Associations Between Breast Cancer Susceptibility Gene Polymorphisms and Clinicopathological Features

124 Vol. 10, 124–130, January 1, 2004 Clinical Cancer Research

Associations between Breast Cancer Susceptibility Gene Polymorphisms and Clinicopathological Features

Wonshik Han,1 Daehee Kang,2 In Ae Park,3 Conclusions: These results indicate that polymorphisms Seok Won Kim,1 Ji Yeon Bae,4 Ki-Wook Chung,5 of some selected breast cancer susceptibility genes are associated with the clinicopathological phenotypes of breast and Dong-Young Noh1,4 cancer. Departments of 1Surgery, 2Preventive Medicine, and 3Pathology, and 4Cancer Research Institute, Seoul National University College of 5 Medicine, Seoul; and Center for Breast Cancer, National Cancer INTRODUCTION Center, Goyang, Korea Hereditary breast cancer accounts for 5–9% of all breast cancers (1). It was recently estimated that a combination of ABSTRACT BRCA1 and BRCA2 gene mutations is responsible for only Purpose: Genetic polymorphisms may affect not only ϳ30% of hereditary breast cancer cases (2) and Ͻ2% of all cancer development but also cancer progression, and as a breast cancer (3), suggesting that there may be other, low- result could influence cancer phenotypes. The aim of this penetrance genes that also increase an individual’s susceptibility study was to examine the relationship between breast cancer to breast cancer. Candidate low-penetrance breast cancer susceptibility genes include those in pathways involved in DNA susceptibility gene polymorphisms and clinicopathological repair, steroid hormone metabolism and signaling, and carcin- features. ogen metabolism (4). Because of the accessibility of enormously Experimental Design: We genotyped 664 Korean pri- expanding genomic databases and the rapid development of mary breast cancer patients for 17 single-nucleotide poly- high-throughput automated genotyping, an ever-increasing morphisms (SNPs) in nine genes, using a high-throughput number of association studies between breast cancer risk and SNP scoring method. genetic polymorphisms, particularly for single-nucleotide poly- Results: CYP1A1 codon 462 Ile/Val or Val/Val variants morphisms (SNPs), have been published. Many of these have and the CYP1B1 codon 432 Leu/Val variant were found demonstrated significant associations, although few studies more in breast cancer patients <35 years of age at onset have elucidated how a germline polymorphism can affect breast than the common homozygote [odds ratio (OR), 1.6 and 1.7, cancer development. Some authors have reported associations respectively]. In combination analysis of these two SNPs, the between various genetic polymorphisms and phenotypic fea- OR was 1.9 when one of them was heterozygous or a rare tures of breast cancer. These include TP53 codon 72 and low- homozygous form, and increased to 2.3 when both were grade histology (5), a TP53 intron 3 16-bp insertion and poor -Cases with Ile/Val at CYP1A1 codon histological grade, p21 codon 31 polymorphism and progester .(0.006 ؍ variants (P 462 were 2.6-fold and those with Val/Val were 5.1-fold more one receptor (PR) status (6), CYP1B1 codon 432 and steroid likely to have first-degree relatives with breast cancer than receptor status (7), ESR1 codon 325 and expression of PR and In the haplotype study of p53 (8), VDR and nodal status (9), PSA promoter and less .(0.002 ؍ those with Ile/Ile (P BRCA1, the 2430C/2731T/3667G/4427C/4956G homozygote aggressive breast cancer (10), and SRD5A2 codon 89 and early- showed less estrogen receptor negativity than the most com- onset, aggressive forms of breast cancer (11). In addition, many mon diplotype (OR, 0.5; 95% confidence interval, 0.26– authors have reported associations between genetic polymor- 0.94). TP53 codon 72 Arg/Pro or Pro/Pro variants were phisms of GSTP1 (12), GSTM1 and GSTT1 (13), SRD5A2 (11), associated with negative axillary lymph node status (OR, SULT1A1 (14), LIG4 (4), and GSTA1 (15) and survival of breast cancer patients. 0.7; 95% confidence interval, 0.49–0.94). The results of these previous reports suggested that genetic polymorphisms may associate with cancer development as well as progression, and as a result may affect cancer phenotype and prognosis. However, these results were anecdotal and have not been reproduced in other studies. More systematic analysis Received 6/3/03; revised 9/2/03; accepted 9/22/03. involving several candidate genes and various clinical parame- Grant support: Korea Health 21 R&D Project, Ministry of Health & ters is required to confirm the hypothesis that genetic polymor- Welfare, R.O.K. (01-PJ3-PG6-01GN07-0004). phisms are associated with cancer phenotype. We selected the The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked genes that have been reported or hypothesized to associate with advertisement in accordance with 18 U.S.C. Section 1734 solely to breast cancer risk in other studies with rare allele frequencies indicate this fact. exceeding 10% in our preliminary study population, focusing Requests for reprints: Dong-Young Noh, M.D., Ph.D., Cancer Re- mainly on the genes involved in estrogen metabolism. They search Institute and Department of Surgery, Seoul National University College of Medicine, 28 Yongon-dong, Chongno-gu, Seoul 110-744, included genetic polymorphisms that we have previously re- Korea. Phone: 82-2-760-2921; Fax: 82-2-766-3975; E-mail: ported to be associated with breast cancer susceptibility (16– [email protected]. 21). These were CYP19 codons 80 and 264; CYP1A1 codon 462;

CYP1B1 codons 48 and 432; COMT codon 158; GSTP1 codon Table 1 Characteristics of breast cancer cases 105; ESR1 codons 325 and 594; TP53 codon 72; TGFBR2 n % codon 389; BRCA1 codons 771, 871, 1183, 1436, and 1613; and No. of subjects 664 BRCA2 codon 372. Median (range) age at diagnosis, years 46.0 (22–82) The purpose of the present work was to evaluate possible Age distribution associations between polymorphisms in these genes and clini- Յ30 years 27 4.1 copathological features of breast cancer by use of a high- 30–40 years 148 22.3 40–50 years 265 39.9 throughput SNP scoring technique. 50–60 years 168 25.3 60–70 years 48 7.2 Ͼ70 years 8 1.2 MATERIALS AND METHODS Family history of breast cancer 50 7.5 Study Subjects. All study subjects were recruited from First-degree relatives 28 4.2 Second-degree relatives 22 3.3 September 2001 to January 2003 at Seoul National University Histological subtype Hospital with the approval of the Institutional Review Board. Ductal 582 87.7 Breast cancer patients (n ϭ 664) underwent surgery for primary Lobular 21 3.2 breast cancer from January 1988 to January 2003, and the Mucinous 17 2.6 Papillary 13 2.0 diagnoses of cancer were confirmed histopathologically at Seoul Tubular 7 1.1 National University Hospital. Informed consent was obtained Medullary 6 0.9 from all participants at the time of blood withdrawal. Patients Metaplastic 6 0.9 who refused to donate blood, who had life-threatening disease Other 12 1.8 Histological grade progression, who had other malignancies than breast cancer, or 1 54 8.1 whose clinical and pathological data were not available were 2 260 39.2 excluded from the study. 3 192 28.9 The characteristics of the breast cancer patients and their Unknown 158 23.8 T stage tumors are shown in Table 1. Age at onset, family history of Tis 39 5.9 breast cancer, tumor size, lymph node status, and histological T1 320 48.2 grade (Scarff–Bloom–Richardson classification) were reviewed. T2 261 39.3 Immunohistochemical studies were performed to determine ex- T3 18 2.7 T4 8 1.2 pression of the tumor markers estrogen receptor (ER) and PR. Unknown 18 2.7 Clinicopathological data were compared among genotypes and Lymph node status analyzed for significant differences. Negative 438 66.0 SNP Genotyping for Allele Frequency. We selected 72 Positive 221 33.3 Unknown 5 0.8 SNPs from a list of low-penetrance breast cancer susceptibility Distant metastasis 8 1.2 genes. After performing preliminary genotyping, we selected 17 Estrogen receptor status SNPs with a rare allele frequency exceeding 10% in the Korean Negative 250 37.7 population (Table 2). All SNPs were in the coding region of Positive 347 52.3 Unknown 67 10.1 each gene, with 11 being nonsynonymous and 6 being synony- Progesterone receptor status mous changes. Negative 344 51.8 SNP genotyping was performed by SNP-IT assay using the Positive 253 38.1 SNPstream 25K System (Orchid Biosciences, Princeton, NJ). Unknown 67 10.1 Briefly, the genomic DNA region spanning the polymorphic site was PCR-amplified using one phosphothiolated primer and one regular PCR primer. The amplified PCR products were then Ј digested with exonuclease. The 5 phosphothioates protected Statistical Analysis. The ␹2 test (Pearson statistic) was one strand of the PCR product from exonuclease digestion, used to determine associations between the frequencies of the generating a single-stranded PCR template. The single-stranded polymorphic alleles and the various clinicopathological features PCR template was overlaid on a 384-well plate that contained a of breast tumors and to ensure the independence of alleles covalently attached SNP-IT extension primer designed to hy- (Hardy–Weinberg equilibrium). The odds ratios (ORs) and 95% bridize immediately adjacent to the polymorphic site. The confidence intervals (CIs) were calculated by use of an uncon- SNP-IT primer was extended for a single base with DNA ditional logistic regression model. To determine the presence of polymerase and a mixture of the appropriate acycloterminators a linear increase in risk with exposure, a linear-by-linear asso- labeled with either FITC or biotin and complementary to the ciation test was performed. All analyses were carried out using polymorphic nucleotide. The identity of the incorporated nucle- SPSS version 10.0 (Chicago, IL). otide was determined by serial colorimetric reactions with anti- BRCA1 haplotypes were constructed from genotype data FITC-alkaline phosphatase and streptavidin-horseradish perox- from five SNPs by use of the HAPLOTYPER2 program (soft- idase, respectively. The resulting yellow and/or blue color was ware for haplotype inference based on the Bayesian algorithm; measured with an ELISA reader, and the final genotype calls Ref. 22). The genetic status of subjects was expressed as the were made using a QCReview program. combination of two haplotypes (diplotype configuration).

Table 2 Gene list and genotype frequencies Common homozygote, Heterozygote, Rare homozygote, Rare allele Gene Variation n n (%) n (%) n (%) frequency CYP19 V80V (GTAϾGTG) 655 203 (31.0) 320 (48.9) 132 (20.1) 0.45 R264C (CGCϾTGC) 664 457 (68.8) 183 (27.6) 24 (3.6) 0.17 CYP1A1 I462V (ATTϾGTT) 661 377 (57.0) 248 (37.5) 36 (5.4) 0.24 CYP1B1 R48G (CGGϾGGG) 657 436 (66.4) 199 (30.3) 22 (3.3) 0.18 L432V (CTGϾGTG) 664 508 (76.5) 143 (21.5) 13 (2.0) 0.13 COMT V158M (GTGϾATG) 661 352 (53.3) 259 (39.2) 50 (7.5) 0.27 GSTP1 I105V (ATCϾGTC) 664 448 (67.5) 192 (28.9) 24 (3.6) 0.18 ESR1 P325P (CCCϾCCG) 664 184 (27.7) 317 (47.7) 163 (24.5) 0.48 T594T (ACGϾACA) 662 421 (63.6) 220 (33.2) 21 (3.2) 0.20 TP53 R72P (CGCϾCCC) 664 288 (43.4) 305 (45.9) 71 (10.7) 0.34 TGFBR2 N389N (AACϾAAT) 661 368 (55.7) 238 (36.0) 55 (8.3) 0.26 BRCA1 L771L (TTGϾCTG) 657 315 (47.9) 279 (42.5) 63 (9.6) 0.31 P871L (CCGϾCTG) 661 316 (47.8) 281 (42.5) 64 (9.7) 0.31 K1183R (AAAϾAGA) 662 318 (48.0) 281 (42.4) 63 (9.5) 0.31 S1436S (TCTϾTCC) 664 318 (47.9) 288 (43.4) 58 (8.7) 0.30 S1613G (AGTϾGGT) 662 317 (47.9) 281 (42.4) 64 (9.7) 0.31 BRCA2 N372H (AATϾCAT) 664 364 (54.8) 258 (38.9) 42 (6.3) 0.26

RESULTS at onset was found more in cases with Leu/Val heterozygos- The allele frequencies of the 17 SNPs are shown in Table ity in CYP1B1 codon 432 than in those with Leu/Leu (OR, 2. The wild-type and variant alleles were in Hardy–Weinberg 1.71; 95% CI, 1.02–2.86). Study subjects were assigned to equilibrium. one of three groups to examine synergistic effects of com- Analysis of the association between each SNP and the bined CYP1A1 codon 462 and CYP1B1 codon 432 polymor- clinicopathological features of breast cancer showed some phisms on onset of breast cancer at a young age (Table 4). We significant associations (Table 3). CYP1A1 codon 462 and found that if one of the two SNPs was heterozygous or a rare CYP1B1 codon 432 polymorphisms were found to be asso- homozygous form, there was a higher correlation with young ciated with young onset (Յ35 years at onset; Table 4). A age at onset of breast cancer (OR, 1.92; 95% CI, 1.14–3.24). young age at onset was found more in cases with the Ile/Val When both SNPs were heterozygous or rare homozygous or Val/Val genotype in CYP1A1 codon 462 than in those with forms, the OR increased to 2.35 (95% CI, 1.14– 4.86; P for Ile/Ile (OR, 1.64; 95% CI, 1.03–2.60). Similarly, a young age trend ϭ 0.006).

Table 3 Association of varianta of each polymorphism with clinicopathological features of breast cancer n (%) CYP19 CYP19 CYP1A1 CYP1B1 CYP1B1 COMT GSTP1 ESR1 ESR1 TP53 TGFBR2 BRCA2 Characteristic n V80V R264C I462V R48G L432V V158M I105V P325P T594T R72P N389N N372H Age Ͼ35 years 582 395 (69) 176 (30) 240 (41) 191 (33) 130 (22) 269 (46) 187 (32) 421 (72) 214 (37) 333 (57) 256 (44) 261 (45) Յ35 years 82 57 (72) 31 (38) 44 (54)b 30 (38) 26 (32) 40 (49) 29 (35) 59 (72) 27 (33) 43 (52) 37 (45) 39 (48) Family history No 614 415 (68) 188 (31) 254 (42) 208 (34) 146 (24) 122 (49) 198 (32) 443 (72) 220 (36) 350 (57) 271 (44) 276 (45) Yesc 50 37 (76) 19 (37) 30 (60)b 13 (25) 10 (20) 153 (45) 16 (31) 37 (73) 21 (41) 26 (51) 22 (43) 24 (47) Histologic grade 1 or 2 314 216 (66) 95 (30) 143 (46) 100 (32) 78 (25) 141 (45) 101 (32) 225 (72) 117 (38) 176 (56) 142 (45) 149 (47) 3 192 133 (71) 58 (30) 73 (38) 64 (34) 43 (23) 93 (48) 56 (29) 141 (73) 68 (35) 109 (57) 69 (36) 81 (42) T stage Tis or T1 359 236 (69) 102 (30) 153 (45) 110 (32) 80 (23) 170 (50) 108 (32) 239 (69) 114 (33) 204 (59) 155 (45) 154 (45) T2–T4 287 191 (69) 94 (33) 118 (42) 97 (35) 66 (24) 118 (42) 89 (31) 211 (75) 113 (40) 151 (53) 116 (41) 127 (45) Lymph node Negative 438 269 (66) 133 (32) 179 (44) 138 (34) 88 (22) 196 (48) 138 (34) 290 (70) 144 (35) 262 (60)b 174 (42) 183 (44) Positive 221 156 (72) 63 (29) 93 (42) 72 (33) 63 (29) 94 (43) 62 (28) 165 (75) 84 (39) 111 (50) 105 (48) 100 (45) Estrogen receptor Positive 347 240 (70) 111 (32) 148 (43) 117 (34) 81 (24) 169 (49) 111 (32) 254 (73) 136 (39) 200 (58) 162 (47) 159 (46) Negative 250 164 (67) 77 (31) 106 (43) 82 (33) 64 (26) 106 (43) 78 (31) 178 (71) 86 (35) 135 (54) 99 (40) 117 (47) a Heterozygote or rare homozygote. b P Ͻ 0.05. c First- or second-degree relatives with breast cancer.

Table 4 Onset of breast cancer at young age (Յ35 years) and genetic polymorphisms in CYP1A1 codon 462 and CYP1B1 codon 432 Genotype Age Ͼ35 years, n (%) Age Յ35 years, n (%) ORa (95% CI) P for trend CYP1A1 codon 462 Ile/Ile 339 (58.5) 38 (46.3) 1.00 (reference) Ile/Val 210 (36.3) 38 (46.3) 1.61 (1.00–2.61) Val/Val 30 (5.2) 6 (7.3) 1.78 (0.70–4.56) 0.043 Ile/Val or Val/Val 240 (41.5) 44 (53.6) 1.64 (1.03–2.60) CYP1B1 codon 432 Leu/Leu 452 (77.7) 56 (68.3) 1.00 (reference) Leu/Val 118 (20.3) 25 (30.5) 1.71 (1.02–2.86) Val/Val 12 (2.1) 1 (1.2) 0.67 (0.09–5.27) Leu/Val or Val/Val 130 (22.4) 26 (31.7) 1.61 (0.98–2.67) CYP1A1462/CYP1B1432 Wild-typeb/wild-type 271 (46.8) 25 (30.5) 1.00 (reference) Wild-type/variantc or variant/wild-type 248 (42.8) 44 (53.7) 1.92 (1.14–3.24) Variant/variant 60 (10.4) 13 (15.9) 2.35 (1.14–4.86) 0.006 a OR, odds ratio; CI, confidence interval. b Wild-type, common homozygote. c Variant, heterozygote or rare homozygote.

Fifty of our cases had a family history of breast cancer. CI, 0.26–0.94) and less PR negativity (OR, 0.56; 95% CI, Twenty-eight of the 50 had first-degree relatives and 22 had 0.31–1.02) than the most common I/I combination (Table 7). second-degree relatives but no first-degree relatives with TP53 codon 72 polymorphisms were associated with axil- breast cancer (Table 1). Subjects with Ile/Val in CYP1A1 lary lymph node metastasis (Table 3). Cases with Arg/Pro or codon 462 were 2.0-fold more likely to have first- or second- Pro/Pro variants in TP53 codon 72 had less axillary lymph node degree relatives with breast cancer, and those with Val/Val metastasis than those homozygous for the Arg/Arg variant (OR, were 2.9-fold more likely (P for trend ϭ 0.008). The signif- 0.68; 95% CI, 0.49–0.94). icance increased when the cases were restricted to those who Given the many polymorphisms examined, we needed to definitely had first-degree relatives with breast cancer (P for adjust for multiple comparisons. We assessed 12 polymor- trend ϭ 0.002; Table 5). When we combined age and family phisms and a haplotype in 10 genes; therefore, a conservative history and analyzed the association with the CYP1A1 codon Bonferroni correction of the P would require the significance 462 polymorphism, younger patients (Յ35 years) with a level to be 0.05/13 ϭ 0.003. When we applied this stringent family history (first- or second-degree relative) were 8.8-fold criterion for significance, the association between the CYP1A1 more likely to have the Ile/Val or Val/Val variants than the codon 462 polymorphism and a family history of breast cancer Ile/Ile, the wild type (OR, 8.89; 95% CI, 1.06–74.35; data not in at least one first-degree relative remained significant (P ϭ shown). 0.002). The five BRCA1 SNPs showed no association with any clinicopathological feature individually. We constructed haplotypes with these five SNPs, using the haplotype inference pro- DISCUSSION gram HAPLOTYPER2 (Table 6). Four types of haplotype were Although many somatic genetic changes correlate with found, and two of them accounted for Ͼ99% of our study cases. phenotypes in breast cancer (23), limited evidence is available The haplotypes found and their frequencies were very similar in about the influence of common germline genetic variations on all 60 controls (data not shown). The association between hap- clinical outcome. Because of possible effects on protein function lotype combination (diplotype) and clinicopathological features or expression, it is reasonable to suspect that polymorphisms in was analyzed. A relatively rare homozygote, type II/II combi- genes involved in carcinogen metabolism, estrogen synthesis/ nation, showed significantly less ER negativity (OR, 0.49; 95% metabolism, DNA repair, and cell-cycle control could predis-

Table 5 Family history of breast cancer and CYP1A1 codon 462 polymorphism First- or No family second-degree First-degree history, relative, ORa P for relative, OR P for Genotype n (%) n (%) (95% CI) trend n (%) (95% CI) trend CYP-A1 codon 462 Ile/Ile 357 (58.4) 20 (40.0) 1.00 (reference) 9 (32.1) 1.00 (reference) Ile/Val 223 (36.5) 25 (50.0) 2.00 (1.09–3.69) 15 (53.6) 2.67 (1.15–6.20) Val/Val 31 (5.1) 5 (10.0) 2.88 (1.01–8.20) 0.008 4 (14.3) 5.12 (1.49–17.58) 0.002 Ile/Val or Val/Val 254 (41.6) 30 (60.0) 2.11 (1.17–3.80) 19 (67.9) 2.97 (1.32–6.67) a OR, odds ratio; CI, confidence interval.

Table 6 Haplotypes of BRCA1 in 657 breast cancer cases phisms on the development of cancer is more apparent in this Exon 11 genetically labile population. Our results showed that the effect Exon 13 Exon 16 was maximized in breast cancer patients who were young at a Haplotype 2430 2731 3667 4427 4956 n (%) disease onset who also had a family history of breast cancer in I T C A T A 906 (68.9) close relatives. II C T G C G 403 (30.7) It has been reported that cancers associated with BRCA1 III C T G T G 4 (0.30) IV T C A T G 1 (0.08) mutations are less often positive for ER and PR and are more frequently medullary carcinoma and higher grade invasive a Nucleotide position (reference sequence U14680.1). ductal carcinomas than are sporadic breast cancers (31, 32). We tried to investigate whether a patient with sporadic breast cancer with a certain BRCA1 haplotype would have a different phenotype or age of onset from others. We found that pose individuals to breast cancer and could also influence the 2430C/2731T/3667G/4427C/4956G homozygotes showed clinical phenotype of the tumor. We looked for relationships more ER and PR expression than the most common diplo- between polymorphisms of candidate genes for breast cancer type. Fan et al. (33) demonstrated that the wild-type BRCA1 susceptibility and cancer phenotypes. gene inhibits signaling by ligand-activated ER-␣ through the The most remarkable finding was the association between estrogen receptor element and blocks the transcriptional ac- germline genetic polymorphisms in CYP1A1 and CYP1B1 and tivation function of activity function-2 of ER-␣; they postu- age at onset of breast cancer. Several investigators have dem- lated that loss of this ability contributes to mammary carci- onstrated that breast cancer in young women, compared with nogenesis. Our results suggest that alterations in BRCA1, their older counterparts, differs in terms of pathological features such as alternative haplotypes, although they may not cause and clinical outcomes. Moreover, age has been shown to be an the truncation mutation or change the breast cancer risk at all independent prognostic factor, suggesting that early- and late- can affect the interaction of BRCA1 with ER. onset breast cancers have different biological origins (24, 25). It Axillary lymph node status is the most significant prog- is possible that some unknown genetic factor may contribute to nostic factor in breast cancer. Therefore, any factor associ- these different propensities with age. In the literature, GSTM1, ated with lymph node metastasis is likely to be associated GSTT1, and CYP17 polymorphisms have been proposed as with survival. In our study, the TP53 codon 72 Pro allele was associated with a young age at onset of breast cancer (26, 27). found to be associated with a negative lymph node status. CYP1A1 and CYP1B1 are enzymes involved in the production Goode et al. (4) found a protective effect against death after of carcinogenic estrogen metabolites and the activation of en- breast cancer among patients who carried the Pro allele of the vironmental carcinogens. The association between polymor- TP53 R72P polymorphism, but inclusion of known prognos- phisms in these genes and breast cancer risk is controversial (7, tic variables in the model reduced the apparent protective 28–30). Our results suggest that germline polymorphisms in effect of this TP53 polymorphism. From these and our re- CYP1A1 and CYP1B1 might affect development of cancer in sults, it appears that the TP53 codon 72 polymorphism may younger patients, whereas other factors may have more influ- affect lymph node metastasis in breast cancer and, conse- ence in carcinogenesis in older subjects. Although the difference quently, survival. Dumont et al. (34) indicated that the Arg72 between age groups regarding exposure to carcinogens could variant of TP53 induces apoptosis markedly better than the explain the difference in cancer risk and tumor characteristics Pro72 variant and suggested that these variants may alter according to age of onset, it is more reasonable to postulate that cancer risk and that Arg72 homozygotes may respond more genetic insults may have greater influence on earlier onset favorably to radiation or chemotherapy. diseases. In conclusion, the present work confirms the hypothesis We also showed an association between CYP1A1 and fam- that polymorphisms in candidate breast cancer susceptibility ily history of breast cancer. Hypothetically, CYP1A1 is a low- genes can influence the clinicopathological features of the dis- penetrance gene for breast cancer, and its polymorphisms would ease. In addition, the study demonstrates the feasibility of high- not show strong familial aggregation. Nevertheless, a woman throughput SNP scoring for mass screening studies. Points of with a family history has a higher risk for developing breast particular note are that the present study is large scale in terms cancer, suggesting that other factors may be already working in of the number of subjects, SNPs, and genes examined; involved this population. It is possible that the effect of genetic polymor- a very homogeneous ethnic population recruited from one in-

Table 7 BRCA1 haplotypes and ERa and PR status ER PR Haplotype combination (diplotype) Positive, n (%) Negative, n (%) OR (95% CI) Positive, n (%) Negative, n (%) OR (95% CI) I/I 155 (45.7) 124 (50.2) 1.00 (reference) 113 (45.9) 166 (48.8) 1.00 (reference) I/II 146 (43.1) 108 (43.7) 0.93 (0.66–1.30) 104 (42.3) 150 (44.1) 0.98 (0.69–1.39) II/II 38 (11.3) 15 (6.1) 0.49 (0.26–0.94) 29 (11.8) 24 (7.0) 0.56 (0.31–1.02) a ER, estrogen receptor; PR, progesterone receptor; OR, odds ratio; CI, confidence interval.

stitute; and involved an accurate genotyping methodology. This patients receiving tamoxifen therapy. J. Natl. Cancer Inst. (Bethesda), study has identified novel associations that provide insight into 94: 1635–1640, 2002. the role of germline polymorphisms in cancer biology and 15. Sweeney, C., Ambrosone, C. B., Joseph, L., Stone, A., Hutchins, clinical outcome. L. F., Kadlubar, F. F., and Coles, B. F. Association between a glutathione S-transferase A1 promoter polymorphism and survival after breast cancer treatment. Int. J. Cancer, 103: 810–814, 2003. ACKNOWLEDGMENTS 16. Shin, A., Kang, D., Nishio, H., Lee, M. J., Park, S. K., Kim, S. U., We thank Dr. Jong Eun Lee at DNA Link, Inc. (Seoul, Korea) for Noh, D. Y., Choe, K. J., Ahn, S. H., Hirvonen, A., Kim, J. H., and Yoo, valuable discussions and excellent genotyping work, and Eun-Kyoung K. Y. Estrogen receptor ␣ gene polymorphisms and breast cancer risk. Hwang for database and sample management. Breast Cancer Res. Treat, 80: 127–131, 2003. 17. Lee, K. M., Abel, J., Ko, Y., Harth, V., Park, W. Y., Seo, J. S., Yoo, K. Y., Choi, J. Y., Shin, A., Ahn, S. H., Noh, D. Y., Hirvonen, A., and REFERENCES Kang, D. Genetic polymorphisms of cytochrome P450 19 and 1B1, 1. Ford, D., and Easton, D. F. The genetics of breast and ovarian cancer. alcohol use, and breast cancer risk in Korean women. Br. J. Cancer, 88: Br. J. Cancer, 72: 805–812, 1995. 675–678, 2003. 2. Ligtenberg, M. J., Hogervorst, F. B., Willems, H. W., Arts, P. J., 18. Park, S. K., Kang, D., Noh, D. Y., Lee, K. M., Kim, S. U., Choi, Brink, G., Hageman, S., Bosgoed, E. A., Van der Looij, E., Rookus, J. Y., Choi, I. M., Ahn, S. H., Choe, K. J., Hirvonen, A., Strickland, M. A., Devilee, P., Vos, E. M., Wigbout, G., Struycken, P. M., Menko, P. T., and Yoo, K. Y. Reproductive factors, glutathione S-transferase F. H., Rutgers, E. J., Hoefsloot, E. H., Mariman, E. C., Brunner, H. G., M1 and T1 genetic polymorphism and breast cancer risk. Breast Cancer and Van’t Veer L. J. Characteristics of small breast and/or ovarian Res. Treat., 78: 89–96, 2003. cancer families with germline mutations in BRCA1 and BRCA2. Br. J. 19. Choi, J. Y., Abel, J., Neuhaus, T., Ko, Y., Harth, V., Hamajima, N., Cancer, 79: 1475–1478, 1999. Tajima, K., Yoo, K. Y., Park, S. K., Noh, D. Y., Han, W., Choe, K. J., 3. Anglian Breast Cancer Study Group. Prevalence and penetrance of Ahn, S. H., Kim, S. U., Hirvonen, A., Kang, D. Role of alcohol and BRCA1 and BRCA2 mutations in a population-based series of breast genetic polymorphisms of CYP2E1 and ALDH2 in breast cancer devel- cancer cases. Br. J. Cancer, 83: 1301–1308, 2000. opment. Pharmacogenetics, 13: 67–72, 2003. 4. Goode, E. L., Dunning, A. M., Kuschel, B., Healey, C. S., Day, N. E., 20. Yim, D. S., Park, S. K., Yoo, K. Y., Yoon, K. S., Chung, H. H., Ponder, B. A., Easton, D. F., and Pharoah, P. P. Effect of germ-line Kang, H. L., Ahn, S. H., Noh, D. Y., Choe, K. J., Jang, I. J., Shin, S. G., genetic variation on breast cancer survival in a population-based study. Strickland, P. T., Hirvonen, A., and Kang, D. Relationship between the Cancer Res., 62: 3052–3057, 2002. Val158Met polymorphism of catechol O-methyl transferase and breast 5. Sjalander, A., Birgander, R., Hallmans, G., Cajander, S., Lenner, P., cancer. Pharmacogenetics, 11: 279–286, 2001. Athlin, L., Beckman, G., and Beckman, L. p53 polymorphisms and 21. Park, S. K., Yoo, K. Y., Lee, S. J., Kim, S. U., Ahn, S. H., Noh, haplotypes in breast cancer. Carcinogenesis (Lond.), 17: 1313–1316, D. Y., Choe, K. J., Strickland, P. T., Hirvonen, A., and Kang, D. 1996. Alcohol consumption, glutathione S-transferase M1 and T1 genetic 6. Powell, B. L., van Staveren, I. L., Roosken, P., Grieu, F., Berns, polymorphisms and breast cancer risk. Pharmacogenetics, 10: 301– E. M., and Iacopetta, B. Associations between common polymorphisms 309, 2000. in TP53 and p21WAF1/Cip1 and phenotypic features of breast cancer. Carcinogenesis (Lond.), 23: 311–315, 2002. 22. Niu, T., Qin, Z. S., Xu, X., and Liu, J. S. Bayesian haplotype inference for multiple linked single-nucleotide polymorphisms. Am. J. 7. Bailey, L. R., Roodi, N., Dupont, W. D., and Parl, F. F. Association Hum. Genet., 70: 157–169, 2002. of cytochrome P450 1B1 (CYP1B1) polymorphism with steroid receptor status in breast cancer. Cancer Res., 58: 5038–5041, 1998. 23. Cuny, M., Kramar, A., Courjal, F., Johannsdottir, V., Iacopetta, B., Fontaine, H., Grenier, J., Culine, S., and Theillet, C. Relating genotype 8. Kang, H. J., Kim, S. W., Kim, H. J., Ahn, S. J., Bae, J. Y., Park, S. K., and phenotype in breast cancer: an analysis of the prognostic signifi- Kang, D., Hirvonen, A., Choe, K. J., and Noh, D. Y. Polymorphisms in the estrogen receptor-␣ gene and breast cancer risk. Cancer Lett., 178: cance of amplification at eight different genes or loci and of p53 175–180, 2002. mutations. Cancer Res., 60: 1077–1083, 2000. 9. Ruggiero, M., Pacini, S., Aterini, S., Fallai, C., Ruggiero, C., and 24. Bonnier, P., Romain, S., Charpin, C., Lejeune, C., Tubiana, N., Pacini, P. Vitamin D receptor gene polymorphism is associated with Martin, P. M., and Piana, L. Age as a prognostic factor in breast cancer: metastatic breast cancer. Oncol. Res., 10: 43–46, 1998. relationship to pathologic and biologic features. Int. J. Cancer, 62: 138–144, 1995. 10. Yang, Q., Nakamura, M., Nakamura, Y., Yoshimura, G., Suzuma, T., Umemura, T., Tamaki, T., Mori, I., Sakurai, T., and Kakudo, K. 25. Nixon, A. J., Neuberg, D., Hayes, D. F., Gelman, R., Connolly, Correlation of prostate-specific antigen promoter polymorphisms with J. L., Schnitt, S., Abner, A., Recht, A., Vicini, F., and Harris, J. R. clinicopathological characteristics in breast cancer. Anticancer Res., 22: Relationship of patient age to pathologic features of the tumor and 1825–1828, 2002. prognosis for patients with stage I or II breast cancer. J. Clin. Oncol., 12: 11. Scorilas, A., Bharaj, B., Giai, M., and Diamandis, E. P. Codon 89 888–894, 1994. polymorphism in the human 5␣-reductase gene in primary breast cancer. 26. Xiong, P., Bondy, M. L., Li, D., Shen, H., Wang, L. E., Singletary, Br. J. Cancer, 84: 760–767, 2001. S. E., Spitz, M. R., and Wei, Q. Sensitivity to benzo(a)pyrene diol- 12. Sweeney, C., McClure, G. Y., Fares, M. Y., Stone, A., Coles, B. F., epoxide associated with risk of breast cancer in young women and Thompson, P. A., Korourian, S., Hutchins, L. F., Kadlubar, F. F., and modulation by glutathione S-transferase polymorphisms: a case-control Ambrosone, C. B. Association between survival after treatment for study. Cancer Res., 61: 8465–8469, 2001. breast cancer and glutathione S-transferase P1 Ile105Val polymorphism. 27. Bergman-Jungestrom, M., Gentile, M., Lundin, A. C., and Wingren, Cancer Res., 60: 5621–5624, 2000. S. Association between CYP17 gene polymorphism and risk of breast 13. Ambrosone, C. B., Sweeney, C., Coles, B. F., Thompson, P. A., cancer in young women. Int. J. Cancer, 84: 350–353, 1999. McClure, G. Y., Korourian, S., Fares, M. Y., Stone, A., Kadlubar, F. F., 28. Zheng, W., Xie, D. W., Jin, F., Cheng, J. R., Dai, Q., Wen, W. Q., and Hutchins, L. F. Polymorphisms in glutathione S-transferases Shu, X. O., and Gao, Y. T. Genetic polymorphism of cytochrome (GSTM1 and GSTT1) and survival after treatment for breast cancer. P450-1B1 and risk of breast cancer. Cancer Epidemiol. Biomark. Prev., Cancer Res., 61: 7130–7135, 2001. 9: 147–150, 2000. 14. Nowell, S., Sweeney, C., Winters, M., Stone, A., Lang, N. P., 29. Ambrosone, C. B., Freudenheim, J. L., Graham, S., Marshall, J. R., Hutchins, L. F., Kadlubar, F. F., and Ambrosone, C. B. Association Vena, J. E., Brasure, J. R., Laughlin, R., Nemoto, T., Michalek, A. M., between sulfotransferase 1A1 genotype and survival of breast cancer and Harrington, A. Cytochrome P4501A1 and glutathione S-transferase

(M1) polymorphisms and postmenopausal breast cancer risk. Cancer 32. Robson, M., Gilewski, T., Haas, B., Levin, D., Borgen, P., Rajan, Res., 55: 3483–3485, 1995. P., Hirschaut, Y., Pressman, P., Rosen, P. P., Lesser, M. L., Norton, L., 30. Ishibe, N., Hankinson, S. E., Colditz, G. A., Spiegelman, D., and Offit, K. BRCA-associated breast cancer in young women. J. Clin. Wil-lett, W. C., Speizer, F. E., Kelsey, K. T., and Hunter, D. J. Oncol., 16: 1642–1649, 1998. Cigarette smoking, cytochrome P450 1A1 polymorphisms, and 33. Fan S., Wang J., Yuan R., Ma Y., Meng Q., Erdos M. R., Pestell breast cancer risk in the Nurses’ Health Study. Cancer Res., 58: R. G., Yuan F., Auborn K. J., Goldberg I. D., and Rosen E. M. BRCA1 667– 671, 1998. inhibition of estrogen receptor signaling in transfected cells. Science 31. Verhoog, L. C., Brekelmans, C. T., Seynaeve, C., van den Bosch, (Wash. DC), 284: 1354–1356, 1999. L. M., Dahmen, G., van Geel, A. N., Tilanus-Linthorst, M. M., Bartels, 34. Dumont, P., Leu, J. I., Della Pietra A. C. 3rd, George, D. L., C. C., Wagner, A., van den Ouweland, A., Devilee, P., Meijers-Heijboer, and Murphy, M. The codon 72 polymorphic variants of p53 have E. J., and Klijn, J. G. Survival and tumour characteristics of breast-cancer markedly different apoptotic potential. Nat. Genet., 33: 357–365, patients with germline mutations of. BRCA1. Lancet, 351: 316–321, 1998. 2003.

Downloaded from clincancerres.aacrjournals.org on September 29, 2021. © 2004 American Association for Cancer Research. Associations between Breast Cancer Susceptibility Gene Polymorphisms and Clinicopathological Features

Wonshik Han, Daehee Kang, In Ae Park, et al.

Clin Cancer Res 2004;10:124-130.

Updated version Access the most recent version of this article at: http://clincancerres.aacrjournals.org/content/10/1/124

Cited articles This article cites 34 articles, 12 of which you can access for free at: http://clincancerres.aacrjournals.org/content/10/1/124.full#ref-list-1

Citing articles This article has been cited by 4 HighWire-hosted articles. Access the articles at: http://clincancerres.aacrjournals.org/content/10/1/124.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://clincancerres.aacrjournals.org/content/10/1/124. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.