Association of common BRCA1 variants with predisposition to breast tumors in Pakistan

BRCA1 variants are extensively associated with increased risk of breast cancer. Early detection and screening of variants is still rare in developing countries. Here, we investigated six BRCA1 variants in 300 subjects from Pakistani population using tetra amplification‐refractory mutation system (T‐ARMS) PCR. Our results indicate significant association of BRCA1 variants rs8176237 (AA; OR 8.2, 95% CI 3.02–22.64, p < 0.0001), rs1060915 (CC; OR 4.29, 95% CI 1.94–9.48, p = 0.0003), and rs799912 (TT; OR 3.16, 95% CI 1.44–6.94, p = 0.004) with up to 8‐fold increased odds of breast cancer under recessive model. Furthermore, BRCA1 haplotypes AGCACG and AGCCCT were associated with up to 18% breast cancer cases (p < 0.05). Additionally, we found association of these variants with up to 11‐fold increased odds of benign breast tumors. Linkage disequilibrium (LD) block‐wise analysis revealed haplotypes GCAC and ATAC were associated with significantly increased risk. To our knowledge, this is the first study that identifies the association of these BRCA1 variants with breast tumors in Pakistani population. In conclusion, BRCA1 variants investigated in the present study are associated with high odds of benign‐ and malignant breast tumors. Studies with bigger sample size may help early detection and screening to reduce the odds of breast cancer.

annually attributed to different genetic variants. Majority of the genes involved in breast cancer code for DNA damage response (DDR) proteins that repairs double-strand breaks (DSBs) by homology-directed DNA repair (HDDR) (Aparicio et al., 2014). Among all genes, BRCA1 (Breast Cancer Susceptibility gene 1) and BRCA2 (Breast Cancer Susceptibility gene 2) variants are reported in 17% of breast cancer cases (Asif et al., 2014). Earlier studies have also reported higher risk of breast-and ovarian cancer in individuals carrying BRCA1 and BRCA2 lethal variants (Vega et al., 2002). Furthermore, alteration in wild type allele of BRCA1 is frequently observed in breast tumors (Martins et al., 2012). BRCA1 is a tumor suppressor gene that plays significant role in DNA repair and cell cycle regulation (Huen et al., 2010). Several genetic alterations associated with heterozygous BRCA1 includes loss of wildtype BRCA1 allele, loss of TP53 (tumor protein P53) and functions of ATM (Ataxia-telangiectasia mutated) or CHEK2 (Checkpoint kinase 2) (Roy et al., 2012). Loss of wild type BRCA1 allele has frequently been reported in patients of hereditary breast and ovarian cancer (HBOC) syndrome. Breast tumors including benign lesions are of considerable importance due to the associated risk of breast cancer (Dyrstad et al., 2015). Benign breast lesions such as lobular carcinoma, atypical ductal or lobular hyperplasia, and proliferative hyperplasia increase the risk of breast cancer up to 10 times (Hartmann et al., 2005).
Along with genetic alterations, interaction of several other factors play crucial role in breast cancer (Bhinder et al., 2019;Went et al., 2019). Although there are several studies on association of BRCA1 variants in different Asian populations as mentioned above, the identification of BRCA1 variants in breast cancer and benign breast tumor cases is never investigated in Pakistani population. Therefore, we aimed to investigate the potential risk of breast tumors associated with BRCA1 variants in female population of Pakistan. Furthermore, we did haplotype analysis with six BRCA1 variants to analyze haplotypes associated with the disease.

Study participants and sampling
A total of 300 age and sex matched controls were recruited in the study including 100 malignant, 100 benign breast tumors patients and 100 healthy controls (Table 1). All the participants were genetically unrelated women from different cities of Pakistan. Sampling was done from September 2016 to February 2017. Subjects recruited in the study were properly diagnosed from Services Hospital, Institute of Nuclear Medicine and Oncology (INMOL) and Jinnah Hospital, Lahore. Demographic, clinical and pathological characteristics of the subjects are given in Table 1.
Tumor-free women without any family history of cancer were taken from out-patient department as controls. They were declared as cancer-free after a complete examination.

Ethics statement
The present study was approved by the review board and ethics committee of the University of the Punjab, Pakistan (SBS/17/18), and all the recruited subjects provided written informed consent for their participation in research.

Variant selection
We selected variants of BRCA1 gene with the following criteria for evaluation: (1) variants located in intron-exon boundaries and 3′ UTR, spanning 52 kb region of the gene with minor allele frequency (MAF) greater than 0.05 and less than 0.5 globally and also in Punjab, Pakistan.
(2) Tagged variants based on data provided by HapMap. The International HapMap Project had genotyped a large number of variants in different populations and provided a set of tagged variants which efficiently represent evolutionally linked genetic variants. Accordingly, we selected six tag SNPs rs4793194, rs8176237, rs1060915, rs2070834, rs799912, and rs8176087 of BRCA1 (capturing 63 variants, Table S1) with most likely associated phenotype HBOC syndrome worldwide. Sequences of the variants were obtained from the National Center for Biotechnology Information (NCBI) database (www.ncbi.nlm.nih.gov.). Five hundred base pairs up-and down-stream sequence of the variants were selected, and primers were designed using "PRIMER 1" online tool (http://primer1.soton.ac.uk/primer1.html). In order to select best primers, different primer sets were analyzed through "OligoAnalyzer" (http://test.idtdna.com/ analyzer/Applications/OligoAnalyzer/).

DNA extraction and genotyping
Approximately 2-3 mL venous blood sample was collected in ethylenediaminetetraacetic acid (EDTA) vials and stored at 4 • C. Genomic DNA was isolated using modified chloroform-isoamyalcohol protocol and stored at −20 • C. Samples were diluted and quantified to obtain the required DNA concentration of 10 ng/μL. Genotyping was done by tetra ARMS-PCR following the protocol described earlier (Ye et al., 2001) with minor modifications in components and concentrations of PCR reaction mixture(Supporting Information, Table S2). Primers sequences and PCR conditions used to obtain genotype data are given in Table S3. PCR products were visualized under gel documentation system after gel electrophoresis. Bands indicating normal and variant alleles were recorded and analyzed using different statistical methods.

Statistical analysis
Hardy-Weinberg equilibrium (HWE) was applied to evaluate the association between selected variants and the studied groups. Allele and genotype frequencies were calculated for each variant in each group. Odds ratio were calculated to analyze the risk associated with minor alleles and genotypes. SNPstats was used to analyze single variant association, genetic model for each variant, haplotype frequency and for the association of haplotypes with breast cancer risk. The D' and r 2 analysis were used to analyze pairwise LD (linkage disequilibrium) between variants, and LD plots were drawn using the Haploview 4.2.

Association analysis of BRCA1 variants with benign breast tumors
All of the six BRCA1 variants showed deviation from HWE (p < 0.05) in benign breast tumor cases. However, we observed no significant deviation in controls (Supporting Information, Table S4). We found statistically significant association of four out of six variants in benign cases. The minor allele frequency of rs4793194, rs8176237, rs2070834 and rs8176087 was significantly higher in benign cases (p < 0.05) compared with controls (Table 2) as these variants occur more frequently in benign tumors as compared to malignant tumors. Furthermore, comparison of genotype frequencies of benign tumor cases and controls showed that genotype AA of rs4793194 (OR 4.71, 95 % CI 1.86-11.92, p < 0.0001) and rs8176237 (OR 11.52, 95% CI 3.64-36.47, p), CC of rs1060915 (OR 1.66, 95% CI 0.61-4.49, p < 0.0001) and rs2070834 (OR 4.66, 95% CI 1.79-12.12, p < 0.0001),and TT of rs799912 (OR 2.72, 95 % CI 1.09-6.79, p = 0.01) and rs8176087 (OR 4.26, 95 % CI 1.62-11.20, p < 0.0001) were associated with increased odds of benign tumors (Table 1). Whereas, the heterozygous genotype was mostly associated with decrease in the odds of benign tumors. Under recessive model, rs4793194, rs8176237, rs2070834, rs799912, and rs8176087 also showed significant association with increased odds of benign breast tumors (p < 0.05, Table S5).

Association analysis of BRCA1 variants in breast cancer patients
In the present study, tetra primer ARMS PCR was used for genotyping of BRCA1 variants. All the six variants showed deviation from HWE in breast cancer samples indicating the possible association of variants with breast cancer in homozygous form (Supporting Information, Table S4S3).
We found statistically significant association of two out of six variants in malignant cases. The minor allele frequency of rs8176237 (minor allele A) and rs1060915 (minor allele C) was significantly higher in breast cancer patients as compared to controls and was associated with increased odds of breast cancer (rs8176237; OR 2.41, 95 % CI 1.30-4.37, p = 0.004, rs1060915; OR 2.13, 95 % CI 1.16-3.89, p = 0.01).
The comparison of genotype frequency revealed that the homozygous recessive genotype A/A of rs8176237 (OR 8.2, 95 % CI 3.02-22.64, p < 0.001), C/C of rs1060915 (OR 4.29, 95 % CI 1.94-9.48, p = 0.0003), and T/T of rs799912 (OR 3.16, 95 % CI 1.44-6.94, p = 0.004) were significantly associated with increased odds of breast cancer (Table 2). However, the heterozygous genotypes were appeared to significantly decrease the breast cancer odds when present in the individuals indicating the major alleles did not increase the odds of breast cancer (Table 3).
In malignant tumor cases, the haplotype H1 was most frequent and recorded in 13% of chromosomes breast

LD Block-wise haplotype analysis
In benign breast tumor cases, all of the variants were located in the same LD block; therefore, block-wise haplotype association with benign cases was similar as measured by simple haplotyping. In breast cancer patients, four variants (rs8176237, rs1060915, rs2070834, and rs799912) were located in the same LD block and the haplotype H4 (GCAC; OR 1.47, 95% CI 0.56-3.82, p = 0.04) and H7 (ATAC; OR 5.96, 95% CI 1.14-31.28, p = 0.03) was associated with significantly increased odds (Table 6).

DISCUSSION
Breast cancer is a serious threat to human health; however, identification of responsible genetic and environmental factors could be helpful in treatment and prevention. Breast cancer susceptibility gene BRCA1 is essential for genomic stability and tumor suppression. Thus it is likely  that BRCA1 variants impair its function and contribute to breast tumors risk (van den Broek et al., 2015). We observed that the breast tumor risk may be modified by BRCA1 variants. The minor allele frequency of rs4793194, rs8176237, rs2070834, and rs8176087 was significantly higher in benign cases. Furthermore, homozygous recessive genotype of rs4793194, rs8176237, rs1060915, rs2070834, rs799912, and rs8176087 were associated with up to 11-fold increased odds of benign tumors. In malignant cases, the frequency of minor allele of variants rs8176237 and rs1060915 was significantly different in patients and controls indicating association with breast cancer. Similarly, the homozygous recessive genotype of BRCA1 variants rs8176237, rs1060915, rs799912 were significantly associated with up to 8-fold high odds of breast cancer among patients. The major allele for all the six variants was not associated with the increasing odds of benign-and malignant breast tumors as the heterozygous genotypes were associated with decreased in the odds of breast cancer. Furthermore, four out of six studied variants were in relatively high LD located in the same LD plot indicating the possibility of these variants to be inherited together. The haplotype AGCACT was present in 18% and AGCCCT was present in 5% of studied cases. BRCA1 plays significant role in the detection and repair of damaged DNA (Powell & Kachnic, 2003). Although, several variants in this gene have been categorized as lethal, the role of common variants is not clear and has been investigated only in relation to the breast cancer risk (Erzurumluoglu et al., 2016). Our results suggest that the TA B L E 6 LD Block-wise haplotypes of BRCA1 variants and association with malignant breast tumors  (Cox et al., 2005). Furthermore, Freedman et al. investigated nine tag SNPs of coding and non-coding regions of BRCA1 and found no association with the risk of breast cancer in multiethnic study (Freedman et al., 2005). Moreover, another study reported no association between individual or combination of the five variants in BRCA1 and the risk of breast cancer (Baynes et al., 2007). A multi-ethnic cohort study found haplotype associated with breast cancer (Lee et al., 2012). Unlike these investigations, we studied benign-and malignant breast tumor cases and identified several haplotypes associated directly with the increased odds, whereas, a few were inversely associated with the odds of breast tumors and need confirmation by further investigations. In conclusion, our study revealed that the odds of breast tumors strongly associated with BRCA1 variants. We suggest that the screening of BRCA1 variants associated with breast cancer should be considered in predication and prevention of breast tumor among females in Pakistan. To our best knowledge, this is the first study that investigated the role of common BRCA1 variants and haplotypes associated with breast tumors in female population of Pakistan. We used tag variants to account for the majority of common variants in BRCA1. In spite the fact that the study represents a severely under studied population, the smaller sample size in each group is the major limitation of the study which may bare more comprehensive analysis. Nevertheless, this study may provide preliminary basis for association analysis in much larger sample sizes and may help in early screening in a population with high incidence of breast cancer.

A U T H O R S C O N T R I B U T I O N
Ayesha Siddique assisted in study design, participated in study design, collected data, interpreted and analyzed the data, wrote first draft of manuscript, had access to all the data in the study and takes responsibility for data integrity and accuracy, critically reviewed and approved final manuscript. Warda Fatima led study design, provided the resources for the study, reviewed and approved final manuscript. Naeem Shahid interpreted data, contributed to data analysis, critically reviewed and approved final manuscript.

A C K N O W L E D G E M E N T S
We would like to thank Dr. Samina Khokhar for assistance in blood sampling.

C O N F L I C T O F I N T E R E S T S TAT E M E N T
The authors declare no conflict of interests.

D ATA AVA I L A B I L I T Y S TAT E M E N T
The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request. O R C I D Ayesha Siddique https://orcid.org/0000-0002-2073-5253