Association of CDH1 and TERT Single-Nucleotide Polymorphisms with Susceptibility to Familial Breast Cancer Risk

Abstract

Breast cancer (BC) is a multistep disease that is thought to result from an interaction between genetic background and environmental factors. In Iran, one of the strongest risk factors for developing BC is a positive family history of the disease. Recently, various polymorphisms of E-cadherin (CDH1) and TERT have been found to be associated with increased BC risk worldwide. This study aimed to analyze the association of CDH1 and TERT single-nucleotide polymorphisms with susceptibility to familial BC (FBC) risk in the Iranian patients. One hundred five patients with FBC and 110 non-FBC (NFBC) were genotyped to elucidate the potential association between CDH1 rs5030625 polymorphism and TERT rs2736098 polymorphism by polymerase chain reaction-restriction fragment length polymorphism. Then, results were evaluated by electrophoresis and Epi Info(™) 2012 software. A significant association was found between CDH1 rs5030625 GAGA genotype and FBC risk. Compared with the control group, the FBC patients had a lower frequency of GG genotype (69% vs. 85%) and a higher frequency of GAGA (5% vs. 2%, P < 0.02). Furthermore, the patients with FBC had a lower frequency of TERT rs2736098 GG genotype (38% vs. 49%, P = 0.001) and a higher frequency of rs2736098 AA genotype (12% vs. 5%, P = 0.001) compared with the NFBC. In contrast, the TERT rs2736098 GG genotype potentially increased the recurring risk of FBC (odds ratio = 3.17, P < 0.01). Allele genotypic frequencies in the FBC patients differed from those of the controls. Interestingly, tumors in FBC patients with rs2736098 GG genotype and rs5030625 GAGA exhibited higher mitotic activity, higher grade, lower estrogen receptor, and progesterone receptor than the other genotypes. In conclusion, CDH1 rs5030625 GAGA genotype and TERT rs2736098 GG genotype in combination with clinical parameters may be prognostic factors rather than susceptibility factors during the progression of FBC.

Introduction

Breast cancer (BC) is a multistep disease that is thought to result from an interaction between genetic background and environmental factors. According to global statistics, 30 out of every 100,000 women under standardized age are affected by BC in Iran. Therefore, it can be expected that about 11,000 women are annually afflicted with this disease.^(1,2) There are two main types of BC, the sporadic variety, which accounts for 90% of patients, and the familial type, which accounts for the remaining 10%. In Iran, one of the strongest risk factors for developing BC is a positive family history of the disease.^(3,4) Different genes such as BRCA1/2, P53, CDH1, and TERT are involved in development of BC. Studying each of these genes and their relationship with BC can be helpful in finding a better and more accurate treatment.^(5–7)

BC development depends on the capacity of tumor cells to invade and metastasize to distant sites, and loss of tumor cell adhesion is an important factor in this process.^(8,9) Epithelial-cadherin (E-cadherin) is a tumor suppressor involved in epithelial cell–cell interactions.^(10–12) It is a 97-kDa transmembrane glycoprotein encoded by the E-cadherin gene (CDH1) located on chromosome 16q22.1 and several single-nucleotide polymorphisms (SNP) in this gene such as rs16260 (−160C/A), rs5030625 (−347G/GA), and rs1801026 (+54C/T) associated with tumor development and progression via modifying transcriptional activity, mRNA stability, or protein expression.^(13–16) The rs5030625GA allele has weak transcriptional factor-binding strength and transcriptional activity compared with that of the rs5030625G allele, while the rs16260A allele decreases the transcriptional efficiency compared with that of the rs16260C allele, and rs1801026T allele decreases mRNA stability.^{(13,17–19,20,21)}

Furthermore, telomerase is also responsible for chromosomal stability and cellular immortality, and it contains a protein catalytic subunit with reverse transcriptase activity, TERT, and an RNA subunit, TERC.^(22–24) The TERT is a limiting factor of telomerase activity and its regulation mainly occurs at the transcriptional level. The TERT gene is essential in maintaining the length of telomer, and TERT is not expressed in most human somatic cells, but defect expression of TERT is associated with development of various cancers.^(25–27) Several SNP in hTERT such as Ex2-659G > A (rs2736098) polymorphism increased the risk of BC in Iranian population.^(28,29) Few investigations have been carried out on the association between CDH1 and TERT polymorphisms in BC severity or progression. This is why in this retrospective study conducted on 215 BC cases, two functional SNPs from of CDH1 and TERT were genotyped, and the possible prognostic values of these genetic variations were investigated.

Materials and Methods

Collection of cancer tissue samples

Tissue samples fixed in formalin and embedded in paraffin (FFPE) were taken from 105 Iranian patients with familial BC (FBC) at the Tehran Khatam and Karaj Emam Hospital. All the cases were reviewed using a special questionnaire, which allowed taking into account the presence of family history of BC and other pathology information. At the same time, FFPE samples were taken from 115 patients with absence of family history of BC with the same age, gender, and ethnicity as the non-FBC (NFBC) patients who were recruited as controls.

Genotyping of CDH1 rs5030625

Paraffin blocks containing tumor tissue were mixed with 200 μL of TBE 0.5 × buffer and put in boiling point for 5 minutes. Then, the resulting mix was microfuged at 12,000 rpm for 15 minutes to get tumor tissue separate from paraffin. The separated tissue was mixed with 100 μL of protease buffer and 5 μL of proteinase k at 54°C for 5 hours. Then, using the kit of DNA extraction obtained from CinnaGen Company and according to the company's protocol, DNA was extracted from tumor tissues. The CDH1 rs5030625 genotyping using polymerase chain reaction-restriction fragment length polymorphism (PCR-RFLP) method was determined by enzymes BanII. A 447-bp fragment containing the −347G→GA polymorphism in the CDH1 promoter was amplified with the following primers (Table 1). The PCR was conducted at 1 cycle of 95°C for 4 minutes; 35 cycles of 94°C for 1 minute, 61°C for 1 minute, 72°C for 1 minute, and a final cycle of 72°C for 10 minutes. The enzymatic digestion mix included 10 μL of PCR product, 17.5 μL distilled water, 2 μL 10 × buffer, and 0.5 μL BanII enzyme, which was put at 37°C for 3 hours and the results were observed on 3% agarose gel. The rs5030625 GA/GA homozygous cases were represented by DNA bands of 332 and 116 bp. The rs5030625 G/G homozygous cases were represented by DNA bands of 263, 116, and 68 bp. But, the rs5030625 GA/G heterozygous cases displayed a combination of both alleles (332, 263, 116, and 68 bp).

Table 1.

Polymerase Chain Reaction Primers Used for CDH1 rs5030625 and TERT rs2736098 Polymorphisms

Polymorphism	Primer sequence	Length (bp)
CDH1	Forward 5′-GCCCCGACTTGTCTCTCTAC-3′	447
−347G→GA	R5′-GGCCACAGCCAATCAGCA-3′	447
TERT	Forward 5′-GCCAGACCCGCCGAAGAAG-3′	379
659G > A	Reverse 5′-GCGCGTGGTTCCCAAGCAG-3′	379

Genotyping of TERT rs2736098 polymorphisms

For TERT rs2736098 genotyping, the PCR conditions were set as follows: 95°C for 5 minutes, 38 cycles of 95°C for 50 seconds, 67°C for 50 seconds, and 72°C for 55 seconds, and a final extension step of 72°C for 10 minutes. The PCR fragments were run in 3% agarose gel and were visualized by SYBR Green staining. The PCR product (10 μL) was digested using Bsp120I restriction enzyme. The G allele was digested and produced 90 and 289 bp fragments, while the A allele was undigested and produced a 379 bp fragment.

Clinical parameters

Clinical parameters of BCs were retrieved from their hospital records and the diagnosis was confirmed by the pathologist. The tubules, mitotic activity, necrosis, polymorphism, and grade of BC were staged by Nottingham histological grading. Immunohistochemical staining of the sections from 202 samples for the expression of estrogen receptor (ER), progesterone receptor (PR), and p53 was carried out using a standard method.

Statistical methods

Epi Info(™) 2012 software was used for statistical analysis of data. Associations between genotypes, clinical diseases, and virulence markers were evaluated. Statistical analyses were performed by chi-square test (χ²) and P-values below 0.05 were considered statistically significant.

Results

Detection of CDH1 rs5030625 genotypes by PCR-RFLP

The genotype at rs5030625 was identified in the FBC and NFBC groups. No significant difference in the parameter of mean age were observed in between these two groups (both P > 0.05). The allele and genotype distributions are shown in Table 2.

Table 2.

Frequency of CDH1 rs5030625 Genotyping in Patients with Familial Breast Cancer and Nonfamilial Breast Cancer

Genotypes	FBC (n = 105)	NFBC (n = 110)	OR
Genotypes	n (%)	n (%)	OR
GG	73 (69)	94 (85)	1.00
GGA	27 (26)	14 (13)	2.46
GAGA	5 (5)	2 (2)	3.08

(χ² = 6.08, P < 0.02).

FBC, familial breast cancer; NFBC, nonfamilial breast cancer; OR, odds ratio.

Genotype distributions at rs2736098 conformed to the Hardy–Weinberg equilibrium in two groups. Compared with the control group, the FBC patients had a lower frequency of GG genotype (69% vs. 85%) and a higher frequency of GAGA (5% vs. 2%, P < 0.02). The frequency of the G allele in the FBC group was estimated to be 0.82, while the variant GA allele was 0.18. In the NFBC patients, the frequency of the G allele was 0.92 while the A allele was 0.08. Thus, A allele genotypic frequencies in the FBC patients were significantly different from those of the NFBC group (P = 0.05).

Detection of TERT rs2736098 genotypes by PCR-RFLP

The genotyping of TERT rs2736098 was identified using enzyme digestion (Table 3). There were two bands for the GG genotype (90 and 289 bp), one band for the AA genotype (379 bp), and three bands for the GA genotype (90, 289, and 379 bp), as shown in Figure 1.

FIG. 1.

Electrophoresis of PCR-RFLP products for TERT rs2736098 genotyping. Lane M, 100-bp ladder; lane 1, 2, and 6, positive for AA genotypes; lane 3, positive for GG genotypes; lane 4, 5, and 7, positive for GA genotypes. PCR-RFLP, polymerase chain reaction-restriction fragment length polymorphism.

Table 3.

Frequency of TERT rs2736098 Genotyping in Patients with Familial Breast Cancer and Nonfamilial Breast Cancer

Genotypes	FBC (n = 105)	NFBC (n = 110)	OR
Genotypes	n (%)	n (%)	OR
AA	19 (18)	35 (32)	1.00
GA	52 (50)	56 (51)	1.65
GG	34 (32)	19 (17)	3.17

(χ² = 7.87, P < 0.01).

The allele and genotype distribution of TERT rs2736098 corresponded to the Hardy–Weinberg equilibrium. Therefore, the samples in our study were random and representative.

The TERT rs2736098 frequency of G allele was 57% in the FBC group and 43% in the NFBC group, and the frequency of A allele was 43% versus 57%; which differed significantly between the FBC group and the NFBC group (P = 0.05). In the NFBC group, the GG, GA, and AA genotype frequencies of hTERT rs2736098 were 32%, 51%, and 17%, respectively. In the FBC group, the GG, GA, and AA genotype frequencies were 18%, 50%, and 32%, respectively. The GG genotype of hTERT rs2736098 can potentially increase the recurring risk of FBC (odds ratio [OR] = 3.17, P < 0.01).

Comparison of clinical parameters in FBC and NFBC groups

In total, breast tissues of 105 patients with FBC and 110 patients with NFBC were investigated. The ovarian cancer was observed in 21 FBC patients. The age of the patients ranged from 18 to 66 years (mean 38.52 years), and most of the patients recognized with tumor of stage II and III. Tumors in patients with FBC exhibited higher mitotic activity, higher grade, higher necrosis, and higher polymorphism compared patients with NFBC (Table 4). In contrast, ER and PR were observed less frequently in the patients with FBC. Furthermore, a significantly higher level of TP53 expression was observed in tumor tissues in the patients with FBC.

Table 4.

Clinical Characteristics of Patients with Familial Breast Cancer and Patients with Nonfamilial Breast Cancer

Parameters	FBC (n = 105)	NFBC (n = 110)	OR
Parameters	n (%)	n (%)	OR
Necrosis
Present	28 (27)	13 (12)	1.00
Absent	77 (73)	97 (89)	0.36
Absent	77 (73)	97 (89)	χ² ⁼ 8.30 (P < 0.004)
Grade
I	22 (21)	40 (36)	1.00
II	36 (34)	33 (30)	1.94
III	47 (45)	37 (34)	2.26
III	47 (45)	37 (34)	χ² ⁼ 4.65 (P < 0.03)
Tubules
3 (<10%)	82 (78)	74 (67)	1.00
2 (10–75)	21 (20)	35 (32)	0.53
1 (>75%)	2 (2)	1 (1)	1.71
1 (>75%)	2 (2)	1 (1)	χ² ⁼ 2.55 (P ∼ 0.1)
Mitotic activity
Low	38 (36)	80 (73)	1.00
Moderate	37 (35)	19 (17)	4.17
High	30 (29)	11 (10)	5.88
High	30 (29)	11 (10)	χ² ⁼ 24.37 (P < 0.0001)
Polymorphisms
1	55 (52)	71 (65)	1.00
2	29 (28)	31 (28)	1.25
3	21 (20)	8 (7)	3.57
3	21 (20)	8 (7)	χ² ⁼ 6.00 (P < 0.02)
ER
+	26 (25)	68 (62)	1.00
−	79 (75)	42 (38)	4.89
−	79 (75)	42 (38)	χ² ⁼ 26.23 (P < 0.0001)
PR
+	38 (36)	79 (72)	1.00
−	67 (64)	31 (28)	4.57
−	67 (64)	31 (28)	χ² ⁼ 24.53 (P < 0.0001)
TP53
+	69 (66)	33 (30)	1.00
−	36 (34)	77 (70)	0.221
−	36 (34)	77 (70)	χ² ⁼ 27.28 (P < 0.0001)

ER, estrogen receptor; PR, progesterone receptor.

According to the statistical analysis, rs2736098 GG genotype and rs5030625 GAGA genotype in combination with clinical parameters were significantly associated with increased risk of BC in patients with FBC. Tumors in FBC patients with rs2736098 GG genotype and rs5030625 GAGA exhibited higher mitotic activity and higher grade compared patients with NFBC. Furthermore, FBC patients who are ER and PR-positive were found to have a significant number of AA and GG genotype, respectively, compared with ER and PR-negative patients. No other parameter was found to be significantly associated with TERT rs2736098 and CDH1 rs5030625 polymorphisms.

Discussion

BC is a multistep disease that is thought to result from an interaction between genetic background and environmental factors. One of the strongest of risk factors is the history of BC in a relative.^(1,2) About 15% to 20% of women with BC have such a family history of the disease, clearly reflecting the participation of inherited (genetic) components in the development of some BCs.⁽³⁾ FBC is a result of collective alterations of oncogenes and tumor suppressor genes. The FBC susceptibility genes BRCA1 and BRCA2 appear responsible for about 5% to 10% of all BCs.⁽⁴⁾ In addition to mutations in BRCA1 and BRCA2 (BRCA1/2) genes, mutations in the following genes can increase the risk of FBC: ATM, CDH1, CHEK2, NBN, NF1, PALB2, PTEN, and STK11.⁽⁵⁾ However, in many families with BC, no predisposing gene mutation can be identified. Recently, various polymorphisms of E-cadherin (CDH1) and TERT have been found to be associated with increased FBC risk worldwide.^(6,7)

In FBC, development and the progression are related to the loss or the reduced expression of the main intercellular adhesion molecule of epithelial cells, the E-cadherin.^(8,9) The loss of cell-to-cell adhesion is an early event in metastatic colonization, leading to the detachment of the cell from her tissue of origin to colonize other sites.^(10–12) Germline mutations in CDH1 are well established as the defects underlying hereditary diffuse gastric cancer (HDGC) syndrome, and an increased risk of lobular BC has been described in HDGC kindreds. In contrast, germline CDH1 mutations can be associated with invasive FBC in the absence of diffuse gastric cancer.⁽⁵⁾

Several studies have reported that the rs503062, rs16260, and rs1801026 polymorphisms in CDH1 gene are associated with a risk for the development of BC.^(13–15) The authors of these studies proposed that the CDH1 rs503062 polymorphism may be functional, and that the GA-allele could lead to transcriptional downregulation of CDH1 and low expression of E-cadherin compared with the G-allele, thereby increasing the risk of BC. To further investigate the association between the functional CDH1 rs5030625 polymorphism and FBC, we conducted the present case–control study in an Iranian population. The results indicated that rs5030625 GAGA genotype by PCR-RFLP was significantly associated with increased risk of BC in the FBC groups (P < 0.02).

Shabnaz et al. did not detect any association between CDH1 rs16260 with clinical parameters of BC patients.⁽⁶⁾ But Tipirisetti et al. found a positive correlation between the CDH1 rs16260A allele occurrences in patients with advanced stage.⁽⁷⁾ In another study, the occurrence of the CDH1 rs16260A allele was significantly associated with more severe clinical stages in hepatocellular carcinoma patients.⁽⁸⁾ Further studies should be conducted among different ethnic groups to help understanding these results.

In Memni study, the rs5030625G/GA and rs1801026C/T, but not the rs16260C/A genotype, associated with BC overall survival, and the CDH1 rs5030625G/G genotype confers risk of death in patients with more aggressive BC progression.⁽¹⁰⁾

In contrast, studies on E-cadherin tissue expression were conflicting. Most of them showed that reduced or loss of E-cadherin expression correlates with high histological grade.^(15–17) In another study, the clinical significance of functional forms of E-cadherin, a full-length membrane form and an extracellular proteolytic soluble form (serum sE-cad) levels, detected in BC patients.⁽¹⁸⁾ They observed a significant correlation of sE-cad levels with tumor stage, grade, lymph node metastasis, and also survival.

We found that the G-allele did not increase the risk of FBC compared with the GA-allele in this population. Shin et al. reported that the GA-allele was associated with a significantly increased risk of colorectal cancer in Korea.⁽¹³⁾ These results may be caused by racial differences. However, in our study, the frequency of the GA-allele was significantly higher in FBC patients than in the NFBC controls. Considering the other studies, the CDH1 rs5030625 polymorphism did not appear to influence E-cadherin expression in normal controls. However, it may play a role in the expression of E-cadherin after BC formation, and may also be associated with the cell differentiation of BC. Cano et al. found that the Snail gene family members were expressed in human carcinoma cells could repress the transcriptional activity of CDH1 through binding to a specified structure or zone in CDH1.⁽²⁰⁾

However, low expression of E-cadherin can occur after the formation of a tumor in response to alterations in the internal environment, such as hypoxic conditions, growth factor expression, and tumor suppressor protein actions.⁽²¹⁾ Therefore, it is possible that the CDH1 rs5030625 polymorphism may change the transcriptional activity of CDH1 after BC tumor formation by combining with other factors, such as telomerase expression, and thereby influencing the differentiation of tumor cells.

In contrast, telomeres are involved in maintaining genomic stability.⁽²²⁾ In the current study, we investigated the impact of TERT rs2736098 polymorphism on BC risk in a sample of the Iranian population. Our data demonstrated that TERT rs2736098 GG genotype increased the risk of FBC and the rs2736098 G allele was associated with an increased risk of FBC. The FBC risk association is strongly related to the geographic area of the study and the selection of the patient population.^(22–24) Several studies have also confirmed that the TERT rs2736098 association with the risk of cancer incidence is highly dependent on the population's ethnicity.⁽²⁵⁾

Haiman et al. observed a positive association between the 5p15 locus and the increased risk of BC.⁽²⁸⁾ Savage et al. suggested a protective effect of three correlated SNPs, including rs2736098, in Polish women with a positive family history.⁽²⁶⁾ There are few reports about the correlation between the TERT rs2735940 variant and BC. Recently, Pellatt et al. found no association between the TERT rs2735940 polymorphism and BC risk.⁽²⁹⁾ They found that this variant was associated with ER-negative/PR-positive tumors (OR = 0.73, 95% CI = 0.59–0.91). In the present study, we found that the rs2736098 GG genotypes increased the risk of FBC in our population, and it can be concluded that an association between BC risk and hTERT rs2736098 is generally related to ethnicity of the study population and the geographical location of the sample.

In conclusion, we found that the CDH1 rs5030625 GAGA genotype and TERT rs2735940 GG genotype were associated with specific FBC features. Tumors in FBC patients with rs2736098 GG genotype and rs5030625 GAGA exhibited higher mitotic activity, higher grade, lower ER, and PR than the other genotypes (Table 4). This also implies that the CDH1 rs5030625 GAGA genotype and TERT rs2735940 GG genotype in combination with clinical parameters may be prognostic factors rather than susceptibility factors during the progression of FBC.

Footnotes

Author Disclosure Statement

No competing financial interests exist.

References

Torre

, Bray

, Siegel

, Ferlay

, Lortet-Tieulent

, and Jemal

: Global cancer statistics, 2012. CA Cancer J Clin, 2015; 65:87–108.

Mohtasebi

, Rassi

, Maleki

, Hajimohammadi

, Bagheri

, Fakhar Miandoab

, and Naserbakht

: Detection of human papillomavirus genotypes and major BRCA mutations in familial breast cancer. Monoclon Antib Immunodiagn Immunother, 2016; 35:19–22.

Rassi

, Houshmand

, Hashemi

, Majidzadeh

, Akbari

, and Panahi

: Application of multiplex PCR with histopathologic features for detection of familial breast cancer in formalin-fixed, paraffin-embedded histologic specimens. Cyto Genet, 2008; 42:55–62.

Rezaei

, Rassi

, and Mansur

: Investigation of methylenetetrahydrofolate reductase C677T polymorphism and human papilloma virus genotypes in Iranian breast cancer. Monoclon Antib Immunodiagn Immunother, 2017; 36:124–128.

Corso

, Intra

, Trentin

, Veronesi

, and Galimberti

: CDH1 germline mutations and hereditary lobular breast cancer. Fam Cancer, 2016; 15:215–219.

Shabnaz

, Ahmed

, Islam

, Al-Mamun

, Islam

, and Hasnat

: Breast cancer risk in relation to TP53 codon 72 and CDH1 gene polymorphisms in the Bangladeshi women. Tumour Biol, 2016; 37:7229–7237.

Tipirisetti

, Govatati

, Kandukuri

, Cingeetham

, Singh

, Digumarti

, Bhanoori

, and Satti

: Association of E-cadherin single-nucleotide polymorphisms with the increased risk of breast cancer: A study in South Indian women. Genet Test Mol Biomarkers, 2013; 17:494–500.

Chien

, Yeh

, Li

, Hsieh

, Lin

, Weng

, Kuo

, and Yang

: Effects of E-cadherin (CDH1) gene promoter polymorphisms on the risk and clinicopathological development of hepatocellular carcinoma. J Surg Oncol, 2011; 104:299–304.

Zhao

, Wang

, Xi

, Liu

, Zhang

, Luo

, and Liu

: Association between E-cadherin (CDH1) polymorphisms and pancreatic cancer risk in Han Chinese population. Int J Clin Exp Pathol, 2015; 8:5753–5760.

10.

Memni

, Macherki

, Klayech

, Ben-Haj-Ayed

, Farhat

, Remadi

, Gabbouj

, Mahfoudh

, Bouzid

, Bouaouina

, Chouchane

, Zakhama

, and Hassen

: E-cadherin genetic variants predict survival outcome in breast cancer patients. J Transl Med, 2016; 14:320.

11.

Kuraoka

, Oue

, Yokozaki

, Kitadai

, Ito

, Nakayama

, and Yasui

: Correlation of a single nucleotide polymorphism in the E-cadherin gene promoter with tumorigenesis and progression of gastric carcinoma in Japan. Int J Oncol, 2003; 23:421–427.

12.

Thiery

: Epithelial-mesenchymal transitions in tumour progression. Nat Rev Cancer, 2002; 2:442–454.

13.

Shin

, Kim

, Kang

, Park

, Lee

, Yoon

, Ku

, and Park

: The E-cadherin −347G− > GA promoter polymorphism and its effect on transcriptional regulation. Carcinogenesis, 2004; 25:895–899.

14.

Zhang

, Wang

, Ge

, Cao

, Chen

, Wen

, Guo

, Wang

, Li

, and Zhang

: Association of CDH1 single nucleotide polymorphisms with susceptibility toesophageal squamous cell carcinomas and gastric cardia carcinomas. Dis Esophagus, 2008; 21:21–29.

15.

Oka

, Shiozaki

, Kobayashi

, Inoue

, Tahara

, Kobayashi

, Takatsuka

, Matsuyoshi

, Hirano

, Takeichi

, and Mori

: Expression of E-cadherin cell adhesion molecules in human breast cancer tissues and its relationship to metastasis. Cancer Res, 1993; 53:1696–1701.

16.

Siitonen

, Kononen

, Helin

, Rantala

, Holli

, and Isola

: ReducedE-cadherin expression is associated with invasiveness and unfavorable prognosis in breast cancer. Am J Clin Pathol, 1996; 105:394–402.

17.

Hunt

, Douglas-Jones

, Jasani

, Morgan

, and Pignatelli

: Loss of E-cadherin expression associated with lymph node metastases in small breast carcinomas. Virchows Arch, 1997; 430:285–289.

18.

Liang

, Sun

, Xu

, and Fu

: Abnormal expression of serum soluble E-cadherin is correlated with clinicopathological features and prognosis of breast cancer. Med Sci Monit, 2014; 20:2776–2782.

19.

Hofmann

, Balic

, Dandachi

, Resel

, Schippinger

, Regitnig

, Samonigg

, and Bauernhofer

: The predictive value of serum soluble E-cadherin levels in breast cancer patients undergoing preoperative systemic chemotherapy. Clin Biochem, 2013; 46:1585–1589.

20.

Cano

, Pérez-Moreno

, Rodrigo

, Locascio

, Blanco

, del Barrio

, Portillo

, and Nieto

: The transcription factor snail controls epithelial-mesenchymal transitions by repressing E-cadherin expression. Nat Cell Biol, 2000; 2:76–83.

21.

Baranwal

, and Alahari

: Molecular mechanisms controlling E-cadherin expression in breast cancer. Biochem Biophys Res Commun, 2009; 384:6–11.

22.

Gomez

, Armando

, Farina

, Menna

, Cerrudo

, Ghiringhelli

, and Alonso

: Telomere structure and telomerase in health and disease (review). Int J Oncol, 2012; 41:1561–1569.

23.

Bacchetti

, and Counter

: Telomeres and telomerase in human cancer (review). Int J Oncol, 1995; 7:423–432.

24.

Andersson

, Osterman

, Sjostrom

, Johansen

, Henriksson

, Brannstrom

, Broholm

, Christensen

, Ahlbom

, Auvinen

, Feychting

, Lönn

, Kiuru

, Swerdlow

, Schoemaker

, Roos

, and Malmer

: MNS16A minisatellite genotypes in relation to risk of glioma and meningioma and to glioblastoma outcome. Int J Cancer, 2009; 125:968–972.

25.

Wang

, Hu

, Liang

, Wang

, Tang

, Wang

, Qin

, Wang

, and Shen

: A tandem repeat of human telomerase reverse transcriptase (hTERT) and risk of breast cancer development and metastasis in Chinese women. Carcinogenesis, 2008; 29:1197–1201.

26.

Savage

, Chanock

, Lissowska

, Brinton

, Richesson

, Peplonska

, Bardin-Mikolajczak

, Zatonski

, Szeszenia-Dabrowska

, and Garcia-Closas

: Genetic variation in five genes important in telomere biology and risk for breast cancer. Br J Cancer, 2007; 97:832–836.

27.

Mocellin

, Verdi

, Pooley

, Landi

, Egan

, Baird

, Prescott

, De Vivo

, and Nitti

: Telomerase reverse transcriptase locus polymorphisms and cancer risk: A field synopsis and meta-analysis. J Natl Cancer Inst, 2012; 104:840–854.

28.

Haiman

, Chen

, Vachon

, Canzian

, Dunning

, Millikan

, Wang

, Ademuyiwa

, Ahmed

, Ambrosone

, Baglietto

, Balleine

, Bandera

, Beckmann

, Berg

, Bernstein

, Blomqvist

, Blot

, Brauch

, Buring

, Carey

, Carpenter

, Chang-Claude

, Chanock

, Chasman

, Clarke

, Cox

, Cross

, Deming

, Diasio

, Dimopoulos

, Driver

, Dünnebier

, Durcan

, Eccles

, Edlund

, Ekici

, Fasching

, Feigelson

, Flesch-Janys

, Fostira

, Försti

, Fountzilas

, Gerty

; Gene Environment Interaction and Breast Cancer in Germany (GENICA) Consortium, Giles

, Godwin

, Goodfellow

, Graham

, Greco

, Hamann

, Hankinson

, Hartmann

, Hein

, Heinz

, Holbrook

, Hoover

, Hu

, Hunter

, Ingles

, Irwanto

, Ivanovich

, John

, Johnson

, Jukkola-Vuorinen

, Kaaks

, Ko

, Kolonel

, Konstantopoulou

, Kosma

, Kulkarni

, Lambrechts

, Lee

, Marchand

, Lesnick

, Liu

, Lindstrom

, Mannermaa

, Margolin

, Martin

, Miron

, Montgomery

, Nevanlinna

, Nickels

, Nyante

, Olswold

, Palmer

, Pathak

, Pectasides

, Perou

, Peto

, Pharoah

, Pooler

, Press

, Pylkäs

, Rebbeck

, Rodriguez-Gil

, Rosenberg

, Ross

, Rüdiger

, Silva Idos

, Sawyer

, Schmidt

, Schulz-Wendtland

, Schumacher

, Severi

, Sheng

, Signorello

, Sinn

, Stevens

, Southey

, Tapper

, Tomlinson

, Hogervorst

, Wauters

, Weaver

, Wildiers

, Winqvist

, Van Den Berg

, Wan

, Xia

, Yannoukakos

, Zheng

, Ziegler

, Siddiq

, Slager

, Stram

, Easton

, Kraft

, Henderson

, and Couch

: A common variant at the TERT-CLPTM1L locus is associated with estrogen receptor-negative breast cancer. Nat Genet, 2011; 43:1210–1214.

29.

Pellatt

, Wolff

, Torres-Mejia

, John

, Herrick

, Lundgreen

, Baumgartner

, Giuliano

, Hines

, Fejerman

, Cawthon

, and Slattery

: Telomere length, telomere-related genes, and breast cancer risk: The breast cancer health disparities study. Genes Chromosomes Cancer, 2013; 52:595–609.