Targeted sequencing to discover germline variants in the BRCA1 and BRCA2 genes in a Russian population and their association with breast cancer risk

High- and Moderate-Risk Variants Among Breast Cancer Patients and Healthy Donors Enrolled in Multigene Panel Testing in a Population of Central Russia

Assessments of breast cancer (BC) risk in carriers of pathogenic variants identified by gene panel testing in different populations are highly in demand worldwide. We performed target sequencing of 78 genes involved in DNA repair in 860 females with BC and 520 age- and family history-matched controls from Central Russia. Among BC patients, 562/860 (65.3%) were aged 50 years or less at the time of diagnosis. In total, 190/860 (22%) BC patients were carriers of 198 pathogenic/likely pathogenic (P/LP) variants in 30 genes, while among controls, 32/520 (6.2%) carriers of P/LP variants in 17 genes were identified. The odds ratio [95% confidence interval] was 16.3 [4.0-66.7] for BRCA1; 12.0 [2.9-45.9] for BRCA2; and 7.3 [0.9-56.7] for ATM (p < 0.05). Previously undescribed BRCA1/2, ATM, and PALB2 variants, as well as novel recurrent mutations, were identified. The contribution to BC susceptibility of truncating variants in the genes BARD1, RAD50, RAD51C, NBEAL1 (p. E1155*), and XRCC2 (p. P32fs) was evaluated. The BLM, NBN, and MUTYH genes did not demonstrate associations with BC risk. Finding deleterious mutations in BC patients is important for diagnosis and management; in controls, it opens up the possibility of prevention and early diagnostics.

Untapped Potential of Poly(ADP-Ribose) Polymerase Inhibitors: Lessons Learned From the Real-World Clinical Homologous Recombination Repair Mutation Testing

Background

Testing for homologous recombination deficiency (HRD) mutations is pivotal to assess individual risk, to proact preventive measures in healthy carriers and to tailor treatments for cancer patients. Increasing prominence of poly(ADP-ribose) polymerase (PARP) inhibitors with remarkable impact on molecular-selected patient survival across diverse nosologies, ingrains testing for BRCA genes and beyond in clinical practice. Nevertheless, testing strategies remain a question of debate. While several pathogenic BRCA1/2 gene variants have been described as founder pathogenic mutations frequently found in patients from Russia, other homologous recombination repair (HRR) genes have not been sufficiently explored. In this study, we present real-world data of routine HRR gene testing in Russia.

Methods

We evaluated clinical and sequencing data from cancer patients who had germline/somatic next-generation sequencing (NGS) HRR gene testing in Russia (BRCA1/2/ATM/CHEK2, or 15 HRR genes). The primary objectives of this study were to evaluate the frequency of BRCA1/2 and non-BRCA gene mutations in real-world unselected patients from Russia, and to determine whether testing beyond BRCA1/2 is feasible.

Results

Data of 2,032 patients were collected from February 2021 to February 2023. Most had breast (n = 715, 35.2%), ovarian (n = 259, 12.7%), pancreatic (n = 85, 4.2%), or prostate cancer (n = 58, 2.9%). We observed 586 variants of uncertain significance (VUS) and 372 deleterious variants (DVs) across 487 patients, with 17.6% HRR-mutation positivity. HRR testing identified 120 (11.8%) BRCA1/2-positive, and 172 (16.9%) HRR-positive patients. With 51 DVs identified in 242 formalin-fixed paraffin-embedded (FFPE), testing for variant origin clarification was required in one case (0.4%). Most BRCA1/2 germline variants were DV (121 DVs, 26 VUS); in non-BRCA1/2 genes, VUS were ubiquitous (53 DVs, 132 VUS). In silico prediction identified additional 4.9% HRR and 1.2% BRCA1/2/ATM/CHEK2 mutation patients.

Conclusions

Our study represents one of the first reports about the incidence of DV and VUS in HRR genes, including genes beyond BRCA1/2, identified in cancer patients from Russia, assessed by NGS. In silico predictions of the observed HRR gene variants suggest that non-BRCA gene testing is likely to result in higher frequency of patients who are candidates for PARP inhibitor therapy. Continuing sequencing efforts should clarify interpretation of frequently observed non-BRCA VUS.

BRACNAC: A BRCA1 and BRCA2 Copy Number Alteration Caller from Next-Generation Sequencing Data

Detecting copy number variations (CNVs) and alterations (CNAs) in the BRCA1 and BRCA2 genes is essential for testing patients for targeted therapy applicability. However, the available bioinformatics tools were initially designed for identifying CNVs/CNAs in whole-genome or -exome (WES) NGS data or targeted NGS data without adaptation to the BRCA1/2 genes. Most of these tools were tested on sample cohorts of limited size, with their use restricted to specific library preparation kits or sequencing platforms. We developed BRACNAC, a new tool for detecting CNVs and CNAs in the BRCA1 and BRCA2 genes in NGS data of different origin. The underlying mechanism of this tool involves various coverage normalization steps complemented by CNV probability evaluation. We estimated the sensitivity and specificity of our tool to be 100% and 94%, respectively, with an area under the curve (AUC) of 94%. The estimation was performed using the NGS data obtained from 213 ovarian and prostate cancer samples tested with in-house and commercially available library preparation kits and additionally using multiplex ligation-dependent probe amplification (MLPA) (12 CNV-positive samples). Using freely available WES and targeted NGS data from other research groups, we demonstrated that BRACNAC could also be used for these two types of data, with an AUC of up to 99.9%. In addition, we determined the limitations of the tool in terms of the minimum number of samples per NGS run (≥20 samples) and the minimum expected percentage of CNV-negative samples (≥80%). We expect that our findings will improve the efficacy of BRCA1/2 diagnostics. BRACNAC is freely available at the GitHub server.

EpiMC: Detecting Epistatic Interactions Using Multiple Clusterings

Detecting single nucleotide polymorphisms (SNPs) interactions is crucial to identify susceptibility genes associated with complex human diseases in genome-wide association studies. Clustering-based approaches are widely used in reducing search space and exploring potential relationships between SNPs in epistasis analysis. However, these approaches all only use a single measure to filter out nonsignificant SNP combinations, which may be significant ones from another perspective. In this paper, we propose a two-stage approach named EpiMC (Epistatic Interactions detection based on Multiple Clusterings) that employs multiple clusterings to obtain more precise candidate sets and more comprehensively detect high-order interactions based on these sets. In the first stage, EpiMC proposes a matrix factorization based multiple clusterings algorithm to generate multiple diverse clusterings, each of which divide all SNPs into different clusters. This stage aims to reduce the chance of filtering out potential candidates overlooked by a single clustering and groups associated SNPs together from different clustering perspectives. In the next stage, EpiMC considers both the single-locus effects and interaction effects to select high-quality disease associated SNPs, and then uses Jaccard similarity to get candidate sets. Finally, EpiMC uses exhaustive search on the obtained small candidate sets to precisely detect epsitatic interactions. Extensive simulation experiments show that EpiMC has a better performance in detecting high-order interactions than state-of-the-art solutions. On the Wellcome Trust Case Control Consortium (WTCCC) dataset, EpiMC detects several significant epistatic interactions associated with breast cancer (BC) and age-related macular degeneration (AMD), which again corroborate the effectiveness of EpiMC.

Targeted Sequencing of 242 Clinically Important Genes in the Russian Population From the Ivanovo Region

We performed a targeted sequencing of 242 clinically important genes mostly associated with cardiovascular diseases in a representative population sample of 1,658 individuals from the Ivanovo region northeast of Moscow. Approximately 11% of 11,876 detected variants were not found in the Single Nucleotide Polymorphism Database (dbSNP) or reported earlier in the Russian population. Most novel variants were singletons and doubletons in our sample, and virtually no novel alleles presumably specific for the Russian population were able to reach the frequencies above 0.1-0.2%. The overwhelming majority (99.3%) of variants detected in this study in three or more copies were shared with other populations. We found two dominant and seven recessive known pathogenic variants with allele frequencies significantly increased compared to those in the gnomAD non-Finnish Europeans. Of the 242 targeted genes, 28 were in the list of 59 genes for which the American College of Medical Genetics and Genomics (ACMG) recommended the reporting of incidental findings. Based on the number of variants detected in the sequenced subset of ACMG59 genes, we approximated the prevalence of known pathogenic and novel or rare protein-truncating variants in the complete set of ACMG59 genes in the Ivanovo population at 1.4 and 2.8%, respectively. We analyzed the available clinical data and observed the incomplete penetrance of known pathogenic variants in the 28 ACMG59 genes: only 1 individual out of 12 with such variants had the phenotype most likely related to the variant. When known pathogenic and novel or rare protein-truncating variants were considered together, the overall rate of confirmed phenotypes was about 19%, with maximum in the subset of novel protein-truncating variants. We report three novel protein truncating variants in APOB and one in MYH7 observed in individuals with hypobetalipoproteinemia and hypertrophic cardiomyopathy, respectively. Our results provide a valuable reference for the clinical interpretation of gene sequencing in Russian and other populations.

A novel frequent BRCA1 recurrent variant c.5117G > A (p.Gly1206Glu) identified after 20 years of BRCA1/2 research in the Baltic region: cohort study and literature review

Background

Several recent studies in the Baltic region have found extended spectrum of pathogenic variants (PV) of the BRCA1/2 genes. The aim of current study is to analyze the spectrum of the BRCA1/2 PV in population of Latvia and to compare common PV between populations of the Baltic region.

Methods

We present a cohort of 9543 unrelated individuals including ones with cancer and unaffected individuals from population of Latvia, who were tested for three most common BRCA1 founder PV. In second line testing, 164 founder negative high-risk individuals were tested for PV of the BRCA1/2 using next generation sequencing (NGS). Local spectrum of the BRCA1/2 PV was compared with the Baltic region by performing a literature review.

Results

Founder PV c.5266dupC, c.4035delA or c.181 T > G was detected in 369/9543 (3.9%) cases. Other BRCA1/2 PV were found in 44/164 (26.8%) of NGS cases. Four recurrent BRCA1 variants c.5117G > A (p.Gly1706Glu), c.4675G > A (p.Glu1559Lys), c.5503C > T (p.Arg1835*) and c.1961delA (p.Lys654fs) were detected in 18/44 (41.0%), 5/44 (11.4%), 2/44 (4.5%) and 2/44 (4.5%) cases respectively. Additionally, 11 BRCA1 PV and six BRCA2 PV were each found in single family.

Conclusions

By combining three studies by our group of the same cohort in Latvia, frequency of the BRCA1/2 PV for unselected breast and ovarian cancer cases is 241/5060 (4.8%) and 162/1067 (15.2%) respectively. The frequency of three “historical” founder PV is up to 87.0% (369/424). Other non-founder PV contribute to at least 13.0% (55/424) and this proportion probably will rise by increasing numbers of the BRCA1/2 sequencing. In relative numbers, c.5117G > A is currently the third most frequent PV of the BRCA1 in population of Latvia, overcoming previously known third most common founder variant c.181 T > G. In addition to three BRCA1 founder PV, a total of five recurrent BRCA1 and two recurrent BRCA2 PV have been reported in population of Latvia so far. Many of the BRCA1/2 PV reported in Latvia are shared among other populations of the Baltic region.

Characterization and in silico analyses of the BRCA1/2 variants identified in individuals with personal and/or family history of BRCA-related cancers

Pathogenic variants in the coding regions of the BRCA1/2 lead dysfunctional or nonfunctional BRCA proteins however the contribution of non-coding BRCA1/2 variants to BRCA-related disease risk has not been fully elucidated. Thus, we characterized the functional impact of both coding and non-coding BRCA1/2 variants identified in individuals with personal and/or family history of BRCA-related cancers. The data were produced by resequencing the exons and exon-intron junctions of the BRCA1/2 in 125 individuals and were comprehensively analyzed by using bioinformatics tools and databases. A total of 96 variants (59 coding and 37 non-coding) including 7 novel variants were identified and analyzed for their functional importance. We identified 11 missense variants that potentially affect protein function; 22 variants were likely to alter different types of posttranslational modifications. Also, multiple non-coding BRCA1/2 variants were found to reside in the critical regulatory regions that have the potential to act as eQTLs and affect alternative splicing. The results of our study shed light on the possible contributions of not only coding variants but also non-coding BRCA1/2 variants in BRCRA-related cancers. Further investigation is required to fully understand their potential associations with phenotypes which may ultimately lead their utilization on cancer management as a biomarker.