CEBP Meeting Calendar Frontiers in Basic Cancer Research
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online
Cancer Epidemiology Biomarkers & Prevention 16, 1536, August 1, 2007. doi: 10.1158/1055-9965.EPI-07-0117
© 2007 American Association for Cancer Research
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Daugherty, S. E.
Right arrow Articles by Hayes, R. B.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Daugherty, S. E.
Right arrow Articles by Hayes, R. B.
Related Collections
Right arrow Epidemiology and Prevention
Right arrow Epidemiology and Prevention: Genetics and Biomarkers

Review

Variants in the {alpha}-Methylacyl-CoA Racemase Gene and the Association with Advanced Distal Colorectal Adenoma

Sarah E. Daugherty1,3, Elizabeth A. Platz3,5,6, Yin Yao Shugart3,4, M. Daniele Fallin3,4, William B. Isaacs5,6, Nilanjin Chatterjee1, Robert Welch2, Wen-Yi Huang1 and Richard B. Hayes1

1 Division of Cancer Epidemiology and Genetics, Department of Health and Human Services, and 2 Core Genotyping Facility, Division of Cancer Epidemiology and Genetics, Advanced Technology Program, SAIC Frederick Inc., NCI-Frederick Frederick, Maryland; Departments of 3 Epidemiology and 4 Biostatistics, Johns Hopkins Bloomberg School of Public Health; 5 The James Buchanan Brady Urological Institute and the Department of Urology, Johns Hopkins School of Medicine; and 6 The Sidney Kimmel Comprehensive Cancer Center, Johns Hopkins Medical Institutions, Baltimore, Maryland

Requests for reprints: Sarah E. Daugherty, Divison of Cancer Epidemiology and Genetics, National Cancer Institute, 6120 Executive Boulevard, EPS Rm 8113, Bethesda, MD 20892. Phone: 301-451-8789; Fax: 301-402-1819. E-mail: daughers{at}mail.nih.gov


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Background: {alpha}-Methylacyl-CoA racemase (AMACR), an enzyme involved in oxidation of branched chain fatty acids and cholesterol metabolites, as well as ibuprofen metabolism, is overexpressed in colorectal adenomas and cancer. AMACR gene variants have been associated with hereditary prostate cancer, but no studies have evaluated their etiologic role in colorectal carcinogenesis.

Methods: We conducted a case-control study of 725 advanced distal colorectal adenoma cases and 729 frequency-matched controls from the screening arm of the Prostate, Lung, Colorectal, and Ovarian Cancer Screening Trial. Seven AMACR polymorphisms were genotyped. Unconditional logistic regression models were used to evaluate the associations adjusting for age at randomization and gender.

Results: The 201L allele of S201L [TT versus CC: odds ratio (OR), 1.74; 95% confidence interval (95% CI), 1.15-2.62; TC versus CC: OR, 1.17; 95% CI, 0.93-1.49] and the 277E allele of K277E (GG versus AA: OR, 1.66; 95% CI, 1.03-2.68; GA versus AA: OR, 1.21; 95% CI, 0.96-1.53) were associated with increased risk of advanced distal colorectal adenoma (both Ptrend ≤ 0.02); the TGTGCG haplotype of six informative single nucleotide polymorphisms was also associated with increased risk (OR, 1.27; 95% CI, 1.03-1.55). Regular ibuprofen users who were homozygous for the variant allele at either M9V or D175G were at reduced risk for adenoma (both Pinteraction < 0.05).

Conclusion: Our study identified variants in AMACR associated with advanced distal colorectal adenoma and pointed to potential interactions with ibuprofen use. (Cancer Epidemiol Biomarkers Prev 2007;16(8):1536–42)


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
{alpha}-Methylacyl-CoA racemase (AMACR) plays a critical role in the oxidation of pristanic acid, a branched chain fatty acid, the metabolism of ibuprofen, and the processing of bile acid intermediates from cholesterol (1, 2). AMACR mRNA and protein levels are overexpressed in a variety of cancers, including colorectal and prostate cancer, as well as their precursor lesions (3). Although much of the recent literature has focused on the use of AMACR expression as a sensitive and specific biomarker for prostate cancer, less attention has been directed at the association between AMACR and colorectal cancer.

Three studies identified elevated AMACR mRNA levels (10-fold elevation compared with normal tissue) and overexpression of AMACR protein in adenomas and colorectal cancer (3-5). In contrast, minimal expression has been reported for hyperplastic polyps (4), which are not thought to be precursors of colorectal cancer. Although AMACR protein overexpression has been associated with the early stages of colorectal cancer (4), reduced expression has been reported for poorly differentiated colorectal adenocarcinomas (3-5).

Despite its potential role in colorectal carcinogenesis, no epidemiologic studies have evaluated the AMACR gene variants or the possible interaction between AMACR and its substrates in relation to risk for colorectal neoplasia. Although less well studied than aspirin, several cohort studies have found an inverse association (40-50% reduction) with long-term use (>1 year) of nonaspirin nonsteroidal anti-inflammatory drugs (NSAID) and colorectal adenomas (6) and cancer (7, 8). Ibuprofen is a 1:1 racemic mixture of active and inactive cyclooxygenase (COX)-inhibiting enantiomers (9). In vitro studies suggest that AMACR has the ability to racemize ibuprofen (2). Thus, the metabolic functioning of AMACR may contribute to the amount of ibuprofen available for COX inhibition (10). Branched chain fatty acids, found primarily in red meats and dairy, have been identified as important ligands for nuclear transcription factors retinoid X receptor (RXR) and peroxisome proliferator–activated receptor {alpha}. Although no research has been done evaluating the direct association between branched chain fatty acids and colorectal cancer, AMACR protein expression has been shown to be enhanced by branched chain fatty acids in some prostate cancer cell lines (11).

We conducted a large case-control study of advanced distal colorectal adenoma, nested within the screening arm of the Prostate, Lung, Colorectal, and Ovarian (PLCO) Cancer Screening Trial, testing for the association between seven AMACR genetic variants and risk for this disease. We also investigated the potential modifying role of two AMACR substrates, ibuprofen and branched chain fatty acids (the latter measured by reported intake of red meat and dairy foods, their primary sources). This study furthers our understanding of the spectrum of cancers in which AMACR plays a role, as well as identifies potentially modifiable environmental factors that alter the overall risk for colorectal adenoma among individuals with a particular genetic susceptibility.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Study Population
The PLCO Cancer Screening Trial is a multicenter randomized trial that is evaluating the effectiveness of prostate, lung, colorectal, and ovarian cancer screening modalities on disease-specific mortality. The details of this trial and its epidemiologic component have been described elsewhere (12, 13). The trial enrolled 154,952 subjects (49.5% men and 50.5% women), ages 55 to 74 years, at 10 screening centers (Washington, DC; Detroit, MI; Salt Lake City, UT; Denver, CO; Honolulu, HI; Minneapolis, MN; Marshfield, WI; Pittsburgh, PA; St. Louis, MO; and Birmingham, AL) between September 1993 and July 2001. All centers offered an initial 60-cm flexible sigmoidoscopy examination at study entry to those individuals who had been randomized to the screening arm of the trial. If abnormalities were detected during the screening, individuals were referred to their personal physicians for diagnostic follow-up. Information documented during the follow-up examination was abstracted from medical records, including data on polyp location, size, multiplicity, and histology.

Cases and Matched Controls
Cases and controls for this study were selected from 42,037 individuals who had a successful sigmoidoscopy examination (insertion to at least 50 cm with >90% of mucosa visible or a suspect lesion identified) between September 1993 and September 1999, completed the baseline risk factor questionnaire, and provided a blood sample. We excluded 4,834 individuals who reported a history of ulcerative colitis, Crohn's disease, familial polyposis, any previous colorectal polyps, Gardner's syndrome, or any cancer (except basal cell and squamous cell skin cancer).

We identified 1,234 individuals diagnosed at the baseline screening examination with at least one advanced colorectal adenoma (adenoma ≥1 cm or containing high-grade dysplasia or villous, including tubulovillous, elements) in the distal colon (descending colon, sigmoid, or rectum) and randomly selected, due to cost constraints, 772 of the cases for this study. For comparison, 777 control subjects who screened negative (i.e., no polyp or other suspected lesion) on the baseline sigmoidoscopy were frequency matched by gender (male, female) and ethnicity (non-Hispanic white, non-Hispanic black, Hispanic, and other) to the cases. Because of allele frequency differences and the small number of cases from ethnic groups other than whites (blacks, n = 45; asian, n = 26; hispanic, n = 14; and other, n = 10), we only included whites in this analysis (cases, n = 725; controls, n = 729).

Genotyping
DNA was extracted from whole blood or buffy coat using the Qiagen Column Kit (Valencia, CA). All genotyping for this study was done at the Core Genotyping Facility (CGF), National Cancer Institute. We evaluated colorectal adenoma risk in relation to five nonsynonymous single nucleotide polymorphisms (SNP) previously identified (14) and two intronic SNPs (Table 1 ). Although the SNPs used in this study were selected before the completion of the HapMap phase 1 project, we obtained reasonable coverage of the 20-kb gene (located on 5p13.2-q11.1) based on a review of the phase II HapMap data for Caucasians. Two of our seven SNPs (S201L and K277E) were identified as tagSNPs according to the tagger algorithm used in Haploview (15), with Gabriel's block definition (16). The remaining SNPs selected for this study show moderate to high correlation (r2 = 0.67-1.0) with the additional tagSNPs in each block identified in Haploview.


View this table:
[in this window]
[in a new window]

  		Table 1. Seven AMACR SNPs evaluated in the advanced distal colorectal adenoma study, PLCO

 
As part of the CGF assay validation effort, TaqMan and MGBeclipse assay results are compared with sequencing results in a multiethnic population of 102 individuals7 (17). The platform with the highest concordance rate was selected for genotyping the individual SNPs. The M9V and IVS1 + 169G>T polymorphisms were genotyped using a TaqMan platform, whereas the other AMACR polymorphisms were genotyped using MGBeclipse (Applied Biosystems). The nomenclature used to describe the intronic SNPs follows recommendations by den Dunnen et al. (18).

Coriell DNA samples containing the homozygous wild-type, heterozygous, and homozygous variant were used for internal laboratory quality control (17). Four samples of each genotype and four no-template samples were included in each plate. Additionally, external blinded DNA quality control samples from 40 different individuals were repeatedly placed among the study samples. Laboratory personnel were blinded to the identity of the external quality control samples and to case-control status. Interassay concordance was >97.5% for all genotypes. Of the total sample, 8.3% of DNA was insufficient for genotyping due to either low yield or poor quality. Among the samples that were genotyped for the seven AMACR SNPs, the failure rate varied between 0.9% and 5.7%. Nearly 1% of the samples were excluded due to inconsistencies in DNA fingerprint profiles.

Exposure Information
At baseline, all participants completed a risk factor questionnaire that included questions about age, ethnicity, education, occupation, current and past smoking behavior, history of cancer and other diseases, use of selected drugs, including aspirin and ibuprofen-containing products, and recent history of screening exams. Regular use of ibuprofen or aspirin was defined as self-reported intake of ≥4 pills per month over the 12 months before baseline. Nonregular users were individuals who did not report regular use (<4 per month) of the medications over the 12 months before baseline. For exploratory purposes, regular ibuprofen use was further defined by narrower categories (moderate use: 1-4 per week; and heavy use: more than 1 per day). Because the use of ibuprofen and aspirin was correlated, we created a specific variable for aspirin use only by excluding those individuals who reported using both ibuprofen and aspirin regularly. Due to the smaller percentage of ibuprofen users in this study, we were not able to create a similar variable for ibuprofen. A positive family history was defined as having at least one first-degree relative diagnosed with colorectal cancer.

We used a 137-item food-frequency questionnaire (FFQ), adapted from the Willett and Block food frequency questionnaires,8 to estimate the usual consumption of food over the 12-month period before enrollment, including the frequency of consumption and portion size for red meat and dairy foods, the primary sources of branched chain fatty acids. Nutrient and food group values (grams per day) were derived from the reported daily frequency of food items multiplied by the gender-specific portion size, determined by the U.S. Department of Agriculture's 1994 to 1996 Continuing Survey of Food Intake among Individuals (CSFII) database (19). Red meat intake was defined as the sum in grams per day of reported roast beef, meat in stew, hamburger, meat loaf, and steak consumption (pork was excluded because this meat does not contain notable amounts of branched chain fatty acids). Dairy intake was defined as the sum in grams per day of whole milk, cheese, ice cream, butter, cottage cheese, sour cream, and sweet cream consumption. Cut points for quartiles of red meat (grams per day) and dairy (grams per day) consumption were based on the distribution among the controls. Individuals who were missing more than seven items on their food frequency questionnaire or were in the top or bottom 1% for energy intake were excluded from the analysis stratified by branched chain fatty acid consumption (n = 67).

Statistical Analysis
The Pearson's {chi}2 test and the Fisher's exact test were used to compare differences in demographic and clinical characteristics among cases and controls. Hardy-Weinberg proportions were tested among cases and controls separately, using the Pearson {chi}2 test. Pairwise linkage disequilibrium was evaluated by two linkage disequilibrium parameters, Lewontin's D' (20) and r2 (21), which were estimated for controls using Haploview (22). Six of the seven SNPs were used to construct an AMACR haplotype (IVS1 + 169G>T, D175G, S201L, Q239H, IVS4 + 3803C>G, K277E,); M9V was dropped from the haplotype analysis because it was redundant (r2 = 0.99) with D175G and had slightly more missing genotype data (10.3% versus 9.9%).

Single SNP analyses were conducted using unconditional logistic regression, adjusting for age at randomization (55-59, 60-64, 65-69, and 70-74 years) and gender. Allele frequencies and parameter estimates did not differ significantly by study center; therefore, study center was not included in the final models. In addition to evaluating the overall association with advanced adenomas, we investigated the association between AMACR genotypes and adenoma size (≥1 cm, <1 cm), multiplicity (multiple, single), and histology (villous component, no villous).

We also assessed the association between AMACR genetic variants and colorectal adenoma within strata of regular ibuprofen use (yes/no), regular aspirin use (yes/no), median consumption of red meat or dairy (above, below), gender (male, female), and family history (yes/no). In addition to stratum-specific estimates, we evaluated the joint effects of ibuprofen and the AMACR genotypes. Individuals who did not report the regular use of ibuprofen and were wild-type homozygotes for the respective polymorphisms served as the comparison group for the analysis of joint effects. We tested heterogeneity between strata by the likelihood ratio test, comparing the likelihood of the models containing the stratum variable and the genotype with the likelihood of the models that included the stratum variable, the genotype, and a cross-product term.

The expectation maximization (EM) algorithm in Haplo Stats for the R programming language9 was used to estimate the haplotype frequencies. A global score test (n = 10,000 permutations), adjusting for age, was used to test the null hypothesis of no difference in haplotype frequencies between cases and controls (23). To reduce the degrees of freedom, all haplotypes with a frequency of <1% were dropped from the score test.

In Haplo Stats, we used generalized linear models for a binomial trait to evaluate individual haplotype associations with colorectal adenoma under the null hypothesis of no haplotype effect (23). Posterior probabilities of subject-specific haplotype pairs, generated from the EM algorithm, were used as weights in an iterative process in which the weights updated the regression coefficients, whereas the regression coefficients were simultaneously used to update the posterior probabilities (24). All individual haplotypes were simultaneously entered into an additive model, adjusting for gender and age. Individuals were dropped from the haplotype analysis if they were missing information on at least five of the seven genotypes (n = 141).


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Cases tended to be older, were more likely to have at least one first-degree relative diagnosed with colorectal cancer, have less formal education, report being a current smoker, and are less likely to report regular use of ibuprofen compared with controls (Table 2 ). After adjusting for age, the statistically significant difference for ibuprofen use among cases and controls disappeared. Of the 725 advanced distal colorectal adenoma cases, 411 (57%) had adenomas with villous components, 536 (74%) had adenomas at least 1 cm in size, and 228 (31.5%) had multiple adenomas.


View this table:
[in this window]
[in a new window]

  		Table 2. Demographic characteristics of advanced distal colorectal adenoma cases and controls, PLCO

 
All genotypes followed Hardy Weinberg proportions among the cases and controls, except for the polymorphism S201L among cases (P = 0.05). The pairwise linkage disequilibrium for the seven SNPs was similar to the LD available from Hapmap (25). The pairwise D' ranged between 0.93 and 0.99 for all markers, except for K277E (pairwise D' = 0.49-0.84), whereas the r2 was modest between most pairs (ranging from 0.01 to 0.76), except for M9V and D175G, which were almost completely correlated (r2 = 0.99; see Table 3 ).


View this table:
[in this window]
[in a new window]

  		Table 3. Pairwise linkage disequilibrium among white controls assessed by D' and r2 values

 
The 201L homozygous genotype for S201L and the 277E homozygous genotype for K277E were both associated with a statistically significant increase in risk of advanced distal colorectal adenoma (see Table 4 ). For these two polymorphisms, the risk increased with increasing number of variant alleles (both Ptrend ≤ 0.02). Two suggestive associations were identified among individuals who carried the homozygote variant genotype at M9V [odds ratio (OR), 1.31; 95% confidence interval (95% CI), 0.95-1.79] and D175G (OR, 1.32; 95% CI, 0.96-1.81). Similar associations were observed after mutual adjustment for SNPs that had low to moderate pairwise correlation (r2 = 0.0-0.5). No notable patterns were observed in the associations between the AMACR genotypes and adenoma by number, size, or histology (data not shown). The global score test did not reveal overall differences between cases and controls in the frequency profile of the six-SNP haplotype (score = 14.16, P = 0.29); however, the TGTGCG haplotype, which incorporates all the risk alleles, was associated with a statistically significant elevation in risk (OR, 1.27; 95% CI, 1.03-1.55), compared with the most common haplotype, TACGCA (see Table 5 ). A similar positive association and a suggestive global score test (score, 6.74, P = 0.08) was observed with the haplotype that included the two risk alleles from the tagSNPs only (K277E and S201L).


View this table:
[in this window]
[in a new window]

  		Table 4. Association between AMACR genotypes and advanced distal colorectal adenoma, PLCO

 

View this table:
[in this window]
[in a new window]

  		Table 5. Haplotype global score test, individual haplotype frequencies, and ORs of advanced distal colorectal adenoma, PLCO

 
When examining joint effects, we found statistically significant increases in adenoma risk related to the homozygous variant genotype at D175G (and M9V), S201L, and K277E among nonregular users of ibuprofen (Ptrend ≤ 0.05; Table 6 ). In contrast, compared with the reference group who were wild-type homozygous and nonregular ibuprofen users, regular ibuprofen users who were homozygous for the variant allele at D175G (and M9V) had a reduced risk of adenoma (Pinteraction ≤ 0.05). Considering subjects homozygous for the at-risk variants at these loci, ibuprofen use was associated with substantial risk reductions for advanced distal colorectal adenoma (for M9V, ibuprofen yes/no: OR, 0.35; 95% CI, 0.18-0.70; and for D175G, ibuprofen yes/no: OR, 0.32; 95% CI, 0.16-0.65). No association was observed between ibuprofen use and colorectal adenoma among individuals who were wild type or heterozygous for these SNPs. Similar protective associations were seen for the more narrowly defined ibuprofen categories, with slightly stronger protective effects observed among daily users who were homozygous or heterozygous for the variant allele at D175G and M9V (data not shown). These subgroup associations are in contrast to the overall association between ibuprofen and colorectal adenomas in our study (ibuprofen yes/no: OR, 0.82; 95% CI, 0.62-1.08). The global score tests were not significant within the strata of ibuprofen (regular ibuprofen users: score = 11.90, P = 0.16; and nonregular users: score = 9.57, P = 0.21), however, the TGTGCG haplotype was associated with risk in nonregular users (OR, 1.33; 95% CI, 1.06-1.66), but not in regular users (OR, 0.98; 95% CI, 0.59-1.62). No significant heterogeneity was noticed among the AMACR genetic variants by strata of aspirin use, family history, or gender (all Pinteraction > 0.22).


View this table:
[in this window]
[in a new window]

  		Table 6. Joint effects of regular ibuprofen use and AMACR genotype among advanced distal colorectal adenoma cases and controls, PLCO

 
With respect to red meat and dairy products, no statistically significant interactions were observed with AMACR genotypes, although individuals who were homozygous for the rarer alleles at S201L and K277E and reported consuming less than the median value of dairy products had an elevated risk of advanced distal colorectal adenoma (K277E: OR, 2.14; 95% CI, 1.15-3.98; and S201L: OR, 1.91; 95% CI, 1.06-3.45). No significant associations with these polymorphisms were observed among individuals who were homozygous for the variant alleles and reported consuming more than the median value of dairy products (S201L: OR, 1.40; 95% CI, 0.78-2.50; and K277E: OR, 0.93; 95% CI, 0.42-2.04; both Pinteraction > 0.20).


   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
This first epidemiologic investigation of AMACR gene variants and colorectal neoplasia indicates that inherent variation in this gene may be related to colorectal adenoma risk. Although a specific mechanism remains to be determined, AMACR-related oxidative stress created by hydrogen peroxide, a by-product of the peroxisomal B-oxidation pathway, may be a contributing factor (10).

We found that several polymorphic markers in AMACR were associated with increased risk of advanced distal colorectal adenoma, with significant associations noted for the 201L allele at S201L and 277E at K277E and a tendency for increased risk with the highly correlated 9V at M9V and 175G at D175G. The haplotype analysis was consistent with the genotype associations, but did not provide further information as a composite allele. These polymorphisms (M9V, D175G, S201L, K277E) have previously been related to prostate cancer in hereditary prostate cancer families (14); the direction of association was the same for two loci (S201L and K277E), but in the opposite direction for the other two (M9V and D175G; 14). The biological significance of these cross-disease associations is unclear.

The interaction between AMACR variants and ibuprofen use suggests that selected AMACR gene variants may modify the chemopreventive nature of ibuprofen (10, 26). Ibuprofen is manufactured as a racemic mixture of R and S enantiomers, but only the S enantiomer inhibits COX enzymes (27). Experimental data suggest that AMACR is a key enzyme in converting the R form of ibuprofen to the active COX-inhibiting S form (26). COX-2 is overexpressed in colorectal adenomas and adenocarcinoma (28) and has been associated with the development of angiogenesis as well as a resistance to apoptotic mechanisms (29). The protective association observed in our study among individuals with specific AMACR variants may be due to a greater amount of the COX-inactive ibuprofen being converted into the active COX-inhibiting form, reducing the potential for the COX enzymes to contribute to tumorigenesis. Thus, the same inherent susceptibility that may increase risk for colorectal adenomas through oxidative stress may modify ibuprofen in such a way as to ameliorate the risk conferred by these genetic variants.

AMACR is not, however, involved in aspirin metabolism. In contrast to the interrelationships observed in this study with ibuprofen, aspirin's effects on adenoma risk were not differential with respect to the AMACR genotypes, supporting the specificity of the AMACR-ibuprofen association.

Although we identified a few statistically significant associations within strata of branched chain fatty acids, we found no evidence for interaction. Moreover, these stratified associations were not consistent across food sources (red meat and dairy) and did not reveal a dose-response relationship when the data were divided into quartiles. The determination of branched chain fatty acid in foods, from food frequency questionnaires, is crude, with modest correlations between reported consumption of red meat or dairy and serum measures of branched chain fatty acids, r2 = 0.16 for meat and r2 = 0.24 for dairy (30). Thus, the interrelationships we observed between AMACR SNPs and dietary branched chain fatty acids contain uncertainty.

Although the SNPs included here were selected before the completion of HapMap, these SNPs capture much of the inherent variability in the gene. Strong similarities were observed between the haplotype frequency patterns (>1%) derived from HapMap and haplotypes generated in our study. Thus, to the extent that HapMap provides a thorough characterization of the variation in this gene, our approach would be comparable.

Little is known about the functional relevance of these markers; however, D175G is conserved in rodents (14); and the nonsynonymous amino acid change in M9V is predicted to have a deleterious impact on the protein functioning, according to the SIFT algorithm (31). Interestingly, these two polymorphisms (which are highly correlated) provided the strongest evidence for the interaction with ibuprofen.

Our study included a large sample, comprehensive information on potential modifiers of the genetic association, and the ability to minimize selection and surveillance bias, due to cases and controls both being selected from the screening arm of the PLCO trial. The associations we observed in this study are specific to advanced adenomas that arise in the distal colon and, therefore, may not be generalizable to the proximal colon. To the extent that the associations would be similar with advanced adenomas across the colon, the results in this study would be attenuated due to disease misclassification among some controls having undetected proximal polyps, which were not identified through sigmoidoscopy. Although the sequence of genetic events in the adenoma-adenocarcinoma transition is well established, our results may not be generalizable to colorectal adenocarcinoma because some adenomas do not develop into adenocarcinoma. Our study was limited to non-Hispanic white men and women. Other racial groups were represented in our sample by only a small number of individuals that precluded meaningful analysis. The etiologic relevance of the AMACR causal alleles should be similar across racial groups; however, linkage disequilibrium patterns may vary, and therefore, the markers evaluated in this study may not be equally powerful across all races.

Our assessment of NSAIDs was broad, including specific use of ibuprofen-containing drugs (frequency and duration of use in the year before study), which was important for investigating our a priori hypothesis about AMACR and ibuprofen interrelationships in colon adenoma risk. In addition to confirmatory epidemiology with expanded attention to ibuprofen dose and etiologically relevant time points of use, functional studies are needed to identify the specific effects of AMACR polymorphisms on protein activity and expression. Also, the etiologic scope of AMACR and its role in carcinogenesis needs to be investigated for other cancer sites in which AMACR overexpression has been reported (3, 5, 32).

Our study provides evidence that variants in the AMACR gene may be associated with an increased risk of advanced distal colorectal adenoma, the major precursor of colorectal cancer. This study provides preliminary evidence that ibuprofen use may best be tailored to individuals who are at higher risk due to their genetic susceptibility for colorectal carcinogenesis.


   Acknowledgments
 
The content of this publication does not necessarily reflect the views or policies of the Department of Health ad Human Services, nor does mention of trade names, commercial products, or organizations imply endorsement by the U.S. Government. The authors thank Drs. Christine Berg and Philip Prorok, Division of Cancer Prevention, National Cancer Institute; the Screening Center investigators and staff or the Prostate, Lung, Colorectal, and Ovarian (PLCO) Cancer Screening Trial; Tom Riley and staff, Information Management Services, Inc.; Barbara O'Brien and staff, Westat, Inc.; and Drs. Bill Kopp and Wen Shao, and staff, Science Applications International Corporation-Frederick for their contributions to making this study possible.


   Footnotes
 
Grant support: Dr. Daugherty was supported by the Institutional National Research Service Award (T32 CA09314) and the Cancer Research Training Award as part of the Intramural Research Program of the National Cancer Institute, NIH. Funded in whole or inpart with federal funds from the National Cancer Institute, NIH, under contract NO1-CO-12400.

7 http://SNP500cancer.nci.nih.gov Back

8 http://www3.cancer.giv/prevention/plco/DQX.pdf Back

9 http://www.mayo.edu/hsr/people/schaid.html Back

Received 2/14/07; revised 4/17/07; accepted 5/21/07.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Ferdinandusse S, Denis S, Clayton PT, et al. Mutations in the gene encoding peroxisomal {alpha}-methylacyl-CoA racemase cause adult-onset sensory motor neuropathy. Nat Genet 2000;24:188–91.[CrossRef][Medline]
  2. Knihinicki RD, Day RO, Williams KM. Chiral inversion of 2-arylpropionic acid non-steroidal anti-inflammatory drugs—II. Racemization and hydrolysis of (R)- and (S)-ibuprofen-CoA thioesters. Biochem Pharmacol 1991;42:1905–11.[CrossRef][Medline]
  3. Zhou M, Chinnaiyan AM, Kleer CG, Lucas PC, Rubin MA. {alpha}-Methylacyl-CoA racemase: a novel tumor marker overexpressed in several human cancers and their precursor lesions. Am J Surg Pathol 2002;26:926–31.[CrossRef][Medline]
  4. Jiang Z, Fanger GR, Banner BF, et al. A dietary enzyme: {alpha}-methylacyl-CoA racemase/P504S is overexpressed in colon carcinoma. Cancer Detect Prev 2003;27:422–6.[CrossRef][Medline]
  5. Chen ZM, Ritter JH, Wang HL. Differential expression of {alpha}-methylacyl coenzyme A racemase in adenocarcinomas of the small and large intestines. Am J Surg Pathol 2005;29:890–6.[Medline]
  6. Garcia Rodriguez LA, Huerta-Alvarez C. Reduced incidence of colorectal adenoma among long-term users of nonsteroidal antiinflammatory drugs: a pooled analysis of published studies and a new population-based study. Epidemiology 2000;11:376–81.[CrossRef][Medline]
  7. Garcia-Rodriguez LA, Huerta-Alvarez C. Reduced risk of colorectal cancer among long-term users of aspirin and nonaspirin nonsteroidal antiinflammatory drugs. Epidemiology 2001;12:88–93.[CrossRef][Medline]
  8. Smalley W, Ray WA, Daugherty J, Griffin MR. Use of nonsteroidal anti-inflammatory drugs and incidence of colorectal cancer: a population-based study. Arch Intern Med 1999;159:161–6.[Abstract/Free Full Text]
  9. Tan SC, Patel BK, Jackson SH, Swift CG, Hutt AJ. Ibuprofen stereochemistry: double-the-trouble? Enantiomer 1999;4:195–203.[Medline]
  10. Zha S, Ferdinandusse S, Denis S, et al. {alpha}-methylacyl-CoA racemase as an androgen-independent growth modifier in prostate cancer. Cancer Res 2003;63:7365–76.[Abstract/Free Full Text]
  11. Mobley JA, Leav I, Zielie P, et al. Branched fatty acids in dairy and beef products markedly enhance {alpha}-methylacyl-CoA racemase expression in prostate cancer cells in vitro. Cancer Epidemiol Biomarkers Prev 2003;12:775–83.[Abstract/Free Full Text]
  12. Prorok PC, Andriole GL, Bresalier RS, et al. Design of the Prostate, Lung, Colorectal and Ovarian (PLCO) Cancer Screening Trial. Control Clin Trials 2000;21:273–309S.
  13. Hayes RB, Sigurdson A, Moore L, et al. Methods for etiologic and early marker investigations in the PLCO trial. Mutat Res 2005;592:147–54.[Medline]
  14. Zheng SL, Chang BL, Faith DA, et al. Sequence variants of {alpha}-methylacyl-CoA racemase are associated with prostate cancer risk. Cancer Res 2002;62:6485–8.[Abstract/Free Full Text]
  15. De Bakker PI, Yelensky R, Pe'er I, Gabriel SB, Daly MJ, Altshuler D. Efficiency and power in genetic association studies. Nat Genet 2005;37:1217–23.[CrossRef][Medline]
  16. Gabriel SB, Schaffner SF, Nguyen H, et al. The structure of haplotype blocks in the human genome. Science 2002;296:2225–9.[Abstract/Free Full Text]
  17. Packer BR, Yeager M, Staats B, et al. SNP500Cancer: a public resource for sequence validation and assay development for genetic variation in candidate genes. Nucleic Acids Res 2004;32 Database issue:D528–32.[Abstract/Free Full Text]
  18. den Dunnen JT, Antonarakis SE. Nomenclature for the description of human sequence variations. Hum Genet 2001;109:121–4.[CrossRef][Medline]
  19. Tippett K, Cypel Y. Design and operation: the continuing survey of food intakes by individuals and the diet and health knowledge survey, 1994–1996. In: Continuing survey of food intakes by individuals 1994–1996. Nationwide Food Surveys Rep No. 96–1. 1998. US Department of Agriculture, Agricultural Research Service. Ref Type: Report.
  20. Lewontin RC, Feldman MW. A general asymptotic property of two-locus selection models. Theor Popul Biol 1988;34:177–93.[CrossRef][Medline]
  21. Devlin B, Risch N. A comparison of linkage disequilibrium measures for fine-scale mapping. Genomics 1995;29:311–22.[CrossRef][Medline]
  22. Barrett JC, Fry B, Maller J, Daly MJ. Haploview: analysis and visualization of LD and haplotype maps. Bioinformatics 2005;21:263–5.[Abstract/Free Full Text]
  23. Schaid DJ, Rowland CM, Tines DE, Jacobson RM, Poland GA. Score tests for association between traits and haplotypes when linkage phase is ambiguous. Am J Hum Genet 2002;70:425–34.[CrossRef][Medline]
  24. Lake SL, Lyon H, Tantisira K, et al. Estimation and tests of haplotype-environment interaction when linkage phase is ambiguous. Hum Hered 2003;55:56–65.[CrossRef][Medline]
  25. The International HapMap Project. Nature 2003;426:789–96.[CrossRef][Medline]
  26. Reichel C, Brugger R, Bang H, Geisslinger G, Brune K. Molecular cloning and expression of a 2-arylpropionyl-coenzyme A epimerase: a key enzyme in the inversion metabolism of ibuprofen. Mol Pharmacol 1997;51:576–82.[Abstract/Free Full Text]
  27. Neupert W, Brugger R, Euchenhofer C, Brune K, Geisslinger G. Effects of ibuprofen enantiomers and its coenzyme A thioesters on human prostaglandin endoperoxide synthases. Br J Pharmacol 1997;122:487–92.[Medline]
  28. Eberhart CE, Coffey RJ, Radhika A, Giardiello FM, Ferrenbach S, DuBois RN. Up-regulation of cyclooxygenase 2 gene expression in human colorectal adenomas and adenocarcinomas. Gastroenterology 1994;107:1183–8.[Medline]
  29. Zha S, Yegnasubramanian V, Nelson WG, Isaacs WB, De Marzo AM. Cyclooxygenases in cancer: progress and perspective. Cancer Lett 2004;215:1–20.[CrossRef][Medline]
  30. Xu J, Thornburg T, Turner AR, et al. Serum levels of phytanic acid are associated with prostate cancer risk. Prostate 2005;63:209–14.[CrossRef][Medline]
  31. Ng PC, Henikoff S. SIFT: Predicting amino acid changes that affect protein function. Nucleic Acids Res 2003;31:3812–4.[Abstract/Free Full Text]
  32. Witkiewicz AK, Varambally S, Shen R, et al. {alpha}-Methylacyl-CoA racemase protein expression is associated with the degree of differentiation in breast cancer using quantitative image analysis. Cancer Epidemiol Biomarkers Prev 2005;14:1418–23.[Abstract/Free Full Text]




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Daugherty, S. E.
Right arrow Articles by Hayes, R. B.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Daugherty, S. E.
Right arrow Articles by Hayes, R. B.
Related Collections
Right arrow Epidemiology and Prevention
Right arrow Epidemiology and Prevention: Genetics and Biomarkers

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online