Tagging Single Nucleotide Polymorphisms in Cell Cycle Control Genes and Susceptibility to Invasive Epithelial Ovarian Cancer

Research Article

Simon A. Gayther,1 Honglin Song,2 Susan J. Ramus,1 Susan Kru¨ger Kjaer,4 Alice S. Whittemore,5 Lydia Quaye,1 Jonathan Tyrer,2 Danielle Shadforth,2 Estrid Hogdall,4 Claus Hogdall,6 Jan Blaeker,7 Richard DiCioccio,8 Valerie McGuire,5 Penelope M. Webb,9 Jonathan Beesley,9 Adele C. Green,9 David C. Whiteman,9 The Australian Ovarian Cancer Study Group,9 The Australian Cancer Study (Ovarian Cancer),20 Marc T. Goodman,10 Galina Lurie,10 Michael E. Carney,10 Francesmary Modugno,10 Roberta B. Ness,11 Robert P. Edwards,12 Kirsten B. Moysich,7 Ellen L. Goode,13 Fergus J. Couch,13 Julie M. Cunningham,13 Thomas A. Sellers,14 Anna H. Wu,15 Malcolm C. Pike,15 Edwin S. Iversen,16 Jeffrey R. Marks,16 Montserrat Garcia-Closas,17 Louise Brinton,17 Jolanta Lissowska,18 Beata Peplonska,19 Douglas F. Easton,3 Ian Jacobs,1 Bruce A.J. Ponder,2 Joellen Schildkraut,16 C. Leigh Pearce,15 Georgia Chenevix-Trench,9 Andrew Berchuck,16 Paul D.P. Pharoah,2 and on behalf of the Ovarian Cancer Association Consortium

1Translational Research Laboratories, University College London, London, United Kingdom; 2Cancer Research United Kingdom Department of Oncology and 3Cancer Research United Kingdom Genetic Epidemiology Unit, University of Cambridge, Strangeways Research Laboratory, Cambridge, United Kingdom; 4Danish Cancer Society, Copenhagen, Denmark; 5Stanford University School of Medicine, Stanford, California; 6University of Copenhagen, Copenhagen, Denmark; 7Aarhus University Hospital, Skejby, Aarhus, Denmark; 8Roswell Park Cancer Institute, Buffalo, New York; 9The Queensland Institute of Medical Research, Post Office Royal Brisbane Hospital, Australia; 10Cancer Research Center, University of Hawaii, Honolulu, Hawaii; 11Department of Epidemiology and University of Pittsburgh Cancer Institute, 12Department of Obstetrics, Gynecology and Reproductive Health, University of Pittsburgh Cancer Institute, Pittsburgh, Pennsylvania; 13Mayo Clinic College of Medicine, Rochester, Minnesota; 14H. Lee Moffitt Cancer Center and Research Institute, Tampa, Florida; 15University of Southern California, Keck School of Medicine, Los Angeles, California; 16Duke University Medical Center, Durham, North Carolina; 17Division of Cancer Epidemiology and Genetics, National Cancer Institute, Rockville, Maryland; 18Department of Cancer Epidemiology and Prevention, M. Sklodowska-Curie Institute of Oncology and Cancer Center, Warsaw, Poland; 19Nofer Institute of Occupational Medicine, Lodz, Poland; and 20Peter MacCallum Cancer Institute, Melbourne, Australia

Abstract the minor allele of these SNPs was foundto be associated with reduced risk [OR, 0.91 (0.85–0.98) for rs3731257; and OR, High-risk susceptibility genes explain <40% of the excess risk 0.93 (0.87–0.995) for rs2066827]. In conclusion, we have of familial ovarian cancer. Therefore, other ovarian cancer foundevidencethat a single taggedSNP in both the CDKN2A susceptibility genes are likely to exist. We have useda single and CDKN1B genes may be associatedwith reducedovarian nucleotide polymorphism (SNP)–tagging approach to evaluate cancer risk. This study highlights the need for multicenter common variants in 13 genes involvedin cell cycle control— CCND1, CCND2, CCND3, CCNE1, CDK2, CDK4, CDK6, CDKN1A, collaborations for genetic association studies. [Cancer Res CDKN1B, CDKN2A, CDKN2B, CDKN2C, and CDKN2D—andrisk 2007;67(7):3027–35] of invasive epithelial ovarian cancer. We useda two-stage, multicenter, case-control study. In stage 1, 88 SNPs that tag Introduction common variation in these genes were genotypedin three Several genes have been identified that confer high-penetrance studies from the United Kingdom, United States, and Denmark f susceptibility to epithelial ovarian cancer. The main protagonists ( 1,500 cases and2,500 controls). Genotype frequencies in BRCA1 BRCA2 cases andcontrols were comparedusing logistic regression. are and . These genes are responsible for about half In stage 2, eight other studies from Australia, Poland, and the of all families containing two or more ovarian cancer cases in first- f f degree relatives and most families containing several ovarian UnitedStates ( 2,000 cases and 3,200 controls) were z genotypedfor the five most significant SNPs from stage 1. cancer cases ( 3) or in which ovarian and breast cancers occur No SNP was significant in the stage 2 data alone. Using the together (1, 2). combinedstages 1 and2 dataset, CDKN2A rs3731257 and However, the known susceptibility genes explain <40% of the CDKN1B rs2066827 were associated with disease risk (unad- excess familial ovarian cancer risk (3). If other highly penetrant justed P trend= 0.008 and0.036, respectively), but these were ovarian cancer susceptibility genes exist, they are likely to be rare. One possible explanation for the residual risks is the existence of not significant after adjusting for multiple testing. Carrying several common but only moderately or low-penetrance susceptibility alleles in the population (4). Note: Supplementary data for this article are available at Cancer Research Online There have been several studies that have attempted to identify (http://cancerres.aacrjournals.org/). S.A. Gayther and H. Song contributed equally to this work. common variants in genes that may be associated with increased Requests for reprints: Simon Gayther, Translational Research Laboratories, Windeyer risk of ovarian cancer. Candidate genes have usually been selected Institute, University College London, 46 Cleveland Street, London W1T 4JF, United based on biological plausibility. These include genes in pathways Kingdom. Phone: 44-20-7679-9204; Fax: 44-20-7679-9687; E-mail: [email protected]. I2007 American Association for Cancer Research. controlling steroid hormone metabolism (e.g., progesterone recep- doi:10.1158/0008-5472.CAN-06-3261 tor, androgen receptor, CYP17, prohibitin; refs. 5–14), DNA repair www.aacrjournals.org 3027 Cancer Res 2007; 67: (7). April 1, 2007

Table 1A. Characteristics of the 11 ovarian cancer case control studies (listed alphabetically) included in the Ovarian Cancer Association Consortium pooled study

Study name Location Total Total Age Age Reference DNA source Days from Number number number range range diagnosis to of SNPs c of cases* of controls* (cases) (controls) blood draw genotyped

AOCS and ACS Multiregional, 802 (71) 933 (42) 23–80 19–81 9 Blood 0–1,663(103) 5 (ovarian Australia lymphocytes cancer) FROC (Family Stanford, CA 327 (39) 429 (59) 23–64 19–66 16, 17, 21, 23 Blood 55–1,477 (281) 88 Registry for lymphocytes; Ovarian mouthwash Cancer) Study Hawaii Ovarian Hawaii 313 (241) 574 (426) 18–87 19–88 33 Blood 16–2,026 (375) 5 Cancer Study lymphocytes; mouthwash HOPE Pittsburgh 57 (1) 152 (8) 34–80 44–79 – Blood 0–280 (149) 5 (hormones lymphocytes; and ovarian mouthwash cancer prediction) study LAC-CCOC Los Angeles, CA 659 (211) 819 (198) 18–84 21–78 8 Blood 46–1,340 (252) 5 lymphocytes MALOVA Copenhagen, 446 1221 (0) 32–80 35–79 16, 17, 21, 23 Blood All preoperative 88 (Malignant Denmark lymphocytes Ovarian Cancer study) Mayo Clinic Minnesota 317 (13) 462 (35) 28–91 23–90 32 Blood 0–365 (31) 5 Rochester, lymphocytes ovarian cancer case- control study NCOCS North Carolina 610 (91) 843(156) 22–74 20–75 9 Blood 29–1,259 (187) 5 lymphocytes POCS Warsaw and 264 625 (0) 24–74 24–74 – Blood 0–525 (82) 5 Lodz lymphocytes; mouthwash SEARCH Cambridge, 731 (42) 855 (4) 21–74 39–77 16, 17, 21, 23 Blood 140–4,850 (1,231) 88 ovarian United lymphocytes cancer Kingdom

*Numbers of subjects that are non-white are given in parentheses. cThe mean is given in parentheses.

(e.g., BRCA1, BRCA2, RAD51, MSH2; refs. 15–17) and in putative the mechanisms that regulate cell division. In mammalian cells, cell oncogenes and tumor suppressor genes (TP53, RB1, HRAS1, STK15; division is controlled by the activity of cyclin-dependent kinases refs. 18–23). Thus far, reported positive associations include (CDK) and their essential activating coenzymes, the cyclins and an increased ovarian cancer risk for two PGR haplotypes (8); a CDK inhibitors (reviewed in refs. 24, 25). CDK activity is regulated protective effect for the PGR promoter +331A allele in endome- on several levels, including cyclin synthesis and degradation, trioid ovarian tumors (9); an increased risk of borderline ovarian phosphorylation, and dephosphorylation. The interaction between cancer associated with the Pro72Arg polymorphism in the TP53 CDKs and cyclins is tightly controlled to ensure an ordered gene (20); and an increased risk associated with the polymorphism progression through the cell cycle from G1 to S to G2. The events in the STK15 oncogene (23). However, none of these associations occurring before DNA synthesis are probably the most thoroughly are conclusive due to the small size of the initial studies, and there understood. The key event in cell cycle regulation is transversion are no reports indicating that they have been validated in of the restriction (R) point late in G1, which is crucial to the cell’s additional populations. destiny toward division, differentiation, senescence, or apoptosis. It There are no published reports describing the association is believed that once the R point has been overcome, cell cycle between variants in cell-cycle control genes and ovarian cancer progression occurs almost automatically. In the event of cellular susceptibility. In general, cancer is associated with a breakdown in stress, there are several proteins that can inhibit the cell cycle in G1;

Cancer Res 2007; 67: (7). April 1, 2007 3028 www.aacrjournals.org

Table 1B. Additional characteristics of the 11 ovarian cancer case-control studies included in the Ovarian Cancer Association Consortium pooled study

c Study Case ascertainment Control ascertainment* Participation rates

AOCS and ACS Diagnosed from 2002 onward; recruited Randomly selected from Commonwealth Cases: 84%; Australia through surgical treatment centers electoral roll. Frequency matched to controls: 47% throughout Australia and cancer registries cases for geographic region of Queensland South Australia and West Australia (AOCS) and cancer registries of New South Wales and Victoria (ACS) FROC, USA Consecutive cases diagnosed from 1997–2002 Random-digit dial identification from Cases: 75%; in Greater Bay Area Cancer Registry study area controls: 75% San Francisco Hawaii, USA Rapid case ascertainment through Hawaii Randomly selected from Hawaii Department Cases: 58.1%; Tumor Registry from 1993onward of Health Annual Survey of the controls: 62.8% representative households HOPE, USA Variable source including physician offices Identified in same regions. All participants Cases: 69%; cancer registries and pathology databases undergo home interviews controls: 81% from counties of western Pennsylvania, eastern Ohio, and western New York LAC-CCOC, USA Rapid case ascertainment through Los Neighborhood-recruited controls Cases: 73%; Angeles Cancer Surveillance program controls: 73% from 1993onward MALOVA, Incident cases (35–79 yrs) diagnosed Random sample of general female Cases: 79%; Denmark 1994–1999 from municipalities of population (35–79 yrs) in study controls: 67% Copenhagen and Frederiksberg and area selected using computerized surrounding counties Central Population Register Mayo, USA Cases attending Mayo Clinic diagnosed Identified through Mayo Clinic; healthy Cases: 84%; from 2000 onward identified in a women seeking general medical controls: 65% six-state surrounding region examination NCOCS, USA Cases from 1999 onward identified from Controls identified from the same Cases: 70%; 48 counties within the region by region as cases controls: 63% rapid-case ascertainment POCS, Poland Cases collected from cities of Warsaw Identified at random through the Polish Cases: 71%; and Lodz 2001–2003by rapid electronic system; stratified by city controls: 67% ascertainment at participating hospitals and 5-yr age categories SEARCH, UK Cases <70 yrs from East Anglian West Selected from the EPIC-Norfolk cohort of 25,000 Cases: 67%; Midlands and Trent regions of England; individuals aged 45–74 based in the same controls: 84% prevalent cases diagnosed 1991–1998; geographic regions as the cases incident cases diagnosed 1998 onward

*All controls were frequency matched to cases for age and ethnicity. cParticipation rates are the percentage of those taking part as proportion of those invited to take part.

for example, in response to DNA damage, P53accumulates in expression, and hypermethylation in ovarian cancer cell lines the cell and induces P21/CDKN1A–mediated inhibition of cyclin and primary tumors (29, 30); and in some primary ovarian D/CDK. cancers, homozygous deletions of P16INK4a involving the neigh- INK4b Somatic alterations of genes involved in the G1 phase of the boring P15 (CDKN2B) gene are associated with a poor prog- cell cycle, including the cyclins, CDKs, and CDK inhibitors, are nosis for ovarian cancer cases (31). Together, these data suggest common events in neoplastic development for multiple tumor an important role for CDKs, cyclins, and CDK inhibitors in ovarian types; but different cell cycle proteins seem to be targets for cancer development. different cancers. A reason for this could be that although the The aim of this study was to evaluate the association between basic mechanisms of proliferative control are identical in common variants in 13genes coding for cyclins, CDKs, and CDK different tissues, the maintenance of normal cell-cycle control inhibitors and susceptibility to epithelial ovarian cancer. We used a is also partly regulated in a tissue-specific manner. For ovarian two-stage case-control study design in which single nucleotide cancer, overexpression of cyclin D1 (CCND1) has been reported polymorphisms (SNP) that tag all known common variants in these in most borderline and invasive ovarian cancers (26, 27); cyclin genes were first genotyped in three studies. The most significant D2 was found to be overexpressed in about 40% of ovarian hits were then genotyped in eight additional ovarian cancer case- tumors compared with normal ovarian tissues (28); P16INK4a control studies from the international Ovarian Cancer Association (CDKN2A) is relatively frequently lost due to deletion, loss of Consortium (OCAC). www.aacrjournals.org 3029 Cancer Res 2007; 67: (7). April 1, 2007

Table 2. Details of tSNP selection for cell cycle control genes

Gene Size Data % of No. No. tSNPs Additional Mean r2 Number r2 for LD (kb) gene common SNPs with poorly blocks resequenced variants Successfully Additional genotyped* r2 > 0.8 tagged by EGP genotyped assays failed SNPs

c CCND1 16.0 EGP 100 12 5 2 2 0.82 10 0.08–0.24 1 CCND2 34.0 HapMap 93 24 14 3 0 0.89 21 0.18–0.74 2 CCND3 9.0 EGP 88 137 1 0 0.92 12 0.09 1 CCNE1 13.5 EGP 84 26 4 11 0 0.57 6 0.10–0.73 1 CDK2 7.6 EGP 98 5 2 30 0.54 2 0.19–0.3 0 1 CDK4 6.7 EGP 96 5 2 1 0 0.86 4 0.29 1 CDK6 220.8 HapMap 18 41 130 0 0.96 41 N/A 3 b CDKN1A 10.9 EGP 85 27 10 7 1 0.85 20 0.07–0.69 3 CDKN1B 5.7 HapMap 100 7 6 1 1 0.90 6 0.231 CDKN2A 28.8 EGP 81 12 6 2 30.97 12 N/A 1 CDKN2B 13.3 EGP 94 20 8 0 0 0.96 20 N/A 1 CDKN2C 8.9 EGP/HapMap 87 32 0 0 1.00 3N/A 1 CDKN2D 4.1 EGP 634 2 0 0 1.00 4 N/A 1

Abbreviation: N/A, not applicable. *Additional SNPs that were genotyped that were not common in the data set used for tagging. These SNPs do not contribute to the estimates of tagging efficiency. cOne common SNP in EGP (environmental genome project; rs3212867) was monomorphic in the CEU HapMap population. bAssay for rs3176359 was too rare in our populations (MAF = 0.002).

Materials and Methods genetic diversity between these groups, and they cannot be excluded with any certainty. Study subjects. In total, 11 different ovarian cancer case-control studies When resequencing data were not available, we used data from the (summarized in Table 1A and 1B) contributed data to this study, three for International HapMap Project (release 34/5 21-06-2005: last public release stage 1 and eight for stage 2 (see also refs. 8–10, 16, 17, 21, 23, 32, 33 for used in this study) to select tSNPs. The HapMap Project has genotyped a further details). Most subjects were white Europeans in origin. Ten studies large number of SNPs in several populations, including 30 parent-offspring used population-based ascertainment for cases and controls, and one study trios collected in 1980 from U.S. residents with northern and western was clinic based. All studies have received ethical committee approval, and European ancestry by the Centre d’Etude du Polymorphisme Humain all study subjects provided informed consent. (CEPH). Tag SNP selection. A set of tagging SNPs (tSNP) was identified for 13 The best measure of the extent to which one SNP tags another SNP is CCND1, CCND2, CCND3, CCNE1 genes comprising four cyclins ( ); three CDKs the pairwise correlation coefficient (r2) because the loss in power incurred CDK2, CDK4, CDK6 CDKN1A, CDKN1B, CDKN2A, p ( ); six CDK inhibitors ( by using a marker SNP in place of a true causal SNP is directly related to CDKN2B, CDKN2C, CDKN2D). The principal hypothesis underlying the study this measure. We aimed to define a set of tSNPs such that all known was that one or more common SNPs in these genes were associated with an r2 common SNPs had an estimated p > 0.8 with at least one tSNP. However, altered risk of ovarian cancer. We therefore identified a set of tagging SNPs some SNPs are poorly correlated with other single SNPs but may be that efficiently tagged all the known common variants [minor allele efficiently tagged by a haplotype defined by multiple SNPs, so-called frequency (MAF) z0.05] and were likely to tag most of the unknown ‘‘aggressive tagging,’’ thus reducing the number of tSNPs needed. As an common variants. alternative, therefore, we aimed for the correlation between each SNP and a The selection of tSNPs is most reliable where the gene has been 2 21 haplotype of tSNPs (rs) to be at >0.8. The program Tagger uses a strategy resequenced in a sample of individuals sufficiently large to identify all that combines the simplicity of pairwise methods with the potential common variants. The National Institute of Environmental Health efficiency of multimarker approaches (3). It begins by selecting a minimal Sciences (NIEHS) Environmental Genome Project (EGP) is currently set of markers such that all alleles to be captured are correlated at an resequencing candidate genes for cancer across a panel of 90/95 r2 z 0.8 with a marker in that set. Certain markers can be forced into the individuals representative of U.S. ethnicities. The original panel (P1- p tag list or explicitly prohibited from being chosen as tags. After this, it tries PDR90) of 90 individuals consists of 24 European Americans, 24 African to ‘‘peel back’’ the tag list by replacing certain tags with multimarker tests. Americans, 12 Mexican Americans, 6 Native Americans, and 24 Asian Tagger avoids overfitting by only constructing multimarker tests from SNPs, Americans, but the ethnic group identifiers are not available. Because it is which are in strong linkage disequilibrium with each other, as measured by known that there is greater genetic and haplotype diversity in individuals a pairwise LD score. If an assay for one of the chosen tSNPs could not be of African origin, we have identified and excluded 28 of the samples with designed, the SNP tagging process was repeated with that SNP excluded the greatest African ancestry in this population by comparing the from the possible set of tSNPs. Table 2 shows details of the tSNP selection genotypes of the resequenced sample with genotypes for the same SNPs for each of the genes analyzed in this study. from the National Heart, Lung, and Blood Institute Variation Discovery Resource Project African American panel (http://pga.gs.washington.edu/ finished_genes.html). Data from the remaining 62 individuals were used to identify tSNPs. Ideally, samples from Native American, Hispanic American, and Asian American individuals should also be removed; but there is less 21 http://www.broad.mit.edu/mpg/tagger/

Cancer Res 2007; 67: (7). April 1, 2007 3030 www.aacrjournals.org

Genotyping. Genotyping was carried out for 9 of the 11 studies using assessed by m2 tests [1 degree of freedom (d.f.)]. We treated the design as TaqMan (Applied Biosystems, Foster City, CA) according to the manufac- a single, two-stage multicenter study rather than hypothesis-generating turer’s instructions. Primers and probes were Assays-by-Design (Applied and replication studies because joint analysis has been shown to be more Biosystems). Following PCR amplification, end reactions were read on the efficient than replication analysis in a staged design (34). The primary tests ABI Prism 7900 using Sequence Detection Software (Applied Biosystems). of association were the univariate analyses between each tSNP and ovarian Two studies [Australian Ovarian Cancer Study (AOCS) and Australian cancer. Genotype frequencies in cases and controls were compared for each Cancer Study (ACS); Table 1A and B] used iPlex technology (Sequenom) to study separately using m2 tests for heterogeneity (2 d.f.). The data were then perform genotyping. Failed genotypes were not repeated. Plates with pooled, and a genetic model free test was carried out by comparing genotype genotype calling <95% were discarded. If there was genotyping discordance frequencies in cases and controls using unconditional logistic regression between duplicates within an experiment, the data were discarded and with terms for genotype and study and an appropriate likelihood ratio test either repeated or the assay replaced with a different assay with similar (Phet, 2 d.f.). We tested for heterogeneity between studies by comparing tagging properties. logistic regression models with and without a genotype-study interaction Each group carried out genotyping on their own samples. Therefore, we term using likelihood ratio tests. Genotypic specific risks with the common tested the quality of genotyping between laboratories for the five tSNPs homozygote as the baseline comparator were estimated as odds ratios analyzed by all groups using the HAPMAPPT01 panel of CEPH-Utah trios- with associated 95% confidence limits by unconditional logistic regression. standard plate supplied by Coriell.22 This panel includes 90 different DNA We also tested for rare allele dose effect (assuming a multiplicative genetic samples, five duplicate samples, and a negative template control in a 96-well model) using m2 tests for each study separately and unconditional logistic P plate format. We compared genotype call rates and concordance between regression for the pooled data ( het, 1 d.f.). Primary analyses were restricted studies. Call rates on these plates ranged from 96% to 99%. Overall to study subjects of white European origin. Secondary analyses included all concordance on these plates was >99%. study subjects with additional adjustment for ethnic group. Statistical methods. Deviations of genotype frequencies in the controls In addition to the univariate analyses, we carried out specific haplotype from those expected under Hardy-Weinberg equilibrium (HWE) were tests for those combinations of alleles that tagged specific SNPs. We also

Table 3. Genotype-specific risks for tSNPs with P < 0.2: combined analysis of 88 SNPs using SEARCH-MALOVA-Stanford studies

c b x Gene (no. tSNPs)* SNP IDdbSNP MAF HetOR (95% CI) HomOR (95% CI) P2df P trend

CCND1 (7) ccnd107 rs602652 0.46 1.14 (0.98–1.33) 1.24 (1.03–1.50) 0.057 0.018 ccnd103rs603965 0.44 1.06 (0.91–1.24) 1.29 (1.07–1.55) 0.021 0.010 ccnd108 rs3212879 0.49 0.85 (0.73–0.99) 0.82 (0.68–0.98) 0.057 0.028 ccnd110 rs3212891 0.46 0.86 (0.74–1.00) 0.83 (0.69–0.997) 0.076 0.034 ccnd101 rs7178 0.07 1.24 (1.03–1.49) 1.22 (0.52–2.88) 0.065 0.021 CCND2 (14) ccnd202 rs3217805 0.40 1.07 (0.93–1.23) 0.86 (0.70–1.05) 0.078 0.33 ccnd215 rs3217916 0.28 1.00 (0.88–1.15) 0.76 (0.59–0.99) 0.11 0.17 ccnd216 rs3217925 0.26 0.98 (0.85–1.12) 0.74 (0.55–0.99) 0.130.13 ccnd207 rs3217936 0.33 1.01 (0.88–1.16) 0.77 (0.61–0.98) 0.069 0.15 CCND3 (7) ccnd303 rs1051130 0.48 1.13 (0.97–1.32) 0.88 (0.73–1.06) 0.0090 0.23 ccnd306 rs9529 0.29 1.00 (0.87–1.15) 0.79 (0.62–1.02) 0.14 0.18 ccnd307 rs3218110 0.24 1.11 (0.97–1.28) 1.07 (0.80–1.42) 0.32 0.19 CCNE1 (4) ccne102 rs997669 0.38 1.10 (0.95–1.27) 1.14 (0.94–1.40) 0.31 0.13 ccne112 rs3218036 0.31 1.07 (0.93–1.23) 1.26 (1.00–1.57) 0.12 0.052 CDK2 (2) cdk202 rs2069408 0.34 0.90 (0.78–1.03) 0.90 (0.72–1.12) 0.27 0.16 cdk204 rs1045435 0.09 1.10 (0.93–1.31) 1.97 (0.89–4.34) 0.14 0.093 CDK6 (13) cdk612 rs2285332 0.25 0.92 (0.80–1.06) 0.86 (0.65–1.14) 0.38 0.16 cdk611 rs3731348 0.06 1.07 (0.86–1.34) 0.28 (0.07–1.21) 0.19 0.93 cdk613 rs2237570 0.12 0.89 (0.76–1.05) 0.70 (0.37–1.32) 0.23 0.094 cdk610 rs2282991 0.10 0.84 (0.71–1.00) 0.85 (0.43–1.67) 0.14 0.054 cdk605 rs8 0.21 1.18 (1.02–1.35) 1.41 (1.01–1.96) 0.017 0.0042 CDKN1A (11) cdkn1a06 rs2395655 0.29 1.12 (0.97–1.29) 1.16 (0.96–1.41) 0.19 0.082 cdkn1a01 rs1801270 0.08 0.91 (0.76–1.11) 0.52 (0.18–1.50) 0.32 0.20 CDKN1B (7) cdkn1b05 rs3759217 0.12 1.13 (0.97–1.33) 1.03 (0.61–1.74) 0.30 0.17 cdkn1b02 rs2066827 0.26 0.88 (0.77–1.01) 0.68 (0.51–0.90) 0.0098 0.0035 CDKN2A/2B (17) cdkn2a04 rs2811712 0.10 0.98 (0.83–1.16) 1.91 (1.07–3.42) 0.086 0.42 cdkn2a10 rs11515 0.14 1.0 (0.86–1.16) 1.55 (0.97–2.47) 0.19 0.38 cdkn2a13 rs3731257 0.26 0.90 (0.78–1.03) 0.80 (0.59–1.07) 0.14 0.046

*We also analyzed tSNPs in the genes CDK4 (two SNPs); CDKN2C (two SNPs); CDKN2D (two SNPs), but none of these reached the required level of statistical significance. Data from the combined analysis of all SNPs are given in Supplementary Tables S1 and S2. Confidence intervals that do not reach or cross 1.00 and P values <0.05 are in bold type. cMinor allele frequency. bHeterozygous odds ratios (95% confidence intervals). x Homozygous odds ratios (95% confidence intervals).

www.aacrjournals.org 3031 Cancer Res 2007; 67: (7). April 1, 2007

Table 4. Age-adjusted ovarian cancer risks associated with five tSNPs in cell cycle control genes for white cases and white controls only

Gene dbSNP MAF No. cases No. controls HetOR (95% CI) HomOR (95% CI) P2df Ptrend

A. Combined analysis of all stage 2 studies CCND1 rs7178 0.08 2,194 3,294 1.01 (0.86–1.18) 1.01 (0.50–2.06) 0.99 0.91 CCND1 rs603965 0.45 1,874 2,833 1.00 (0.87–1.15) 1.01 (0.86–1.20) 0.98 0.90 CDK6 rs8 0.21 2,188 3,298 1.02 (0.91–1.15) 0.91 (0.69–1.19) 0.69 0.83 CDKN1B rs2066827 0.23 2,202 3,297 1.05 (0.93–1.18) 0.88 (0.69–1.12) 0.36 0.82 CDKN2A/2B rs3731257 0.27 2,190 3,290 0.92 (0.82–1.04) 0.89 (0.72–1.10) 0.29 0.13

B. Combined analysis for all studies, white subjects only CCND1 rs7178 0.08 3,607 5,725 1.09 (0.96–1.22) 1.12 (0.64–1.97) 0.37 0.12 CCND1 rs603965 0.45 3,285 5,236 1.02 (0.92–1.13) 1.13 (0.99–1.28) 0.15 0.084 CDK6 rs8 0.21 3,597 5,720 1.09 (1.00–1.19) 1.06 (0.86–1.31) 0.17 0.082 CDKN1B rs2066827 0.25 3,618 5,719 0.98 (0.89–1.07) 0.79 (0.65–0.95) 0.044 0.036 CDKN2A/2B rs3731257 0.27 3,601 5,705 0.89 (0.81–0.97) 0.87 (0.73–1.03) 0.021 0.0080

2 carried out a general comparison of common haplotype frequencies in each rp > 0.8. For two genes (CDK2 and CCNE1), the best tagging we gene haplotype block using the data from all the tSNPs in that block. r2 were able to achieve was p 0.54 and 0.57, respectively. These data Haplotype blocks were defined such that the common haplotypes (>5% are summarized in Table 2. frequency) accounted for at least 80% of the haplotype diversity. We considered haplotypes with >2% frequency in at least one study to be Genotyping in Ovarian Cancer Cases andControls common. Rare haplotypes were pooled. For both specific haplotype marker Genotyping was done in two stages. In stage 1, the 88 tSNPs tests and the general comparison of haplotype frequencies by haplotype described above were genotyped in three case-control populations block, haplotype frequencies and subject-specific expected haplotype from the United Kingdom (SEARCH), Denmark [Malignant Ovarian indicators were calculated separately for each study using the program Cancer (MALOVA)], and the United States (Stanford). Combined, TagSNPs. This implements an expectation substitution approach to account f for haplotype uncertainty given unphased genotype data (35). Subjects these studies comprise 1,500 invasive ovarian cancer cases and missing >50% genotype data in each block were excluded from haplotype 2,500 controls. In stage 2, we genotyped five SNPs from four genes analysis. We used unconditional logistic regression to test the null that showed the strongest evidence of association in stage 1, in hypothesis of no association between specific tSNP haplotype frequency f3,000 additional cases and 4,400 controls from another eight and ovarian cancer stratified by study by comparing a model with terms for case-control studies. subject-specific haplotype indicator and study with a model including study Stage 1. Nineteen out of 264 genotype distributions in controls term only. The global null hypothesis of no association between haplotype deviated from HWE (P < 0.05). In the majority of cases, deviations frequency (by haplotype block) and ovarian cancer was tested by comparing from HWE occurred in only one of the studies. The discrimination a model with multiplicative effects for each common haplotype (treating of genotypes for these assays was good. For rs3176319 (CDKN1A), the most common haplotype as the reference) to the intercept-only model. the deviations from HWE were substantial in all three studies Haplotype-specific odds ratios were also estimated with their associated P 7 confidence intervals. ( <10 ), and so this SNP was excluded from further analyses. rs3176359 in CDKN1A was also excluded from further analyses because it was found to be very rare in all three populations Results (MAF = 0.002). P V Identifying Tagging SNPs in Cell Cycle Control Genes We identified 28 SNPs with a 0.2, of which 9 SNPs had a P V 0.05 in tests for association. These data are summarized in Using a combination of NIEHS resequencing data and data from Table 3. The complete data for all SNPs are given in Supplementary the international HapMap project, we identified a total of 199 Tables S1 and S2. There was no evidence for association of geno- common SNPs (MAF z0.05) in the 13genes. One-hundred and type with age in controls, and as expected, age-adjusted risks were sixty-seven SNPs were tagged with r2 > 0.8, of which 10 were tagged similar to unadjusted risks (data not shown). The haplotype fre- by 5 different multimarker haplotypes. The remaining SNPs were quencies and associated risks for the five multimarker haplotype tagged with r2 between 0.07 and 0.74. These could not be tagged tags are presented in Supplementary Table S3. There was no evi- more efficiently because assays could not be designed for them. dence for association at a significance level P V 0.05 for SNPs in 6 of Eighty-one tSNPs were selected for this series. We also selected a the 13genes ( CDK2, CDK4, CCND2, CDKN1A, CDKN2C, CDKN2D). further seven SNPs that were not found in the EGP or HapMap data Evidence of association at the P = 0.05 level was found for the sets used for tSNP selection. These SNPs do not contribute to the following: five of seven SNPs in CCND1 (strongest association estimates of tagging efficiency. Thus, we identified 88 SNPs from 13 P trend, 0.01); one of seven SNPs in CCND3 (P = 0.0 09); 1 of 13 genes for genotyping. Eleven genes were tagged with an average 2df SNPs in CDK6 (Ptrend = 0.004); one of eight SNPs in CDKN1B (P2df = 0.01); and 1 of 17 SNPs in CDKN2A/2B (P2df = 0.046). Further details are provided in Table 3. The global test of association with common haplotypes was not 22 http://locus.umdnj.edu/nigms/nigms_cgi/panel.cgi?id=2&query=HAPMAP01 significant for any of the 13genes (data not shown). One haplotype

Cancer Res 2007; 67: (7). April 1, 2007 3032 www.aacrjournals.org

Downloaded from cancerres.aacrjournals.org on September 28, 2021. © 2007 American Association for Cancer Research. SNPs in Cell Cycle Control Genes and Ovarian Cancer Risk that showed a significant frequency difference between cases and Consortium, but none was significant in the second stage. However, controls was observed in CDK6 block 3(haplotype frequency, 0.2 in the power to detect an effect in the replication studies was just 70% controls). This haplotype, comprising the rare allele of rs8 and to detect an allele with frequency and effect size similar to CDKN1B common alleles of rs479049, rs445, and rs992519, showed increased rs2066827 (the most significant association from stage 1) at a type I risk for ovarian cancer compared with the most common haplotype in this block (OR, 1.19 [1.06–1.35]; P = 0.004). Stage 2. Five SNPs were genotyped in the stage 2 studies. These were rs7178 and rs603965 (both in CCND1), rs8 (CDK6), rs2066827 (CDKN1B), and rs3731257 (CDKN2A/2B). The evidence for association for these SNPs from the stage 1 analyses (adjusted for age and ethnicity) ranged from P trend 0.003to 0.046. There was some evidence for between-study heterogeneity for rs8 (P = 0.036). None of the SNPs were significantly associated with ovarian cancer when considering the stage 2 studies as a replication set (Table 4A); rs3731257 was the most significant (P trend = 0.13). Combinedanalysis. As joint analysis with appropriate adjustment for multiple testing has been shown to be more powerful than replication analysis (34), we also carried out a combined analysis of all the data. Figure 1 shows the genotype-specific risks (estimated as odds ratios) for each study and for the combined data for the analyses restricted to white subjects. Only two of the five SNPs were significant at the 0.05 level (Table 4B). These were rs2066827 (CDKN1B) and rs3731257 (CDKN2A-2B; Ptrend = 0.036 and 0.008, respectively). The rare alleles for both SNPs were associated with reduced ovarian cancer risks [rs2066827 HetOR, 0.98 (0.89– 1.07); HomOR, 0.79 (0.65–0.95); and rs3731257 HetOR, 0.89 (0.81– 0.97); HomOR, 0.87 (0.73–1.03)]. There was little change in the association of rs2066827 when the data for all ethnic groups were included (Ptrend = 0.006); but for rs3731257, the association was almost completely attenuated (P = 0.16). Standard methods for correcting for multiple testing, such as the Bonferroni correction, are too conservative when hypotheses are correlated, as is the case here. We have therefore used a permutation procedure modified to account for the staged design (36). A P value smaller than the most significant result was observed in 368 out of 1,000 permutations, which is equivalent to an adjusted Ptrend of 0.37. Statistical power to identify subgroup effects from these studies is limited; but there were sufficient data to carry out an analysis with cases restricted to serous and nonserous histopathologic subtypes. In total, there were f2,000 cases of serous histology in the 11 studies combined and 1,600 nonserous cases. The analysis based on this histologic subgroup stratification suggested marginal associations for rs3731257 with both serous (Ptrend = 0.030) and P P nonserous cancers ( trend = 0.042) and for rs7178 ( trend = 0.034), rs8 (P2df = 0.046), and rs2066827 (Ptrend = 0.023) with nonserous cancers only (Table 5).

Discussion We have assessed the effect of tSNPs in 13cell cycle control genes on the risks of invasive epithelial ovarian cancer. We chose to analyze genes involved in the control of cell division (cyclins, CDK, Figure 1. Genotype-specific risks by study. Heterozygous and homozygous and CDK inhibitors) because several of these genes have previously genotype-specific odds ratios are illustrated for White subjects only for all five been implicated in the somatic development of multiple tumor SNPs analyzed in all studies. Lines, 95% confidence intervals; size of the box, proportionate size of the study (number of white subjects genotyped) compared types. To our knowledge, this is the first report describing the with other studies. There were no homozygous odds ratios for some studies association between these genes and ovarian cancer susceptibility. because the minor allele was too rare in the population. Key to studies in Table 1A and B: ACS, ACS Australia; AOC, AOCS Australia; HAW, Hawaii A two-stage design was used for this study. In the first stage, 88 Ovarian Cancer Study Hawaii; HOP, HOPE study, Pittsburgh, PA; MAY, Mayo tSNPs were genotyped in three ovarian cancer case-control studies. Clinic, Minnesota; MAL, MALOVA study, Denmark; NCO, North Carolina Ovarian The best five SNPs that were marginally associated with epithelial Cancer Study (NCOCS), North Carolina; POL, Polish Ovarian Cancer Study (POCS), Poland; SEA, SEARCH (Studies of Epidemiology and Risk Factors ovarian cancer were then genotyped in a further eight case-control in Cancer Heredity) study, United Kingdom; USC, Los Angeles County studies from the international Ovarian Cancer Association Case-Control Studies of Ovarian Cancer (LAC-COCC), southern California. www.aacrjournals.org 3033 Cancer Res 2007; 67: (7). April 1, 2007

Table 5. The effects of histologic subtype on ovarian cancer risks associated with five tSNPs in cell cycle control genes

Histologic type Gene dbSNP MAF Number Number HetOR (95% CI) HomOR (95% CI) P2df Ptrend of cases of controls

Serous CCND1 rs7178 0.08 2,014 5,725 0.99 (0.87–1.12) 1.11 (0.95–1.29) 0.06 0.74 CCND1 rs603965 0.45 1,807 5,236 0.99 (0.87–1.12) 1.11 (0.95–1.29) 0.26 0.27 CDK6 rs8 0.21 2,009 5,720 1.04 (0.93–1.16) 1.14 (0.88–1.47) 0.53 0.28 CDKN1B rs2066827 0.25 2,022 5,719 1.02 (0.92–1.14) 0.82 (0.65–1.03) 0.18 0.30 CDKN2A/2B rs3731257 0.27 2,012 5,705 0.90 (0.81–1.01) 0.86 (0.69–1.06) 0.11 0.03 Nonserous CCND1 rs7178 0.08 1,5935,725 1.09 (0.93–1.28) 2.16 (1.18–3.95) 0.034 0.034 CCND1 rs603965 0.45 1,478 5,236 1.07 (0.93–1.22) 1.15 (0.97–1.36) 0.26 0.12 CDK6 rs8 0.21 1,588 5,720 1.16 (1.03–1.31) 1.00 (0.75–1.34) 0.046 0.063 CDKN1B rs2066827 0.25 1,596 5,719 0.93(0.82–1.05) 0.75 (0.58–0.97) 0.058 0.023 CDKN2A/2B rs3731257 0.27 1,589 5,705 0.87 (0.78–0.99) 0.88 (0.69–1.11) 0.073 0.042

NOTE: White invasive cases (serous/non serous) are compared with white controls.

error rate of 0.01. Because a combined analysis with adjustment for cies differ between population subgroups, and cases and controls multiple testing has been shown to have greater power than a are drawn differentially from those subgroups. However, this study replication analysis, we also carried out a joint analysis of the comprised 11 different case-control populations from England, stages 1 and 2 data. Two SNPs, CDKN1B-rs2066827 and CDKN2A/ Denmark, Poland, Australia, and several states throughout the 2B-rs3731257, showed weak evidence for association, but neither United States. It is unlikely that population stratification is was highly statistically significant (naı¨ve Ptrend = 0.036 and 0.008, important because any population stratification will be study respectively), but these were not significant after adjustment for specific. Furthermore, the combined study was of sufficient size to multiple testing using a permutation procedure that allows for the allow the analyses to be restricted to white-only cases and controls, correlation between the SNPs. As an alternative to correcting for which represented the vast majority of all subjects. multiple testing, some authors have suggested that simple but Disease heterogeneity could also lead to false-positive reporting stringent criteria should be applied to statistical tests for genetic or mask the presence of true associations. In this study, we association; e.g., P <104 for candidate gene studies or even stratified cases according to histologic subtype and did the analysis P <10 7 for genomewide significance (37). Using these criteria, it for the combined data after dividing cases into serous and can be estimated that 26,448 cases and an equal number of nonserous subgroups. Inevitably, this leads to a loss in statistical controls would be needed to confirm the effect for rs2066827 to power, and although some marginal associations were found, these detect a codominant allele with a minor allele frequency of data should be treated with caution. Cases were collected from 0.25 that confers a risk of 1.08 with 90% power at a type I error rate multiples of different centers, and there are likely to be variations of 10 4. between studies in the completeness of these data and in the Because the selection of SNPs in this study was based on a reporting of different histopathologic subtypes. tagging approach rather than on putative function, we are unable There was no evidence of association with disease risk for to comment in great detail on the possible functional significance polymorphisms in CCND2, CCNE1, CDK2, CDK4, CDKN1A, of these findings. The rs2066827 variant in CDKN1B is potentially CDKN2C, and CDKN2D at the 0.05 level. The SNPs tested in this functional. It encodes a nonsynonymous amino acid change study were selected to tag all the common variants in each gene Val109Gly in CDKN1B and is situated in the interaction surface of and not because of their putative functional effects. However, we CDKN1B and its negative regulator p38jab1, in the region spanning were unable to design assays for 31 of the selected tSNPs, and 35 of 2 amino acid residues 97 to 151. This SNP could therefore alter the the 199 common SNPs were tagged with rp < 0.8. Furthermore, interaction between CDKN1B and p38jab1. The Val109Gly variant has although tSNPs based on HapMap and EGP data are likely to tag previously been reported in association with advanced prostate most of the common SNPs, there is a possibility that other cancer risk (38). unknown common variants were poorly tagged. The rs3731257 SNP in CDKN2A/2B is of no obvious functional There have been many candidate SNP/gene association studies significance, but it could be in linkage disequilibrium with another for ovarian cancer published over the past few years. Few of the functional SNP within the gene. CDKN2A is a G1 CDK inhibitor that published studies report results that are statistically significant; but binds to CDK4 and CDK6 and prevents their association with most of these studies have had insufficient statistical power to D-type cyclins, thereby facilitating CDK4/6–cyclin D–mediated detect moderate risks even for common genetic variants. phosphorylation and inactivation of retinoblastoma protein and Furthermore, very few studies have used comprehensive tagging entry into S-phase. CDKN2B inhibits MDM2-mediated degradation approaches to capture all the common variation in a gene. Where of P53; thus, its loss would lead to a reduction in levels of the P53 associations have been found, the existence of a susceptibility allele protein (39, 40). The known function of both proteins suggests a remains unproven because results await confirmation, or there are plausible role in tumor development. conflicting results in follow-up studies. The current study illustrates An alternative explanation for a false positive association is the value of large consortia to follow-up putative positive hidden population stratification. This occurs when allele frequen- associations for common polymorphisms in complex diseases to

Cancer Res 2007; 67: (7). April 1, 2007 3034 www.aacrjournals.org

Downloaded from cancerres.aacrjournals.org on September 28, 2021. © 2007 American Association for Cancer Research. SNPs in Cell Cycle Control Genes and Ovarian Cancer Risk clarify their risks. In the future, a consortium approach to genetic 0729, the Cancer Council Tasmania and Cancer Foundation of Western Australia (AOCS study); The National Health and Medical Research Council of Australia (199600; epidemiology studies might also enable the analysis of rarer genetic ACS study); National Cancer Institute grants K07-CA80668 and R01CA095023; and variants and gene-gene/gene-environment interactions where Department of Defence grant DAMD17-02-1-0669 [the Hormones and Ovarian Cancer individual studies have inadequate power. Prediction (HOPE) project]; the California Cancer Research Program grants 00- 01389V-20170 and 2110200; U.S. Public Health Service grants CA14089, CA17054, CA61132, CA63464, N01-PC-67010, and R03-CA113148, and the California Department Acknowledgments of Health Services subcontract 050-E8709 as part of its statewide cancer reporting program (University of Southern California). Received 9/1/2006; revised 10/27/2006; accepted 2/1/2007. The costs of publication of this article were defrayed in part by the payment of page Grant support: Funded by a grant from WellBeing (H. Song); B.A.J. Ponder is a charges. This article must therefore be hereby marked advertisement in accordance Gibb Fellow; P.D.P. Pharoah is a Senior Clinical Research Fellow; S.A. Gayther is a with 18 U.S.C. Section 1734 solely to indicate this fact. Higher Education Funding Council for England–funded senior lecturer; the Mermaid We thank Joan MacIntosh, Hannah Munday, Barbara Perkins, Clare Jordan, Kristy component of the Eve Appeal (S.J. Ramus); D.F. Easton is a Principal Research Fellow Driver, Mitul Shah, the local general practices and nurses and the East Anglian Cancer of Cancer Research United Kingdom. E.L. Goode is a Fraternal Order of Eagles Fellow. Registry for recruitment of the United Kingdom cases: the EPIC-Norfolk investigators G. Chenevix-Trench and D. Whiteman are supported by Fellowships from the National for recruitment of the United Kingdom controls; Craig Luccarini and Don Conroy Health and Medical Research Council; P. Webb is supported by a Fellowship from the for expert technical assistance. Dr. Stephen Chancok of the Core Genotyping Facility, Queensland Cancer Fund. Division of Cancer Epidemiology and Genetics of the National Cancer Institute Stage 1 genotyping was funded by a grant from Cancer Research United Kingdom. (United States), Dr. Neonila Szeszenia-Dabrowska of the Nofer Institute of Stage 2 genotyping and the Ovarian Cancer Association Consortium are funded by the Occupational Medicine (Lodz, Poland), and Dr. Witold Zatonski of the M. Ovarian Cancer Research Fund, thanks to generous donations from the family and Sklodowska-Curie Institute of Oncology and Cancer Center (Warsaw, Poland) friends of Kathryn Sladek Smith. for their contribution to the Polish Breast Cancer Study; Debby Bass, Shari Hutchison, Additional support was provided by the Roswell Park Alliance and the National Carlynn Jackson, Jessica Kopsic, and Mary Hartley for the HOPE project; Karin Cancer Institute (CA71766 and core grant CA16056; Stanford study); the Mayo Goodman and Renee Weatherly for subject recruitment; and Robert Vierkant, Zachary Foundation (Mayo study); Intramural Research Funds of the National Cancer Institute, Fredericksen, and Ludi Fan for data management in the Mayo Clinic study. Department of Health and Human Services, United States (Polish Ovarian Cancer Finally, we express our profound thanks to all the study participants who Study); U.S. Army Medical Research and Materiel Command under DAMD17-01-1- contributed to this research.

References 13. Spurdle AB, Chen X, Abbazadegan M, et al. CYP17 epithelial ovarian cancer. Int J Gynecol Cancer 2005;15: promotor polymorphism and ovarian cancer risk. Int J 878–83. 1. Gayther SA, Russell P, Harrington P, et al. The Cancer 2000;86:436–9. 29. Milde-Langosch K, Ocon E, Becker G, Loning T. p16/ BRCA1 BRCA2 contribution of germline and mutations 14. Spurdle AB, Purdie D, Chen X, Chenevix-Trench G. MTS1 inactivation in ovarian carcinomas: high frequen- to familial ovarian cancer: No evidence for other ovarian The prohibitin 3¶ untranslated region polymorphism is cy of reduced protein expression associated with hyper- cancer-susceptibility genes. Am J Hum Genet 1999;65: not associated with risk of ovarian cancer. Cancer Res methylation or mutation in endometrioid and mucinous 1021–9. 2003;90:145–9. tumors. Int J Cancer 1998;79:61–5. 2. FordD,EastonDF,StrattonM,etal.Genetic 15. Auranen A, Spurdle AB, Chen X, et al. BRCA2 30. Kusume T, Tsuda H, Kawabata M, et al. p16-cyclin BRCA1 heterogeneity and penetrance analysis of the Arg372His polymorphism and epithelial ovarian cancer D1/CDK4-pRb pathway and clinical outcome in epithe- BRCA2 and genes in breast cancer families. Am J Hum risk. Int J Cancer 2003;103:427–30. lial ovarian cancer. Clin Cancer Res 1999;5:4152–7. Genet 1998;62:676–89. 16. Auranen A, Song H, Waterfall C, et al. Polymorphisms 31. Kudoh K, Ichikawa Y, Yoshida S, et al. Inactivation of 3. Antoniou AC, Easton DF. Risk prediction models for in DNA repair genes and epithelial ovarian cancer risk. p16/CDKN2 and p15/MTS2 is associated with prognosis familial breast cancer. Future Oncol 2006;2:257–74. Int J Cancer 2005;117:611–8. and response to chemotherapy in ovarian cancer. Int J 4. Pharoah PD, Antoniou A, Bobrow M, et al. Polygenic 17. Song H, Ramus SJ, Quaye L, et al. Common variants Cancer 2002;99:579–82. susceptibility to breast cancer and implications for in mismatch repair genes and risk of invasive ovarian 32. Sellers TA, Schildkraut JM, Pankratz VS, et al. Estrogen prevention. Nat Genet 2002;31:33–6. cancer. Carcinogenesis 2006;27:2235–42. bioactivation genetic polymorphisms and ovarian can- 5. Rowe SM, Coughlan SJ, McKenna NJ, et al. Ovarian 18. Runnebaum IB, Tong XW, Konig R, et al. p53-based cer. Cancer Epidemiol Biomarkers Prev 2005;14:2536–43. carcinoma-associated TaqI restriction fragment length blood test for p53PIN3 and risk for sporadic ovarian 33. Goodman MT, McDuffie K, Kolonel LN, et al. polymorphism in intron G of the progesterone receptor cancer. Lancet 1995;345:994. Multiethnic case-control study of ovarian cancer and gene is due to an Alu sequence insertion. Cancer Res 19. Lancaster JM, Brownlee HA, Wiseman RW, Taylor J. polymorphisms in genes involved in catechol estrogen 1995;55:2743–5. p53polymorphism in ovarian and bladder cancer. formation and metabolism. Cancer Epidemiol Bio- 6. Spurdle AB, Webb PM, Purdie DM, et al. No significant Lancet 1995;346:182. markers Prev 2001;10:209–16. association between progesterone receptor exon 4 660 20. Buller RE, Sood A, Fullenkamp C, et al. The influence 34. Skol AD, Scott LJ, Abecasis GR, Boehnke M. Joint Val Leu G/T polymorphism and risk of ovarian cancer. of the p53codon 72 polymorphism on ovarian carcino- analysis is more efficient than replication-based analysis Carcinogenesis 2001;22:717–21. genesis and prognosis. Cancer Gene Ther 1997;4:239–45. for two-stage genome-wide association studies. Nat 7. Lancaster JM, Berchuck A, Carney M, Wiseman RW, 21. Song H, Ramus SJ, Shadforth D, et al. Common Genet 2006;38:209–13. Taylor JA. Progesterone receptor gene and risk of breast variants in RB1 gene and risk of invasive ovarian cancer. 35. StramDO,HaimanCA,HirschhornJN,etal. and ovarian cancer. Br J Cancer 1997;78:277. Cancer Res 2006;66:10220–6. Choosing haplotype-tagging SNPS based on unphased 8. Pearce CL, Hirschhorn JN, Wu AH, et al. Clarifying the 22. Weitzel JN, Ding S, Larson GP, et al. The HRAS1 genotype data using a preliminary sample of unrelated PROGINS allele association in ovarian and breast cancer minisatellite locus and risk of ovarian cancer. Cancer subjects with an example from the Multiethnic Cohort risk: a haplotype-based analysis. J Natl Cancer Inst 2005; Res 2000;60:259–61. Study. Hum Hered 2003;55:27–36. 97:51–9. 23. Dicioccio RA, Song H, Waterfall C, et al. STK15 36. Dudbridge F. A note on permutation tests in 9. Berchuck A, Schildkraut JM, Wenham RM, et al. polymorphisms and association with risk of invasive multistage association scans. Am J Hum Genet 2006; Progesterone receptor promoter +331A polymorphism ovarian cancer. Cancer Epidemiol Biomarkers Prev 78:1094–5; author reply 1096. is associated with increased risk of endometrioid and 2004;13:1589–94. 37. Thomas DC, Haile RW, Duggan D. Recent develop- clear cell ovarian cancers. Cancer Epidemiol Biomarkers 24. Funk JO. Cancer cell cycle control. Anticancer Res ments in genome-wide association scans: a workshop Prev 2004;13:2141–7. 1999;19:4772–80. summary and review. Am J Hum Genet 2005;77:337–45. 10. Spurdle AB, Webb PM, Chen X, et al. Androgen 25. Lundberg AS, Weinberg RA. Control of the cell cycle 38. Kibel AS, Suarez BK, Belani J, et al. CDKN1A and receptor exon 1 CAG repeat length and risk of ovarian and apoptosis. Eur J Cancer 1999;35:1886–94. CDKN1B polymorphisms and risk of advanced prostate cancer. Int J Cancer 2000;87:637–43. 26. Barbieri F, Cagnoli M, Ragni N, et al. Increased cyclin carcinoma. Cancer Res 2003;63:2033–6. 11. Levine DA, Boyd J. The androgen receptor and D1 expression is associated with features of malignancy 39. Tomoda K, Kubota Y, Arata Y, et al. The cytoplasmic genetic susceptibility to ovarian cancer: results from a and disease recurrence in ovarian tumors. Clin Cancer shuttling and subsequent degradation of p27Kip1 case series. Cancer Res 2001;61:908–11. Res 1999;5:1837–42. mediated by Jab1/CSN5 and the COP9 signalosome 12. Garner EI, Stokes EE, Berkowitz RS, Mok SC, 27. Shigemasa K, Tanimoto H, Parham GP, et al. Cyclin complex. J Biol Chem 2002;277:2302–10. Cramer DW. Polymorphisms of the estrogen-metabo- D1 overexpression and p53mutation status in epithelial 40. Bardeesy N, Aguirre AJ, Chu GC, et al. Both lizing genes CYP17 and catechol-O-methyltransferase ovarian cancer. J Soc Gynecol Investig 1999;6:102–8. p16(Ink4a) and the p19(Arf)-p53pathway constrain and risk of epithelial ovarian cancer. Cancer Res 2002; 28. Chen CH, Shen J, Lee WJ, Chow SN. Overexpression progression of pancreatic adenocarcinoma in the 62:3058–62. of cyclin D1 and c-Myc gene products in human primary mouse. Proc Natl Acad Sci U S A 2006;103:5947–52.

www.aacrjournals.org 3035 Cancer Res 2007; 67: (7). April 1, 2007

Downloaded from cancerres.aacrjournals.org on September 28, 2021. © 2007 American Association for Cancer Research. Tagging Single Nucleotide Polymorphisms in Cell Cycle Control Genes and Susceptibility to Invasive Epithelial Ovarian Cancer

Simon A. Gayther, Honglin Song, Susan J. Ramus, et al.

Cancer Res 2007;67:3027-3035.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/67/7/3027

Supplementary Access the most recent supplemental material at: Material http://cancerres.aacrjournals.org/content/suppl/2007/03/27/67.7.3027.DC1

Cited articles This article cites 40 articles, 14 of which you can access for free at: http://cancerres.aacrjournals.org/content/67/7/3027.full#ref-list-1

Citing articles This article has been cited by 12 HighWire-hosted articles. Access the articles at: http://cancerres.aacrjournals.org/content/67/7/3027.full#related-urls

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/67/7/3027. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.