Sage Journals: Discover world-class research

Abstract

BACKGROUND:

LncRNA MEG3 rs7158663 has been shown to confer cancer susceptibility, maybe through altering its gene expression level.

OBJECTIVE:

We aimed to weigh the effect of rs7158663 on MEG3 serum level and breast cancer susceptibility.

METHODS:

We genotyped rs7158663 G $>$ A and measured serum MEG3 in 150 breast cancer, 95 fibroadenoma , and 154 controls by the TaqMan method.

RESULTS:

The presence of rs7158663 G $>$ A is a risk factor for breast cancer among fibroadenoma patients and controls, AA vs. GG genotypes (OR $=$ 6.320, 95% CI $=$ 2.587–15.439, $P<$ 0.0001 when compared to controls and OR $=$ 10.825, 95% CI $=$ 1.929–60.742, $P=$ 0.007 when compared to fibroadenoma). Decreased serum MEG3 was observed in breast cancer group when compared with fibroadenoma and/or controls [median (IQR) $=$ 0.43 (0.27–0.55)] ( $P<$ 0.0001). However, increased serum MEG3 was noted in fibroadenoma group when compared with controls ( $P<$ 0.0001). A significance decreased serum MEG3 was found to be associated with mutant A allele than with wild G allele ( $P<$ 0.0001). The results showed that rs7158663 and lower MEG3 were significantly associated with patients with higher TNM staging and larger tumor size $>$ 5 cm.

CONCLUSION:

The presence of both rs7158663 and low MEG3 are diagnostic and unfavorable prognostic factors for BC patients.

Keywords

MEG3 polymorphism rs7158663 breast cancer fibroadenoma

1. Introduction

Breast cancer (BC) is the second common cancer in females and is a leading cause of death between women [1]. Despite great advances in the early diagnostic and treatment regimens of BC cases, there is no significant improvement in the BC survival rate [2]. Breast cancer etiology is heterogeneous in which multiple genetic and epigenetic variations affect its incidence and prognosis. The exact mechanisms that contribute to the occurrence of BC are still obscure [3]. Numerous studies have attempted to identify molecular factors predispose to and underlying BC pathogenesis in order to introduce new effective therapeutic approaches [4].

Long non-coding RNAs (lncRNAs) are RNA polymerase II transcripts without protein-coding capacity. They are longer than 200 nucleotides in length. It has been approved that lncRNAs are implicated in carcinogenesis through regulating several biological activities, including proliferation, apoptosis, autophagy, angiogenesis, and metastasis by influencing histone modification, transcription, and posttranscriptional processes [5].

Maternally expressed gene 3 (MEG3) is lncRNA imprinted from the maternal allele. It is expressed in various normal tissues [6]. LncRNA MEG3 is a tumor suppressor that has been detected to be downregulated in many cancerous tissues such as gastric cancer [7] colorectal cancer [8], ovarian cancer [9] and hepatocellular carcinoma [10].

Several studies investigated the relationship between MEG3 expression level and breast cancer and demonstrated that: First, MEG3 is normally expressed in normal breast tissues [6]. Next, Loss of MEG3 expression has been identified in breast cancer tissues and cell lines. Moreover, the decreased lncRNA MEG3 expression was significantly associated with the lymph node metastasis, TNM stage, and molecular subtypes. Furthermore, Low lncRNA MEG3 expression level predicts a shorter overall survival and relapse-free survival in breast cancer Last, overexpressed MEG3 suppresses tumor growth, angiogenesis and induces apoptosis in breast cancer cells both invivo and invitro [11, 12, 13, 14]. Collectively the previous findings suggested that MEG3 may act as a possible novel diagnostic, therapeutic, and even prognostic target of breast cancer.

Recent studies reported that abnormal MEG3 expression levels identified in the serum samples from bladder cancer and colorectal cancer patients suggesting the potential use of MEG3 as a biomarker for cancer early diagnosis, prognosis, and prediction of recurrence of the tumor as well as drug resistance [15, 16, 17]. No study measures the expression level of MEG3 in the serum of BC patients and examines its utility as a non-invasive biomarker for BC.

Single nucleotide polymorphisms (SNPs) are one of the most common types of genetic variations in the human genome. SNPs in genes affect gene expression by different mechanisms and are associated with genetic susceptibility to different diseases, including cancer [18, 19]. MEG3 rs7158663 G $>$ A was found to be associated with increased colorectal cancer risk [20]. Also, Ghaedi et al. reported that AA genotype of MEG3 rs7158663 was related to a significant increase in type 2 diabetes mellitus susceptibility [21]. Finally, the silica and bioinformatic analysis showed that the rs7158663 SNP is situated in the binding sites of transcription factors and can bind to different transcriptional factors such as [C/EBP $\alpha$ , Oct-1, and TATA-binding protein (TBP)] [22]. Besides, this SNP could modify the local RNA folding structure, subsequently, the miRNA-lncRNA interactions [20, 21].

Fibroadenoma (FA) is an expression that refers to a wide range of solid, benign breast lesions that commonly affect women in their reproductive years. Moreover, there is not any risk of breast cancer development on top of fibroadenoma tissue. The standard treatment procedure of fibroadenomas by surgical excision has recently been challenged because if confident diagnosis and exclusion of malignancy is available with noninvasive investigations. Then the excision biopsy can be avoided with avoidance of surgical hazards, including disruption of shape and texture of the breast [23]. So we aimed to examine the utility of serum MEG3 in differentiation between BC patients and those with fibroadenomas.

By taking in consideration the preceding findings, including the pivotal role of dysregulated MEG3 in breast cancer, along with the potential impact of rs7158663 on MEG3 expression level and/or its structure, we suppose that rs7158663 SNP of MEG3 may modify the susceptibility to breast cancer through affecting its expression level. To test this supposition, we performed an association study between rs7158663 SNP in MEG3 and breast cancer risk as well as the correlation between this SNP and MEG3 serum expression level in a case-control study involving 150 BC patients, 95 fibroadenoma patients and 154 control subjects from Egypt.

2. Materials and methods

2.1 Subjects

This case-control study was done on 245 Egyptian patients. They were enrolled from the General Surgery Department at the Fayoum University Hospital at Fayoum University and the General Surgery Department at the General Fayoum Hospital. Initial diagnoses depended on history taking, clinical examination. However, final diagnoses were confirmed by mammography and surgical biopsies. Accordingly, patients were classified into two groups: Group I: 150 patients with breast cancer. All breast cancer patients were newly diagnosed postmenopausal women with no past history of breast cancer. All patients experienced previous oral contraception usage for variable periods throughout their lives, also all of them with parity status for more than two times. None of them received antihormonal, chemotherapy, or radiotherapy treatment before participation in the study. Since the age, tumor size, tumor type, TNM staging (tumor size-the number of lymph nodes involved-metastasis) [24] are essential issues in cancer, they were involved in this study. Also, estrogen receptors and progesterone receptors status (ER/PR status) were determined; ER/PR+ve means ER/PR-positive, while ER/PR-ve means ER/PR-negative. ER/PR+ve tumors are much more likely to respond to hormone therapy than tumors that are ER/PR-ve. Group II: 95 patients with fibroadenoma. Clinical information (Age, Parity status, Family history (FH), and/or past history, contraceptive usage, and the state of menstrual cycles) were obtained from patients reports.

While group III included 154 age-matched healthy females (controls) with no family history of breast cancer or fibroadenoma, with no palpable breast masses, received no contraceptives, and they do not suffer from hypertension or diabetes mellitus were recruited during a routine checkup.

The current study was permitted by the ethical committee of the Faculty of Medicine, Fayoum University. Consequently, all subjects signed a written consent after the declaration of the general aim and procedure of the study. The ethical principals of Helsinki were respected and followed in this study.

2.2 Blood sample collection and storage

Blood samples were obtained from all contributors and divided into two tubes. The first tube was used for serum separation and estimation of MEG3. The second tube contained EDTA and was used for DNA extraction and MEG3 genotyping analysis. The samples were frozen at $-$ 80 ${}^{\circ}$ C until processed.

2.3 DNA extraction and genotyping

Genomic DNA was extracted from whole EDTA blood samples of all participants via using QIA-amplification extraction kit (Qiagen, Venlo, Limburg, Netherlands) as per the manufacturer’s instructions. Quantitation and assessment of DNA purity were measured using NanoDrop ${}^{\text{\textregistered}}$ (ND)-1000 spectrophotometer (Nano Drop Technologies, Inc. Wilmington, USA). The purified DNA was used for genotyping in the real-time polymerase chain reaction with the TaqMan allelic discrimination assay (Applied Biosystems, USA) using predesigned specific primer/probe sets for lncRNA MEG3 rs7158663 (G $>$ A) [c_9693465_10, PN 4351379, Lot: P171206_012E08]. Real-time PCR was executed using a Rotor-gene Q Real-Time PCR System (Qiagen, Valencia, CA, USA). PCR program was as follows: 95 ${}^{\circ}$ C for 10 min for an initial denaturation followed by 45 cycles at 92 ${}^{\circ}$ C for 15 s, then 60 ${}^{\circ}$ C for 90 s for annealing and extension, and fluorescence was measured at the end of each cycle and at the endpoint.

2.4 Measurement of serum lncRNA MEG3 fold change

The MEG3 Expression level was detected in the serum. Recent studies measured MEG3 in sera of colorectal cancer and bladder cancer patients [12, 19].

Total RNA (with preserved lncRNAs) was extracted from Serum by miRNeasy extraction kit (Qiagen, Valencia, CA, USA) by following the manufacturer’s protocol. Quantitation and purity of RNA samples were evaluated by using the NanoDrop (ND)-1000 spectrophotometer (NanoDrop Technologies, Inc. Wilmington, USA).

Reverse transcription (RT) was achieved on 60 ng of total RNA in a final volume of 20 $\mu$ L RT reactions using RT2 first strand Kit (Qiagen, Valencia, CA, USA) according to the manufacturer’s instructions.

The serum expression level of the lncRNA MEG3 was determined by using GAPDH as an internal control [25] using customized primers and Maxima SYBR Green PCR kit (Thermo, USA) according to the manufacturer’s protocol. The primer sequences for GAPDH were as follows: 5 ${}^{\prime}$ -C C C T T C A T T G A C C T C A A C T A-3 ${}^{\prime}$ , reverse, 5 ${}^{\prime}$ -T G G A A G A T G G T G A T G G G A T T-3 ${}^{\prime}$ [The MEG3 LPH 02974A-200 and Lot No: 201710030001].

Table 1
Baseline characteristics of studied groups (breast cancer, fibroadenoma, and control groups)

Parameter	BC ( $N=$ 150)	FA ( $N=$ 95)	Control ( $N=$ 154)	Breast cancer vs. FA
Parameter	BC ( $N=$ 150)	FA ( $N=$ 95)	Control ( $N=$ 154)	Unadjusted OR (95% CI) $P$ -value	Adjusted ${}^{\rm{a}}$ OR (95% CI) $P$ -value
Age; mean $\pm$ SD	53.3 $\pm$ 7.26	30.68 $\pm$ 13.23	52.87 $\pm$ 7.44	$<$ 0.0001 ${}^{\rm{\#}}$
DM (N %)
No	120 (80.0%)	95 (100.0%)	154 (100.0%)	–	–
Yes	30 (20.0%)	0	0
HTN (N %)
No	117 (78.0%)	90 (94.7%)	154 (100.0%)	5.077 (1.906–13.526) $<$ 0.0001 ${}^{\rm{\#\#}}$	0.098 (0.024–0.397)0.001
Yes	33 (22.0%)	5 (5.3%)	0
FH (N %)
No	66 (44.0%)	75 (78.9%)	154 (100.0%)	4.773 (2.648–8.604) $<$ 0.0001 ${}^{\rm{\#\#}}$	1.951 (0.822–4.628)0.130
Yes	84 (56.0%)	20 (21.1%)	0

BC, Breast Cancer, FA, Fibroadenoma. DM, Diabetes Mellitus. HTN, Hypertension. FH, Family History. # One way ANOVA ## Chi squared-test ${}^{\rm a}$ Adjusted for age.

Real-time PCR was done on 20 $\mu$ l reaction mixture prepared by mixing 10 $\mu$ l master mix, 1 $\mu$ l forward primer, 1 $\mu$ l reverse primer, 2.5 $\mu$ l diluted cDNA, and 5.5 $\mu$ l RNAase-free water using Rotor-gene Q System (Qiagen) with the the subsequent conditions: 95 ${}^{\circ}$ C for 10 min, 45 cycles at 95 ${}^{\circ}$ C for 15 s and 60 ${}^{\circ}$ C for 60 s.

The 2 ${}^{-\Delta\Delta\text{Ct}}$ equation estimated the fold change of lncRNA MEG3. The cycle threshold (Ct) value is the number of qPCR cycles required for the fluorescent signal to cross a specified threshold. $\Delta$ Ct was obtained by subtracting the Ct values of internal control from those of MEG3. $\Delta\Delta$ Ct was calculated by subtracting the $\Delta$ Ct of the control samples from the $\Delta$ Ct of the patients’ samples. For the control samples, $\Delta\Delta$ Ct equals zero, and 2 ${}^{0}$ equals one [26].

2.5 Statistical analysis of data

Deviation from Hardy-Weinberg equilibrium (HWE) was tested for polymorphism using a Chi-square test by a specific calculator, available online at http://www. oege.org/software/hardy-weinberg.html. The gathered data were organized, tabulated, and statistically analyzed using SPSS software statistical computer package version 22 (SPSS Inc, USA). For quantitative data, the mean, median, standard deviation (SD), and Interquartile range (IQR) were calculated. Kolmogorov-Smirnov test (KS) test was done as a test of normality. If the variable was not normally distributed, the Mann-Whitney-U test or Kruskal Wallis test was used in comparing between any two groups or three groups, respectively. Otherwise, one way ANOVA was used. Qualitative information was presented as numbers and percentages. Chi-Square ( $\chi^{2}$ ) was used as a test of significance. Age-adjusted odds ratios (ORs) with 95% confidence intervals (CI) of genotypes and allelic frequency of MEG3 rs7158663 associated with different clinicopathological features in breast cancer and FA groups were estimated using multivariate logistic regression analysis. Also, adjusted Odds Ratios (ORs) for the association of the study groups with different models and allelic frequency of MEG3 rs7158663 were calculated. ROC curve was performed to determine the cut-off point in which the highest sensitivity and specificity of MEG-3 fold change (FC) in differentiating between different conditions. For interpretation of the results of tests, significance was adopted at $P\leqslant$ 0.05.

3. Results

3.1 Characteristics of the study population

The baseline characteristics (demographic, clinical, and pathological) of the 150 BC, 95 FA cases, and 154 healthy controls are shown in (Tables 1 and 2).

Table 2
Clinical and pathological characteristics of studied groups

Parameter	BC ( $N=$ 150)		FA ( $N=$ 95)		Controls ( $N=$ 154)
$\leqslant$ 65	24	16.0%	95	100%	30	19.5%
$>$ 65	126	84.0%	0	0	124	80.5%
ER/PR status
Negative	126	84.0%
Positive	24	16.0%
Tumor type
Invasive ductal	141	94.0%
Invasive lobular	9	6.0%
Tumor grade
1	3	2.0%
2	117	78.0%
3	30	20.0%
TNM staging
T
2	90	60.0%
3	57	38.0%
4	3	2.0%
N
1	24	16.0%
2	84	56.0%
3	42	28.0%
M
0	150	100.0%
Tumor size
$\leqslant$ 5	90	60.0%
$>$ 5 cm	60	40.0%
U/S
Normal	24	16.0%
Fatty liver	126	84.0%
Menstruation
Premenopausal (regular)			15	15.8%
Premenopausal (irregular)			70	73.7%
Postmenopausal			10	10.5%
Contraceptive use
No			90	94.7%
Yes			5	5.3%
Past history
No			80	84.2%
Yes			15	15.8%
Parity
$\leqslant$ 2			65	68.4%
$>$ 2			30	31.6%

BC, Breast Cancer, FA, Fibroadenoma. ER/PR, Estrogen Receptor/Progesterone Receptor. U/S, Ultrasound. All breast cancer patients were newly diagnosed postmenopausal women with no past history of breast cancer. All patients experienced previous oral contraceptive usage for variable periods throughout their lives, also all of them with parity status for more than two times. All contols were healthy postmenopausal women with no history of usage of contraception. Liver ultrasound is missing data for fibroadenoma patients and controls.

There was a significant difference between breast cancer and FA patients as well as between controls and FA group as regard age ( $P<$ 0.0001) with the mean ages were 53.3 $\pm$ 7.26, 30.68 $\pm$ 13.23, and 52.87 $\pm$ 7.44 years for BC cases, FA, and controls respectively. Significant higher proportion of hypertensive patients was present in BC group (22.0%) than FA group (5.3%) ( $P=$ 0.001). Also, the number of breast cancer patients with positive family history 84 (56%) was significantly higher than the number of fibroadenoma patients with positive family history 20 (21.1%) ( $P<$ 0.0001). As expected, the mean age for BC and control groups paired quite well (Table 1).

As for clinicopathological data of the breast cancer patients, Table 2 shows that patients with weight $\leqslant$ 65 Kg were 24 (16%) and that with weight $>$ 65 were 126 (84%). Eighty-four percent (126 cases) show-ve ER/PR status, while sixteen percent (24 patients) show +ve ER/PR status. The number of invasive ductal breast cancer patients was 141 (94%), and that of invasive lobular type subjects was 9 (6%). As regard tumor grade, 3 subjects (2%) were grade one, 117 subjects (78%) were grade 2, and 30 subjects (20%) were grade 3. Regarding TNM staging, percentage of patients with T 2, 3, 4 were (60, 38, and 2% respectively) whereas the percentage of patients with N1, 2, 3 were (16, 56 and 28% respectively), All patients in the current study had no metastasis. Forty percent of patients had tumor size > 5cm. By ultrasound (U/S) examination, 84% of BC cases had fatty liver.

As for clinicopathological data of the FA patients, 89.5% of FA patients were premenopausal (73.7% had irregular menstrual cycles, 15.8% had regular menstruation), while 10.5% were postmenopausal. 5.3% use contraceptive pills. 15.8% were with a past history of FA (recurrent fibroadenoma) and 31.6% of FA patients with parity more than two times.

3.2 Associations between MEG3 rs7158663 genotypes and the risk of BC and FA

Rs7158663 G $>$ A genotype frequency was in accordance with Hardy-Weinberg equilibrium (HWE) in the control subjects ( $P=$ 0.259).

The genotype and allele distributions of MEG3 SNP (rs7158663) in cases (BC, FA) and controls are shown in (Table 3). A significant difference was determined in the frequency distribution of genotypes and alleles of rs7158663 between breast cancer patients and healthy controls with the following details; in all models having a mutant genotype AA and/or GA or A allele were risk factors for the development of BC.

Table 3
Genotypes and allelic distribution of MEG3 rs7158663 G/A in breast cancer, fibroadenoma cases and control group

MEG3 rs7158663 (G/A)	BC ( $N=$ 150)	Breast cancer vs. control	Breast cancer vs. FA	FA ( $N=$ 95)	FA vs. control	Control ( $N=$ 154)
	$N$ (%)	OR (95% CI), $p^{\rm a}$ -value	OR (95% CI), $p^{\rm a}$ -value	$N$ (%)	OR (95% CI), $p^{\rm a}$ -value	$N$ (%)
Genotypic model
GG	57 (38)	1	1	50 (52.6)	1	84 (54.5)
GA	63 (42)	1.473 (0.906–2.398), 0.118	3.849 (1.497–9.900), 0.005	40 (42.1)	1.067 (0.629–1.810), 0.811	63 (40.9)
AA	30 (20)	6.320 (2.587–15.439), $<$ 0.0001	10.825 (1.929–60.742), 0.007	5 (5.3)	1.200 (0.361–3.984), 0.766	7 (4.5)
Dominant model
GG	57 (38)	1	1	50 (52.6)	1	84 (54.5)
GA+AA	93 (62)	1.948 (1.230–3.086), 0.004	4.706 (1.916–11.562), 0.001	45 (47.4)	1.080 (0.647–1.803), 0.769	70 (45.5)
Recessive model
GG+GA	120 (80)	1	1	90 (94.7)	1	147 (95.5)
AA	30 (20)	5.229 (2.213–12.357), $<$ 0.0001	5.736 (1.191–27.613), 0.029	5 (5.3)	1.167 (0.359–3.786), 0.797	7 (4.5)
Allele
G	177 (59)	1	1	140 (73.7)	1	231 (75)
A	123 (41)	2.073 (1.466–2.933), $<$ 0.0001	3.653 (1.837–7.264), $<$ 0.0001	50 (26.3)	1.071 (0.709–1.620), 0.744	77 (25)

BC, Breast Cancer, FA, Fibroadenoma. ${}^{\rm a}$ Adjusted for age, Hardy-Weinberg $P$ -value for the controls was not a statistically significant, Chi squared value was 1.27 and $P=$ 0.259.

Table 4

MEG-3 ct values and MEG-3 fold change in different studied groups

Parameter	MEG-3 ct value		MEG-3 fold change
	Median (IQR)	$P$ -value	Median (IQR)	$P$ -value
Breast cancer	25.18 (24.09–28.75)	0.008 ${}^{\rm a}$	0.43 (0.27–0.55)	$<$ 0.0001 ${}^{\rm a}$
Fibroadenoma	24.89 (23.25–28.21)	0.257 ${}^{\rm b}$	1.15 (0.99–1.28)	$<$ 0.0001 ${}^{\rm b}$
Controls	25 (19.53–29.93)	1.000 ${}^{\rm c}$	1 (0.9–1.1)	$<$ 0.0001 ${}^{\rm c}$

${}^{\rm a}$ , difference from control group, ${}^{\rm b}$ difference from breast cancer, ${}^{\rm c}$ difference from fibroadenoma.

In genotypic model AA vs. GG genotype (OR $=$ 6.320, 95% CI $=$ 2.587–15.439, $P<$ 0.0001). In dominant model GA+AA vs. GG (OR $=$ 1.948, 95% CI $=$ 1.230–3.086, $P=$ 0.004). In recessive model AA vs. GG $+$ GA (OR $=$ 5.229, 95% CI $=$ 2.213–12.357, $P<$ 0.0001), and allelic model A vs. G (OR $=$ 2.073, 95% CI $=$ 1.466–2.933, $P<$ 0.0001).

Besides, a significant difference was observed in the frequency distribution of rs7158663 polymorphism between breast cancer patients and fibroadenoma patients. Again in all models the presence of mutant genotypes or allele is associated with increased risk of BC; Presence of GA or AA in genotypic model ( $P=$ 0.005, $P=$ 0.007 respectively), GA+AA In dominant model ( $P=$ 0.001), AA in recessive model ( $P=$ 0.029), and A allele in allelic model ( $P<$ 0.0001). Odds ratios were adjusted for age.

No significant differences were detected as regard MEG3 rs7158663 genotypes distribution between fibroadenomas and healthy controls.

3.3 MEG3 fold change in the sera of BC and FA patients and controls

On examination of the serum expression level of MEG3 in different studied groups (Table 4), Decreased serum MEG3 was observed in BC group when compared with FA and or controls [median (IQR) $=$ 0.43 (0.27–0.55)] ( $P<$ 0.0001). However, increased serum MEG3 was noted in FA group when compared with controls [median (IQR) $=$ 1.15 (0.99–1.28)] ( $P<$ 0.0001).

3.4 Association between rs7158663 genotypes and MEG3 serum expression level

We further investigated the correlations between rs7158663 genotypes and MEG3 serum expression levels. As shown in Table 5, lncRNA MEG3 serum expression levels in BC group were significantly higher for the GG (median $=$ 0.52) when compared with GA (median $=$ 0.41) or AA genotype (Median $=$ 0.37) ( $P=$ 0.001 for both). A significance decreased MEG3 serum expression was found to be associated with mutant A allele (median $=$ 0.39) than with wild G allele (median $=$ 0.46) ( $P<$ 0.0001).

No significant differences were found between rs715 8663 genotypes and alleles as regards the serum expression level of MEG3 in FA group.

Table 5
Association between rs7158663 genotypes and MEG3 serum expression level

		MEG-3 ct value Median (IQR)	MEG-3 fold change median (IQR)
Breast cancer ( $n=$ 150)	GG	25.18 (24.51–26.07)	0.52 (0.31–0.72)
	GA	25.11 (24.41–26.22)	0.41 (0.27–0.52)
	AA	25.76 (24.69–26.59)	0.37 (0.21–0.49)
	$P$ -value	0.267	0.001 ${}^{\rm ab}$
	G	25.10 (24.46–26.16)	0.46 (0.31–0.59)
	A	25.29 (24.5–26.43)	0.39 (0.21–0.49)
	$P$ -value	0.436	$<$ 0.0001
FA ( $n=$ 95)	GG	25.06 (24.70–25.94)	1.06 (0.93–1.31)
	GA	24.75 (24.66–25.19)	1.21 (1.12–1.25)
	AA	24.89 (24.88–24.90)	1.10 (1.0–1.2)
	$P$ -value	0.086	0.156
	G	24.99 (24.70–25.68)	1.14 (0.97–1.293)
	A	24.83 (24.70–24.97)	1.20 (1.08–1.22)
	$P$ -value	0.084	1.000
Controls ( $n=$ 154)	GG	25.00 (24.43–25.64)
	GA	24.53 (24.30–25.53)
	AA	25.00 (24.53–25.64)
	$P$ -value	0.438
	G	25.00 (24.43–25.64)
	A	24.93 (24.40–25.53)
	$P$ -value	0.575

FA, Fibroadenoma. ${}^{\rm a}$ significant differences between GG vs. AG, ${}^{\rm b}$ significant differences between GG vs. AA.

3.5 Association between MEG3 serum expression level and clinicopathological characteristics of BC patients

We further examined the association between MEG3 expression levels and clinicopathological characteristics of BC. As shown in Table 6, the results showed that lower MEG3 level was significantly associated with patients with hypertension ( $P=$ 0.027), patients with diabetes mellitus ( $P=$ 0.042), higher T stage of TNM staging ( $P=$ 0.001) and larger tumor size $>$ 5 cm ( $P=$ 0.008). However, no significant correlations between MEG3 expression levels and other clinicopathological variables, including age, FH, weight, ER_PR status, tumor type, tumor grade, N staging of TNM staging, and the presence of fatty liver as detected by ultrasound examination (All $p>$ 0.05).

Table 6
Association between MEG3 fold change and clinical data of breast cancer group

Parameter	MEG-3 ct value		MEG-3 fold change
	Median (IQR)	$P$ -value	Median (IQR)	$P$ -value
Age (Years)
$\leqslant$ 50	25.45 (24.76–26.59)	0.221	0.40 (0.27–0.56)	0.474
$>$ 50	25.10 (24.48–26.12)		0.44 (0.25–0.55)
HTN
No	25.18 (24.51–26.22)	0.450	0.63 (0.39–1.15)	0.027
Yes	24.55 (24.39–26.56)		0.55 (0.44–0.79)
DM
No	25.24 (24.60–26.22)	0.176	0.63 (0.39–1.13)	0.042
Yes	24.53 (24.39–26.56)		0.53 (0.27–0.61)
FH
No	25.76 (2.01–26.5)	0.003	0.41 (0.46–1.22)	0.125
Yes	2.81 (24.42–26.9)		0.44 (0.27–0.72)
Weight (kg)
$\leqslant$ 65	25.45 (24.69–26.01)	0.872	0.47 (0.25–0.59)	0.579
$>$ 65	25.14 (24.50–26.9)		0.43 (0.27–0.55)
ER_PR
Negative	25.98 (25.02–26.5)	0.017	0.42 (0.27–0.55)	0.488
Positive	25.10 (24.43–26.16)		0.49 (0.19–0.61)
Tumor type
Invasive ductal	25.29 (24.51–26.9)	0.003	0.43 (0.27–0.55)	0.859
Invasive lobular	24.40 (24.11–2.94)		0.39 (0.19–0.72)
Tumor grade
I $+$ II	25.24 (24.48–26.9)	0.597	0.43 (0.29–0.55)	0.069
III	2.93 (24.51–2.45)		0.31 (0.13–0.55)
TNM staging
T
2	25.06 (24.43–26.16)	0.138	0.46 (0.31–0.56)	0.001 ${}^{\rm abc}$
3	25.78 (24.55–26.98)		0.35 (0.23–0.55)
4	25.79 (25.78–25.80)		0.09 (0.08–0.1)
N
1	24.96 (24.63–25.20)	0.001 ${}^{\rm{bc}}$	0.50 (0.29–0.56)	0.151
2	24.98 (24.42–26.19)		0.42 (0.29–0.57)
3	26.20 (25.18–27.12)		0.32 (0.21–0.52)
Tumor size
$\leqslant$ 5	25.06 (24.43–26.16)	0.051	0.46 (0.31–0.56)	0.008
$>$ 5 cm	25.79 (24.62–26.79)		0.34 (0.22–0.52)
U/S
Normal	24.73 (24.34–25.48)	0.004	0.40 (0.31–0.57)	0.83
Fatty liver	25.29 (24.51–26.43)		0.43 (0.23–0.55)

DM, Diabetes Mellitus. HTN, Hypertension. FH, Family History. U/S, Ultrasound. ${}^{\rm a}$ Significant difference between T2 vs. T3 or between N1 vs. N2, ${}^{\rm b}$ Significant difference between T2 vs. T4 or between N1 vs. N3, ${}^{\rm c}$ Significant difference between T3 vs. T 4 or between N2 vs. N3.

3.6 Effect of MEG3 rs7158663 on clinicopathological characteristics of BC patients

We evaluated the association between rs7158663 genotypes and alleles and clinical features of BC (Table 7).

Table 7
Relation of genotypes and allelic frequency of MEG3 rs7158663 with different characteristics in breast cancer group

	Genotypes			Allele
	GG	GA+AA	OR (95% CI), $p^{\rm{a}}$ -value	G	A	OR (95% CI), $p^{\rm{a}}$ -value
	$N$ (%)			$N$ (%)
Age (years)
$\leqslant$ 50	15 (26.3)	27 (29.0)	1	51 (28.8)	33 (26.8)	1
$>$ 50	42 (73.7)	66 (71.0)	0.873 (0.416–1.830), 0.719 ${}^{\rm b}$	126 (71.2)	90 (73.2)	1.104 (0.660–1.847), 0.707 ${}^{\rm b}$
HTN
No	48 (84.2)	69 (74.2)	1	141 (79.7)	93 (75.6)	1
Yes	9 (15.8)	24 (25.8)	1.368 (0.495–3.879), 0.534	36 (20.3)	30 (24.4)	0.873 (0.441–1.728), 0.69
DM
No	48 (84.2)	72 (77.4)	1	141 (79.7)	99 (80.5)	1
Yes	9 (15.8)	21 (22.6)	1.102 (0.401–3.024), 0.851	36 (20.3)	24 (19.5)	0.614 (0.311–1.21), 0.16
FH
No	39 (68.4)	27 (29.0)	1	96 (54.2)	36 (29.3)	1
Yes	18 (31.6)	66 (71.0)	5.52 (0.52–12.11), $<$ 0.0001	81 (45.8)	87 (70.7)	0.78 (0.64–4.73), $<$ 0.0001
Weight (Kg)
$\leqslant$ 65	9 (15.8)	15 (16.1)	1	24 (13.6)	24 (19.5)	1
$>$ 65	48 (84.2)	78 (83.9)	0.748 (0.286–1.957), 0.554	153 (86.4)	99 (80.5)	0.647 (0.348–1.202), 0.169
ER/PR status
Negative	48 (84.2)	78 (83.9)	1	150 (84.7)	102 (82.9)	1
Positive	9 (15.8)	15 (16.1)	3.026 (0.839–10.923), 0.091	27 (15.3)	21 (17.1)	3.162 (1.274–7.848), 0.013
Tumor type
Ductal	57 (100.0)	84 (90.3)	1	171 (96.6)	111 (90.2)	1
Lobular	0 (0.0)	9 (9.7)	–	6 (3.4)	12 (9.8)	4.064 (1.423–11.609), 0.009
Tumor grade
I $+$ II	45 (78.9)	75 (80.6)	1	141 (79.7)	99 (80.5)	1
III	12 (21.1)	18 (19.4)	0.863 (0.376–1.981), 0.729	36 (20.3)	24 (19.5)	0.912 (0.509–1.634), 0.756
T
2	45 (78.9)	45 (48.4)	1	129 (72.9)	51 (41.5)	1
3	12 (21.1)	45 (48.4)	3.535 (1.643–7.608), 0.001	48 (27.1)	66 (53.7)	3.322 (2.016–5.473), $<$ 0.0001
4	0 (0.0)	3 (3.2)	–	0 (0.0)	6 (4.9)	–
N
1	3 (5.3)	21 (22.6)	1	24 (13.6)	24 (19.5)	1
2	42 (73.7)	43 (45.2)	0.126 (0.034–0.467), 0.002	117 (66.1)	51 (41.5)	0.417 (0.215–0.808), 0.010
3	12 (21.1)	30 (32.3)	0.292 (0.071–1.212), 0.090	36 (20.3)	48 (39.0)	1.217 (0.590–2.511), 0.595
Tumor size
$\leqslant$ 5	45 (78.9)	45 (48.4)	1	129 (72.9)	51 (41.5)	1
$>$ 5 cm	12 (21.1)	48 (51.6)	3.783 (1.765–8.107), 0.001	48 (27.1)	72 (58.5)	3.64 (2.22–5.96), $<$ 0.0001
U/S
Normal	15 (26.3)	9 (9.7)	1	33 (18.6)	15 (12.2)	1
Fatty liver	42 (73.7)	84 (90.3)	3.318 (1.328–8.290), 0.010	144 (81.4)	108 (87.8)	1.625 (0.837–3.156), 0.152

DM, Diabetes Mellitus. HTN, Hypertension. FH, Family History. U/S, Ultrasound. ${}^{\rm a}$ Adjusted for age. ${}^{\rm b}$ Unadjusted $P$ value.

The presence of mutant GA+AA and/or A allele showed significant association with positive FH ( $P<$ 0.0001 for both GA+AA and A alleles), positive ER/PR status ( $P=$ 0.013 for A allele), presence of invasive lobular type of BC ( $P=$ 0.009 for A allele), advanced T stage of TNM staging T3 when compared with T2 ( $P=$ 0.001 and $P<$ 0.0001 for GA+AA and A allele respectively), larger tumor size ( $>$ 5 cm) ( $P=$ 0.001 and $P<$ 0.0001 for GA+AA and A allele respectively) and with the presence of fatty liver in ultrasound examination ( $P=$ 0.010 for GA+AA).

Also, the presence of mutant GA+AA and/or A allele were significantly associated with N1 of TNM staging when compared with N2 ( $P=$ 0.002 and $P=$ 0.010 for GA+AA and A allele respectively).

No significant correlation was found between rs715 8663 SNP and age, HTN, DM, weight, and tumor grade (all $P>$ 0.05).

3.7 Association between MEG3 serum expression level and clinicopathological characteristics of FA patients

The relationship between MEG3 expression levels and clinicopathological characteristics of FA are shown in (Table 8), lower MEG3 level was significantly associated with older age of patients ( $P>$ 0.0001), hypertensive patients ( $P=$ 0.001), patients with Positive FH ( $P=$ 0.003), postmenopausal patients ( $P=$ 0.002), patients used contraception ( $P=$ 0.005) and past history of fibroadenoma ( $P=$ 0.006). No significant correlations between MEG3 expression levels and Parity numbers ( $P=$ 0.279).

Table 8
Association between MEG3 fold change and clinical data of fibroadenoma group

Parameter	MEG-3 ct value		MEG-3 fold change
	Median (IQR)	$P$ -value	Median (IQR)	$P$ -value
Age (years)
$\leqslant$ 25	24.73 (24.70–25.02)	0.009	1.22 (1.08–1.36)	$<$ 0.0001
$>$ 25	25.04 (24.76–25.94)		1.06 (0.93–1.2)
HTN
No	24.86 (24.70–25.40)	0.001	1.2 (1–1.28)	0.001
Yes	26.60 (26.59–26.61)		0.93 (0.92–0.94)
DM
No	24.89 (24.70–25.41)	–	1.15 (0.99–1.28)	–
Yes	–		–
FH
No	25.02 (24.70–25.94)	0.022	1.22 (0.99–1.31)	0.003
Yes	24.80 (24.53–24.86)		1 (0.93–1.15)
Menstruation
Pre (regular)	24.76 (24.73–25.40)	0.934	1.15 (1.08–1.22)	0.002 ${}^{\rm{ab}}$
Pre (irregular)	24.93 (24.70–25.41)		1.22 (0.99–1.31)
Post	25.45 (24.30–26.60)		0.96 (0.93–1)
Contraceptive use
No	24.91 (24.60–25.43)	0.676	1.2 (1–1.28)	0.005
Yes	24.83 (24.81–24.84)		0.93 (0.92–0.94)
Past history of FA
No	24.93 (24.70–25.68)	0.074	1.2 (0.99–1.3)	0.006
Yes	24.73 (24.30–25.11)		1 (0.91–1.22)
Parity
$\leqslant$ 2	24.83 (24.70–25.11)	0.229	1.2 (1–1.28)	0.279
$>$ 2	25.22 (24.30–25.94)		1.06 (0.99–1.28)

DM, Diabetes Mellitus. HTN, Hypertension. FH, Family History. U/s, Ultra sound. ${}^{\rm a}$ Significant difference between Pre (regular) vs. Post. ${}^{\rm b}$ Significant difference between Pre (irregular) vs. Post.

3.8 Effect of MEG3 rs7158663 on clinicopathological characteristics of FA patients

We evaluated the correlations between rs7158663 genotypes and alleles and clinical features of FA (Table 9).

Table 9
Relation between genotypes and allelic frequency of MEG3 rs7158663 and different characteristics in fibroadenoma group

Parameter	Genotypes			Allele
	GG	GA+AA	OR (95% CI),	G	A	OR (95% CI),
	$N$ (%)		$p^{\rm{a}}$ -value	$N$ (%)		$p^{\rm{a}}$ -value
Age (years)
$\leqslant$ 25	15 (30.0)	30 (66.7)	1	55 (39.3)	35 (70.0)	1
$>$ 25	35 (70.0)	15 (33.3)	0.214 (0.090–0.509), $<$ 0.0001 ${}^{\rm{b}}$	85 (60.7)	15 (30.0)	0.277 (0.139–0.555), 0.038 ${}^{\rm{b}}$
HTN
No	45 (90.0)	45 (100.0)	1	130 (92.9)	50 (100.0)	1
Yes	5 (10.0)	0 (0.0)	–	10 (7.1)	0 (0.0)	–
DM
No	50 (100.0)	45 (100.0)	1	140 (100.0)	50 (100.0)	1
Yes	0 (0.0)	0 (0.0)	–	0 (0.0)	0 (0.0)	–
FH
No	40 (80.0)	35 (77.8)	1	115 (82.1)	35 (70.0)	1
Yes	10 (20.0)	10 (22.2)	1.489 (0.505–4.396), 0.471	25 (17.9)	15 (30.0)	2.678 (1.197–5.992), 0.017
Menstruation
Premenopausal (regular)	0 (0.0)	15 (33.3)	1	15 (10.7)	15 (30.0)	1
Premenopausal (irrigular)	40 (80.0)	30 (66.7)	–	105 (75.0)	35 (70.0)	0.398 (0.173–0.913), 0.030
Postmenopausal	10 (20.0)	0 (0.0)	–	20 (14.3)	0 (0.0)	–
Contraceptive use
No	45 (90.0)	45 (100.0)	1	130 (92.9)	50 (100.0)	1
Yes	5 (10.0)	0 (0.0)	–	10 (7.1)	0 (0.0)	–
Past History
No	40 (80.0)	40 (88.9)	1	115 (82.1)	45 (90.0)	1
Yes	10 (20.0)	5 (11.1)	1.008 (0.260–3.915), 0.990	25 (17.9)	5 (10.0)	0.841 (0.282–2.510), 0.756
Parity
$\leqslant$ 2	25 (50.0)	40 (88.9)	1	85 (60.7)	45 (90.0)	1
$>$ 2	25 (50.0)	5 (11.1)	0.294 (0.072–1.206), 0.089	55 (39.3)	5 (10.0)	0.312 (0.088–1.100), 0.070

DM, Diabetes Mellitus. HTN, Hypertension. FH, Family History. ${}^{\rm a}$ Adjusted for age. ${}^{\rm b}$ Unadjusted $P$ value.

The presence of mutant GA+AA and/or A allele showed a significant association with younger age of patients ( $\leqslant$ 25) ( $P<$ 0.0001, $P=$ 0.038 for GA+AA and A allele respectively), positive FH ( $P=$ 0.017 A alleles), and with regular menstrual cycles in premenopausal women ( $P=$ 0.030 for A allele).

No significant correlation was found between rs715 8663 SNP and past history of fibroadenoma or parity ( $P>$ 0.05).

3.9 Sensitivity and specificity of MEG3 fold change in differentiation between BC patients and non-cancerous group (FA+ controls)

In order to determine the MEG3 diagnostic potential, as a specific biomarker for diagnosis of breast cancer among FA patients and controls, a ROC curve was drawn. ROC analysis revealed that serum MEG3 discriminated BC from non-cancerous group (FA+ controls) with best cut-off $=$ 0.88, AUC $=$ 0.927, 95% CI $=$ 0.890–0.964, $P<$ 0.0001, with sensitivity $=$ 94.3%, specificity $=$ 97.7% (Fig. 1).

Figure 1.

A-ROC curve of MEG-3 ct value revealed that serum MEG3 discriminated BC from non-cancerous group (FA+ controls) with best cut-off $=$ 25.01, AUC $=$ 0.607, 95% CI $=$ (0.528–0.645), $P=$ 0.004, with sensitivity $=$ 58%, specificity $=$ 55.8%. B-ROC curve of MEG-3 fold change revealed that serum MEG3 discriminated BC from non-cancerous group (FA+ controls) with best cut-off $=$ 0.88, AUC $=$ 0.927, 95% CI $=$ 0.890–0.964, $P<$ 0.0001, with sensitivity $=$ 94.3%, specificity $=$ 97.7%.

3.10 Sensitivity and specificity of MEG3 fold change in differentiation between different grades of BC

MEG3 fold change couldn’t differentiate grade III BC from lower grades (I $+$ II) ( $P=$ 0.069) (Fig. 2).

Figure 2.

ROC curve of significance of MEG-3 ct value and MEG3 serum fold change differentiating grade III tumor from other grades (I $+$ II) determined that MEG3 ct value (A) and MEG3 fold change (B) could not differentiate grade III of BC from lower grades (I $+$ II) ( $P=$ 0.596, and $P=$ 0.069 respectively).

4. Discussion

Breast cancer (BC) is considered one of the most prevalent cancers worldwide. In 2018, according to GLOBOCAN Statistics, there were 2.1 million newly diagnosed BC cases among females all over the world. That year, nearby 626,679 patients died of the disease, accounting for 6.6% of all cancer deaths, and ranking BC the second most common cause of death from cancer [27].

Long non-coding RNAs (lncRNAs) were defined as genomic noncoding transcripts longer than 200 nucleotides. Accumulating evidence suggested that lncRNAs have been involved in the initiation and propagation of BC. Furthermore, recent studies have indicated that SNPs in lncRNAs could potentially impact BC susceptibility by influencing gene expression level and biological pathways [5]. For example, Luo et al. reported that there is a robust relationship between rs4801078 and the risk of BC by altering the expression of Linc-ROR mRNA [28]. Also, Yan et al. identified an allelic regulation of rs10463297 on lncRNA SRA expression level and breast cancer susceptibility [29].

Science, lncRNA MEG3 was highlighted as a tumor suppressor gene in breast cancer tissues [11, 12, 13, 14], aberrantly expressed MEG3 can be detected in serum [15, 16], and SNP function prediction showed that rs7158663 might affect transcription factor binding sites [22]. So based on the evidence above, we proposed a hypothesis that the polymorphism of lncRNA MEG3 may be a regulatory SNP and contribute to the risk of BC through altering MEG3 serum expression level. We conducted a case-control study to weigh the serum expression level of MEG3 as a non-invasive biomarker in BC patients and to elucidate the effect of rs7158663 on this level and breast cancer susceptibility.

Our study was conducted on 150 BC, 95 FA patients, and 154 healthy subjects collected from the Egyptian population.

In our study, there were statistically significant differences between the age of FA (30.68 $\pm$ 13.23) and that of BC (53.3 $\pm$ 7.26) or controls (52.87 $\pm$ 7.44) ( $P<$ 0.0001). It was found that breast cancer is most commonly found in women who are 50 years or more [30]. However, it was found that fibroadenoma is the most common benign breast tumor in women at their reproductive age, especially under the age of 30 [31]. Our results also revealed that the BC group was significantly associated with hypertension ( $P<$ 0.0001). Some studies cited that hypertension is a risk factor for breast cancer, especially in women after menopause [32]. In addition, we found that there was a significant difference between BC and FA regarding FH. This result agreed with Brewer et al., who reported that breast cancer risk increased significantly with greater FH [33].

When we studied the distribution of rs7158663 G $>$ A genotypes and alleles among studied groups, we determined for the first time that the presence of mutant A allele in all genetic models was a robust risk factor for BC when compared with FA or controls. Preceding studies declared the association between rs7158663 and other types of cancer, in a study done by Cao et al. which reported that MEG3 rs7158663 AA genotype significantly increased colorectal cancer risk by 96% compared with GG genotype [20]. However, the other two studies cited that rs7158663 SNP was not associated neither to neuroblastoma risk in children nor to lung cancer susceptibility in the Chinese population [34, 35]. Previous studies also examined the association of MEG3 rs7158663 and other diseases than cancer and demonstrated that the dominant model of rs7158663 (GG vs. AG+AA) is associated with increased risk of Ischemic Stroke [22]. Moreover, AA genotype of MEG3 rs7158663 was associated with significantly increased risks of T2D compared with the GG genotype [21].

Former studies explained the association between rs7158663 and risk of cancer or other diseases by performing silica analysis which predicted that SNP located in the binding sites of transcription factors and the mutant A allele of rs7158663 can bind to transcriptional factors [C/EBP $\alpha$ , Oct-1, and TATA-binding protein (TBP)] [22] and altering their binding affinity to a response element resulting in differences in gene expression [36]. Also, the polymorphism of rs7158663 might alter the secondary structure and minimum free energy (MFE), changing the folding of MEG3 [20, 21] and its function [37]. Furthermore, Ghaedi et al. found that rs7158663 polymorphism could disrupt binding sites for miR-4307 and miR-1265 on MEG3 [21].

However, no study elucidate the correlation between MEG3 expression level and different rs7158663 genotypes and alleles.

To explore this point first, we measure serum MEG3 in studied groups, and we demonstrated that decreased serum MEG3 could be used as a non-invasive biomarker differentiating BC patients from FA and/or controls ( $P<$ 0.0001). Our data agreed with prior studies, which declared that MEG3 expression level was decreased in breast cancer tissues compared to adjacent normal tissues [11, 12, 13, 14]. Until now the exact mechanism of dysregulated MEG3 in cancer was not clear, however preceding researchers demonstrated that hypermethylation of differentially methylated regions (DMRs) in MEG3 was associated with its decreased expression [38].

Regarding FA patients, significant upregulation of MEG3 was detected in the sera of FA patients when compared to controls ( $P<$ 0.0001). As far as we know, no study examined the expression of MEG3 in FA however in a recent study done by Chuang et al. who determined that MEG3 was upregulated in leiomyomas tissues compared to normal myometrium tissues with taking in consideration the robust relationship between FA and uterine fibroid [39]. Increased serum MEG3 in FA could be explained by that overexpressed MEG3 increased the expression of p53 that was increased at both mRNA and protein level [40, 41] and in turn, p53 gene and protein has been reported to be increased in fibroadenoma, and it may participate in its formation [42].

Second, we performed a correlation study between rs7158663 genotypes, alleles, and serum MEG3 and elucidated that medians of MEG3 were 0.52, 0.41, and 0.37 for GG, GA and AA genotypes respectively ( $P=$ 0.001). A significance decreased serum MEG3 expression was found for the effect of the A allele. Our study is the first study confirmed the regulatory effect of rs7158663 on serum MEG3 expression level.

On analyzing the relationship between MEG3 expression levels and clinicopathological characteristics of BC, a lower MEG3 level was significantly associated with patients with hypertension ( $P=$ 0.027) (also the same significant correlation was found in FA group) and patients with diabetes mellitus ( $P=$ 0.042). This result agreed with Yu et al., who cited that MEG3 levels were down-regulated by 80% in placental samples collected from preeclamptic patients compared to samples collected from normotensive patients [43]. Also, Zhang et al. identified that serum levels of MEG3 were significantly lower in patients with diabetic patients compared with healthy controls [44].

As well, lower MEG3 was significantly linked to higher T stage of TNM staging ( $P=$ 0.001) and larger tumor size $>$ 5 cm ( $P=$ 0.008). It was detected that decreased lncRNA MEG3 expression was significantly associated with the lymph node metastasis, higher TNM stage, poor differentiation grade, shorter overall survival rate, and shorter progression-free survival rate. Generally, decreased expression of lncRNA MEG3 in breast cancer is associated with a worse prognosis and serves as a critical risk factor for the survival of patients with breast cancer [11, 12, 13, 14].

The current study is the first one that examined the relationship between rs7158663 SNP and clinical characters of BC and FA and by referral of these associations to MEG3 expression level which affected by the presence of different rs7158663 alleles, Our data revealed that rs7158663 was strongly associated with positive FH (in both BC and FA patients), younger age group (in FA), sequentially lower MEG3 was associated with positive FH and positive past history in the FA group. These findings were consistent with Hemminki et al., who reported that single nucleotide polymorphisms (SNPs) are inherited from parents, and they measure heritable events [45].

Also, we detected a significant correlation between A allele (and in turn lower MEG3) and positive ER/PR status and though this finding consented with their synergistic effects on BC cells proliferation, propagation and prognosis [46], it was against results demonstrated by Cui et al. who cited that higher MEG3 mRNA level was linked to the $+$ ER and $+$ PR [38].

Besides, A allele was significantly associated with the invasive lobular subtype of BC, higher T but with lower N. For interpretation of this result, we have to take into consideration that TNM is a complicated staging with different subcategories depending on multiple factories, including tumor size, ER, PR, HER2 status [24]. So the tumor may be N1or of lobular subtype but with higher stage according to other TNM factors. However, we failed to find significant differences between different tumor grades.

Moreover, there was a significant positive association between A allele (lower MEG3) and fatty liver in the BC group, which detected by ultrasound. It was demonstrated that high-fat diet treatment of mice (induced Non-Alcoholic Fatty Liver Disease model in mice) significantly increased serum lipid levels and reduced MEG3 expression [47].

Meanwhile, in the FA group, higher MEG3 was significantly associated with the younger age of patients, premenopausal patients having irregular menstruation, and with patients who do not use contraception. We noticed that these factors represent the most frequent risk factors for FA; however, for the definite explanation, larger scaled functional studies were needed [48].

Limitations faced the present study were: Firstly, these results were relied on genotyping of a single gene with a relatively small size of Breast cancer and Fibroadenoma patients, which need to be validated using an independent functional study. Secondly, our study was depended on a hospital-based study; therefore, potentially valuable sources of selection bias may be present. Thirdly, reproductive factors are major determinants of risk for breast cancer and fibroadenoma [48]; however, our study had incomplete data on these reproductive issues. Further studies are needed to distinguish the association between genetic variants in MEG3 and its expression level and reproductive period details (age at menarche, number of pregnancies, menopausal status and age,).

5. Conclusion

In summary, this study demonstrates that rs7158663 differentiate between BC patients from FA patients and/or healthy individuals. Also, MEG3 rs7158663 G $>$ A mutant A allele is correlated to lower serum MEG3 expression levels. Furthermore, the presence of both rs7158663 and low MEG3 are unfavorable prognostic factors for BC patients.

Higher serum MEG3 expression level was significantly associated with FA and with its risk factors.

Footnotes

Acknowledgments

All authors acknowledge help from all paramedical personals to accomplish and finalize the clinical research work.

Conflict of interest

The authors have no conflict of interest related to this study.

References

Siegel

R.L.

Miller

K.D.

and Jemal

, Cancer statistics, CA Cancer J Clin 66 (2016), 7–30.

Hong

and Dong

, The past, present and future of breast cancer research in China, Cancer Lett 351 (2014), 1–5.

Polyak

, Heterogeneity in breast cancer, J Clin Invest 121 (2011), 3786–3788.

Richard

J.L.C.

and Eichhorn

P.J.A.

, Deciphering the roles of lncRNAs in breast development and disease, Oncotarget 9 (2018), 20179–20212.

Bhat

S.A.

Ahmad

S.M.

Mumtaz

P.T.

Malik

A.A.

Dar

M.A.

Urwat

Shah

R.A.

and Ganaib

N.A.

, Long non-coding RNAs: Mechanism of action and functional utility, Non-coding RNA Res 1 (2016), 43–50.

Zhou

Zhang

and Klibanski

, MEG3 non-coding RNA: A tumor suppressor, J Mol Endocrinol 48 (2012), 45–53

Dan

Wang

Zhu

Yang

Peng

Jiang

and Chen

, LncRNA-MEG3 inhibits proliferation and metastasis by regulating miRNA-21 in gastric cancer, Biomed Pharmacother 99 (2018), 931–938.

Yin

D.D.

Liu

Z.J.

Zhang

Kong

Zhang

Z.H.

and Guo

R.H.

, Decreased expression of long non-coding RNA MEG3 affects cell proliferation and predicts a poor prognosis in patients with colorectal cancer, Tumour Biol 36 (2015), 4851–4859.

Xiu

Y.L.

Sun

K.X.

Chen

Zhao

Guo

Q.G.

and Zong

Z.H.

, Upregulation of the lncRNA Meg3 induces autophagy to inhibit tumorigenesis and progression of epithelial ovarian carcinoma by regulating activity of ATG3, Oncotarget 8 (2017), 31714–31725.

10.

J.H.

Han

Z.P.

M Liu

B Zhou

Zou

M.X.

Y.B.

Y.G.

and Cao

M.R.

, Overexpression of long non-coding RNA MEG3 inhibits proliferation of hepatocellular carcinoma Huh7 cells via negative modulation of miRNA-664, Cell Biochem 118 (2017), 3713–3721.

11.

Zhang

J.J.

Guo

S.H.

and Jia

B.Q.

, Down-regulation of long non-coding RNA MEG3 serves as an unfavorable risk factor for survival of patients with breast cancer, Review for Medical and Pharmacological Sciences 20 (2016), 5143–5147.

12.

Shi

Xia

Yin

and Xing

, Decreased expression of lncRNA MEG3 in breast cancer is associated with poor prognosis, Int J Clin Exp Pathol 9 (2016), 5327–5333.

13.

Soleimanpour

Hosseinpourfeizi

M.A.

Babaei

and Montazeri

, The study of long noncoding RNA, Meg3, expression level and its association with clinicopathologic features in breast cancer, J Babol Univ Med Sci 19 (2017), 14–20.

14.

Zhang

C.Y.

M.S.

Zhang

Han

C.R.

and Yan

, Overexpression of long non-coding RNA MEG3 suppresses breast cancer cell proliferation, invasion, and angiogenesis through AKT pathway, Tumour Biol 39 (2017), 1010428317701311.

15.

Duan

Jiang

Wang

Yan

Xie

Yan

Wang

Zhang

Pan

Yang

and Wang

, Identification of a serum circulating lncRNA panel for the diagnosis and recurrence prediction of bladder cancer, Oncotarget 7 (2016), 78850–78858.

16.

Shang

Zhang

Liu

Peng

Zhou

Pan

Wang

Chen

and Zhao

, MEG3 is a prognostic factor for CRC and promotes chemosensitivity by enhancing onxaliplatin-induced cell apoptosis, Oncol Rep 38 (2017), 1383–1392.

17.

Dong

Wang

Y.P.

and Li

S.Y.

, Down regulation of lncRNA MEG3 promotes colorectal adenocarcinoma cell proliferation and inhibits the apoptosis by upregulating TGF-β1 and its downstream sphingosine kinase 1, Eur Rev Med Pharmacol Sci 22 (2018), 8265–8272.

18.

Deng

Zhou

Fan

and Yuan

, Single nucleotide polymorphisms and cancer susceptibility, Oncotarget 8 (2017), 110635–110649.

19.

Ramírez-Bello

and Jiménez-Morales

, Functional implications of single nucleotide polymorphisms (SNPs) in protein-coding and non-coding RNA genes in multifactorial diseases, Gac Med Mex 153 (2017), 238–250.

20.

Cao

Zhuang

Deng

Sheng

and Shen

, Associations between polymorphisms of long non-coding RNA MEG3 and risk of colorectal cancer in Chinese, Oncotarget 7 (2016), 19054–19059.

21.

Ghaedi

Zare

Omrani

M.D.

Doustimotlagh

A.H.

Meshkani

Alipoor

and Alipoor

, Genetic variants in long non-coding RNA H19 and MEG3 confer risk of type 2 diabetes in an Iranian population, Gene 30 (2018), 265–271.

22.

Han

Zheng

Wang

and Wang

, Association between MEG3/miR-181b polymorphisms and risk of ischemic stroke, Lipids Health Dis 17 (2018), 292.

23.

Begum

Thomus

and Babu

, Chances of breast cancer with fibroadenoma-review, Journal of Scientific and Innovative Research 6 (2017), 84–86.

24.

Edge

S.B.

and Compton

C.C.

, The American joint committee on cancer: The 7th edition of the AJCC cancer staging manual and the future of TNM, Ann Surg Oncol 17 (2010), 1471–1474.

25.

Shaker

O.G.

Ali

M.A.

Ahmed

T.I.

Zaki

O.M.

Ali

D.Y.

Hassan

E.A.

Hemeda

N.F.

and AbdelHafez

M.N.

, Association of LINC00657 & miR-106a serum expression levels and susceptibility to colorectal cancer, adenomatous polyposis and ulcerative colitis in Egyptian population, IUBMB Life 71 (2019), 1322–1335.

26.

Shaker

O.G.

Abdelaleem

O.O.

Mahmoud

R.H.

Abdelghaffar

N.K.

Ahmed

T.I.

Said

O.M.

and Zaki

O.M.

, Diagnostic and prognostic role of serum miR-20b, miR-17-3p, HOTAIR, and MALAT1 in diabetic retinopathy, IUBMB Life 71 (2019), 310–320.

27.

Bray

Ferlay

Soerjomataram

Siegel

R.L.

Torre

L.A.

and Jemal

, Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries, CA Cancer J Clin 68 (2018), 394–424.

28.

Luo

Cao

Peng

Guo

Wang

and Song

, Functional variants in Linc-ROR are associated with mRNA expression of Linc-ROR and breast cancer susceptibility, Sci Rep 8 (2018), 4680.

29.

Yan

Wang

Peng

Wang

Cao

Wang

and Song

, Genetic variants in lncRNA SRA and risk of breast cancer, Oncotarget 7 (2016), 22486–22496.

30.

Kamińska

Ciszewski

Łopacka-Szatan

Miotła

and Starosławska

, Breast cancer risk factors, Prz Menopausal 14 (2015), 196–202.

31.

Lee

and Soltanian

H.T.

, Breast fibroadenomas in adolescents: Current perspectives, Adolesc Health Med Ther 6 (2015), 159–163.

32.

Han

Guo

Shi

Zhang

and He

, Hypertension and breast cancer risk: A systematic review and meta-analysis, Sci Rep 7 (2017), 44877.

33.

Brewer

H.R.

Jones

M.E.

Schoemaker

M.J.

Ashworth

and Swerdlow

A.J.

, Family history and risk of breast cancer: An analysis accounting for family structure, Breast Cancer Res Treat 165 (2017), 193–200.

34.

Zhuo

Z.J.

Zhang

Zhu

Yang

Zou

and Xia

, Associations between lncRNA MEG3 polymorphisms and neuroblastoma risk in Chinese children, AGING 10 (2018), 481–491.

35.

Yang

Gao

Zhang

Wang

Cui

Zhou

and Yin

, Association between long noncoding RNA MEG3 polymorphisms and lung cancer susceptibility in chinese northeast population, DNA Cell Biol 10 (2018), 812–820.

36.

E Buroker

, Regulatory SNPs and transcriptional factor binding sites in ADRBK1, AKT3, ATF3, DIO2, TBXA2R and VEGFA, Transcription 5 (2014), e964559.

37.

Sabarinathan

Tafer

Seemann

S.E.

Hofacker

I.L.

Stadler

P.F.

and Gorodkin

, The RNAsnp web server: Predicting SNP effects on local RNA secondary structure, Nucleic Acids Research 41 (2013), 475–479.

38.

Cui

Yia

Jing

Huang

Tian

Long

Xiang

Liu

Zhang

Tan

and Zhu

, Mining prognostic significance of MEG3 in human breast cancer using bioinformatics analysis, Cell Physiol Biochem 50 (2018), 41–51.

39.

Chuang

T.D.

and Khorram

, Expression profiling of lncRNAs, miRNAs, and mRNAs and their differential expression in leiomyoma using next-generation RNA sequencing, Reprod Sci 25 (2018), 246–255.

40.

Zhou

Zhong

Wang

Zhang

Batista

D.L.

Gejman

Ansell

P.J.

Zhao

Weng

and Klibanski

, Activation of p53 by MEG3 non-coding RNA, J Biol Chem 282 (2007), 24731–24742.

41.

Shi

and Tu

, MEG3 inhibits proliferation and invasion and promotes apoptosis of human osteosarcoma cells, Oncology Letters 15 (2018), 1917–1923.

42.

Schneider

Branchini

Cericatto

Capp

and Brum

I.S.

, Gene and protein expression of p53 and p21 in fibroadenomas and adjacent normal mammary tissue, Endocrine 35 (2009), 118–122.

43.

Kuang

L.Y.

L.L.

Q.L.

Sun

Zhou

Y.M.

X.M.

X.Y.

and Chen

D.J.

, The role and molecular mechanism of long nocoding RNA-MEG3 in the pathogenesis of preeclampsia, Reprod Sci 25 (2018), 1619–1628.

44.

Zhang

Qin

Leng

Zhang

Bai

Meng

and Wang

, LncRNA MEG3 overexpression inhibits the development of diabetic retinopathy by regulating TGF β 1 and VEGF, Experimental and Therapeutic Medicine 16 (2018), 2337–2342.

45.

Hemminki

Försti

and Bermejo

J.L.

, Single nucleotide polymorphisms (SNPs) are inherited from parents and they measure heritable events, J Carcinog 4 (2005), 2.

46.

Yue

Wang

J.P.

Fan

, G Liu Zhang

Conaway

Wang

Korach

K.S.

Bocchinfuso

and Santen

, Effects of estrogen on breast cancer development: Role of estrogen receptor independent mechanisms, Int J Cancer 127 (2010), 1748–1757.

47.

Wang

and Wang

, High-content hydrogen water-induced downregulation of miR-136 alleviates non-alcoholic fatty liver disease by regulating Nrf2 via targeting MEG3, Biol Chem 399 (2018), 397–406.

48.

Humphreys

P.J.

Eriksson

Darai-Ramqvist

Lindstrom

L.S.

Hall

and Czene

, Family history, reproductive, and lifestyle risk factors for fibroadenoma and breast cancer, JNCI Cancer Spectrum 2, pky051 (2018).

Association analyses of a genetic variant in long non-coding RNA MEG3 with breast cancer susceptibility and serum MEG3 expression level in the Egyptian population

Abstract

BACKGROUND:

OBJECTIVE:

METHODS:

RESULTS:

CONCLUSION:

Keywords

1. Introduction

2. Materials and methods

2.1 Subjects

2.2 Blood sample collection and storage

2.3 DNA extraction and genotyping

2.4 Measurement of serum lncRNA MEG3 fold change

Table 1 Baseline characteristics of studied groups (breast cancer, fibroadenoma, and control groups)

3. Results

3.1 Characteristics of the study population

Table 2 Clinical and pathological characteristics of studied groups

Table 3 Genotypes and allelic distribution of MEG3 rs7158663 G/A in breast cancer, fibroadenoma cases and control group

3.4 Association between rs7158663 genotypes and MEG3 serum expression level

Table 5 Association between rs7158663 genotypes and MEG3 serum expression level

Table 6 Association between MEG3 fold change and clinical data of breast cancer group

Table 7 Relation of genotypes and allelic frequency of MEG3 rs7158663 with different characteristics in breast cancer group

Table 8 Association between MEG3 fold change and clinical data of fibroadenoma group

Table 9 Relation between genotypes and allelic frequency of MEG3 rs7158663 and different characteristics in fibroadenoma group

5. Conclusion

Footnotes

Acknowledgments

Conflict of interest

References

Table 1
Baseline characteristics of studied groups (breast cancer, fibroadenoma, and control groups)

Table 2
Clinical and pathological characteristics of studied groups

Table 3
Genotypes and allelic distribution of MEG3 rs7158663 G/A in breast cancer, fibroadenoma cases and control group

Table 5
Association between rs7158663 genotypes and MEG3 serum expression level

Table 6
Association between MEG3 fold change and clinical data of breast cancer group

Table 7
Relation of genotypes and allelic frequency of MEG3 rs7158663 with different characteristics in breast cancer group

Table 8
Association between MEG3 fold change and clinical data of fibroadenoma group

Table 9
Relation between genotypes and allelic frequency of MEG3 rs7158663 and different characteristics in fibroadenoma group