Sage Journals: Discover world-class research

Abstract

Variable association of transforming growth factor beta 1 (TGFβ1) in breast cancer (BC) pathogenesis was documented, and the contribution of specific TGFB1 polymorphisms to the progression of BC and associated features remains poorly understood. We investigated the contribution of TGFB1 rs1800469, rs1800470, rs1800471, and rs1800472 variants and 4-locus TGFB1 haplotypes on BC susceptibility, and pathological presentation of BC subtypes. Study subjects comprised 430 female BC cases, and 498 cancer-free control women. BC-associated pathological parameters were also evaluated for correlation with TGFB1 variants. Results obtained showed that the minor allele frequency (MAF) of rs1800471 (+74G>C) was higher seen in BC cases than in control subjects, and was associated with increased risk of BC. Significant differences in rs1800471 and rs1800469 (−509C>T) genotype distribution were noted between BC cases and controls, which persisted after controlling for key covariates. TGFB1 rs1800472 was positively, while rs1800470 was negatively associated with triple negativity, while rs1800470 positively correlated with menarche, but negatively with tumor size and molecular type, and rs1800469 correlated positively with menstrual irregularity, distant metastasis, nodal status, and hormonotherapy. Heterogeneity in LD pattern was noted between the tested TGFB1 variants. Four-locus (rs1800472-rs1800471-rs1800470-rs1800469) Haploview analysis identified haplotype TGCT to be negatively associated, and haplotypes CGTT and CCCC to be positively associated with BC. This association of CGTT and CCCC, but not TGCT, with BC remained significant after controlling for key covariates. In conclusion, TGFB1 alleles and specific genotypes, and 4-locus TGFB1 haplotypes influence BC susceptibility, suggesting dual association imparted by specific SNP, consistent with dual role for TGFB1 in BC pathogenesis.

Keywords

Breast cancer genotypes haplotypes transforming growth factor–β1

Introduction

Breast cancer (BC) is the most prevalent cancer among women, with more than 2.08-million new cases of BC identified globally in 2018.¹ BC is the leading cancer in women in Tunisia, with reported high incidence (15.61%) and mortality (8.43%) rates, and currently ranks as the second cause of female mortality in Tunisia.¹ The pathogenesis and progression of BC requires participation of modifiable factors such as lifestyle, treatment with xenobiotics, and environmental factors, and nonmodifiable factors which include age, racial, and genetic background.¹ Current screening tools include radiological (mammography), ultrasonography, genetic (BRCA), and biochemical techniques.^2,3 This prompts search for novel diagnostic and prognostic biomarkers, with sufficient sensitivity and/or specificity.

Earlier studies identified common germline variants associated with altered BC susceptibility.^4,5 Of these, transforming growth factor beta 1 (TGFβ1), a pleiotropic anti-inflammatory cytokine secreted by cancerous and non-cancerous cells, was extensively studied for its role in modulating the risk of BC.^6,7 TGFβ1 is the most abundant member of the TGFβ family of growth factors which also comprises TGFβ2 and TGFβ3, and is expressed by endothelial cells, connective tissues, and hematopoietic cells.⁸ TGFβ1 plays a crucial role in embryogenesis and development of mammary glands,⁹ and in many physiological and disease states, largely by controlling cellular differentiation and cell cycle progression.^10,11 Paradoxically, TGFβ1 has a dual role in cancer.^11,12 It was described to act as tumor suppressor in early carcinogenesis by inducing apoptosis or inhibiting cell growth, while promoting tumorigenicity and metastasis through suppression of local immunosurveillance and supporting tumor progression in advanced stages of cancer.^13,14

Located on chromosome 19 (19q13.1), TGFB1 gene contains several functional and non-functional polymorphisms, some of which were shown to contribute to altered BC susceptibility.^7,15–18 Of these, rs1800469 (–1347C>T) promoter, rs1800470 (+29T>C) codon 10, and rs1800471 (+74G>C) signal peptide sequence variants were widely studied in BC.^15,19–22 However, the association of TGFB1 variants with BC remains controversial,^12,15,17 and apparently varies in different ethnic groups.^12,23,24 No association^25,26 and weak/moderate association^25,27 were reported. These apparently contradictory findings were explained by ethnicity-related variations, along with the likely interactions with related and distant variants.

We previously demonstrated the association of genetic variants in progesterone receptor (PR),²⁸ estrogen receptor (ER),²⁹ and NF-κB gene³⁰ with altered risk of BC. Given the central role of TGFβ1 as inflammatory mediator in modulating processes associated with BC, and as TGFβ1 signaling is linked with NF-κB signaling,³¹ here we analyze the contribution of genetic variation in TGFB1 to BC susceptibility in Arab-speaking Tunisian women.

Subjects and methods

Study subjects

Between June 2017 and November 2018, 430 women with BC and 498 cancer-free control women were recruited into this study from the outpatient oncology and surgery oncology services of Salah Azaïz Institute Hospital (Tunis, Tunisia). All participants consented in writing to be included in the study. Assessment of BC was done according to American Cancer Society guidelines.³ These included mammography and breast biopsy testing for confirmation of BC; all patients had these tests done. Control women comprised volunteer from the community or women reporting for annual physical checkup, and none had personal or a family history of BC. Control women were matched to BC patients according to age and self-declared origin.

Demographic and clinical biodata were obtained from medical records and through personal interview by the referring physician or senior resident, using a unified structured questionnaire. Collected data included the age at entry of the study, age at primary diagnosis of BC, status of menopause, and metastatic disease at presentation. Histological assessment (disease stage, nuclear grade), ER and PR status, and treatment regimen (surgery, endocrine-based therapy, radiotherapy, chemotherapy) were collected for all BC cases. The study was done per Helsinki II declaration and was approved by the Research & Ethic Committee of Salah Azaïz Institute in Tunis (registration number: ISA/2018/19, granted on 25 June 2017).

Informed consent

Informed consent was obtained from all individual participants included in the study.

TGFB1 genotyping

Four single nucleotide polymorphisms (SNPs) in TGFB1 gene with clinical relevance and minor allele frequency (MAF) ≥4% in control Tunisian subjects were identified, using National Center for Biotechnology Information (NCBI) Entrez Gene SNP Geneview. These comprised the missense variants rs1800472 (+791C>T; position 41341955; assay number C___8708464_20), rs1800471 (+74G>C; position 41352971; assay number C__11464118_30), and rs1800470 (+29T>C; position 41353016; assay number C__22272997_10), and the 5′ near-gene variant rs1800469 (–1347C>T; position 41354391; assay number C___8708473_10). TGFB1 genotyping was performed by assay-on-demand (VIC- and FAM-labeled primers) TaqMan assays (Applied Biosystems, Dubai, UAE). The reaction was performed on StepOnePlus real-time PCR system (Applied Biosystems). The reproducibility of genotyping was ascertained by inclusion of replicate blinded quality control samples; concordance exceeded 99%, and average successful sample and SNP genotyping rate was 98.9%.

Statistical analysis

Statistical analysis was performed on SPSS version 24 (IBM, Armonk, NY). Continuous (quantitative) data which were normally distributed are shown as mean (±SD), while categorical (qualitative) variables are expressed as percent of total. Student’s t-test and Pearson χ² test were employed in testing differences in means and inter-group significance, respectively. Genetic Power Calculator (http://pngu.mgh.harvard.edu) was employed in calculation of study power, using the following parameters: 430 BC cases and 498 control women, MAF of tested variants (in BC cases), estimated BC prevalence in Tunisia, and genotypic relative risk for heterozygote (1/2), and minor allele homozygous (2/2). Accordingly, the overall power (87.3%) was determined as the average of the power of the four tested variants.

Haploview 4.2 (Broad Institute, Cambridge, MA) was utilized in the evaluation Hardy–Weinberg equilibrium (HWE), and to check for linkage disequilibrium (LD) between TGFB1 variants, and for estimation of haplotype patterns, which were reconstructed by the expectation maximization method. Taking control women as reference group (odds ratio (OR) = 1.00), logistic regression analysis was used in determination of the ORs and 95% confidence intervals (CIs) associated with the risk of BC. Statistical significance set at p < 0.05.

Results

Study subjects

Table 1 presents the demographic and clinical characteristics of study subjects. Significant differences between BC cases and control women were noted in body mass index (BMI; p < 0.001), menarche (p = 0.021), and number of smokers (p = 0.022). In addition, irregular menses (p < 0.001) and previous oral contraceptives use (p < 0.001) were more frequent in BC cases, and breast feeding frequency was lower in BC cases than in control women (p < 0.001). Accordingly, we selected as these as the main covariates that were controlled for in later analysis.

Table 1.

Characteristics of study subjects.

	Cases (430)	Controls (498)	p ^a
Age (years)^b	45.6 ± 9.3	46.8 ± 11.1	0.066
BMI (kg/m¹)^b	28.5 ± 4.8	27.1 ± 5.0	<0.001
Obesity (BMI > 30 kg/m²)^c	133 (30.9)	157 (31.9)	0.776
Menarche (years)^b	12.5 ± 1.4	12.2 ± 1.1	0.021
Smokers^c	26 (6.0)	14 (2.8)	0.022
Breast feeding^c	310 (72.1)	449 (90.2)	1.1 × 10⁻¹¹
Menstrual history^c
Regular	266 (61.9)	364 (73.1)	3.2 × 10⁻⁴
Irregular	164 (38.1)	134 (26.9)
Menopausal status^c
Pre-menopausal	220 (51.2)	264 (53.0)	0.598
Post-menopausal	210 (48.8)	234 (47.0)
Oral contraception users^c	128 (29.8)	74 (14.9)	4.8 × 10⁻⁷
Triple negative^c	102 (23.7)	N/A	N/A
ER positive^c	291 (67.7)	N/A	N/A
PR positive^c	224 (52.1)	N/A	N/A
ER positive/PR positive^c	205 (41.2)	N/A	N/A
HER-2 positive^c	117 (27.2)	N/A	N/A
ER positive/Her-2 negative^c	211 (49.1)	N/A	N/A

BMI: body mass index.

Student’s t-test (continuous variables), Pearson’s χ² (categorical variables).

Mean ± SD.

Number of subjects (percent total).

Association studies

MAF of the tested TGFB1 variants in BC cases and cancer-free control women is presented in Table 2. The genotype distributions of the four tested TGFB1 variants were in HWE among study subjects. Higher frequency of rs1800471 minor (“C”) allele (p = 0.003) was seen in BC cases than in control women and was associated with increased risk of BC (OR (95% CI) = 1.84 (1.22–2.78)). MAF of the remaining TGFB1 variants was comparable between BC cases and control subjects.

Table 2.

Distribution of TGFB1 alleles in breast cancer cases and control women.

SNP	Position^a	Alleles	HWE	Cases^b	Controls^b	χ²	p	OR (95% CI)
rs1800472	41341955	C > T	0.42	49 (0.06)	66 (0.07)	0.96	0.33	0.83 (0.56–1.21)
rs1800471	41352971	G > C	0.10	61 (0.08)	40 (0.04)	8.73	0.003	1.84 (1.22–2.78)
rs1800470	41353016	T > C	0.13	377 (0.44)	433 (0.45)	0.25	0.62	0.95 (0.79–1.15)
rs1800569	41354391	C > T	0.09	326 (0.39)	381 (0.40)	0.09	0.76	0.97 (0.80–1.17)

TGFB1: transforming growth factor beta 1; SNP: single nucleotide polymorphism; HWE: Hardy–Weinberg equilibrium; OR: odds ratio; CI: confidence interval.

Boldface indicates minor allele.

Location on chromosome based on dbSNP build 125.

Minor allele number (frequency).

Table 3 illustrates the distribution of TGFB1 genotypes in BC cases and control women. The distribution of rs1800469 C/T (p = 0.05) and rs1800471 G/C (p = 0.003) genotypes was significantly different between BC cases and control women. This association remained significant after controlling for the main covariates (age, BMI, menarche and menses pattern, previous use of oral contraceptives, smoking, and breast feeding). The overall BC risk associated with a specific TGFB1 genotype was further confirmed by testing the effect of BMI, menarche, smoking, and oral contraceptive use as confounders. Results from Table 4 confirmed the association of rs1800471/+74G>C (p = 2.3 × 10⁻³; OR (95% CI) = 1.92 (1.25–2.92)) with increased risk of BC, and to a lesser extent rs1800469/–1347C>T (p = 0.048) with decreased risk of BC under the codominant genetic model.

Table 3.

TGFB1 genotype frequencies.

SNP	Nucleotide	1/1^a		1/2^a		2/2^a
SNP	Nucleotide	Cases	Controls	Cases	Controls	Cases	Controls	p ^b
rs1800472	+788C>T	378 (0.89) ^c	415 (0.86)	49 (0.11)	66 (0.14)	0 (0.00)	0 (0.00)	0.31
rs1800471	+74G>C	348 (0.85)	437 (0.92)	61 (0.15)	40 (0.08)	0 (0.00)	0 (0.00)	0.002
rs1800470	+29T>C	134 (0.31)	128 (0.27)	215 (0.50)	273 (0.57)	81 (0.19)	80 (0.17)	0.12
rs1800469	−509C>T	159 (0.38)	160 (0.33)	194 (0.46)	261 (0.54)	66 (0.16)	60 (0.13)	0.05

SNP: single nucleotide polymorphism.

Genotypes were coded as “1” = major allele and “2” = minor allele. The corresponding genotypes were C/C, C/T, and T/T for rs1800472; G/G, G/C, and C/C for rs1800471; T/T, T/C, and C/C for rs1800470; and C/C, C/T, and T/C for rs1800469.

Chi-square test.

Number of subjects (frequency).

Table 4.

Effects of TGFB1 SNP genotypes on the risk of CRC according to the different genetic models.

		Codominant model		Dominant model		Recessive model		Log additive
	Genotype	p ^a	OR (95% CI)^a	p ^a	OR (95% CI)	p ^a	OR (95% CI)	p ^a	OR (95% CI)
rs1800472	C/C	0.31	1.00 (Reference)
	C/T		0.82 (0.55–1.21)
rs1800471	G/G	2.3 × 10⁻³	1.00 (Reference)
	G/C		1.92 (1.25–2.92)
rs1800470	T/T	0.12	1.00 (Reference)	T/T vs T/C + C/C		T/T + T/C vs C/C		0.60	0.95 (0.78–1.15)
	T/C		0.75 (0.56–1.02)	0.13	0.80 (0.60–1.07)	0.38	1.16 (0.83–1.64)
	C/C		0.97 (0.65–1.43)
rs1800469	C/C	0.05	1.00 (Reference)	C/C vs C/T + T/T		C/C + C/T vs T/T		0.75	0.97 (0.80–1.18)
	C/T		0.75 (0.56–1.00)	0.14	0.82 (0.62–1.07)	0.16	1.31 (0.90–1.91)
	T/T		1.11 (0.73–1.67)

TGFB1: transforming growth factor beta 1; SNP: single nucleotide polymorphism; CRC: colorectal cancer; OR: odds ratio; CI: confidence interval; BMI: body mass index.

Boldface indicates statistically significant differences.

Models controlled for BMI, menarche, smoking, and oral contraceptive use.

Association of TGFB1 genotypes with BC features

The association of the tested TGFB1 variants with BC features was investigated. Results from Table 5 show that rs1800472 was positively associated with triple negativity (ρ = 0.165; p = 0.001), molecular type (ρ = 0.143; p = 0.001), ER/HER-2 status (ρ = 0.132; p = 0.005), and menarche (ρ = 0.960; p = 0.002). On the other hand, rs1800470 was positively associated with menarche (ρ = 0.100, p = 0.015), but negatively with triple negativity (ρ = –0.127; p = 0.018), tumor size (ρ = –0.119, p = 0.003), nodal status (ρ = –0.094, p = 0.038), and ER/HER-2 status (ρ = –0.073, p = 0.022). Furthermore, rs1800469 correlated positively with menstrual irregularity (ρ = 0.190, p < 0.001), menarche (ρ = 0.108; p = 0.010), distant metastasis (ρ = 0.184, p < 0.001), nodal status (ρ = 0.158, p = 0.001), molecular type (ρ = 0.194, p = 0.007), ER/HER-2 status (ρ = 0.121, p = 0.032), and hormonotherapy (ρ = 0.127, p = 0.011).

Table 5.

Matrix of correlation between TGFB1 variants and breast cancer feature and outcome.

Parameter	rs1800472		rs1800471		rs1800470		rs1800469
Parameter		ρ^a	p ^b	ρ	p	ρ	p	ρ	p
Menstrual irregularity	−0.039	0.451	−0.018	0.786	0.084	0.178	0.190	2.1 × 10 ⁻⁴
Menarche (years)	0.960	0.002	0.003	0.937	0.100	0.015	0.108	0.010
Triple negative	0.165	0.001	−0.001	0.975	–0.127	0.018	0.048	0.104
Histological type (Ductal/lobular/mixed)	0.021	0.720	−0.006	1.000	0.051	0.530	0.097	0.107
Tumor size	−0.081	0.071	−0.029	0.758	–0.119	0.003	0.157	0.071
Distant metastasis	0.051	0.278	0.001	0.936	−0.041	0.370	0.184	3.5 × 10 ⁻⁴
Nodal status (N0, N1, N2)	−0.038	0.605	0.058	0.461	–0.094	0.038	0.156	0.001
Molecular type (HR (+/–)/HER-2 (+/–))	0.143	0.001	0.009	0.134	−0.097	0.202	0.194	0.007
ER/HER-2 status	0.132	0.005	0.005	1.000	–0.073	0.022	0.121	0.032
Chemotherapy	0.028	0.709	0.034	1.000	−0.017	0.220	−0.027	0.174
Hormonotherapy	0.031	0.428	−0.032	0.622	−0.041	0.321	0.127	0.011

Boldface indicates statistically significant differences.

Spearman correlation coefficient.

Pearson χ² test/Fisher’s exact test for categorical variables, or ANOVA/Student’s t-test for continuous variables.

Haploview analysis

Haploview analysis revealed mixed LD pattern between the tested TGFB1 variants (Figure 1). Four-locus haplotypes were constructed based on the prevalence of individual TGFB1 SNP and LD pattern. Of the possible 16 TGFB1 haplotypes, 6 were found to be common (frequency > 2%), capturing 98.2% of all possible haplotypes (97.0% in cases and 98.8% in controls). Haploview analysis identified haplotype TGCT (p = 0.02) to be associated with reduced risk of BC, and haplotypes CGTT (p = 0.001) and CCCC (p = 0.04) to be associated with increased risk of BC (Table 6). This association of CGTT (p = 7.0 × 10⁻⁴) and CCCC (p = 0.02), but not TGCT (p = 0.10), with BC remained statistically significant after controlling for BMI, menarche, smoking, and oral contraceptive use.

Table 6.

Haplotype frequencies across TGFB1 SNPs analyzed.

Haplotype^a	Frequency	Case: Control frequencies	χ²	p	OR (95% CI)	ap^b	aOR^b (95% CI)
C G T C	0.499	0.473; 0.521	2.51	0.11	1.00 (Reference)	0.07	1.00 (Reference)
C G C T	0.285	0.265; 0.299	2.63	0.50	0.92 (0.73–1.16)	0.60	0.94 (0.74–1.19)
C G C C	0.057	0.070; 0.048	3.99	0.27	1.25 (0.84–1.85)	0.27	1.25 (0.84–1.87)
T G C T	0.055	0.039; 0.069	7.40	0.02	0.61 (0.38–0.98)	0.10	0.67 (0.41–1.08)
C G T T	0.044	0.072; 0.019	30.82	0.001	1.96 (1.28–3.00)	7.0 × 10⁻⁴	2.12 (1.38–3.26)
C C C C	0.042	0.051; 0.032	4.13	0.04	1.65 (1.00–2.71)	0.02	1.84 (1.11–3.08)

TGFB1 haplotypes: rs1800472–rs1800471–rs1800470–rs1800469. OR: odds ratio; aOR: adjusted odds ratio; CI: confidence interval.

Underlined indicates minor allele.

Adjusted for BMI, menarche, smoking, and oral contraceptive use.

Figure 1.

Haploview plot of TGFB1 SNP analyzed. The relative positions of TGB1 SNPs (build 37.3) are displayed, along with the basic gene structure, above the Haploview diagram. The relative LD between pairs of TGFB1 SNPs is color-indicated. This was based on D′, that is, normalized linkage disequilibrium measure or D divided by the theoretical maximum for the observed allele frequencies, multiplied by 100. Values close to zero indicate no LD, while values approaching 100 indicate full LD. The red-colored square represents varying degrees of LD < 1 and LOD (logarithm of odds) > 2 scores; darker shades indicating stronger LD.

Discussion

TGFβ1 plays a central, but paradoxical role in BC development.^18,32,33 While it controls cell cycle progression and apoptosis in early carcinogenesis, it induces a state of immunosuppression in advanced malignancies, which promotes aggressive carcinogenesis and metastasis.^8,10,11,13 As the secretion of TGFβ1 is genetically determined, we tested the association between common variants in TGFB1 and BC susceptibility. Previous studies documented an association between TGFB1 polymorphisms and susceptibility to BC.^{15,23,25,33,34} However, controversial and inconclusive associations were also reported.^26,35 In the present study, we documented the positive association of TGFB1 rs1800471 and rs1800469 with BC, and identified TGFB1 haplotypes that were negatively (TGCT) and positively (CGTT and CCCC) associated with BC. Our article is the first to identify specific TGFB1 polymorphism related to BC risk in Tunisia.

A case-control study design was employed in investigating the association between the risk of BC and the presence of TGFB1 polymorphisms. The control group consisted of 498 healthy women, who reported no positive family history of BC, and no personal history of any cancer types. This in turn increased the population power in evaluating the role of genetic factors to the risk of BC. These cancer-free control women were matched to BC cases according to age, menopause status, and ethnic origin (only Arabic-Speaking Tunisian women were included). Significant differences between controls and BC cases were noted in the prevalence of smoking, in agreement with findings elsewhere.^36,37 Higher prevalence of women with irregular menses, and previous/current users of oral contraceptives were seen in BC cases than in control women. This was reminiscent of findings which documented association of irregular menses^38,39 and oral contraceptive use^38,40 with increased incidence of BC.

The polymorphisms included in this study were the missense rs1800472 (+791C>T, T264I), rs1800471 (+74G>C, R25Q), and rs1800470 (+29T>C, P10L), and the 5′ near-gene rs1800469 (–1347C>T) variants.^{12,18,19,27,34} While MAF and genotype distribution of only rs1800471 was significantly different between BC cases and controls, the genotype distribution of rs1800469 homozygous variant was significantly higher in BC cases. The rs1800469 variant lies within a binding site for YY1 (Yin Yang 1), and the transition of C with T leads to reducing the ligation.⁴¹ While not tested here, functionally carriage of the minor (T) allele was linked with augmented TGF-β1 levels, compared to C/C genotype carriers.⁴²

TGFB1 rs1800470 is the most extensively studied polymorphism, located in codon 10 (CTG→CCG), resulting in leucine-to-proline substitution, and increased TGFβ1 secretion.⁴³ The allele and genotype distribution of rs1800470 was not significantly different between BC cases and control women. TGFB1 rs1800470 minor allele was previously reported to increase the risk of BC in Indians,²¹ Koreans,⁴³ and Caucasians,¹⁹ but paradoxically was associated with reduced risk of BC in White US⁴⁴ and Indian⁶ populations. A recent Iranian study involving 100 BC cases and 100 control women⁴⁵ reported no significant difference in MAF between two groups, but demonstrated strong positive association between T/T genotype and BC risk (OR (95% CI) = 2.409 (1.087–5.337)). Furthermore, an earlier meta-analysis demonstrated that T allele from this variant was modestly protective against BC.²⁰ These apparently conflicting results may be explained by heterogenicity due to sampling variation and background ethnicity.

Compared to rs1800470, rs1800471 (+74G>C) is less studied. This variant is associated with arginine-to-proline substitution in codon 25,¹⁹ and with lower TGFβ1 production. While rs1800471 MAF and genotype distribution was comparable between BC cases and control women, carriage of rs1800471 minor (C) allele correlated positively with menarche, but negatively with triple negativity, tumor size, and hormone receptor (HR) and HER-2 status among BC cases. An earlier Indian study involving 456 BC cases and 239 control women also reported no association between rs1800471 and BC risk.¹⁷ Interestingly, a study on North Indian and South Indian women reported that this variant has a significant protective effect against BC.²¹ Additional studies involving additional ethnic groups are needed to confirm or alternatively rule out the association of rs1800471 (+74G>C) with altered risk of BC, or with BC-associated features.

The rs1800472 located at exon 5 codon 263, and the corresponding residue lies near the site where the latency-associated peptide is cleaved from the active peptide, and thus appears to be related to TGFB1 activation.⁴⁶ MAF of rs1800472 seen in Tunisians (7.0%) was comparable to frequencies established for Caucasians (7.4%) and Hispanics (7.1%), but was higher than frequencies seen in Pan Europeans (2.5%–5.0%) and Africans (2.2%) (https://www.ncbi.nlm.nih.gov/projects/SNP/snp_ref.cgi?do_not_redirect&rs=rs1800472). While rs1800472 was not associated with BC per se, it was positively linked with triple negative status among BC cases. To the best of our knowledge, this was the first evidence documenting association of this variant with BC-associated features.

Haplotype association analysis revealed that CCCC and CGTT haplotypes were positively associated with BC, while TGCT haplotype was negatively associated with BC. This assigns BC-susceptible and protective nature to these haplotypes. Our results are reminiscent of a recent Brazilian study, which analyzed the positive association of 4-locus (rs1800468–rs1800469–rs1800470–rs1800471) CTCG TGFB1 haplotype with BC and with Her2+ status in the dominant model, but negative association of GCCG haplotype with BC but not BC-associated features.¹² This highlights the significance of haplotype analysis in disease-association studies, as suggested elsewhere.⁴⁷

Our study had several strengths. First, the number of BC cases and control women was substantial, and thus the study was sufficiently powered. Second, BC cases and control women were matched according to ethnicity (only Arabic-speaking Tunisian women included), which minimizes the problems of ethnic differences inherent in genetic association studies. Another strength of the study is in controlling for number of potential covariates. However, our study has also shortcomings. We could not measure serum TGFβ1 levels in BC cases and control women, which did not allow for addressing genotype–phenotype correlations. Another limitation lies in its retrospective nature, which prompts speculation on cause–effect relationship. The small number of TGFB1 SNPs studied prompts the speculation of the potential association of additional variants with BC and its associated features. Future studies involving other ethnic groups and additional variants are needed to confirm these findings.

Conclusion

TGFB1 constitutes a genetic risk locus of BC and rs1800471 and rs1800469 variants, and CGTT and CCCC 4-locus haplotypes are linked with altered risk of BC. TGFB1 polymorphisms affect key BC phenotypic features: rs1800472 was positively associated with menarche, triple negativity, molecular type, and ER/HER-2 status, while rs1800470 was positively associated with menarche, but negatively associated with triple negativity, nodal status, and ER/HER-2 status. Furthermore, rs1800469 correlated positively with menstrual irregularity, menarche, distant metastasis, distant metastasis, nodal status, molecular type, ER/HER-2, and hormonotherapy. This indicates that TGFB1 variants may constitute potential biomarkers for BC susceptibility, and for early identification of individuals at high risk of developing BC.

Footnotes

Acknowledgements

The authors wish to thank Ms. Zainab Malalla for her technical assistance.

Author contributions

M.H.-A. and R.M.G. contributed to data entry and drafting of the manuscript. H.B. contributed to patient screening, selection, and referral. A.H. contributed to literature search and data analysis. M.S. contributed to drafting of the manuscript. M.A., M.H., and K.R. contributed to patient screening, selection, and referral. B.Y.-L. contributed to data analysis. W.Y.A., project leader, contributed to statistical analysis.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Ethical approval

The study was done according to Helsinki II declaration and approved by the Research & Ethic Committee of Salah Azaïz Institute (IRB number: ISA/2018/19, granted on 25 June 2017); all participants provided written informed consent.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Wassim Y Almawi

References

Porter

. Global trends in breast cancer incidence and mortality. Salud Publica Mex 2009; 51: s141–s146.

Fiorica

. Breast cancer screening, mammography, and other modalities. Clin Obstet Gynecol 2016; 59(4): 688–709.

Jorgensen

Kalager

Barratt

et al. Overview of guidelines on breast screening: why recommendations differ and what to do about it. Breast 2017; 31: 261–269.

Buys

Sandbach

Gammon

et al. A study of over 35,000 women with breast cancer tested with a 25-gene panel of hereditary cancer genes. Cancer 2017; 123(10): 1721–1730.

Couch

Shimelis

et al. Associations between cancer predisposition testing panel genes and breast cancer. JAMA Oncol 2017; 3(9): 1190–1196.

Joshi

Kale

Hake

et al. Transforming growth factor β signaling pathway associated gene polymorphisms may explain lower breast cancer risk in western Indian women. PLoS ONE 2011; 6: e21866.

Chen

Xiong

et al. Transforming growth factorβ1 L10P variant plays an active role on the breast cancer susceptibility in Caucasian: evidence from 10,392 cases and 11,697 controls. Breast Cancer Res Treat 2010; 124: 453–457.

Komai

Okamura

Yamamoto

et al. The effects of TGF-βs on immune responses. Nihon Rinsho Meneki Gakkai Kaishi 2016; 39: 51–58.

Bakiri

Macho-Maschler

Custic

et al. Fra-1/AP-1 induces EMT in mammary epithelial cells by modulating Zeb1/2 and TGFβ expression. Cell Death Differ 2015; 22: 336–350.

10.

Jakowlew

. Transforming growth factor-beta in cancer and metastasis. Cancer Metastasis Rev 2006; 25: 435–457.

11.

Pardali

Moustakas

. Actions of TGF-beta as tumor suppressor and pro-metastatic factor in human cancer. Biochim Biophys Acta 2007; 1775(1): 21–62.

12.

Vitiello

GAF

Guembarovski

Hirata

BKB

et al. Transforming growth factor beta 1 (TGFβ1) polymorphisms and haplotype structures have dual roles in BC pathogenesis. J Cancer Res Clin 2018; 144: 645–655.

13.

Seoane

Gomis

. TGF-β family signaling in tumor suppression and cancer progression. Cold Spring Harb Perspect Biol 2017; 9: a022277.

14.

Xia

Xiao

Liu

et al. Coactosin-like protein CLP/Cotl1 suppresses breast cancer growth through activation of IL-24/PERP and inhibition of non-canonical TGFβ signaling. Oncogene 2018; 37: 323–331.

15.

Hishida

Iwata

Hamajima

et al. Transforming growth factor B1 T29C polymorphism and breast cancer risk in Japanese women. Breast Cancer 2003; 10(1): 63–69.

16.

Rajkumar

Samson

Rama

et al. TGFbeta1 (Leu10Pro), p53 (Arg72Pro) can predict for increased risk for BC in south Indian women and TGFbeta1 Pro (Leu10Pro) allele predicts response to neo-adjuvant chemo-radiotherapy. Breast Cancer Res Treat 2008; 112: 81–87.

17.

Saha

Gupta

Bairwa

et al. Transforming growth factor-beta1 genotype in sporadic BC patients from India: status of enhancer, promoter, 59-untranslated-region and exon-1 polymorphisms. Eur J Immunogenet 2004; 31: 37–42.

18.

Shin

Shu

Cai

et al. Genetic polymorphisms of the transforming growth factor-beta1 gene and breast cancer risk: a possible dual role at different cancer stages. Cancer Epidemiol Biomarkers Prev 2005; 14(6): 1567–1570.

19.

Dunning

Ellis

McBride

et al. A transforming growth factor beta1 signal peptide variant increases secretion in vitro and is associated with increased incidence of invasive breast cancer. Cancer Res 2003; 63: 2610–2615.

20.

Niu

Gao

et al. Association of TGFB1–509 C>T polymorphism with breast cancer: evidence from a meta-analysis involving 23,579 subjects. Breast Cancer Res Treat 2010; 124: 243–249.

21.

Pooja

Francis

Rajender

et al. Strong impact of TGF-β1 gene polymorphisms on breast cancer risk in Indian women: a case-control and population-based study. PLoS ONE 2013; 8: e75979.

22.

Qiu

Yao

Mao

et al. TGFB1 L10P polymorphism is associated with breast cancer susceptibility: evidence from a meta-analysis involving 47,817 subjects. Breast Cancer Res Treat 2010; 123(2): 563–567.

23.

Boone

Baumgartner

et al. Associations between genetic variants in the TGF-β signaling pathway and breast cancer risk among Hispanic and non-Hispanic white women. Breast Cancer Res Treat 2013; 141: 287–297.

24.

Slattery

Lundgreen

Stern

et al. The influence of genetic ancestry and ethnicity on breast cancer survival associated with genetic variation in the TGF-β-signaling pathway: the breast cancer health disparities study. Cancer Causes Control 2014; 25: 293–307.

25.

Cox

Penney

Guo

et al. TGFB1 and TGFBR1 polymorphisms and breast cancer risk in the Nurses’ Health Study. BMC Cancer 2007; 7: 175.

26.

Kripple

Langsenlehner

Renner

et al. The L10P polymorphism of the transforming growth factor-beta 1 gene is not associated with breast cancer risk. Cancer Lett 2003; 201: 181–184.

27.

Scollen

Luccarini

Baynes

et al. TGF-β signaling pathway and breast cancer susceptibility. Cancer Epidemiol Biomarkers Prev 2011; 20: 1112–1119.

28.

Ghali

Al-Mutawa

Ebrahim

et al. Progesterone receptor (PGR) gene variants associated with breast cancer and associated features: a case-control study. Pathol Oncol Res. Epub ahead of print 4 January 2018. DOI: 10.1007/s12253-017-0379-z.

29.

Ghali

Al-Mutawa

Al-Ansari

et al. Differential association of ESR1 and ESR2 gene variants with the risk of breast cancer and associated features: a case-control study. Gene 2018; 651: 194–199.

30.

Ghali

Mahjoub

Zaied

et al. Association of genetic variants in NF-kB with susceptibility to breast cancer: a case control study. Pathol Oncol Res. Epub ahead of print 19 July 2018. DOI: 10.1007/s12253-018-0452-2.

31.

Ray

Dhar

Ray

. Transforming growth factor-beta1-mediated activation of NF-kappaB contributes to enhanced ADAM-12 expression in mammary carcinoma cells. Mol Cancer Res 2010; 8(9): 1261–1270.

32.

Bierie

Chung

Parker

et al. Abrogation of TGF-β signaling enhances chemokine production and correlates with prognosis in human breast cancer. J Clin Invest 2009; 119: 1571–1582.

33.

Liu

Lin

et al. Transforming growth factor beta-1 C-509T polymorphism and cancer risk: a meta-analysis of 55 case-control studies. Asian Pac J Cancer Prev 2012; 13(9): 4683–4688.

34.

Jin

Hemminki

Grzybowska

et al. Polymorphisms and haplotype structures in genes for transforming growth factor beta1 and its receptors in familial and unselected breast cancers. Int J Cancer 2004; 112(1): 94–99.

35.

Alqumber

Dar

Haque

et al. No association of the TGF-β1 29T/C polymorphism with breast cancer risk in Caucasian and Asian populations: evidence from a meta-analysis involving 55,841 subjects. Asian Pac J Cancer Prev 2014; 15: 8725–8734.

36.

Catsburg

Miller

Rohan

. Active cigarette smoking and risk of breast cancer. Int J Cancer 2015; 136: 2204–2209.

37.

Jones

Schoemaker

Wright

et al. Smoking and risk of breast cancer in the generations study cohort. Breast Cancer Res 2017; 19(1): 118.

38.

Nindrea

Aryandono

Lazuardi

. Breast cancer risk from modifiable and non-modifiable risk factors among women in Southeast Asia: a meta-analysis. Asian Pac J Cancer Prev 2017; 18(12): 3201–3206.

39.

Orgeas

Hall

Rosenberg

et al. The influence of menstrual risk factors on tumor characteristics and survival in postmenopausal breast cancer. Breast Cancer Res 2008; 10(6): R107.

40.

Moorman

Havrilesky

Gierisch

et al. Oral contraceptives and risk of ovarian cancer and breast cancer among high-risk women: a systematic review and meta-analysis. J Clin Oncol 2013; 31(33): 4188–4198.

41.

Silverman

Palmer

Subramaniam

et al. Transforming growth factor-beta1 promoter polymorphism C-509T is associated with asthma. Am J Respir Crit Care Med 2004; 169(2): 214–219.

42.

Eickmeier

Boom

Schreiner

et al. Transforming growth factor β1 genotypes in relation to TGFβ1, interleukin-8, and tumor necrosis factor alpha in induced sputum and blood in cystic fibrosis. Mediators Inflamm 2013; 2013; 913135.

43.

Lee

Park

Hamajima

et al. Genetic polymorphisms of TGF-beta1 & TNF-beta and breast cancer risk. Breast Cancer Res Treat 2005; 90: 149–155.

44.

Ziv

Cauley

Morin

et al. Association between the T29->C polymorphism in the transforming growth factor beta1 gene and breast cancer among elderly white women: the study of osteoporotic fractures. JAMA 2001; 285: 2859–2856.

45.

Parvizi

Mohammadzadeh

Karimi

et al. Effects of two common promoter polymorphisms of transforming growth factor-β1 on breast cancer risks in Ahvaz, West South of Iran. Iran J Cancer Prev 2016; 9: e5266.

46.

Syrris

Carter

Metcalfe

et al. Transforming growth factor b1 gene polymorphisms and coronary artery disease. Clin Sci 1998; 95: 659–667.

47.

Broeks

Schmidt

Sherman

et al. Low penetrance breast cancer susceptibility loci are associated with specific breast tumor subtypes: findings from the Breast Cancer Association Consortium. Hum Mol Genet 2011; 20(16): 3289–3303.

Transforming growth factor beta 1 polymorphisms and haplotypes associated with breast cancer susceptibility: A case-control study in Tunisian women

Abstract

Keywords

Introduction

Subjects and methods

Study subjects

Informed consent

TGFB1 genotyping

Statistical analysis

Results

Study subjects

Association studies

Association of TGFB1 genotypes with BC features

Haploview analysis

Discussion

Conclusion

Footnotes

Acknowledgements

Author contributions

Declaration of conflicting interests

Ethical approval

Funding

ORCID iD

References