CYP19A1 gene polymorphisms and risk of lung cancer

Abstract

Objectives

To investigate the association between polymorphisms of aromatase (encoded by the CYP19A1 gene and a key enzyme in biosynthesis of oestradiol) and the risk of lung cancer, and whether there were differences stratified by sex and smoking history.

Methods

This case–control study included consecutive, nonselected and pathologically-confirmed lung cancer patients and healthy people. Participants were classed as nonsmokers or smokers by questionnaire. Peripheral blood samples from all participants were genotyped for three single-nucleotide polymorphism (SNPs; rs727479, rs730154 and rs10046); allelic frequencies were compared across genotype and clinical records.

Results

A total of 529 patients with lung cancer and 567 age- and sex-matched controls were included. After adjustment for age, sex and smoking history, rs727479 was significantly associated with the incidence of lung cancer (for alleles AC vs AA). There was also a significant difference between patients and controls in haplotype CCA, while haplotype ACA was only significantly associated with nonsmokers and female nonsmokers.

Conclusions

Polymorphisms of CYP19A1 may be related to the increased risk of lung cancer; in particular, haplotype ACA may contribute to lung-cancer progression in nonsmokers. Further validation with larger populations is required.

Keywords

Lung cancer aromatase single-nucleotide polymorphism genotype haplotype allele frequency

Introduction

Lung cancer has the highest death rate worldwide among malignancies, with prevalence and mortality rates seemingly increasing more rapidly in females than in males.^1,2 Lung cancer in women is more likely to be adenocarcinoma (ADC); women who have never smoked have a higher risk of lung cancer than men who have never smoked.³ The proportion of lung cancers occurring among people who have never smoked is 15–25% among women, but only 10% among men.⁴ These facts are attracting research attention to risk factors for lung cancer in nonsmokers, especially the involvement of the oestrogen system.

Studies indicate that the oestrogen system is involved in the biological behaviour of lung cancer. A higher level of oestrogens in the blood is found in females with nonsmall-cell lung cancer (NSCLC) compared with healthy women of the same age.⁵ Epidemiological and laboratory-based studies have also suggested that tumour expression of oestrogen may influence NSCLC pathogenesis, and the prognosis of all lung cancer patients.^1, ⁶ Furthermore, there is a positive correlation between high tumour aromatase and oestrogen levels: tumour aromatase levels may predict survival in women with early-stage NSCLC.⁷

The cytochrome P (CYP) 450, family 19, subfamily A, polypeptide 1 (CYP19A1) encodes aromatase, which is a key enzyme in oestradiol biosynthesis^8–9 and a member of the CYP 450 superfamily of enzymes. This protein localizes in the endoplasmic reticulum and catalyses three successive hydroxylations of the ring of androgen, which is the last step of oestrogen biosynthesis. Mutations in the CYP19A1 gene can result in either increased or decreased aromatase activity.¹⁰

The present study explored genetic risk factors in lung-cancer development by analysing three single-nucleotide polymorphisms (SNPs) of the CYP19A1 gene, in consecutive lung-cancer cases and healthy controls from the same geographic area during the same time period, paying particular attention to smoking status and sex.

Patients and methods

Study population

Consecutive, nonselected, pathologically-confirmed lung cancer patients attending the Shanghai Cancer Hospital, Fudan University, Shanghai, China, between January 2005 and December 2008, were included in the study. All cases were diagnosed as ADC, squamous cell carcinoma or small-cell lung cancer by histological examination, according to the World Health Organization 2004 tumour classification system;¹¹ diagnoses were confirmed by two independent pathologists. There were no other inclusion or exclusion criteria in this study. Healthy residents with no history of cancer, living in 11 communities in Pudong New Area and Baoshan District, Shanghai, China, were recruited as controls during the same time period.

Patients and controls provided written informed consent and the study was approved by the Ethics Committee of the Shanghai Institute for Biological Sciences, Shanghai, China.

Smoking history of patients and controls was obtained via questionnaire interview during clinic visits or at hospital admission. The smoking level of each participant was expressed as ‘cigarettes per day’, i.e. the number of cigarettes smoked per day on most days. Smoking exposure was expressed as ‘pack-years’, defined as number of packs per day (i.e. cigarettes per day divided by 20) multiplied by years of smoking. ‘Smokers’ were defined as those who had smoked regularly for ≥12 months at any time in their life; ‘nonsmokers’ were defined as those who had never smoked or who had stopped smoking for a period of 12 months at some time, and were not current smokers.

Genotyping

Three SNPs of CYP19A1 (rs727479, rs730154, and rs10046) were used in the study: these had minor allele frequencies (MAFs) > 10% in the Chinese Han population, according to the dbSNP database (http://www.ncbi.nlm.nih.gov/SNP/). Two SNP (rs727479, rs730154) have been reported previously.¹² Genotype frequencies for each polymorphism were calculated for patients and controls, stratifying by the three SNPs.

Standard blood samples were collected from all participants during clinic visits or at hospital admission. Samples were treated with 1.8 mg/ml ethylenediamine tetra-acetic acid to prevent coagulation and stored at −80℃ until further analysis.

Genomic DNA was extracted from 2-ml peripheral blood samples using TIANamp Blood DNA Kit (DP318; TIANGEN Biotech [Beijing] Co. Ltd, Beijing, China) according to the manufacturer’s instructions. Genomic DNA was amplified and the CYP19A1 polymorphisms were genotyped, using a gene-specific TaqMan® real-time PCR assay kit (Invitrogen, Carlsbad, CA, USA) and the ABI Prism® 7900HT Sequence Detection System (Applied Biosystems, Foster City, CA). Primers for genotyping and the cycling programme were undertaken, as recommended by the manufacturers.

Genotyping was performed by laboratory personnel who were blinded to patient status, and a random 5% of the samples were repeated to validate genotyping procedures. The authors independently reviewed genotyping results, data entry and statistical analyses. Genotype frequencies were calculated using the Hardy–Weinberg equilibrium test, which was performed using the SHEsis software (available at http://analysis2.bio-x.cn/myAnalysis.php). Calculation of linkage disequilibrium (D′ and r² values), as well as haplotype estimation, was undertaken using the expectation–maximization algorithm of the same software.

Statistical analyses

Data were compared across genotypes and clinical records using Pearson’s χ²-tests (for categorical variables) and Kruskal–Wallis tests (for continuous variables) where appropriate. The strength of association of genotypes was measured as odds ratios (ORs) with 95% confidence intervals (95% CIs), using an unconditional logistic regression analysis adjusted for sex, age and smoking status. These statistical analyses were performed using SPSS® software, version 15.0 (SPSS Inc., Chicago, IL, USA). A level of P < 0.05 was considered statistically significant.

Results

A total of 1096 participants were included in the study: 529 patients with lung cancer and 567 controls (Table 1). Clinical characteristics and smoking status are shown in Table 1. There were 66.1% females in the control group and 32.9% females in the patient group. Compared with controls, patients with lung cancer had a higher rate of smoking and smoked a higher number of cigarettes per day. Among the patients with lung cancer, most females had ADC and those with ADC had the lowest smoking rate. Baseline data were not tested for statistical significance.

Table 1.

Clinical characteristics and smoking status of Chinese patients with lung cancer and healthy controls, included in a study assessing the risk of lung cancer associated with polymorphisms of the cytochrome P450, family 19, subfamily A, polypeptide 1 (CYP19A1, aromatase) gene.

Variable	Controls	Patients
Variable	Controls	All	ADC	SQC	SCLC
n	567	529	356	131	42
Age, years	59 ± 9	58 ± 10	57 ± 10	61 ± 9	60 ± 10
Sex
Male	192 (33.9)	355 (67.1)	191 (53.7)	124 (94.7)	40 (95.2)
Female	375 (66.1)	174 (32.9)	165 (46.3)	7 (5.3)	2 (4.8)
Smoking status
Nonsmoker	480 (84.7)	238 (45.0)	218 (61.2)	14 (10.7)	6 (14.3)
Smoker	87 (15.3)	291 (55.0)	138 (38.8)	117 (89.3)	36 (85.7)
Pack-years	27.9 ± 18.7	38.8 ± 22.5	32.8 ± 20.5	45.0 ± 20.5	41.9 ± 29.6
Cigarettes per day, n	16.3 ± 8.7	23.9 ± 11.6	22.4 ± 11.8	25.5 ± 10.2	24.0 ± 14.3
Smoking duration, years	34.3 ± 10.6	31.9 ± 10.4	28.7 ± 10.3	35.2 ± 8.8	33.6 ± 11.5

Data presented as mean ± SD or n (%) of participants.

ADC, adenocarcinoma; SQC, squamous; SCLC, small-cell lung cancer.

Baseline data were not tested for statistically significant between-group differences.

Gene analysis findings were compared among patients and control groups in nonsmokers and female nonsmokers. No significant differences were identified in allelic differentiation between patients with lung cancer and controls (Table 2). After adjustment for sex and smoking history, there was no significant association between the allele frequencies of CYP19A1 SNPs and lung cancer.

Table 2.

Allele frequency of three single nucleotide polymorphisms (rs727479, rs730154, rs10046) of the cytochrome P450 (CYP), family 19, subfamily A, polypeptide 1 (CYP19A1, aromatase) gene, in Chinese patients with lung cancer (n = 529) and healthy controls (n = 567).

SNP	Allele	Category	Control	Patients	OR 95% confidence interval
rs727479	A	All	800 (71.3)	670 (73.8)	1.13 (0.93, 1.38)
Nonsmokers	677 (71.4)	283 (71.5)	1.00 (0.77, 1.30)
Female non-smokers	517 (71.2)	207 (71.9)	1.03 (0.76, 1.40)
rs730154	T	All	667 (65.6)	694 (67.5)	0.92 (0.77, 1.11)
Nonsmokers	578 (66.4)	303 (65.9)	1.03 (0.81, 1.30)
Female non-smokers	446 (66.0)	220 (66.7)	0.97 (0.73, 1.28)
rs10046	G	All	476 (42.9)	396 (43.0)	1.00 (0.84, 1.19)
Nonsmokers	389 (44.1)	204 (44.0)	1.01 (0.80, 1.26)
Female non-smokers	301 (44.3)	147 (43.8)	1.02 (0.79, 1.33)

Data presented as n (%) allele frequency.

Genotype data were missing for 81/1096 participants for rs727479, 7/1096 participants for rs730154 and 80/1096 participants for rs10046.

No significant between-group differences were observed (P ≥ 0.05).

Estimated haplotype frequencies and associations for patients with lung cancer and controls are presented in Table 3. The difference between patients with lung cancer and controls in haplotype CCA was significant (OR 0.57 [95% CI 0.36, 0.91]), but no differences were observed for nonsmokers or female nonsmokers. Haplotype ACA was significantly associated with lung cancer in nonsmokers (OR 1.46 [95% CI 1.01, 2.11]) and in female nonsmokers (OR 1.55 [95% CI 1.02, 2.35]).

Table 3.

Estimated haplotype frequencies of three single nucleotide polymorphisms (rs727479, rs730154, rs10046) of the cytochrome P450 (CYP), family 19, subfamily A, polypeptide 1 (CYP19A1, aromatase) gene, in Chinese patients with lung cancer (n = 529) and healthy controls (n = 567).

rs727479	rs730154	rs10046	Control	Patients	OR 95% confidence intervals
All participants
A	C	A	141 (14.3)	115 (14.7)	1.04 (0.79, 1.36)
C	C	A	58 (5.9)	27 (3.5)	0.57 (0.36, 0.91)*
C	T	A	100 (10.1)	74 (9.5)	0.94 (0.69, 1.29)
Nonsmokers
A	C	A	81 (10.4)	53 (14.5)	1.46 (1.01, 2.11)*
C	C	A	57 (7.3)	16 (4.4)	0.59 (0.34, 1.04)
C	T	A	64 (8.2)	40 (10.8)	1.35 (0.89, 2.05)
Nonsmoking females
A	C	A	64 (10.6)	42 (15.5)	1.55 (1.02, 2.35)*
C	C	A	44 (7.2)	14 (5.1)	0.69 (0.37, 1.29)
C	T	A	64 (10.5)	29 (10.5)	1.00 (0.63, 1.59)

Data presented as n (%) haplotype frequency.

Those with frequency < 0.03 in both control and case have been excluded.

Genotype data were missing for 81 of 1096 participants for rs727479, 74 of 1096 participants for rs730154 and 80 of 1096 participants for rs10046.

Statistically significant differences observed.

Table 4 shows lung cancer risk associated with each genotype. After adjustment for age, sex and smoking status, multivariate analysis showed that rs727479 was significantly associated with lung cancer incidence (OR 0.73 [95% CI 0.55, 0.98]). This association was not identified among nonsmokers or female nonsmokers. Neither rs730154 nor rs10046 showed any significant differences, whether stratified by sex or smoking history.

Table 4.

Genotype frequency of three single nucleotide polymorphisms (rs727479, rs730154, rs10046) of the cytochrome P450 (CYP), family 19, subfamily A, polypeptide 1 (CYP19A1, aromatase) gene in Chinese patients with lung cancer (n = 529) and healthy controls (n = 567).

Category	Genotype	Controls	Patients	Adjusted OR 95% confidence intervals
rs727479
All^a	AA	273 (48.7)	248 (54.6)	Reference
AC	254 (45.3)	174 (38.3)	0.73 (0.55, 0.98)*
CC	34 (6.1)	32 (7.0)	1.32 (0.75, 2.30)
Nonsmokers^b	AA	233 (49.2)	105 (53.0)	Reference
AC	211 (44.5)	73 (36.9)	0.77 (0.54, 1.10)
CC	30 (6.3)	20 (10.1)	1.50 (0.81, 2.78)
Female nonsmokers^c	AA	177 (48.8)	76 (52.8)	Reference
AC	163 (44.9)	55 (38.2)	0.79 (0.53, 1.19)
CC	23(6.3)	13 (9.0)	1.33 (0.64, 2.77)
rs730154
All^a	CC	52 (9.8)	58 (11.9)	Reference
CT	244 (45.8)	219 (44.8)	0.85 (0.54, 1.35)
TT	237 (44.5)	212 (43.4)	0.78 (0.49, 1.24)
Nonsmokers^b	CC	47 (10.5)	22 (10.2)	Reference
CT	208 (46.3)	103 (47.7)	1.06 (0.60, 1.86)
TT	194 (43.2)	91 (42.1)	0.99 (0.56, 1.75)
Female nonsmokers^c	CC	35 (10.1)	15 (9.7)	Reference
CT	162 (46.6)	78 (50.3)	1.10 (0.57, 2.14)
TT	151 (43.4)	62 (40.0)	0.94 (0.48, 1.85)
rs10046
All^a	AA	173 (30.6)	151 (33.5)	Reference
AG	282 (49.9)	230 (51.0)	0.97 (0.71, 1.32)
GG	110 (19.5)	70 (15.5)	0.78 (0.52, 1.18)
Nonsmokers^b	AA	144 (30.1)	63 (32.3)	Reference
AG	241 (50.4)	98 (50.3)	0.93 (0.64, 1.36)
GG	93 (19.5)	34 (17.4)	0.84 (0.51, 1.37)
Female nonsmokers^c	AA	109 (29.8)	47 (33.1)	Reference
AG	182 (49.7)	74 (52.1)	0.95 (0.61,1.47)
GG	75 (20.5)	21(14.8)	0.65 (0.36,1.18)

Data presented as n (%) genotype frequency.

Adjusted for: ^aage, sex and smoking status; ^bage and sex; ^cage.

Genotype data were missing for 81/1096 participants for rs727479, 74/1096 participants for rs730154 and 80/1096 participants for rs10046.

Statistically significant differences observed.

Discussion

It has been reported that higher levels of aromatase may predict prognosis among female patients with lung cancer.^13,14 Additionally, an aromatase inhibitor, exemestane, has already been shown to reduce proliferation effectively in NSCLC cell lines.¹⁴

The present study assessed the possible association between SNPs of the CYP19A1 gene and the risk of lung cancer by evaluating the allele, haplotype and genotype frequencies of three SNPs (rs727479, rs730154, and rs10046), in smoking and nonsmoking lung cancer patients and healthy controls. The haplotype CCA was associated with the presence of lung cancer whereas the haplotype ACA was significantly associated with lung cancer in all non-smokers, and in nonsmoking females. The study also showed that one locus rs727479 at 15q21.1 may be regarded as a potential disease candidate in lung cancer. It was hypothesized that, when mutation of the three SNPs presented as haplotype ACA, expression of aromatase would be upregulated. This would directly contribute to the upregulation of oestrogen in the patient and contribute to the nonsmoking pathway of lung cancer. The haplotype CCA also possibly contributes to the progression of lung cancer, but has no strong relationship with lung cancer in nonsmokers. As the statistical significance of the correlations between CYP19A1 gene polymorphism and the risk of lung cancer were often marginal, this study needs further validation with larger populations.

Existing epidemiological studies that examine whether oestrogen is a risk factor for lung cancer are inconsistent.^15,16 Associations between SNPs related to oestrogen metabolism and lung cancer have been investigated in numerous studies, with variable results.^17–19 It has been suggested that such inconsistency is possibly due to linkage disequilibrium across these loci. In other words, different choices of SNP locus result in the conflicting findings. Race differences (and the use of different techniques, such as genome-wide association studies) can also contribute to inconsistencies.

In summary, the present study suggests that polymorphisms of CYP19A1 may be associated with an increased risk of lung cancer. In particular, the haplotype ACA may contribute to the progression of lung cancer in nonsmokers.

Footnotes

Declaration of conflicting interests

The authors declare that there are no conflicts of interest.

Funding

This work is supported, in part, by the Pujiang Talent grant, Longhua Medicial Project, Young Investigator grant from Shanghai Municipal Health Bureau, and the Zhuo Shi grant from Fudan University, Ministry of Education Returned Scientific Research Foundation (to J. Z.).

References

Cote

Yoo

Wenzlaff

. Tobacco and estrogen metabolic polymorphisms and risk of non-small cell lung cancer in women. Carcinogenesis 2009; 30: 626–635.

Siegel

Naishadham

Jemal

. Cancer statistics, 2013. CA Cancer J Clin 2013; 63: 11–30.

Zang

Wynder

. Differences in lung cancer risk between men and women: examination of the evidence. J Natl Cancer Inst 1996; 88: 183–192.

Subramanian

Govindan

. Lung cancer in never smokers: a review. J Clin Oncol 2007; 25: 561–570.

Tiutiunova

Chirvina

Mironenko

. Hormonal balance in women with lung cancer and its changes after combined treatment. Vopr Onkol 1986; 32: 26–30. in Russian, English abstract].

Demura

Ameshima

. Changes in aromatase (CYP19) gene promoter usage in non-small cell lung cancer. Lung Cancer 2011; 73: 289–293.

Verma

Miki

Sasano

. Aromatase in human lung carcinoma. Steroids 2011; 76: 759–764.

Drakoulis

Cascorbi

Brockmöller

. Polymorphisms in the human CYP1A1 gene as susceptibility factors for lung cancer: exon-7 mutation (4889 A to G), and a T to C mutation in the 3'-flanking region. Clin Investig 1994; 72: 240–248.

Means

Mahendroo

Corbin

. Structural analysis of the gene encoding human aromatase cytochrome P-450, the enzyme responsible for oestrogen biosynthesis. J Biol Chem 1989; 264: 19385–19391.

10.

Timofeeva

Kropp

Sauter

. CYP450 polymorphisms as risk factors for early-onset lung cancer: gender-specific differences. Carcinogenesis 2009; 30: 1161–1169.

11.

World Health Organization Classification of Tumours (2004). Pathology and Genetics of Tumours of the Lung, Pleura, Thymus and Heart. Available at: http://www.iarc.fr/en/publications/pdfs-online/pat-gen/bb10/index.php (accessed17 April 2013).

12.

Colomer

Monzo

Tusquets

. A single-nucleotide polymorphism in the aromatase gene is associated with the efficacy of the aromatase inhibitor letrozole in advanced breast carcinoma. Clin Cancer Res 2008; 14: 811–816.

13.

Mah

Seligson

. Aromatase expression predicts survival in women with early-stage non-small cell lung cancer. Cancer Res 2007; 67: 10484–10490.

14.

Koutras

Giannopoulou

Kritikou

. Antiproliferative effect of exemestane in lung cancer cells. Mol Cancer 2009; 8: 109–109.

15.

Liu

Inoue

Sobue

. Reproductive factors, hormone use and the risk of lung cancer among middle-aged never-smoking Japanese women: a large-scale population-based cohort study. Int J Cancer 2005; 117: 662–666.

16.

Taioli

Wynder

. Re: Endocrine factors and adenocarcinoma of the lung in women. J Natl Cancer Inst 1994; 86: 869–870.

17.

Hung

Hall

Brennan

. Genetic polymorphisms in the base excision repair pathway and cancer risk: a HuGE review. Am J Epidemiol 2005; 162: 925–942.

18.

Spivack

Hurteau

Fasco

. Phase I and II carcinogen metabolism gene expression in human lung tissue and tumours. Clin Cancer Res 2003; 9: 6002–6011.

19.

Kohno

Kakinuma

Iwasaki

. Association of CYP19A1 polymorphisms with risks for atypical adenomatous hyperplasia and bronchioloalveolar carcinoma in the lungs. Carcinogenesis 2010; 31: 1794–1799.