Although germline copy-number variants (CNVs) are the genetic cause of multiple hereditary diseases, detecting them from targeted next-generation sequencing data (NGS) remains a challenge. Existing tools perform well for large CNVs but struggle with single and multi-exon alterations. The aim of this work is to evaluate CNV calling tools working on gene panel NGS data and their suitability as a screening step before orthogonal confirmation in genetic diagnostics strategies. Five tools (DECoN, CoNVaDING, panelcn.MOPS, ExomeDepth, and CODEX2) were tested against four genetic diagnostics datasets (two in-house and two external) for a total of 495 samples with 231 single and multi-exon validated CNVs. The evaluation was performed using the default and sensitivity-optimized parameters. Results showed that most tools were highly sensitive and specific, but the performance was dataset dependant. When evaluating them in our diagnostics scenario, DECoN and panelcn.MOPS detected all CNVs with the exception of one mosaic CNV missed by DECoN. However, DECoN outperformed panelcn.MOPS specificity achieving values greater than 0.90 when using the optimized parameters. In our inhouse datasets, DECoN and panelcn.MOPS showed the highest performance for CNV screening before orthogonal confirmation. Benchmarking and optimization code is freely available at https://github.com/TranslationalBioinforma ticsIGTP/CNVbenchmarkeR.

Fanconi anemia (FA) is caused by biallelic mutations in FA genes. Monoallelic mutations in five of these genes (BRCA1, BRCA2, PALB2, BRIP1 and RAD51C) increase the susceptibility to breast/ovarian cancer and are used in clinical diagnostics as bona-fide hereditary cancer genes. Increasing evidence suggests that monoallelic mutations in other FA genes could predispose to tumor development, especially breast cancer. The objective of this study is to assess the mutational spectrum of 14 additional FA genes (FANCA, FANCB, FANCC, FANCD2, FANCE, FANCF, FANCG, FANCI, FANCL, FANCM, FANCP, FANCQ, FANCR and FANCU) in a cohort of hereditary cancer patients, to compare with local cancer-free controls as well as GnomAD. A total of 1021 hereditary cancer patients and 194 controls were analyzed using our next generation custom sequencing panel. We identified 35 pathogenic variants in eight genes. A significant association with the risk of breast cancer/breast and ovarian cancer was found for carriers of FANCA mutations (odds ratio (OR) = 3.14 95% confidence interval (CI) 1.4–6.17, p = 0.003). Two patients with early-onset cancer showed a pathogenic FA variant in addition to another germline mutation, suggesting a modifier role for FA variants. Our results encourage a comprehensive analysis of FA genes in larger studies to better assess their role in cancer risk.

Only a small fraction of hereditary breast and/or ovarian cancer (HBOC) cases are caused by germline variants in the high-penetrance breast cancer 1 and 2 genes (BRCA1 and BRCA2). BRCA1-associated ring domain 1 (BARD1), nuclear partner of BRCA1, has been suggested as a potential HBOC risk gene, although its prevalence and penetrance are variable according to populations and type of tumor. We aimed to investigate the prevalence of BARD1 truncating variants in a cohort of patients with clinical suspicion of HBOC. A comprehensive BARD1 screening by multigene panel analysis was performed in 4015 unrelated patients according to our regional guidelines for genetic testing in hereditary cancer. In addition, 51,202 Genome Aggregation Database (gnomAD) non-Finnish, non-cancer European individuals were used as a control population. In our patient cohort, we identified 19 patients with heterozygous BARD1 truncating variants (0.47%), whereas the frequency observed in the gnomAD controls was 0.12%. We found a statistically significant association of truncating BARD1 variants with overall risk (odds ratio (OR) = 3.78; CI = 2.10–6.48; p = 1.16 × 10−5). This association remained significant in the hereditary breast cancer (HBC) group (OR = 4.18; CI = 2.10–7.70; p = 5.45 × 10−5). Furthermore, deleterious BARD1 variants were enriched among triple-negative BC patients (OR = 5.40; CI = 1.77–18.15; p = 0.001) compared to other BC subtypes. Our results support the role of BARD1 as a moderate penetrance BC predisposing gene and highlight a stronger association with triple-negative tumors.

CHEK2 variants are associated with intermediate breast cancer risk, among other cancers. We aimed to comprehensively describe CHEK2 variants in a Spanish hereditary cancer (HC) cohort and adjust the American College of Medical Genetics and Genomics and the Association for Molecular Pathology (ACMG-AMP) guidelines for their classification. First, three CHEK2 frequent variants were screened in a retrospective Hereditary Breast and Ovarian Cancer cohort of 516 patients. After, the whole CHEK2 coding region was analyzed by next-generation sequencing in 1848 prospective patients with HC suspicion. We refined ACMG-AMP criteria and applied different combined rules to classify CHEK2 variants and define risk alleles. We identified 10 CHEK2 null variants, 6 missense variants with discordant interpretation in ClinVar database, and 35 additional variants of unknown significance. Twelve variants were classified as (likely)-pathogenic; two can also be considered "established risk-alleles" and one as "likely risk-allele." The prevalence of (likely)-pathogenic variants in the HC cohort was 0.8% (1.3% in breast cancer patients and 1.0% in hereditary nonpolyposis colorectal cancer patients). Here, we provide ACMG adjustment guidelines to classify CHEK2 variants. We hope that this study would be useful for variant classification of other genes with low effect variants.

No abstract

Motivation:Although germline copy number variants (CNVs) are the genetic cause of multiple hereditary diseases, detecting them from targeted next-generation sequencing data (NGS) remains a challenge. Existing tools perform well for large CNVs but struggle with single and multi-exon alterations. The aim of this work is to evaluate CNV calling tools working on gene panel NGS data with CNVs up to single-exon resolution and their suitability as a screening step before orthogonal confirmation in genetic diagnostics strategies.Results: Five tools (DECoN, CoNVaDING, panelcn.MOPS, ExomeDepth and CODEX2) were tested against four genetic diagnostics datasets (495 samples, 231 CNVs), using the default and sensitivityoptimized parameters. Most tools were highly sensitive and specific, but the performance was datasetdependant. In our in-house datasets, DECoN and panelcn.MOPS with optimized parameters showed enough sensitivity to be used as screening methods in genetic diagnostics. Availability:Benchmarking-optimization code is freely available at https://github.com/TranslationalBioinformaticsIGTP/CNVbenchmarkeR.

Motivation Germline variant classification allows accurate genetic diagnosis and risk assessment. However, it is a tedious iterative process integrating information from several sources and types of evidence. It should follow gene-specific (if available) or general updated international guidelines. Thus, it is the main burden of the incorporation of NGS into the clinical setting. Results We created the vaRHC R package to assist the process of variant classification in hereditary cancer by : 1) collecting information from diverse databases; 2) assigning or denying different types of evidence according to updated ACMG/AMP gene-specific criteria for ATM, CDH1, CHEK2, MLH1, MSH2, MSH6, PMS2, PTEN, and TP53 and general criteria for other genes; 3) providing an automated classification of variants using a Bayesian metastructure and considering CanVIG-UK recommendations; 4) optionally printing the output to an .xlsx file. A validation using 659 classified variants demonstrated the robustness of vaRHC, presenting a better criteria assignment than Cancer SIGVAR, an available similar tool. Availability The source code can be consulted in the GitHub repository (https://github.com/emunte/vaRHC) Additionally, it will be submitted to CRAN soon. Supplementary information Supplementary data are available at Bioinformatics online.

IntroductionGermline CNVs are important contributors to hereditary cancer. In genetic diagnostics, multiplex ligation-dependent probe amplification (MLPA) is commonly used to identify them. However, MLPA is time-consuming and expensive if applied to many genes, hence many routine laboratories test only a subset of genes of interest.Methods and resultsWe evaluated a next-generation sequencing (NGS)-based CNV detection tool (DECoN) as first-tier screening to decrease costs and turnaround time and expand CNV analysis to all genes of clinical interest in our diagnostics routine. We used DECoN in a retrospective cohort of 1860 patients where a limited number of genes were previously analysed by MLPA, and in a prospective cohort of 2041 patients, without MLPA analysis. In the retrospective cohort, 6 new CNVs were identified and confirmed by MLPA. In the prospective cohort, 19 CNVs were identified and confirmed by MLPA, 8 of these would have been lost in our previous MLPA-restricted detection strategy. Also, the number of genes tested by MLPA across all samples decreased by 93.0% in the prospective cohort.ConclusionIncluding an in silico germline NGS CNV detection tool improved our genetic diagnostics strategy in hereditary cancer, both increasing the number of CNVs detected and reducing turnaround time and costs.