This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Poynter, J. N. Search for Related Content PubMed PubMed Citation Articles by Poynter, J. N.

Variants on 9p24 and 8q24 Are Associated with Risk of Colorectal Cancer: Results from the Colon Cancer Family Registry

¹ Department of Preventive Medicine, University of Southern California, Los Angeles, California; ² Translational Genomics Research Institute, Phoenix, Arizona; ³ Cancer Care Ontario, Toronto, Ontario, Canada; ⁴ The Cleveland Clinic Foundation, Cleveland, Ohio; ⁵ Mayo Clinic, Rochester, Minnesota; ⁶ University of Melbourne, Melbourne, Australia; ⁷ Dartmouth Medical School, Lebanon, New Hampshire; ⁸ Epidemiology Program, Cancer Research Center of Hawaii, University of Hawaii, Honolulu, Hawaii; ⁹ Fred Hutchinson Cancer Research Center, Seattle, Washington; ¹⁰ Department of Cellular Pathology, St. Mark's Hospital, Middlesex, United Kingdom; and ¹¹ Familial Cancer Laboratory, Queensland Institute of Medical Research, Queensland, Australia

Requests for reprints: Jenny N. Poynter, Department of Preventive Medicine, University of Southern California, 1441 Eastlake Avenue, NOR4411A, Los Angeles, CA 90089-9175. Phone: 323-865-3985; E-mail: poynter{at}usc.edu.

	Abstract

Top Abstract Introduction Materials and Methods Results Discussion References

 
Recent publications have reported that common variants on 8q24 are associated with both prostate and colorectal cancers (CRC). In addition, one of these studies (the ARCTIC study) initially observed an association with a single nucleotide polymorphism (SNP) on 9p24 that was not confirmed in some of their validation data sets. In the research described here, we conducted a case-unaffected sibling analysis using population- and clinic-based discordant sibships (N = 1,567 sibships) from the Colon Cancer Family Registry (Colon CFR) to investigate the associations between common variants at 9p24 and 8q24 and risk of CRC. We also evaluated whether these associations differed by age, family history, and tumor characteristics, including microsatellite instability and tumor site. Associations were estimated using conditional logistic regression, treating sibship as the matching factor. Analyses were adjusted for age and sex, and stratified by ascertainment source (population versus clinic). We observed an association between a SNP on 9p24 (rs719725) and risk of CRC in the population-based series (AA versus CC: odds ratios, 1.46; 95% confidence interval, 1.06–2.02; AC versus CC: odds ratios, 1.50; 95% confidence interval, 1.14–1.98; P = 0.011 on 2 df). In the population-based series, we also detected statistically significant associations between two SNPs on 8q24, rs10505477 and rs6983267, and risk of CRC (P = 0.005 and P = 0.002, respectively). There was no evidence of statistically significant heterogeneity by age at diagnosis, family history of CRC, microsatellite instability, or tumor site at either locus and no evidence of interaction between SNPs on 8q24 and 9p24. These data suggest that common variants may play important roles in the risk of CRC. [Cancer Res 2007;67(23):11128–32]

	Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

 
Previous studies have identified variants associated with risk of prostate cancer on 8q24 (1–5). Recently, variants on 8q24, including rs6983267 and rs10505477, were reported to be associated with risk of colorectal cancer (CRC; refs. 6–9). In addition, Zanke et al. (7) also identified a potential susceptibility allele for CRC on 9p24 (rs719725), although this was not validated in all replication phases of their study. Here, we used data from the Colon Cancer Family Registry (Colon CFR) to evaluate associations between single nucleotide polymorphisms (SNPs) on 8q24 and 9p24 and risk of CRC, and to assess the heterogeneity of associations by age, family history, and tumor characteristics.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

 
Study samples. Participants were recruited to the Colon CFR from six centers: University of Hawaii, Fred Hutchinson Cancer Research Center, Mayo Clinic, University of Southern California Consortium, Cancer Care Ontario, and University of Melbourne, using population-based and clinic-based ascertainment strategies (10). Some population-based centers recruited all incident cases whereas others oversampled cases with a family history or early age of onset. Standardized procedures were used to collect epidemiologic data and blood samples from all participants and tumor blocks and pathology reports from cases.

In this analysis, we used a case-unaffected sibling control design. Cases were probands and affected relatives with primary invasive CRC interviewed within 5 years of diagnosis. Controls were siblings without a diagnosis of CRC at the time of ascertainment. A total of 2,112 discordant sibships (2,195 cases and 3,581 controls) were available for this analysis. Three individuals with discordant genotypes on blinded replicates and 30 individuals for whom the sex predicted by genotyping did not match the reported sex were excluded. We also excluded 157 individuals because samples were not available for genotyping and 45 who had missing genotypes at all SNPs; resulting in a loss of 164 sibships. In addition, 384 population-based sibships from Seattle and Ontario included in the ARCTIC study (7) were excluded to maintain a data set independent of prior analyses.

Analyses of population-based families included 1,344 cases (1,262 probands and 82 affected relatives) and 2,196 unaffected siblings from 1,304 sibships. Clinic-based analyses included 288 cases (170 probands and 118 affected relatives) and 505 unaffected siblings from 263 sibships. In addition, 359 non–Hispanic white unrelated population controls from the Seattle CFR center were included for genotyping quality control. We obtained informed consent from all participants. The study was approved by the Institutional Review Board at each Colon CFR site.

Genotyping. Genotyping for SNPs on 9p24 (rs719725) and 8q24 (rs6983267 and rs10505477) was performed using Sequenom's iPLEX Gold. PCR and extension primers for these SNPs were designed using the MassARRAY Assay Design 3.0 software (Sequenom, Inc.) and are available upon request. PCR amplification and single base extension reactions were performed according to the manufacturer's instructions. Extension product sizes were determined by mass spectrometry using Sequenom's Compact matrix-assisted laser desorption ionization-time of flight mass spectrometer. The resulting mass spectra were converted to genotype data using SpectroTYPER-RT software.

Genotype data from 30 CEPH trios (Coriell Cell Repository, Camden, NJ) were used to confirm reliability and reproducibility of the genotyping. No errors of mendelian inheritance were detected and genotypes for these individuals showed perfect concordance with genotypes from the International HapMap Project. Intraplate and interplate replicates at a rate of 5% were included on all plates and in all batches with no discordant genotypes in 284 known replicates. Discordant genotypes were detected in 3 of 142 blinded replicates (2%). Genotype data from the Ontario samples included in the ARCTIC study (n = 248; ref. 7) were used as an additional blinded replicate sample and showed 100% concordance. Hardy-Weinberg equilibrium was evaluated in unrelated population controls. All SNPs were in Hardy-Weinberg equilibrium.

Microsatellite instability status. Microsatellite instability (MSI) was evaluated using a panel of 10 markers (BAT25, BAT26, BAT40, MYCL, D5S346, D17S250, ACTC, D18S55, D10S197, and BAT34C4) using standard techniques (11). Results were required for at least five markers to determine MSI status. Tumors were deemed MSI-H if instability was observed at 30% of markers, MSI-L if >0 and <30% of markers were instable, and MSS if all markers were stable.

Mismatch repair mutations. Mutations in the mismatch repair (MMR) genes MSH2, MLH1, and MSH6 were identified in selected participants in the Colon CFR using a combined approach of denaturing high performance liquid chromatography/direct sequencing and multiplex ligation–dependent probe amplification. All clinic-based probands, all MSI-H or MSI-L population-based probands, and a sample of 100 MSS population-based probands were screened.

Statistical analysis. All statistical analyses were performed using SAS v9.1 (SAS Institute). Multivariable conditional logistic regression adjusted for sex and age with sibship as the matching factor was used to estimate associations between variants and risk of CRC. Because we were not certain these SNPs were the causal variants, we used a robust variance estimator to prevent biased estimates from testing association in the presence of linkage (12). Population- and clinic-based data were analyzed separately. We first tested the associations between SNPs and the risk of CRC using a codominant model with a 2 df test of heterogeneity across all genotypes. We then used an allele-dosage model to estimate the per-allele effect and performed a test of deviation from additivity. Where the estimated heterozygote and homozygote odds ratios (OR) were similar, we estimated the OR of these genotypes combined. We estimated stratum-specific ORs among population-based families to evaluate heterogeneity by age at diagnosis, tumor site, MSI, and family history of CRC in a first-degree relative as defined in the proband. Analyses were repeated among non–Hispanic white individuals; we did not have sufficient sample size to evaluate associations for other racial/ethnic categories.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

 
The characteristics of the study samples are presented in Table 1 . Clinic-based cases were more likely to have MSI-H tumors than population-based cases.

8q24 (rs10505477 and rs6983267). In the population-based families, we observed a statistically significant association between rs10505477 and risk of CRC using a 2 df test in a codominant model (CT versus CC: OR, 1.38; 95% CI, 1.09–1.75; TT versus CC: OR, 1.15; 95% CI, 0.85–1.55; P = 0.005; Table 2). However, we saw no statistically significant per-allele association for the T allele (OR, 1.07; 95% CI, 0.93–1.23; P = 0.35). There was evidence of deviation from additivity (P = 0.001). The OR for individuals who carried one or two copies of the T allele was 1.34 (95% CI, 1.06–1.70).

Because rs10505477 and rs6983267 were in almost complete linkage disequilibrium (r² = 0.94), the results were similar for rs6983267. The 2 df test for the codominant model was statistically significant in the population-based families (TG versus TT: OR, 1.45; 95% CI, 1.14–1.85; GG versus TT: OR, 1.23; 95% CI, 0.91–1.85; P = 0.002; Table 2). The allele-dosage model showed no statistically significant association for the G allele (OR, 1.11; 95% CI, 0.96–1.29; P = 0.18). A test of deviation from additivity was statistically significant (P = 0.002) and, when individuals carrying one or two copies of the G allele were combined for stratum-specific analyses, we found an OR of 1.42 (95% CI, 1.12–1.80; P = 0.004).

There was no association between either SNP and risk of CRC in clinic-based families (Table 2). The risk estimates for both rs10505477 and rs6983267 were statistically significantly different between the population-based and the clinic-based series (P = 0.036 and P = 0.037, respectively).

When we excluded 41 known MMR mutation carriers from the population-based series, the ORs did not differ substantially for rs10505477 (OR, 1.35; 95% CI, 1.06–1.71 for CT versus CC; OR, 1.13; 95% CI, 0.84–1.52 for TT versus CC) or for rs6983267 (OR, 1.41; 95% CI, 1.11–1.80 for TG versus TT; OR, 1.20; 95% CI, 0.89–1.62 for GG versus TT). Similarly, when we excluded 43 known MMR mutation carriers from the clinic-based series, the ORs were virtually unchanged (rs10505477: OR, 0.74; 95% CI, 0.43–1.29 for CT versus CC; OR, 0.84; 95% CI, 0.43–1.64 for TT versus CC; rs6983267: OR, 0.83; 95% CI, 0.47–1.47 for TG versus TT; OR, 0.95; 95% CI, 0.48–1.85 for GG versus TT). MMR mutation data were not available for 37 MSI-H–affected relatives because our mutation screening strategy included only probands.

We observed no evidence for heterogeneity by age of onset, family history, or tumor site in population-based families for the SNPs on 8q24 (Table 3). We saw suggestions that the associations of both rs10505477 and rs6983267 with risk of CRC were stronger in MSI-H cases, but the trend of increasing risk across MSI categories (MSS, MSI-L, and MSI-H) was not statistically significant for either SNP (P for trend 0.20 and 0.19, respectively). When we removed individuals with known MMR mutations from the analysis, the ORs for the MSI-H group were slightly attenuated (OR, 1.77; 95% CI, 0.83–3.78 for rs10505477; and OR, 1.67; 95% CI, 0.77–3.62 for rs6983267).

To address potential survival bias due to the inclusion of cases interviewed up to 5 years after diagnosis, we repeated the analyses of main effects at both loci including only probands interviewed within 2 years of diagnosis and their unaffected siblings. The findings were qualitatively very similar to those above (data not shown). When we limited the analysis to non–Hispanic whites, the results also did not change substantially (data not shown).

We evaluated the association between risk alleles at 8q24 and 9p24 jointly. Because the SNPs on 8q24 were in strong linkage disequilibrium, we included only rs6983267 in the joint model. The OR was 1.36 (95% CI, 0.81–2.26) for carrying a risk allele at rs6983267, 1.41 (95% CI, 0.87–2.29) for carrying a risk allele at rs719725, and 1.07 (95% CI, 0.63–1.83) for the interaction term, suggesting that these two loci have independent noninteractive associations with risk of CRC.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

 
In this study, we found an association between rs719725 on 9p24 and risk of CRC consistent with the report of Zanke et al. (7). The association did not seem to differ by MSI status, age of onset, family history, or tumor site (Table 3). An association with this SNP was detected in phase 1 of that study and was observed in three of its replication phases, but was not confirmed using samples from two French data sets or the EPIC cohort (7). It is conceivable that there was a greater degree of heterogeneity when the French and EPIC data sets were analyzed compared with the cases in the data sets included in earlier stages.

The SNP on 9p24 (rs719725) does not lie within a gene; however, several genes are in close proximity. The gene most proximal to this SNP (37 kb telomeric) is protein kinase NYD-SP25 isoform 3 (TPD52L3), in the tumor protein D52 family. Three additional genes are in the region including interleukin 33 (IL33, 124 kb telomeric), ubiquitin-like PHD and RING finger domain–containing protein (UHRF2, 47 kb centromeric), and glycine dehydrogenase (GLDC, 167 kb centromeric). To our knowledge, none of these genes have been evaluated for associations with CRC. More detailed mapping of the region and functional data will help clarify a mechanistic role for a susceptibility allele at this locus.

We also observed a statistically significant association between 8q24 and risk of CRC (6–9). This is the fifth study to find an association between 8q24 SNPs and risk of CRC, although the most likely mode of inheritance has not been consistent across all studies (6–9). Two investigations observed a log-linear per-allele effect at this locus (7, 8). One study observed a statistically significant per-allele association for rs6983267, but found that a nonmultiplicative risk model was a better fit for the data (9). A fourth study observed a dominant effect for the association between rs10505477 and risk of CRC (6). Our data suggest a deviation from additivity consistent with a dominant model. We note that the OR for heterozygous carriers of the risk allele at 8q24 was higher than the OR for homozygous carriers, although the confidence limits largely overlap. Other studies have previously reported that the SNPs we evaluated on 8q24 are not in a known gene (7–9), although these SNPs are located near the oncogene MYC.

Published studies of 8q24 and CRC risk have differed in their sampling strategies for case selection. Our study population has been enriched with individuals diagnosed at an early age and a positive family history of CRC, which may limit the ability to generalize our results if there is heterogeneity in risk by age of onset and family history. It is important to examine possible sources of heterogeneity as this may help us understand the role of these SNPs in risk of CRC.

Heterogeneity in the observed associations with CRC risk by age, family history, and tumor site has not been consistently reported in other studies (6, 8, 9). Our results suggest that the association between SNPs on 8q24 and the risk of CRC does not differ by age at diagnosis, family history, or tumor site (Table 3). Heterogeneity by MSI status was not observed in one published study, although the actual stratum-specific results were not provided (8). Our population-based results suggest that the association may be stronger among MSI-H cases, although the trend of increasing risk across MSI categories was not statistically significant (Table 3).

The stronger ORs among the MSI-H cases at the 8q24 locus suggest that we might expect the association between this locus and CRC to be stronger in the clinic-based series due to the higher prevalence of MSI in that sample. However, we observed no association between this locus and risk of CRC in the clinic-based families. One potential explanation for this finding is that there are multiple pathways that lead to MSI-H CRC, including germline mutation in the MMR genes and MLH1 promoter hypermethylation. It is possible that the MSI-H tumors in the population-based series are more likely to be due to MLH1 methylation than those in the clinic-based series.

We observed associations between these common 9p24 and 8q24 variants and CRC in the population-based families but not in the clinic-based families. One plausible explanation for this finding is that the genetic difference between affected and unaffected siblings in the multiple-case, clinic-based families may for the most part be due to variants which have a strong association with risk, including those in as yet undiscovered colorectal cancer susceptibility genes. Another possibility is that the limited sample size in the clinic-based data set was not sufficient to detect an association. This is unlikely to explain the difference for the SNPs on 8q24 because we observed significant heterogeneity in the risk estimates.

We observed statistically significant associations between SNPs on 9p24 and 8q24 and risk of CRC. The potential susceptibility alleles on both chromosomes are not located within known genes, so further work will be required before the clinical and biological implications of these variants can be understood. The risk estimates for the associations between these SNPs and CRC are modest; however, the high frequency of the variant alleles suggests that these are important susceptibility loci.

	Acknowledgments

 
Grant support: NCI 1 UO1 CA74799, NCI R01 CA112237 (RWH), NCI T32 CA009142 (J.N. Poynter), NCI R01 CA104667 (E.L. Goode), and National Cancer Institute of Canada post-Ph.D. Research Fellowship no. 017602 (J.C. Figueiredo).

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 the participants in the Colon Cancer Family Registry who have generously donated their time for this project.

Received 8/22/07. Revised 9/19/07. Accepted 10/15/07.

	References

Top Abstract Introduction Materials and Methods Results Discussion References

 

1. Haiman CA, Patterson N, Freedman ML, et al. Multiple regions within 8q24 independently affect risk for prostate cancer. Nat Genet 2007;39:638–44.[CrossRef][Medline]
2. Freedman ML, Haiman CA, Patterson N, et al. Admixture mapping identifies 8q24 as a prostate cancer risk locus in African-American men. Proc Natl Acad Sci U S A 2006;103:14068–73.[Abstract/Free Full Text]
3. Gudmundsson J, Sulem P, Manolescu A, et al. Genome-wide association study identifies a second prostate cancer susceptibility variant at 8q24. Nat Genet 2007;39:631–7.[CrossRef][Medline]
4. Amundadottir LT, Sulem P, Gudmundsson J, et al. A common variant associated with prostate cancer in European and African populations. Nat Genet 2006;38:652–8.[CrossRef][Medline]
5. Yeager M, Orr N, Hayes RB, et al. Genome-wide association study of prostate cancer identifies a second risk locus at 8q24. Nat Genet 2007;39:645–9.[CrossRef][Medline]
6. Gruber SB, Moreno V, Rozek LS, et al. Genetic variation in 8q24 associated with risk of colorectal cancer. Cancer Biol Ther 2007;6.
7. Zanke BW, Greenwood CM, Rangrej J, et al. Genome-wide association scan identifies a colorectal cancer susceptibility locus on chromosome 8q24. Nat Genet 2007;39:989–94.[CrossRef][Medline]
8. Tomlinson I, Webb E, Carvajal-Carmona L, et al. A genome-wide association scan of tag SNPs identifies a susceptibility variant for colorectal cancer at 8q24.21. Nat Genet 2007;39:984–8.[CrossRef][Medline]
9. Haiman CA, Le Marchand L, Yamamato J, et al. A common genetic risk factor for colorectal and prostate cancer. Nat Genet 2007;39:954–6.[CrossRef][Medline]
10. Newcomb PA, Baron J, Cotterchio M, et al. Colon Cancer Family Registry: an international resource for studies of the genetic epidemiology of colon cancer. Cancer Epidemiol Biomarkers Prev. In press 2007.
11. Lindor NM, Rabe K, Petersen GM, et al. Lower cancer incidence in Amsterdam-I criteria families without mismatch repair deficiency: familial colorectal cancer type X. JAMA 2005;293:1979–85.[Abstract/Free Full Text]
12. Siegmund KD, Langholz B, Kraft P, Thomas DC. Testing linkage disequilibrium in sibships. Am J Hum Genet 2000;67:244–8.[CrossRef][Medline]

This article has been cited by other articles:

				M. S. Cicek, S. L. Slager, S. J. Achenbach, A. J. French, H. E. Blair, S. R. Fink, N. R. Foster, B. F. Kabat, K. C. Halling, J. M. Cunningham, et al. Functional and Clinical Significance of Variants Localized to 8q24 in Colon Cancer Cancer Epidemiol. Biomarkers Prev., September 1, 2009; 18(9): 2492 - 2500. [Abstract] [Full Text] [PDF]

				S. S. Kupfer, J. B. Torres, S. Hooker, J. R. Anderson, A. D. Skol, N. A. Ellis, and R. A. Kittles Novel single nucleotide polymorphism associations with colorectal cancer on chromosome 8q24 in African and European Americans Carcinogenesis, August 1, 2009; 30(8): 1353 - 1357. [Abstract] [Full Text] [PDF]

				K. Curtin, W.-Y. Lin, R. George, M. Katory, J. Shorto, L. A. Cannon-Albright, D. T. Bishop, A. Cox, N. J. Camp, and Colorectal Cancer Study Group Meta Association of Colorectal Cancer Confirms Risk Alleles at 8q24 and 18q21 Cancer Epidemiol. Biomarkers Prev., February 1, 2009; 18(2): 616 - 621. [Abstract] [Full Text] [PDF]

				S. L. Park, S.-C. Chang, L. Cai, C. Cordon-Cardo, B.-G. Ding, S. Greenland, S. K. Hussain, Q. Jiang, S. Liu, M.-L. Lu, et al. Associations between Variants of the 8q24 Chromosome and Nine Smoking-Related Cancer Sites Cancer Epidemiol. Biomarkers Prev., November 1, 2008; 17(11): 3193 - 3202. [Abstract] [Full Text] [PDF]

				S. I. Berndt, J. D. Potter, A. Hazra, M. Yeager, G. Thomas, K. W. Makar, R. Welch, A. J. Cross, W.-Y. Huang, R. E. Schoen, et al. Pooled analysis of genetic variation at chromosome 8q24 and colorectal neoplasia risk Hum. Mol. Genet., September 1, 2008; 17(17): 2665 - 2672. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Poynter, J. N. Search for Related Content PubMed PubMed Citation Articles by Poynter, J. N.