Venkat Subramaniam Rathinakannan

Background: Breast cancer is the most common malignancy, with a mean age of onset of approximately 60 years. Only a minority of breast cancer patients present with an early onset at or before 40 years of age. An exceptionally young age at diagnosis hints at a possible genetic etiology. Currently, known pathogenic genetic variants only partially explain the disease burden of younger patients. Thus, new knowledge is warranted regarding additional risk variants. In this study, we analyzed DNA repair genes to identify additional variants to shed light on the etiology of early-onset breast cancer. Methods: Germline whole-exome sequencing was conducted in a cohort of 63 patients diagnosed with breast cancer at or before 40 years of age (median 33, mean 33.02, range 23–40 years) with no known pathogenic variants in BRCA genes. After filtering, all detected rare variants were sorted by pathogenicity prediction scores (CADD score and REVEL) to identify the most damaging genetic changes. The remaining variants were then validated by comparison to a validation cohort of 121 breast cancer patients with no preselected age at cancer diagnosis (mean 51.4 years, range 28–80 years). Analysis of novel exonic variants was based on protein structure modeling. Results: Five novel, deleterious variants in the genes WRN, RNF8, TOP3A, ERCC2, and TREX2 were found in addition to a splice acceptor variant in RNF4 and two frameshift variants in EXO1 and POLE genes, respectively. There were also multiple previously reported putative risk variants in other DNA repair genes. Conclusions: Taken together, whole-exome sequencing yielded 72 deleterious variants, including 8 novel variants that may play a pivotal role in the development of early-onset breast cancer. Although more studies are warranted, we demonstrate that young breast cancer patients tend to carry multiple deleterious variants in one or more DNA repair genes.

Annotation of large amounts of generated sequencing data is a demanding task. Most of the currently available robust annotation tools, like ANNOVAR, are command-line based tools which require a certain degree of programming skills. User-friendly tools for variant annotation of sequencing data with graphical interface are under-represented. We have developed an interactive application, which harnesses the easy usability of R Shiny and combines it with the versatile annotation features of ANNOVAR. This application is easy to use and gives comprehensive annotations for user supplied vcf files using multiples databases. The output table contains the list of variants and their corresponding annotation presented within the graphical interface. In addition, the annotation results are downloadable as text file.