Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Mutat Res. Author manuscript; available in PMC 2010 July 21.
Published in final edited form as:
PMCID: PMC2907671

Colon tumor mutations and epigenetic changes associated with genetic polymorphism: Insight into disease pathways


Variation in genes associated with serum levels of proteins may be useful for examining specific disease pathways. Using data from a large study of colon cancer, we examine genetic variants in insulin, inflammation, estrogen, metabolizing enzymes, and energy homeostasis genes to explore associations with microsatellite instability (MSI), CpG Island methylator phenotype (CIMP), mutations of p53 in exons 5 through 8, and mutations in codons 12 and 13 of Ki-ras. Insulin-related genes were associated with CIMP positive and MSI tumors, with the strongest associations among aspirin users. The Fok1 Vitamin D Receptor (VDR) polymorphism was associated with CIMP positive/Ki-ras mutated tumors; the Poly A and CDX2 VDR polymorphisms were associated only with Ki-ras mutated tumors. NAT2 was associated with CIMP positive/Ki-ras mutated tumors but not with MSI tumors. The TCF7L2 rs7903146 polymorphism was associated with p53 mutated tumors. Most associations varied by recent aspirin/NSAID use: IL6 rs1800796 and rs1800795 polymorphisms were associated inversely with tumor mutations in the presence of aspirin/NSAIDs; POMC significantly reduced risk of Ki-ras- mutated tumors when aspirin/NSAIDs were not used; the TCF7L2 rs7903146 was associated with reduced risk of Ki-ras-mutated tumors in the presence of aspirin and increased risk in the absence of aspirin. These data, although exploratory, identify specific tumor subsets that may be associated with specific exposures/polymorphism combinations. The important modifying effects of aspirin/NSAIDs on associations with genetic polymorphisms reinforce the underlying role of inflammation in the etiology of colon cancer.

Keywords: Insulin, polymorphisms, colon cancer, aspirin, non-steroidal anti-inflammatory, CIMP, p53, MSI, Ki-ras

In 1988 it was proposed that colorectal cancers (CRC) followed a series of mutational events, with the majority of tumors initially having an APC mutation that was followed by mutations in the p53 and Ki-ras genes [1]. Existing evidence suggests that CRC pathways that once were believed to be overlapping, are often unique, and that many tumors do not have multiple mutations [2]. It is possible that insight into unique pathways resulting in epigenetic changes and mutations in colorectal tumors can be obtained when evaluated in conjunction with inherited susceptibility, diet, lifestyle, and environmental exposure data.

Although studies are limited, there is evidence that examination of epigenetic and genetic mutations in colon tumors and adenomas are associated with certain exposures that alter the likelihood of developing those mutations. For instance, cigarette smoking has been associated with greater risk of developing tumors that have microsatellite instability (MSI) [3] and CpG Island methylator phenotype (CIMP) [4] and specifically with hyperplastic rather than adenomatous polyps [5]. Dietary folate has been associated with specific Ki-ras mutations in adenomas and cancers [6, 7]. We also have reported that polymorphisms in certain insulin-related genes influence p53 mutations [8]. While several pathways, including insulin, estrogen, and inflammation have been proposed as being important in colon cancer etiology [912], we have not systematically examined how these pathways influence the broad spectrum of acquired genetic and epigenetic changes in tumors.

Genetic variation in genes associated with serum levels of various proteins may be a useful for examining specific pathways. For instance the −202 IGFBP3 genetic variant has been associated with serum levels of IGFBP-3 [13, 14], variants of the IGF1 gene has been associated with serum IGF-1 levels [15]; and variants of the IL6 gene has been associated with IL-6 levels [16]. Thus, assessment of genetic variants with tumor markers may further our understanding of how major pathways lead to specific types of mutations in tumors.

In this study, we examine genetic variants that are linked to the pathways hypothesized to contribute to colon cancer. Because little is known about factors that contribute to specific tumor mutations, we explore variation in genes in pathways including insulin, estrogen, carcinogen metabolism, inflammation, and energy homeostasis to describe exposure-specific pathways that can be linked to specific tumor markers. Although the analyses are exploratory in nature, we believe that the results add to our previous work by simultaneously examining a wide spectrum of epigenetic and genetic changes in tumors, including MSI, CIMP, p53, and Ki-ras, and will help us to gain insight into the carcinogenic process and generate hypotheses that can be tested in other studies.


The Institutional Review Board from all centers approved all aspects of the study. A case-control study of 1839 colon cancer cases and 2014 controls was conducted among black, white, or Hispanic people from either the Kaiser Permanente Medical Care Program (KPMCP) of Northern California, an eight county area in Utah (Davis, Salt Lake, Utah, Weber, Wasatch, Tooele, Morgan, and Summit counties), or the Twin Cities Metropolitan area in Minnesota. Eligibility criteria for cases included diagnosis with first-primary incident colon cancer (ICD-O 2nd edition codes 18.0, 18.2 to 18.9) between October 1, 1991 and September 30, 1994, between 30 and 79 years of age at time of diagnosis, and mentally competent to complete the interview. Cases with adenocarcinoma or carcinoma of the rectosigmoid junction or rectum (defined as the first 15 cm from the anal opening), with known familial adenomatous polyposis, ulcerative colitis, or Crohn’s disease were not eligible. Of all cases asked to participate, 75.6% cooperated. We obtained tumor blocks from 96% of cases identified in Utah, from 81% of cases at KPMCP, and from 38% of cases in Minnesota.

Controls, in addition to the eligibility criteria for cases, could not at the time of recruitment have had a previous colorectal tumor. Controls were selected from eligibility lists for KPMCP, driver’s license lists for Minnesota, and random-digit-dialing, driver’s license lists, or Health Care Finance Administration (HCFA) lists for Utah. These methods have been described in detail [17]. Of all controls selected, 63.7% participated.

All data were collected in-person by trained and certified interviewers [17] using a laptop-administered computerized questionnaire. The referent period for the study was the calendar year approximately two years prior to date of diagnosis for cases and a comparable time period from selection for controls. Information was collected on demographic factors such as age, sex, and center, diet, physical activity [17, 18], adult height and weight two and five years prior to diagnosis, regular use of aspirin and/or non-steroidal anti-inflammatory drugs, cigarette smoking history, and medical history.

Genetic Analysis

Tumor DNA was obtained from paraffin-embedded tissue as described previously. Tumors were previously characterized by their genetic profile that include sequence data for exons 5 through 8, or the mutation hotspots of the p53 gene; sequence data for Ki-ras codons 12 and 13; five CpG Island markers MINT1, MINT2, MINT31, p16, and hMLH1; and microsatellite instability (MSI). This study uses MSI markers Bat-26 (a mononucleotide repeats which is a very good measure of generalized instability), TGFβRII (a coding mononucleotide repeat which is unstable in most colorectal cancers with MSI) and a panel of 10 tetranucleotides repeats [1922] since it precedes the establishment of the Bethesda Panel for microsatellite instability [23]. The CIMP markers used were those used in early studies and have been able to discriminate associations with lifestyle exposures in this study population [4]. Tumor marker data were available for 1212 MSI stable and 212 MSI unstable, 920 CIMP negative and 334 CIMP positive, 732 p53 non-mutated and 642 p53 mutated, and 911 Ki-ras non-mutated and 433 Ki-ras mutated cases that also had germline data for polymorphism testing.

Germline DNA obtained from blood samples provided by study participants or from normal tissue in the paraffin blocks was used to examine variation in genes along important colon cancer disease pathways. The polymorphisms examined are shown in Table 2 and represent major pathways hypothesized to influence colon cancer risk: inflammation; energy homeostasis; hormones including both insulin and sex steroid hormones; and metabolizing genes. Several genes, such as apoE, PPARγ, the vitamin D receptor (VDR), and the adenomatous polyposis coli (APC) gene may influence multiple pathways. The polymorphisms assessed were initially selected as candidate polymorphisms hypothesized to be associated with colon cancer; the main effects of most of these polymorphisms have been reported previously [12, 2434]. All genes examined were in Hardy-Weinberg equilibrium in non-Hispanic white controls except for CYP2C9 I359L and IGF1R.

Table 2
Candidate polymorphisms examined with colon cancer tumors

Statistical Methods

Using the maximum efficiency robust test (MERT), we compared the trend tests for each mode of inheritance to determine the most appropriate model for each gene, as being either recessive, dominant, or additive [35]. Briefly, MERT is a linear combination of the test statistics for the recessive and dominant models that tests genotype trend in cases where the mode of inheritance in unknown. Using this technique, we determined if the best fitting model for each gene was recessive, dominant or additive. However, in some instances the homozygote variant was sufficiently rare that it was only possible to evaluate the dominant model.

The polymorphism associations were run using polytomous logistic regression models that included the outcomes of control, MSI, CIMP, p53, Ki-ras. Odds ratios (OR) and 95% Confidence Intervals (CI) are reported along with the p value for a linear trend. Although we report the exact p value, we refer to the 0.05 level as being statistically significant in the text and consider p values of <0.10 as borderline significant. Some cases were in multiple categories if they had a mutation in more than one of the tumor markers. To account for the fact that some cases were in multiple categories, we treated each subject as a cluster or primary sampling unit rather than each mutation in order to estimate the variance among subjects. In these models, adjustments were made for factors associated with colon cancer, including age, gender, body mass index (kg/m2), long-term vigorous physical activity, use of aspirin/NSAIDs, usual number of cigarettes smoked per day, and dietary calcium and energy. Additionally, we stratified data by use of aspirin and non-steroidal anti-inflammatory drugs within the past two years (labeled as NSAIDs) because previous analysis suggests that NSAIDs are major effect modifiers for many risk factors [36]. The data were analyzed using the procedure SURVEYLOGISTIC in SAS version 9.1 (Cary, NC). We examined all genes described in Table 2 for overall associations and for associations that varied by recent use of aspirin/NSAIDs. We present data for genes that appear to have unique associations with specific tumor markers as determined by either a significant linear trend across genotypes or because of significant association for the homozygote variant genotype.


Characteristics of the population are described in Table 1. The majority of the population was over 60 years of age. Approximately half of tumors were proximal and half were distal. Controls were more likely to report having recently used aspirin/NSAIDs within the past two years than cases. The most common tumor variant was a p53 mutation (46.7% of tumors) and the least common was MSI (14.9%).

Table 1
Description of Study Population

Genes and specific polymorphisms examined are shown in Table 2 along with their minor allele frequency (MAF) among non-Hispanic white controls. Because many genes can function in several pathways, we have noted the pathways that we believe are most relevant to the specific gene examined.

The polymorphism in POMC was associated with Ki-ras mutations and of borderline significance with p53 mutations (Table 3). LEPR was associated with colon cancer overall and had similar levels of association with p53 mutations although the confidence intervals ranged from 0.52 to 1.05. Genes associated with insulin and/or inflammation related pathways were associated with several tumor markers. The CG/GG genotypes of PPARγ were significantly associated with reduced likelihood of having an MSI or p53 tumor and of borderline significance with CIMP-positive tumors. VDR polymorphisms were associated with a variety of tumor markers, depending upon the polymorphism evaluated. For instance, the polyA and Bsm1 (data not shown in table for Bsm1) and CDX2 polymorphisms were associated with Ki-ras mutations. On the other hand, the VDR Fok1 polymorphism was directly associated with Ki-ras and CIMP-positive tumors and inversely associated with MSI tumors. NAT2 intermediate/rapid acetylators have the strongest associations with CIMP-positive and Ki-ras-mutated tumors and no association with MSI tumors. Statistically significant associations were noted for the CA repeat in the IGF1 gene with Ki-ras- mutated tumors; IGF1R, IRS1, and IRS2 were associated mainly with CIMP-positive and MSI tumors. TCF7L2 was directly associated with p5-mutated tumors.

Table 3
Associations between polymorphisms and specific tumor mutations.

Evaluation of tumor markers among those who recently used aspirin/NSAIDs (Table 4) and those who did not (Table 5) showed several associations that were modified by recent use. POMC continued to be significantly inversely associated with Ki-ras mutations among those not recently using aspirin/NSAIDs (Table 5). LEP rs7799039 was inversely associated with MSI tumors among non-aspirin/NSAID users (Table 5) and showed a strong direct association with MSI and CIMP-positive tumors among recent aspirin/NSAIDs users (Table 4). Both IL6 polymorphisms were associated with a statistically significant reduced risk of having a p53 mutation among recent aspirin/NSAID users (Table 4). The VDR polyA polymorphism was associated with reduced risk of developing a Ki-ras mutation only among non-aspirin/NSAID users (Table 5), while the VDR Fok1 polymorphism was associated with an increased likelihood of both Ki-ras and CIMP-positive tumor for users and non-users of aspirin/NSAIDs alike. Unlike the Fok1 polymorphism, the CDX2 polymorphism appeared to be associated strongly with Ki-ras mutations and have weaker associations with CIMP-positive tumors among those who reported recent use of aspirin/NSAIDs (Table 4). Insulin-related polymorphisms appeared to have stronger associations with CIMP-positive and MSI tumors among people who reported recent use of aspirin/NSAIDs (Table 4). IGF1R and IGFBP3 were strongly associated with increased risk of CIMP-positive and MSI tumors (IGF1R only) among those with the variant allele. Having an A allele of the IRS1 gene was associated with increased colon cancer risk regardless of aspirin/NSAID use. The CC/TT genotype of TCF7L2 was significantly inversely associated with Ki-ras-mutated tumors among recent aspirin/NSAID users (Table 4) and was directly associated with CIMP-positive, p53, and Ki-ras tumors (Table 5).

Table 4
Associations between polymorphisms and tumor markers among those who have recently used aspirin/NSAIDs
Table 5
Associations between polymorphisms and tumor markers among those not using aspirin/NSAIDs recently.


Results from this study provide some evidence that specific exposure/susceptibility combinations may lead to unique tumor mutations, thus potentially explaining the heterogeneity of colon cancer. Although the analyses presented here were exploratory in nature, they suggest that recent use of aspirin/NSAIDs modify associations with several pathways, most notably those involving insulin and inflammation-related genes. These findings, although descriptive, provide avenues for scientific inquiry into the cause of various mutations that occur in tumors and support the idea of multiple and sometimes interacting pathways that lead to colon cancer development. Many tumors with positive levels of CIMP have been shown to have MSI, whereas another set of CIMP-positive but MSS tumors frequently are accompanied by a Ki-ras mutation and are sometimes labeled as CIMP-2 [37]. In our data, approximately 27% of tumors were classified as CIMP-positive, while approximately 15% were MSI. We have reported previously that some risk factors for MSI, such as alcohol, are associated only with those tumors that are CIMP-negative and MSI, while others such as cigarette smoking are associated with both CIMP-positive and MSI tumors [3, 4, 38, 39]. Those data along with the current data suggest that there may be multiple pathways to CIMP-positive and MSI tumors which include both insulin and inflammation. The data presented here lend additional support for a CIMP-positive Ki-ras pathway that is independent of MSI. Both CIMP-positive and MSI tumors were associated with IGF1R, especially in the presence of aspirin/NSAIDs and IGFBP3 (202 C>A variant) was associated with increased risk of CIMP-positive tumors among users of aspirin/NSAIDs, while inversely associated with MSI among non-aspirin/NSAID users. These associations suggest that CIMP appears to be influenced by insulin-related factors that may be, at least in part, regulated by the underlying inflammatory state. Although our study was exploratory in nature, other studies have shown that IGF1R influences IGF-1 levels, and that it regulates COX-2 expression via the MAKP/Erk pathway [40, 41], thus providing some support for our observations. Overall, IRS1 and IRS2 appear to have stronger associations with MSI and CIMP-positive tumors than other tumor markers, but these associations appear to be less influenced by aspirin/NSAIDs use.

Genes, such as POMC, LEP, LEPR, and PPARγ have been hypothesized as being involved in multiple pathways including energy homeostasis, inflammation and insulin that could influence colon cancer risk. POMC had a greater influence on colon cancer risk in the absence of NSAIDS. Although POMC has been associated with energy homeostasis [42], one of the functions of POMC is anti-inflammatory. Given the interaction with NSAIDS, POMC appears to have a greater effect in the absence of anti-inflammatory drugs, suggesting the anti-inflammatory mechanisms are operational for the Ki-ras pathway. While LEP and LEPR have been hypothesized mainly as involved in regulation of energy homeostasis, involvement in both insulin and inflammation has been suggested [43, 44]. Our data suggest possibly greater involvement in the latter two pathways given the effect modification by aspirin/NSAID use. LEP (rs7799039) was directly associated with both MSI and CIMP among aspirin/NSAID users and inversely associated with MSI among non-users of aspirin/NSAIDs. PPARγ likewise may to be involved in an inflammation-related pathway that appears to alter risk of both MSI and CIMP tumors among non-aspirin/NSAID users. The underlying inflammatory state of the GI tract may be altered by the recent use of aspirin/NSAID thus modifying the ultimate risk associated with inflammation and insulin-related genes.

The inflammatory process is thought to be a key underlying component of colon cancer that is initiated by pro-inflammatory cytokines such as IL-6 in response to an inflammatory insult. IL-6 has been shown to stimulate secretion of C-reactive protein (CRP), an important biomarker for pro-inflammatory status in several diseases. Polymorphisms in the IL6 gene promoter have been reported to be related to levels of circulating C-reactive protein [45], to be associated with different profiles of plasma IL6 response to immunization [46], and to modify the association between high BMI and incident type-2 diabetes [47]. The rs1800975 (−174) and the rs1800796 polymorphism (−572 G>C) of the IL6 gene have been associated with IL-6 levels, body size, and colorectal cancer [4851]. We have previously reported that IL6 rs1800795 and rs1800796 are associated with reduced risk of colon cancer among people who report recently using aspirin/NSAIDs [29]; the reduced risk for the rs1800795 variant appears to be greatest for p53 mutations while the rs1800796 appears to influence all tumor types among NSAID users with no influence for any tumor type among non-users of aspirin/NSAIDs.

The transcription factor 7-like 2 (TCF7L2) rs7903146 marker was identified in genome-wide association studies as being associated with type-2 diabetes [52]; studies have corroborated these findings with the T allele associated with impaired insulin secretion [53]. In addition to its functional role in insulin regulation, TCF7L2 is involved in the Wnt/β-catenin signaling pathway that is central to colon cancer [2, 54, 55]. Although having a T allele was associated with an increased likelihood of having a p53 mutation, the associations differed by NSAIDs use. We have previously shown that NSAIDs modify the overall colon cancer risk associated with TCF7L2 [56]. In this study, we show that the inverse association among recent aspirin/NSAID users is confined to a reduced risk of having a Ki-ras mutation and to a lesser extent CIMP-positive profile, whereas an increased risk for CIMP-positive, p53, and Ki-ras mutations is observed among non-aspirin/NSAID users.

Not all polymorphisms in one gene may have the same functional effects or even be involved in the same disease pathway. VDR has many functions which may be dependent on the site of the polymorphism examined [57]. Our previous analyses have shown that VDR polymorphisms may influence colon and rectal tumor sites differently [30]. In these analyses, we show consistent associations for Ki-ras mutations for all VDR polymorphisms, however, Fok1 and CDX2 polymorphisms may have additional functions given associations with CIMP-positive tumors. It also is interesting to note that VDR polymorphisms can increase risk of CIMP-positive tumors while reducing risk of having a MSI tumor. These data suggest that a CIMP-positive/Ki-ras-mutated tumor may be relevant for some VDR markers. The fact that these associations are at least in part modified of NSAIDs use suggests an underlying inflammation-regulatory component to the observed associations.

The association between NAT2 and colon cancer is mixed in the literature, with most studies showing no or modest effects [58]. Data evaluating NAT2 with specific tumor markers has not, to our knowledge, been done previously. In our data, intermediate/rapid acetylators have much lower odds of developing CIMP-positive and Ki-ras tumors and to a lesser extent, p53 tumors; this effect is seen for both users and non-users of NSAIDS, suggesting a pathway that is independent of inflammation. This tumor-specific association could contribute to the inconsistent associations between NAT2 and colon cancer that exist in the literature.

There are several strengths to the study. We have a large collection of cases with tumor markers that were derived from a population-based sample that included all colon cancer cases identified over a three-year time period from three separate geographic areas. Thus, we believe our tumor data are representative of the most common tumor mutations that exist in colon cancers. However there are also limitations. Errors in genotyping could contribute to misclassification of exposure or outcome. Polymorphism exposures had elements of quality control to minimize, if not prevent genotyping errors. As part of the quality control, each genotype was included in every plate analyzed; all genotypes were in HWE. There is a greater likelihood of errors in outcomes since as we have described we did not assess the entire gene for p53 and Ki-ras, but looked at the common hotspots. Likewise, while our markers of CIMP have been shown to have validity in terms of associations, it is possible that other CIMP markers may also have been informative. We do not have tumor-mutation data on the APC gene, which given the size of the gene and the cost of sequencing is beyond the scope of this study. Additionally, while we have polymorphism data from many candidate genes that are based on existing data that suggest that these polymorphisms have functional components, other polymorphisms of these genes may have different associations. Thus, while using our existing data, we have only been able to look at a limited set of polymorphisms within the candidate pathways explored. Additional genes within these pathways need to be considered and may be important in defining associations with specific tumor markers. We used MERT to help define the best statistical model -- dominant, recessive, or additive -- to evaluate each polymorphism. Also we have made many comparisons, however, since this is a study exploring pathways we report all associations that were detected using confidence intervals along with exact p values. We were able to define recent aspirin/NSAID use, which appears to be an important component in defining risk for many factors rather than long-term use [59]. However, our recent aspirin/NSAID use variable does not take into account dosage.

Although these analyses should be viewed as exploratory, we believe that they are valuable given that little is known about various exposures and pathways and how they influence specific tumor subsets. We believe that we have added to our understanding of exposure-specific pathways to colon cancer. It is however vital that others replicate these results to help determine factors that lead to specific disease pathways as determined by genetic and epigenetic alterations in tumors.


Grant support: This study was funded by CA48998, CA85846, and CA61757 to Dr. Slattery. This research was supported by the Utah Cancer Registry, which is funded by Contract #N01-PC-67000 from the National Cancer Institute, with additional support from the State of Utah Department of Health and the University of Utah, the Northern California Cancer Registry, and the Sacramento Tumor Registry. The contents of this manuscript are solely the responsibility of the authors and do not necessarily represent the official view of the National Cancer Institute.

We would like to acknowledge the contributions of Dr. Kristen Anderson for her contributions to the data collection and study guidance, Sandra Edwards, Leslie Palmer, and Judy Morse to the data collection and management efforts of this study and to Dr. Hans Albertson, Michael Hoffman, and Erica Wolff for genotyping, and the core facility of the University of Utah Health Sciences Center for sequencing


Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.


1. Vogelstein BFE, Hamilton SR, Kern SE, zPreisinger AC, Leppert M, et al. Genetic alterations during colorectal-tumor development. New England Journal of Medicine. 1988;319:525–532. [PubMed]
2. Samowitz WS, et al. APC mutations and other genetic and epigenetic changes in colon cancer. Mol Cancer Res. 2007;5(2):165–70. [PubMed]
3. Slattery ML, et al. Associations between cigarette smoking, lifestyle factors, and microsatellite instability in colon tumors. J Natl Cancer Inst. 2000;92(22):1831–6. [PubMed]
4. Samowitz WS, et al. Association of smoking, CpG island methylator phenotype, and V600E BRAF mutations in colon cancer. J Natl Cancer Inst. 2006;98(23):1731–8. [PubMed]
5. Morimoto LM, et al. Risk factors for hyperplastic and adenomatous polyps: evidence for malignant potential? Cancer Epidemiol Biomarkers Prev. 2002;11(10 Pt 1):1012–8. [PubMed]
6. Slattery ML, et al. Associations between dietary intake and Ki-ras mutations in colon tumors: a population-based study. Cancer Res. 2000;60(24):6935–41. [PubMed]
7. Martinez ME, et al. Risk factors for Ki-ras protooncogene mutation in sporadic colorectal adenomas. Cancer Res. 1999;59(20):5181–5. [PubMed]
8. Samowitz WS, et al. Polymorphisms in insulin-related genes predispose to specific KRAS2 and TP53 mutations in colon cancer. Mutat Res. 2006;595(1–2):117–24. [PubMed]
9. Yang L, Pei Z. Bacteria, inflammation, and colon cancer. World J Gastroenterol. 2006;12(42):6741–6. [PMC free article] [PubMed]
10. Belfiore A. The role of insulin receptor isoforms and hybrid insulin/IGF-I receptors in human cancer. Curr Pharm Des. 2007;13(7):671–86. [PubMed]
11. Newcomb PA, et al. Estrogen plus progestin use, microsatellite instability, and the risk of colorectal cancer in women. Cancer Res. 2007;67(15):7534–9. [PubMed]
12. Slattery ML, et al. Associations between ERalpha, ERbeta, and AR genotypes and colon and rectal cancer. Cancer Epidemiol Biomarkers Prev. 2005;14(12):2936–42. [PubMed]
13. Deal C, et al. Novel Promoter Polymorphism in Insulin-Like Growth Factor-Binding Protein-3: Correlation with Serum Levels and Interaction with Known Regulators. J Clin Endocrinol Metab. 2001;86(3):1274–1280. [PubMed]
14. Schernhammer ES, et al. Polymorphic variation at the −202 locus in IGFBP3: Influence on serum levels of insulin-like growth factors, interaction with plasma retinol and vitamin D and breast cancer risk. Int J Cancer. 2003;107(1):60–4. [PubMed]
15. DeLellis K, et al. IGF1 genotype, mean plasma level and breast cancer risk in the Hawaii/Los Angeles multiethnic cohort. Br J Cancer. 2003;88(2):277–82. [PMC free article] [PubMed]
16. Georges JL, et al. Impact of pathogen burden in patients with coronary artery disease in relation to systemic inflammation and variation in genes encoding cytokines. Am J Cardiol. 2003;92(5):515–21. [PubMed]
17. Slattery ML, et al. Energy balance and colon cancer--beyond physical activity. Cancer Res. 1997;57(1):75–80. [PubMed]
18. Slattery ML, David J, Jacobs R. Assessment of ability to recall physical activity of several years ago. Ann Epidemiol. 1995;5:292–296. [PubMed]
19. Samowitz WS, et al. Relationship of Ki-ras mutations in colon cancers to tumor location, stage, and survival: a population-based study. Cancer Epidemiol Biomarkers Prev. 2000;9(11):1193–7. [PubMed]
20. Samowitz WS, et al. Evaluation of a large, population-based sample supports a CpG island methylator phenotype in colon cancer. Gastroenterology. 2005;129(3):837–45. [PubMed]
21. Samowitz WS, et al. Poor survival associated with the BRAF V600E mutation in microsatellite-stable colon cancers. Cancer Res. 2005;65(14):6063–9. [PubMed]
22. Slattery ML, et al. Diet activity, and lifestyle associations with p53 mutations in colon tumors. Cancer Epidemiol Biomarkers Prev. 2002;11(6):541–8. [PubMed]
23. Samowitz WS, et al. Inverse relationship between microsatellite instability and K-ras and p53 gene alterations in colon cancer. Am J Pathol. 2001;158(4):1517–24. [PubMed]
24. Curtin K, et al. MTHFR C677T and A1298C polymorphisms: diet, estrogen, and risk of colon cancer. Cancer Epidemiol Biomarkers Prev. 2004;13(2):285–92. [PubMed]
25. Slattery ML, et al. GSTM-1 and NAT2 and genetic alterations in colon tumors. Cancer Causes Control. 2002;13(6):527–34. [PubMed]
26. Slattery ML, et al. A molecular variant of the APC gene at codon 1822: its association with diet, lifestyle, and risk of colon cancer. Cancer Res. 2001;61(3):1000–4. [PubMed]
27. Slattery ML, et al. Associations among IRS1, IRS2, IGF1, and IGFBP3 genetic polymorphisms and colorectal cancer. Cancer Epidemiol Biomarkers Prev. 2004;13(7):1206–14. [PubMed]
28. Slattery ML, et al. PPARgamma and colon and rectal cancer: associations with specific tumor mutations, aspirin, ibuprofen and insulin-related genes (United States) Cancer Causes Control. 2006;17(3):239–49. [PubMed]
29. Slattery ML, et al. IL6 genotypes and colon and rectal cancer. Cancer Causes Control. 2007;18(10):1095–105. [PMC free article] [PubMed]
30. Slattery ML, et al. CDX2 VDR polymorphism and colorectal cancer. Cancer Epidemiol Biomarkers Prev. 2007;16(12):2752–5. [PMC free article] [PubMed]
31. Slattery ML, et al. Leptin and leptin receptor genotypes and colon cancer: gene-gene and gene-lifestyle interactions. Int J Cancer. 2008;122(7):1611–7. [PMC free article] [PubMed]
32. Curtin K, et al. Thymidylate synthase polymorphisms and colon cancer: associations with tumor stage, tumor characteristics and survival. Int J Cancer. 2007;120(10):2226–32. [PubMed]
33. Slattery ML, et al. Associations between apoE genotype and colon and rectal cancer. Carcinogenesis. 2005;26(8):1422–9. [PubMed]
34. Ulrich CM, et al. Polymorphisms in the reduced folate carrier, thymidylate synthase, or methionine synthase and risk of colon cancer. Cancer Epidemiol Biomarkers Prev. 2005;14(11 Pt 1):2509–16. [PubMed]
35. Freidlin B, et al. Trend tests for case-control studies of genetic markers: power, sample size and robustness. Hum Hered. 2002;53(3):146–52. [PubMed]
36. Camp NJ, Slattery ML. Classification tree analysis: a statistical tool to investigate risk factor interactions with an example for colon cancer (United States) Cancer Causes Control. 2002;13(9):813–23. [PubMed]
37. Shen L, et al. Integrated genetic and epigenetic analysis identifies three different subclasses of colon cancer. Proc Natl Acad Sci U S A. 2007;104(47):18654–9. [PubMed]
38. Slattery ML, et al. Dietary intake and microsatellite instability in colon tumors. Int J Cancer. 2001;93(4):601–7. [PubMed]
39. Slattery ML, et al. Diet and lifestyle factor associations with CpG island methylator phenotype and BRAF mutations in colon cancer. Int J Cancer. 2007;120(3):656–63. [PubMed]
40. Stoeltzing O, et al. Regulation of cyclooxygenase-2 (COX-2) expression in human pancreatic carcinoma cells by the insulin-like growth factor-I receptor (IGF-IR) system. Cancer Lett. 2007;258(2):291–300. [PMC free article] [PubMed]
41. Nosho K, et al. Interplay of insulin-like growth factor-II, insulin-like growth factor-I, insulin-like growth factor-I receptor, COX-2, and matrix metalloproteinase-7, play key roles in the early stage of colorectal carcinogenesis. Clin Cancer Res. 2004;10(23):7950–7. [PubMed]
42. Liu GS, et al. Gene transfer of pro-opiomelanocortin prohormone suppressed the growth and metastasis of melanoma: involvement of alpha-melanocyte-stimulating hormone-mediated inhibition of the nuclear factor kappaB/cyclooxygenase-2 pathway. Mol Pharmacol. 2006;69(2):440–51. [PubMed]
43. Bastard JP, et al. Recent advances in the relationship between obesity, inflammation, and insulin resistance. Eur Cytokine Netw. 2006;17(1):4–12. [PubMed]
44. Halle M, Persson PB. Role of leptin and leptin receptor in inflammation. Am J Physiol Regul Integr Comp Physiol. 2003;284(3):R760–2. [PubMed]
45. Ferrari SL, et al. Two promoter polymorphisms regulating interleukin-6 gene expression are associated with circulating levels of C-reactive protein and markers of bone resorption in postmenopausal women. J Clin Endocrinol Metab. 2003;88(1):255–9. [PubMed]
46. Bennermo M, et al. Genetic Predisposition of the Interleukin-6 Response to Inflammation: Implications for a Variety of Major Diseases? Clin Chem. 2004 [PubMed]
47. Mohlig M, et al. Body mass index and C-174G interleukin-6 promoter polymorphism interact in predicting type 2 diabetes. J Clin Endocrinol Metab. 2004;89(4):1885–90. [PubMed]
48. Fishman D, et al. The effect of novel polymorphisms in the interleukin-6 (IL-6) gene on IL-6 transcription and plasma IL-6 levels, and an association with systemic-onset juvenile chronic arthritis. J Clin Invest. 1998;102(7):1369–76. [PMC free article] [PubMed]
49. Theodoropoulos G, et al. Relation between common polymorphisms in genes related to inflammatory response and colorectal cancer. World J Gastroenterol. 2006;12(31):5037–43. [PMC free article] [PubMed]
50. Landi S, et al. Association of common polymorphisms in inflammatory genes interleukin (IL)6, IL8, tumor necrosis factor alpha, NFKB1, and peroxisome proliferator-activated receptor gamma with colorectal cancer. Cancer Res. 2003;63(13):3560–6. [PubMed]
51. Slattery ML, et al. IL6, aspirin, nonsteroidal anti-inflammatory drugs, and breast cancer risk in women living in the southwestern united states. Cancer Epidemiol Biomarkers Prev. 2007;16(4):747–55. [PubMed]
52. Salonen JT, et al. Type 2 diabetes whole-genome association study in four populations: the DiaGen consortium. Am J Hum Genet. 2007;81(2):338–45. [PubMed]
53. Lyssenko V, et al. Mechanisms by which common variants in the TCF7L2 gene increase risk of type 2 diabetes. J Clin Invest. 2007;117(8):2155–63. [PMC free article] [PubMed]
54. Prunier C, Hocevar BA, Howe PH. Wnt signaling: physiology and pathology. Growth Factors. 2004;22(3):141–50. [PubMed]
55. Wong NA, Pignatelli M. Beta-catenin--a linchpin in colorectal carcinogenesis? Am J Pathol. 2002;160(2):389–401. [PubMed]
56. Slattery ML, et al. Transcription Factor 7-like 2 Polymorphism and Colon Cancer. Cancer Epidemiol Biomarkers Prev. 2008;17(4):978–82. [PMC free article] [PubMed]
57. Slattery ML. Vitamin D receptor gene (VDR) associations with cancer. Nutr Rev. 2007;65(8 Pt 2):S102–4. [PMC free article] [PubMed]
58. Brockton N, et al. N-acetyltransferase polymorphisms and colorectal cancer: a HuGE review. Am J Epidemiol. 2000;151(9):846–61. [PubMed]
59. Friedman GD, et al. Drugs and colon cancer. Pharmacoepidemiol Drug Saf. 1998;7(2):99–106. [PubMed]