Identification of Genetic Variants Among Breast Cancer Patients and At-Risk Individuals: A Cohort Study in Sri Lanka

Breast cancer (BC) is characterized by the uncontrolled growth of breast tissue and remains the most frequently diagnosed cancer globally with significant implications for women’s health. According to the International Agency for Research on Cancer, there were 2.3 million new cases and 670 000 deaths due to BC in 2022. In Sri Lanka, the National Cancer Control Program reported approximately 3000 new cases per year in 2020 and 2021, accounting for about 27% of all female cancers.^1,2 As a consequence, numerous risk factors have been studied and found to cause this cancer. Some of the BC-associated risk factors are age, body mass index, hormone replacement therapy, exposure to radiation, reproductive factors, prior history of tumor or hyperplasia in breasts, and genetic background.³ Furthermore, many studies have identified that certain risk factors significantly increase relative risk. Particularly, genetic predisposition is identified as one of the most significant risk factors associated with BC. A study conducted to rate the risk factors for BC identified that genetic predisposition significantly affects the risk of BC, ranging from 3 to 200.^4,5

When genetic predisposition is considered, family history and specific germline mutations of cancer susceptibility genes were well explained in several studies. A previous study³ reported that the degree of the risk associated with family history is a function of the type of affected relative, age of cancer development, and the number of relatives affected. Another study revealed that a twofold increased risk is associated with having a first-degree relative compared with the general population.⁵ However, only 10% to 20% of BCs are found to be family clustered, and only 30% to 40% of familial BCs are known to predispose genetically.⁶ Additionally, BCs can be caused by specific germline mutations of cancer-predisposing genes, and hundreds of genes were evaluated for gene variants^7,8

BCs may be classified as sporadic, familial, or hereditary. Sporadic BC occurs in individuals with no notable family history and accounts for the majority of cases. Familial cases cluster within families but lack a clearly identifiable inherited mutation, while hereditary BC is caused by germline mutations with high penetrance. Familial cases may arise due to a combination of shared genetic factors with low to moderate penetrance and environmental or lifestyle influences.⁹ Recent reviews, such as “Understanding genetic variations in familial breast cancer” (2024), have further emphasized this classification of BC susceptibility genes based on their penetrance and functional roles in DNA repair and cell cycle control. These studies highlight how the integration of multigene panel testing has improved the identification of high-, moderate-, and low-risk genetic variants, aiding in risk assessment and clinical decision-making.¹⁰

Most of the genes responsible for cancer predisposition play a significant role in cell cycle regulation and DNA damage repair. Among them, BRCA1 and BRCA2 are considered as major susceptibility genes, with BRCA1 being the most significant. Mutations in highly penetrant BRCA1 and BRCA2 confer a high risk of BC. It has been identified that female carriers with mutations in BRCA1 and BRCA2 have a 50% to 85% lifetime risk.⁸ Pathogenic variants in the predisposition genes ATM, BRCA1, BRCA2, CHEK2, and PALB2 are associated with increased risks of BC, and pathogenic variants in BARD1, RAD51 C, and RAD51D are associated with increased risks of ER-negative BC.^11,12 PALB2 has recently been confirmed as a high-risk BC gene.^11,12 Many studies have utilized multigene panel testing for cancer screening and assessing risk management strategies for women with pathogenic variants in cancer-predisposing genes in the general population. These studies found that protein-truncating variants in BRCA1, BRCA2, and PALB2 fall into the high-risk category, while those in ATM, BARD1, CHEK2, RAD51C, and RAD51D are categorized as moderate risk. This highlights their clinical utility for inclusion in BC risk prediction panels.^11,12

It is important to scrutinize the genetic variants in a population since it provides an insight into several aspects regarding BC prediction and patient management. Genetic testing of familial BC patients facilitates the identification of population-based genetic variants, which can enhance our understanding of the disease process and lead to more targeted treatments. Exploring the patterns and role of genetic heterogeneity in sporadic BC patients is also important.¹³ Therefore, genetic testing can be used as a powerful predictive tool. Individuals who have a strong family history of BC are more prone to get the disease, and gene panel testing has a growing clinical importance in treatment options such as prophylactic mastectomy and chemopreventive drugs.⁸ However, as BC risk is influenced by other genetic and lifestyle factors, in addition to family history, it is important to consider the genetic test results in risk assessment. Considering all the above facts, this study aims to reveal genetic polymorphisms in cancer-predisposing genes among the BC confirmed patients and individuals at risk of developing BC in a selected group of patients presented to two BC imaging facilities located in the Central Province of Sri Lanka.

This study was approved by the Ethics Review Committee, Faculty of Allied Health Sciences, University of Peradeniya, Sri Lanka (AHS/ERC/2018/001). The guidelines and methodology approved by the Ethics Review Committee were followed to complete the study. This is a descriptive, retrospective study carried out from July 2018 to July 2019. Written informed consent for the study was obtained from the entire study participants after providing pre-test counseling. A detailed questionnaire was given to the study group to obtain data on a voluntary basis. The questionnaire was pilot tested in approximately 12% of participants to ensure clarity and cultural appropriateness. Participants were recruited from two breast imaging units in Kandy, Sri Lanka. Privacy and the confidentiality of all the data collected from the patients are maintained. Altogether, the cohort consisted of 79 subjects. Sample size was determined by the availability of eligible patients within the study period. Guidelines given by the American College of Molecular Genetics and Genomics were adopted when selecting the patients for this study. Study participants were recruited as four groups. (i) familial BC patients with a positive family history (n = 13), (ii) sporadic BC patients without a family history (n = 20), (iii) at risk individuals who had at least one first- or second-degree relative with BC but no personal history of the disease (n = 26), and (iv) healthy controls frequency matched by age and without personal or family history of any cancer (n = 10). The BC patients were selected for the study after reviewing their mammography and histopathology reports. Family history is defined by at least one first- or second-degree female relative reported to have had BC. Those who were in the terminally ill stage and those who had not unconsented were excluded from the study. An additional group of age-matched at-risk individuals (n = 10) was included solely for primer validation. These samples were analyzed using Sanger sequencing to confirm successful amplification of target regions in BRCA2 and PALB2 and to ensure the reliability of the designed PCR assays. Their data were not included in the main cohort analysis.

Each study participant was subjected to venipuncture under strict aseptic condition and 2 ml blood sample was collected into an ethylene diaminetetra acetic acid (EDTA) tube and immediately stored at −20°C. Genomic DNA was extracted from peripheral white blood cells using the Promega genomic DNA purification kit ® (Promega, Madison, USA). Then, these DNA samples were subjected to Next-Generation sequencing (NGS), and 10 samples were analyzed by Sanger sequencing. The inherited predisposition to BC was evaluated using 18-gene panel test on the Credence breast platform to analyze genes having BC predisposition (BRCA1, BRCA2, TP53, ATM, BARD1, BRIP1, CDH1, CHEK2, MRE11A, MUTYH, NBN, NF1, PALB2, PEN, RAD50, RAD51C, RAD51D, and STK11). Genomic regions were amplified using oligonucleotides, prepared into single-ended libraries of 200 bp size followed by Next Generation Sequencing on Ion Torrent PGM. The DNA sequences were mapped to the published Human Genome (GRCh37/hg19) using TMAP aligner version 4.0.6, followed by Variant Calling (TVC: Version 4.2.3.) and annotation (VEP: Version 75) using Credence Breast Cancer Pipeline version 1.1.2 (ClinVar:02: Mar:2015 Update). Variants were classified according to the ACMG/AMP 5-tier system (pathogenic, likely pathogenic, variant of uncertain significance, likely benign, and benign). After analyzing the results, reports were provided to participants along with post-test counseling as a benefit of their participation. The sequences obtained for 10 samples completed by Sanger sequencing were checked for mutations and variants using BioEdit software (version 7.2). Primers were designed to amplify the pathological mutations identified for BRCA2 and PALB2 genes in this cohort to develop a PCR-based assay. Among these 10 samples, six were representative at-risk individuals with a positive family history who had been included in the main cohort, and four were additional age-matched at-risk individuals. All 10 samples were used exclusively for primer validation. PCR products obtained using the designed primers were sequenced by the Sanger method, and the resulting sequences were analyzed through multiple sequence alignments and BLAST searches to verify the presence of single-nucleotide polymorphisms (SNPs) or mutations. To minimize potential sources of bias, standardized data collection methods were used for all participants, and genetic analyses were performed using the same laboratory protocols and blinded sample identifiers. The reporting of this study conforms to the STROBE statement for observational studies;¹⁴ the completed checklist is provided as Supplementary File 4.

The Scope of this study was to identify genetic variants among familial BC patients, sporadic BC patients, and individuals who are at risk of developing BC in a patient cohort of Sri Lanka. Sixty-nine individuals were included in the study for Next Generation Sequencing, and pathogenic variants, frequent missense, and intronic variants were identified. In contrast to earlier reports suggesting that familial BCs present at younger ages, our cohort demonstrated a higher mean age at diagnosis in familial patients (57.2 ± 7.1 years) compared with sporadic patients (52.2 ± 13.3 years). This difference could be influenced by our limited sample size, so broader studies will be important to clarify this pattern.¹²

Four pathogenic variants in highly penetrant genes, BRCA2 and PALB2, were identified in our study. These genes play key roles in DNA repair and maintaining genomic integrity.¹³ All identified variants have been previously reported and are listed in the Breast Cancer Information Core (BIC) database.¹⁵ Consistent with our findings, several studies have shown that a greater number of pathogenic variants are associated with BRCA2 than with BRCA1.^16-19 However, the relative risk of developing BC remains higher for individuals with BRCA1 mutations.¹⁸

Several studies have highlighted the unique spectrum of germline variants in Sri Lankan BC patients. Initial work conducted by De Silva et al identified two novel pathogenic frameshift mutations, c.3086delT and c.5404delG along with several novel intronic and missense variants. This highlighted the genetic heterogeneity within this population.²⁰ A follow-up study on BRCA2 in a similar cohort revealed a higher frequency of potentially pathogenic variants, including c.2403insA and c.2667insT, suggesting that BRCA2 may play a more significant role than BRCA1 in hereditary BC predisposition in Sri Lanka.²¹ In young familial BC patients, 23% carried pathogenic BRCA2 variants, emphasizing the need for early, targeted testing.²² More recent multigene panel testing revealed additional mutations in moderate-risk genes such as CHEK2, ATM, PALB2, and CDKN2A, supporting the value of comprehensive genetic screening beyond BRCA1/2 in the Sri Lankan population.²³

BRCA1 [c.3113A>G; p.Glu1038Gly] variant was the most common variant in our cohort and has also been reported at high frequencies in Algerian and Indian populations.^24,25 In both studies, the variant was found to exert a benign effect. It was found in both BC confirmed patients (familial and sporadic) and at-risk individuals in the present study. BRCA2 deleterious mutation [c.5574_5577delAATT; p.Ile1859LysfsTer3] had originally been reported in a study intended to identify the prevalence and spectrum of BRCA1/2 germline mutations in women with BC in China²⁶ and BRCA2 [c.6509A>G; p.Lys2170Arg] was barely found in general.²⁷ On the contrary, BRCA2 [c.7879A>T; p.Ile2627Phe] was most commonly found in Macedonian and Lithuanian populations.^19,28 However, none of these variants were reported in Sri Lankan studies conducted earlier.^15,20-29 PALB2 [c.1592delT; p.Leu531fs], a high-risk deleterious mutation, was also identified in our study.

Several studies have shown that it plays an important role in BC, particularly in cases with a familial origin.^23,30 However, none of the previous studies involving Asians or Sri Lankan populations have reported this mutation. In addition to pathogenic variants in high-risk genes such as BRCA2 and PALB2, variants have also been identified in BRIP1 and MRE11A. However, pathogenic variants in these genes are generally not considered to confer a significant risk for BC development.³¹

Other than the pathogenic polymorphisms, variants with uncertain significance (VUS) exert an important function in BC. A study stated that most of the variants of uncertain significance are basically missense, silent, and intronic variants or in-frame deletions and insertions.³² Consistent with this, Arachchige et al conducted a comprehensive bioinformatics analysis of selected germline VUS identified in a Sri Lankan hereditary BC cohort, revealing that several variants in BRCA1, BRIP1, and MET may have deleterious structural and functional consequences. These findings highlight the need for follow-up functional work in the local context.³³ Therefore, it was essential to study the pathogenicity and the role of these variants in BC. In this study, frequent missense and intronic variants were observed. Missense variant is a change in the type of amino acid without altering the protein. Though the protein is not truncated, it is hypothesized that this may initiate a pathogenic mechanism which leads to BC.³⁴ Also, most of these missense polymorphisms are SNPs. The current study revealed more than 100 missense variants, and some were found in higher frequencies. Those were reported in BRCA1 [c.3113A > G; p.Glu1038Gly; c.3548A>G; p.Lys1183Arg; c.2612C>T; p.Pro871Leu], BRCA2 [c.7397T>C; p.Val2466Ala], and ATM [c.5948A>G; p.Asn1983Ser] like high-penetrance genes. Similarly, these polymorphisms shared with Algeria,²⁶ India,²⁷ Colombia,³⁵ and Bahrain³⁶ studies have a considerably higher frequency of [c.3113A>G, c.3548A>G, c.2612C>T, c.7397T>C] variants. As reported earlier,³⁶ this study also confirmed that the BRCA1:rs799917 (c.2612C>T; p.Pro871Leu) variant was not associated with BC risk. However, ATM [c.5948A>G; p.Asn1983Ser] was also reported in this cohort. In Sri Lanka, two similar experiments have been conducted on multigene panel testing of BC,^23,30 and frequent missense polymorphisms in the present study were not reported in those experiments except for c.823G>A; p.Gly275Ser in BRCA1 gene. Several other missense polymorphisms found in this study are given in Table 5.

In addition to the effect of pathogenic and other missense variants, it is evident that the intronic variants also play a particular significance in terms of BC risk where the exact mechanism is yet to be explored. It is stated that the intronic variants in intronic and exonic boundaries can disturb the splicing capacity.³⁴ Therefore, it is important to have data on common intronic polymorphisms in a community. A present study revealed a high number of intronic variants while three of them being more frequent. MUTYH is a DNA repairing gene which shows a notable association with colorectal cancer patients. However, it is found that mono-allelic and bi-allelic variants of this gene were also noted in different BC cohorts.^36-38 Based on the National Center for Biotechnology Information (NCBI) database, MUTYH c.1468-40C>G was previously reported as a benign single-nucleotide variant with an unknown disease specificity which was found in this study as well.³⁹ PALB2 c.3114-51T>A was another variant frequently reported in the current study. Similarly, this was identified in almost all individuals in an Australian research with 347 study participants and also in a South African study.^40,41 NF1 acts as a gene producing tumor suppressor protein neurofibromin.⁴² NF1 c.288+41G>A has previously been reported in the NCBI database as a benign single-nucleotide variant.⁴³ However, this variant has not shown a clinical effect in the current study. Similarly, there was no effect in NF1 c.328+37C>G variant that was not reported in previous literature.

Evaluation of germ line variants using 18 gene BC panel testing in this cohort revealed that the pathogenic mutations and other variants with benign or uncertain significance in both familial and sporadic BC patients in many genes of this cohort. Altogether, three pathogenic variants were found, and none of those have been reported from previous studies conducted in Sri Lanka. The benign BRCA1 [c.3113A>G; p.Glu1038Gly] variant identified in this cohort has been found in Algerian and Indian populations as well. Deleterious BRCA2 [c.5574_5577delAATT; p.Ile1859LysfsTer3] found in this study has been previously reported among Chinese. BRCA2 [c.509A>G; p.Lys2170Arg] identified in this study was found in Poland and BRCA2 [c.7879A>T; p.Ile2627Phe] was most commonly found among Macedonians and Lithuanians. The Finnish population was reported to have PALB2 [c.1592delT; p.Leu531fs] deletion. Recent studies in diverse populations support testing beyond BRCA1/2. In Colombia, multi-gene analysis of unselected BC patients identified pathogenic variants in ATM, PALB2, CHEK2, BARD1, and RAD51D in addition to BRCA1/2.⁴⁴ Similarly, in Turkey, BRCA1/2 mutations were linked to earlier onset even in patients without a family history.⁴⁵ These findings highlight the value of broader gene-panel testing for comprehensive risk assessment in genetically diverse populations.

Additionally, in the present study, a diverse array of protein-truncating missense variants and intronic variants were observed in large numbers while a few were more frequent. These variants were mostly SNPs, and the frequency of variants was different within each study group. These results would have been missed if the testing were strictly limited to single-gene or syndrome testing guidelines.

In the Sri Lankan setting, using PCR-based assays to detect the main pathogenic mutations found in BRCA2 and PALB2 offers a practical and affordable alternative to costly NGS-based BC panels. Many people who need genetic risk screening are reluctant to undergo multigene panel testing because of the high expense. As a solution, targeted PCR followed by Sanger sequencing of commonly mutated regions can be used as an initial screening approach. Individuals who test positive can then be referred for full NGS or exome sequencing, helping to use resources more effectively while improving early detection and risk management. In this study, a group of age-matched at-risk individuals (n = 10) was included for primer validation to confirm the reliability of the method. Among these, six were from the 26 at-risk individuals included in the main cohort analysis, and four were additional age-matched individuals. Sanger sequencing of these samples showed clear amplification of the target regions in BRCA2 and PALB2, confirming that the designed primers were specific and efficient. Sequence analysis using BioEdit and BLAST further verified the presence of previously identified mutations and single-nucleotide variants. Overall, the PCR assay successfully amplified the same pathogenic regions detected by NGS in the main cohort, showing its potential as a reliable and low-cost screening tool for BRCA2 and PALB2 mutations in familial BC.

Although this study primarily analyzed BC patients, a group of healthy control individuals was also subjected to NGS analysis, allowing for comparative assessment of variant occurrence. Variants that were shared between patients and healthy controls are likely germline polymorphisms, whereas those uniquely detected in patients may represent disease-associated changes. However, since paired tumor and normal DNA were not analyzed from the same individuals, it is not possible to definitively classify these as somatic or germline mutations. Future studies incorporating matched tumor-normal sequencing would provide greater resolution in determining variant origin.

The relatively small cohort size (N = 79) may limit the generalizability of our findings. However, this study is one of the few attempts to genetically characterize BC-associated variants in a Sri Lankan population, which remains significantly underrepresented in global genomic datasets. It is well established that BRCA1 and BRCA2 mutations in Sri Lankan individuals tend to be private mutations rather than founder mutations. Thus, the identification of even a few rare or previously unreported variants is not unexpected but is still clinically relevant. These findings contribute toward building a population-specific mutation database and may eventually support the design of targeted diagnostic tools for hereditary BC risk prediction in Sri Lanka. Our sample size was limited primarily due to funding constraints and the high cost of NGS. Despite this, the study has laid a valuable foundation for understanding the local mutational landscape and highlights the need for broader, multi-center studies with larger cohorts to validate and expand on these preliminary observations. To improve accessibility and cost-effectiveness in low-resource settings like Sri Lanka, we recommend the development of simplified screening approaches. PCR-based simple tests can be used to amplify common mutated regions of at-risk individuals with positive family history. It revealed that mutations are reported in BRCA1, BRCA2, and PALPB2 genes commonly in the Sri Lankan community based on the results obtained from the previous studies and current study. PCR-based tests with Sanger sequencing can be used to check the common mutated regions, and it may be cost-effective to use specific primers to amplify these regions in high-risk genes using multiplex PCR and to analyze the sequences rather than using single-gene tests. Multigene BC panels available in Sri Lanka are expensive because of the use of NGS platforms. PCR & Sanger sequencing-based screening tests would be economically reasonable for the Sri Lankan community who need to screen for hereditary BC risk prediction.

This study identified three pathogenic germline variants in the high-penetrance genes BRCA2 and PALB2, present at low frequencies (<10%) among a Sri Lankan cohort of BC patients and at-risk individuals. Multiple benign and uncertain variants, including common missense and intronic variants, were also detected. PCR and Sanger sequencing confirmed the reproducibility of key pathogenic mutations, demonstrating the reliability of the designed primers. Our findings emphasize the genetic heterogeneity of BC in Sri Lanka and support the integration of population-specific genetic information into risk assessment and management strategies. Larger studies are needed to confirm these results and to investigate the functional impact of frequent variants.

We would like to acknowledge the financial assistance given by the University of Peradeniya, Sri Lanka. Furthermore, we would like to acknowledge all the participants of this study.