Discordancy in interpretation of a BRCA1 variant in the Belgian population

Pathogenic variants of BRCA1 gene predispose to hereditary breast and ovary cancer. A correct classification of these variants is essential for an accurate genetic counselling and management of the patient in order to prevent over or undertreatment.
However, the classification of BRCA1 c.5071A>G or p.(Thr1691Ala) remains unclear.
This variant is absent from public databases and is predicted to be deleterious to protein function by in silico analyses. Functional studies are contradictory. Bowman and al. (2020) suggest a deleterious effect in DR-GFP assay, but a neutral and intermediate one in Cisplatin and Olaparib assays respectively. Findlay and al. (2018) report an intermediate impact on cell survival. Petitalot and al. (2019) report four different assays (HR, localization, solubility, phosphopeptide-binding assay) in which this variant does not impact protein function. Therefore, we classified this variant as a variant of unknown significance following the ACMG classification.
We then reviewed all personal and familial history of the five cases included in CHU de Liège database. Familial history of breast or other BRCA1 related cancers was not informative in all the pedigrees. Some presented with early breast cancer others with late onset breast cancer. Two patients presented this variant in co-occurrence with another pathogenic variant of a different breast cancer predisposition gene.
In conclusion, despite the rarity of this variant in databases, its frequency seems to be higher than expected in our region questioning a potential founder effect. Clinical data from our own database are not in favor of pathogenicity, although it is not excluded. We hope that future data will help to classify this variant more precisely in order to facilitate the management and the genetic counselling for patients.