas.
N Engl J Med
2003
;
348
:
891
-9.
7
Akre K, Ekstrom AM, Signorello LB, Hansson LE, Nyren O. Aspirin and risk for gastric cancer: a population-based case-control study in Sweden.
Br J Cancer
2001
;
84
:
965
-8.
8
Farrow DC, Vaughan TL, Hansten PD, et al. Use of aspirin and other nonsteroidal anti-inflammatory drugs and risk of esophageal and gastric cancer.
Cancer Epidemiol Biomark Prev
1998
;
7
:
97
-102.
9
Corley DA, Kerlikowske K, Verma R, Buffler P. Protective association of aspirin/NSAIDs and esophageal cancer: a systematic review and meta-analysis.
Gastroenterology
2003
;
124
:
47
-56.
10
Cotterchio M, Kreiger N, Sloan M, Steingart A. Nonsteroidal anti-inflammatory drug use and breast cancer risk.
Cancer Epidemiol Biomark Prev
2001
;
10
:
1213
-7.
11
Khuder SA, Mutgi AB. Breast cancer and NSAID use: a meta-analysis.
Br J Cancer
2001
;
84
:
1188
-92.
12
Muscat JE, Chen SQ, Richie JP Jr, et al. Risk of lung carcinoma among users of nonsteroidal antiinflammatory drugs.
Cancer
2003
;
97
:
1732
-6.
13
Oshima M, Dinchuk JE, Kargman SL, et al. Suppression of intestinal polyposis in Apc δ716 knockout mice by inhibition of cyclooxygenase 2 (COX-2).
Cell
1996
;
87
:
803
-9.
14
Chulada PC, Thompson MB, Mahler JF, et al. Genetic disruption of Ptgs-1, as well as Ptgs-2, reduces intestinal tumorigenesis in Min mice.
Cancer Res
2000
;
60
:
4705
-8.
15
Watanabe K, Kawamori T, Nakatsugi S, et al. Role of the prostaglandin E receptor subtype EP1 in colon carcinogenesis.
Cancer Res
1999
;
59
:
5093
-6.
16
Hansen-Petrik MB, McEntee MF, Jull B, Shi H, Zemel MB, Whelan J. Prostaglandin E(2) protects intestinal tumors from nonsteroidal anti-inflammatory drug-induced regression in Apc(Min/+) mice.
Cancer Res
2002
;
62
:
403
-8.
17
Ulrich CM, Bigler J, Sibert J, et al. Cyclooxygenase 1 (COX1) polymorphisms in African-American and Caucasian populations. Human Mutation (Mutation in Brief) Online: #551, 2002.
18
Potter JD, Bostick RM, Grandits GA, et al. Hormone replacement therapy is associated with lower risk of adenomatous polyps of the large bowel: the Minnesota Cancer Prevention Research Unit Case-Control Study.
Cancer Epidemiol Biomark Prev
1996
;
5
:
779
-84.
19
O'Brien M, Winawer SJ, Zauber AG, et al. The National Polyp Study. Patient and polyp characteristics associated with high-grade dysplasia in colorectal adenomas.
Gastroenterology
1990
;
98
:
371
-9.
20
Xie X, Ott J. Testing linkage disequilibrium between a disease gene and marker loci.
Am J Hum Genet
1993
;
53
:
1107.
21
Zhao JH, Curtis D, Sham PC. Model-free analysis and permutation tests for allelic associations.
Hum Hered
2000
;
50
:
133
-9.
22
Morimoto LM, Newcomb PA, Ulrich CM, Bostick RM, Lais CJ, Potter JD. Risk factors for hyperplastic and adenomatous polyps: evidence for malignant potential?
Cancer Epidemiol Biomark Prev
2002
;
11
:
1012
-8.
23
Ulrich CM, Kampman E, Bigler J, et al. Colorectal adenomas and the C677T MTHFR polymorphism: evidence for gene-environment interaction?
Cancer Epidemiol Biomark Prev
1999
;
8
:
659
-68.
24
Bigler J, Whitton J, Lampe JW, Fosdick L, Bostick RM, Potter JD. CYP2C9 and UGT1A6 genotypes modulate the protective effect of aspirin on colon adenoma risk.
Cancer Res
2001
;
61
:
3566
-9.
25
Ng PC, Henikoff S. Predicting deleterious amino acid substitutions.
Genome Res
2001
;
11
:
863
-74.
26
Taylor NE, Greene EA. PARSESNP: a tool for the analysis of nucleotide polymorphisms.
Nucleic Acids Res
2003
;
31
:
3808
-11.
27
Picot D, Loll PJ, Garavito RM. The X-ray crystal structure of the membrane protein prostaglandin H2 synthase-1.
Nature
1994
;
367
:
243
-9.
28
Sandler RS, Galanko JC, Murray SC, Helm JF, Woosley JT. Aspirin and nonsteroidal anti-inflammatory agents and risk for colorectal adenomas.
Gastroenterology
1998
;
114
:
441
-7.
29
Hamberg M, Svensson J, Samuelsson B. Prostaglandin endoperoxides. A new concept concerning the mode of action and release of prostaglandins.
Proc Natl Acad Sci USA
1974
;
71
:
3824
-8.
30
Hamberg M, Samuelsson B. Prostaglandin endoperoxides. Novel transformations of arachidonic acid in human platelets.
Proc Natl Acad Sci USA
1974
;
71
:
3400
-4.
31
Halushka MK, Walker LP, Halushka PV. Genetic variation in cyclooxygenase 1: effects on response to aspirin.
Clin Pharmacol Ther
2003
;
73
:
122
-30.
32
Li Y, Smith T, Grabski S, DeWitt DL. The membrane association sequences of the prostaglandin endoperoxide synthases-1 and -2 isozymes.
J Biol Chem
1998
;
273
:
29830
-7.
33
Nielsen H, Engelbrecht J, Brunak S, von Heijne G. Identification of prokaryotic and eukaryotic signal peptides and prediction of their cleavage sites.
Protein Eng
1997
;
10
:
1
-6.
34
Martoglio B, Dobberstein B. Signal sequences: more than just greasy peptides.
Trends Cell Biol
1998
;
8
:
410
-5.