CYP2B6*6 is associated with increased breast cancer risk

Abstract

The cytochrome P450 2B6 (CYP2B6) is involved in the metabolism of testosterone. Functional changes in this enzyme may influence endogenous hormone exposure, which has been associated with risk of breast cancer. To assess potential associations between two functional polymorphisms CYP2B6_516_G>T (rs3745274) and CYP2B6_785_A>G (rs2279343) and breast cancer risk we established a specific matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS) assay. The GENICA breast cancer case-control study showed associations between the variant genotypes CYP2B6_516_TT and CYP2B6_785_GG and breast cancer risk with odds ratios (ORs) of 1.34 (p = 0.001) and 1.31 (p = 0.002), respectively. A similar effect was observed for carriers of the CYP2B6_516_T allele in a validation study including four independent studies from Germany, Sweden and USA. In a pooled analysis of all five studies involving 4,638 breast cancer cases and 3,594 controls of European ancestry, carriers of the CYP2B6_516_G and the CYP2B6_785_G variant had an increased breast cancer risk with ORs of 1.10 (p = 0.027) and 1.10 (p = 0.031), respectively. We conclude that the genetic variants CYP2B6_516_G and CYP2B6_785_G (designated CYP2B6*6), which are known to decrease activity of the CYP2B6 enzyme, contribute to an increased breast cancer risk.

Introduction

Women with high serum and urinary concentrations of sex hormones have a significantly increased breast cancer risk.^1-3 The steroid hormones estradiol and testosterone are independently associated with breast cancer risk with testosterone reported to confer a stronger influence on breast cancer risk than estradiol in postmenopausal women.² There are two hypothetical ways for androgens to affect breast cancer risk, they can serve as an important source of estrogens being converted to estrogen in the breast,⁴ or can directly stimulate growth and division of breast cells.² At the level of testosterone metabolism the cytochrome P450 2B6 (CYP2B6) enzyme mediates the 16-alpha- and 16-beta-hydroxylation, a key step in the deactivation of this hormone.⁵ CYP2B6 contributes about 2% to 5% of the total liver CYP content but it exhibits about 300-fold variability of expression and has been suggested to increase with age.⁶ It is expressed in liver as well as in extra-hepatic tissues such as the breast, notably, levels of CYP2B6 vary between normal breast tissue and breast tumors, with higher levels in tumors.^7,8 In line with the in vitro observation of the transcriptional regulation of CYP2B6 by the estrogen receptor (ER), ER positive breast tumors show higher CYP2B6 levels than ER negative tumors.^9,10

The CYP2B6 gene and enzyme activity are highly polymorphic,⁶ raising the possibility that functional polymorphisms in this gene influence testosterone levels and breast cancer risk. Currently 29 distinct star alleles have been identified in CYP2B6, of which the *6 allele including two amino acid changes Gln172His (CYP2B6_516_G>T, rs3745274) and Lys262Arg (CYP2B6_785_A>G, rs2279343) is the most common at a frequency of 15% to 60% across populations and being associated with 50 to 75% decreased hepatic liver expression.^6,11-13 However, CYP2B6 polymorphisms have not been among the loci recently associated with breast cancer susceptibility in genome-wide association studies (GWAS).^14-16 Importantly, this may be in part explained by the absence of polymorphisms in gene regions with high sequence similarities to other genes or pseudo genes from GWAS arrays because specific genotyping requires an adapted primer design and amplification conditions.¹⁴ Thus, discrimination of CYP2B6 and CYP2B7 pseudogene genotypes requires unique low plex genotyping assays or small scale microarrays.¹⁷ Using this technical improvement and following the hypothesis that a genetically determined decreased CYP2B6 activity increases testosterone levels and consequently breast cancer risk, we performed a two-stage breast cancer association study by genotyping the CYP2B6_516_G>T and CYP2B6_785_A>G polymorphisms in the GENICA breast cancer-case control study. This was followed by validation in independent case-control study collections from Germany, Sweden and the USA and a pooled analysis. The study involving 4,638 breast cancer cases and 3,594 controls showed that CYP2B6*6 is associated with increased breast cancer risk.

Material and Methods

Study populations

Five breast cancer case-control study collections from Germany, Sweden and the USA were included in this two-stage association analysis comprising a total of 4,638 breast cancer cases and 3,594 controls of European descent.

The population-based GENICA (Gene Environment Interactions and Breast Cancer study) from Germany recruited breast cancer cases and age-matched control from the Greater Bonn Region, between 08/2000 and 9/2004. Cases and controls were eligible if they were of Caucasian ethnicity, current residents of the study region, and below 80 years of age. Cases were enrolled within six month of diagnosis. Controls were selected from population registries from 31 communities in the study regions.^18,19 The German hospital-based BBCC (Bavarian Breast Cancer Cases and Controls) recruited women of European descent from 2002 to 2006. Invasive breast cancer cases were recruited at the University Breast Centre, Franconia in Northern Bavaria. Controls were invited by a newspaper advertisement in Northern Bavaria. Healthy women aged 55 or older with no diagnosis of cancer were enrolled.²⁰ The HaBCS (Hannover Breast Cancer Study) from Germany recruited breast cancer cases from the Hannover Medical School between 1997 and 2003. Controls were collected randomly from the female blood bank donors at Hannover Medical School between 8/2005 and 12/2005. They were not age-matched but residing in the same geographic region.²¹ The population-based SASBAC (Singapore and Sweden Breast Cancer Study) from Sweden recruited patients and controls of European descent between 10/1993 and 03/1995. Incident cases were identified via the regional cancer registries in Sweden. Age-matched controls were randomly selected from the total population registry.²² The US hospital-based MCBCS (Mayo Clinic Breast Cancer Study) recruited participants of European descent residing in 6 states (MN, WI, IA, IL, ND, SD) between 2002 and 2005. Incident breast cancer cases were enrolled within six month of diagnosis. Controls were recruited from women without a prior history of cancer visiting Mayo Clinic for general medical examination. Cases and controls were matched by age, ethnicity and state. ²³

Genotyping

Blood-derived DNA samples of all breast cancer cases and controls were genotyped by matrix assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS). Due to sequences homologies resulting in more than one hit in the human genome in sequence alignment an automatic assay design was hampered for the CYP2B6_516_G>T and CYP2B6_785_A>G polymorphism. Therefore, specific primers were designed by eye inspection of the sequence. To enhance specific primer binding an adapted PCR program was used: 94°C for 2 min followed by 5 cycles with 95°C for 20 sec, 65°C for 30 sec, and 72°C for 1 min, followed by 40 cycles with 95°C for 20 sec, 62°C for 30 sec, and 72°C for 1 min, followed by 72°C for 10 min. MALDI-TOF MS analysis was performed using the Sequenom MassARRAY platform and homogenous MassEXTEND methodology (Sequenom, San Diego, CA, USA) according to manufacturer’s instructions. For all MALDI-TOF MS analyses, a Sequenom Compact MALDI-TOF MS was used for data acquisitions from the SpectroCHIP. Genotyping calls were made with MASSARRAY RT software v 3.0.0.4 (Sequenom, San Diego, CA, USA). For quality control no template controls and positive controls with known genotypes were included. In addition, repeated analysis was performed for 20% randomly selected samples. Primers were synthesized by Metabion International AG, Martinsried, Germany. Primer sequences are available upon request.

Statistical analyses

Genotype frequencies of all polymorphisms were tested for Hardy-Weinberg Equilibrium. Associations between CYP2B6 genotypes and over all breast cancer risk were analyzed by logistic regression adjusted for age and study using SPSS v 15.0. Risk estimates were given in odds ratio (OR) and 95% confidence interval (95% CI). Analysis for a potential association between genotypes and respective tumor steroid hormone receptor status was done for estrogen receptor (ER) and progesterone receptor (PR) status. To improve the strength of the analysis we applied a multi-step analysis: First a pilot study was performed including the GENICA study only. In a next step a validation study was done comprising four independent study collections: BBCC, HABCS, MCBCS and SASBAC Finally, we performed a pooled analysis including all five studies.

Results

CYP2B6_516_G>T and CYP2B6_785_A>G were genotyped and call rates were >98%. Positive controls and repeated samples showed 100% congruency. No genotype call was observed in the negative controls. Genotype frequencies were in Hardy-Weinberg equilibrium and allele frequencies are in line with published data.²⁴

Risk associations were observed for overall breast cancer risk for both CYP2B6 variants. At the CYP2B6_516 locus we observed a significant association between the TT genotype and breast cancer risk in the pilot study (OR 1.34; 95% CI 1.12-1.59; p = 0.001). An association with breast cancer was also observed for carriers of the GT genotype (OR of 1.12; 95% CI 1.02-1.24; p = 0.020; Table 1) and the combined group of carriers of the GT and TT genotypes (OR 1.10; 95% CI 1.01-1.21; p = 0.042; Table 1) in the validation study. At the CYP2B6_785 locus we observed a significant risk effect for homozygous carriers of the G allele (OR 1.31; 95% CI 1.10-1.57; p = 0.002). This association was not observed in the validation study.

Table 1.

Genotype frequencies and estimated risks of CYP2B6_516_G>T (rs3745274) and CYP2B6_785_A>G (rs2279343) in breast cancer cases and controls

Study	CYP2B6 Genotypes	Cases n(%)	Controls n (%)	ORa (95% CI)	p-value
Pilot study GENICA	516_G>T
	GG	577 (56.7)	590 (58.5)	1.00b
	GT	344 (33.8)	363 (36.0)	0.97 (0.80-1.17)	0.741
	TT	96 (9.5)	55 (5.5)	1.34 (1.12-1.59)	0.001
	GT+TT	440	418	1.08 (0.90-1.24)	0.413
	785_A>G
	AA	577 (56.7)	596 (59.1)	1.00b
	AG	347 (34.2)	357 (35.4)	1.00 (0.83-1.21)	0.967
	GG	92 (9.1)	55 (5.5)	1.31 (1.10-1.57)	0.002
	AG+GG	439	412	1.10 (0.92-1.31)	0.287
Validation study BBCC, HaBCS, MCBCS, SASBAC	516_G>T
	GG	2631 (57.8)	2132 (60.5)	1.00b
	GT	1660 (36.5)	1177 (33.4)	1.12 (1.02-1.24)	0.020
	TT	257 (5.7)	215 (6.1)	0.99 (0.90-1.10)	0.886
	GT+TT	1917	1392	1.10 (1.01-1.21)	0.042
	785_A>G
	AA	2494 (57.2)	1991 (59.2)	1.00b
	AG	1578 (36.2)	1147 (34.1)	1.09 (0.99-1.21)	0.081
	GG	290 (6.6)	226 (6.7)	1.03 (0.93-1.13)	0.589
	AG+GG	1868	1373	1.09 (0.99-1.19)	0.086
Pooled analysis BBCC, HaBCS, MCBCS, SASBAC and GENICA	516_G>T
	GG	3208 (57.7)	2722 (60.0)	1.00b
	GT	2004 (36.0)	1540 (34.0)	1.09 (1.00-1.19)	0.048
	TT	353 (6.3)	270 (6.0)	1.07 (0.98-1.16)	0.143
	GT+TT	2357	1810	1.10 (1.01-1.19)	0.027
	785_A>G
	AA	3071 (57.1)	2587 (59.2)	1.00b
	AG	1925 (35.8)	1504 (34.4)	1.08 (0.99-1.18)	0.078
	GG	382 (7.1)	281 (6.4)	1.09 (1.00-1.18)	0.053
	AG+GG	2307	1785	1.10 (1.01-1.19)	0.031

^aOR adjusted for age and study

^bReference

Abbreviations: CI = confidence interval, OR = odds ratio

The pooled analysis of all five study collections confirmed the association with breast cancer risk for both CYP2B6 variant loci. The heterozygous CYP2B6_516_GT carriers and the combined group of GT and TT carriers showed an increased breast cancer risk (OR 1.09; 95% CI 1.00-1.19; p = 0.048; and OR of 1.10 (95% CI: 1.01-1.19 p = 0.027; Table 1). The association with breast cancer risk was also observed for carriers of the CYP2B6_785_G variant OR of 1.10 (95% CI 1.01-1.19; p = 0.031; Table 1). No significant associations with breast cancer risk were observed when cases were stratified by ER or PR status of breast tumors.

Discussion

In this work we successfully established specific MALDI-TOF MS genotyping assays for the genotyping of CYP2B6 variants located in genomic sequences with high homology. This enabled us to perform breast cancer risk association studies at CYP2B6*6 polymorphic loci known to modify CYP2B6 enzymatic activity. We observed an increased breast cancer risk associated with both variants constituting the CYP2B6*6 polymorphism, e.g. CYP2B6_516_T and CYP2B6_785_G alleles.

Although the increase in breast cancer risk has been observed at both loci, they were not consistent between pilot and validation studies, both with respect to genotype groups and effect size. However, it is important to note, that risk effects were similar at both loci for the total combined groups of carriers of the respective variant alleles, i.e. CYP2B6_516_GT and TT carriers or CYP2B6_785_AG and GG carriers, with OR 1.10 (95% CI 1.01-1.19), respectively. Because both alleles together constitute the CYP2B6*6 polymorphism this suggests for the first time a role of this polymorphism in breast cancer risk. The observed ORs are in the range of the so far established common low-penetrance breast cancer susceptibility loci with ORs from 1.07 to 1.22 including polymorphisms in genes such as FGFR2 and TOX3. As of yet, this set of more than 20 known susceptibility loci accounts for about 8% of the heritability of breast cancer.¹⁶ Therefore, it will be of particular interest to evaluate CYP2B6*6 in concert with all known risk factors in large study collections to establish a more comprehensive breast cancer susceptibility panel.

The observed risk effect is in line with the current understanding of the role of CYP2B6 in steroid hormone metabolism.⁵ Decreased enzyme activity caused by CYP2B6_516_T and CYP2B6_785_G, designated CYP2B6*6, potentially increases steroid hormone levels, a known breast cancer risk factor.²⁵ It is of note, that a similar steroid hormone-related carcinogenic effect of CYP2B6 has been shown for prostate cancer.^26,27 Recently a report on an association between the CYP2B6_516_T variant and leukemia suggested that low CYP2B6 activity decreased degradation of carcinogenic xenobiotoc compounds which promotes tumor cell formation.²⁸ Taking into consideration that CYP2B6 is one of the most polymorphic CYPs in humans but at the same time is one of the less well-studied CYP enzymes,^6,29 our observation of a role in breast cancer risk together with the published literature highlights the need to overcome possible technical issues and further explore its role in cancer.

What’s new?

A multistage study of more than 8,000 breast cancer cases and controls suggested that the CYP2B6*6 allele, which influences CYP2B6 activity, is associated with an increased risk of breast cancer. This finding adds CYP2B6 to the candidate genes for the evolving panel of known breast cancer susceptibility loci and implicates CYP2B6 in the mechanism by which steroid hormone levels influence breast cancer risk.