PS4-05-26: Classivus: a clinical platform for reclassifying uncertain variants in hereditary breast and ovarian cancer in a Brazilian cohort

Abstract Variants of Uncertain Significance (VUS) represent a major challenge in clinical genetics, particularly in the context of hereditary breast and ovarian cancer (HBOC). The underrepresentation of Latin American populations in global genomic databases contributes to a higher prevalence of inconclusive results, hindering genetic counseling and risk-reduction interventions. Efforts to improve variant interpretation through population-specific evidence and clinical integration are urgently needed. To reclassify VUS identified in patients suspected of HBOC by applying the structured ACMG/AMP guidelines, integrating computational prediction, population data, and clinical context; and to develop and implement a dedicated institutional platform (CLASSIVUS) to record, standardize, and track the classification of genetic variants in a Brazilian clinical genetics center. We analyzed 105 patients with a confirmed diagnosis of breast cancer who met the NCCN criteria for HBOC. Genetic testing was performed using next-generation multigene panel sequencing covering key susceptibility genes. VUS were selected based on ACMG criteria and evaluated using in silico predictors (PolyPhen-2, SIFT, MutationTaster), population frequency databases (gnomAD, ABraOM), and clinical annotation platforms (ClinVar, Franklin by Genoox, Varsome). Each variant was scored according to the ACMG criteria, and a final classification was determined. A web-based institutional database, CLASSIVUS, was developed to register variants, document the applied criteria, associate the responsible professionals, and enable filtered queries and reanalyses. Thirteen VUS were identified in 10 patients (9.5%) across seven genes (ATM, BRCA1, BRCA2, BRIP1, BARD1, CHEK2, TP53). Five variants (38.5%) were reclassified as likely benign based on ACMG scoring, while eight (61.5%) remained classified as VUS. The CLASSIVUS platform enabled centralized storage of variant records, real-time retrieval of previous classifications, and standardized documentation of scoring justification. This system reduced analytical redundancy and ensured traceability and reproducibility among professionals, offering a scalable solution for routine clinical workflows. The integration of ACMG guidelines with computational tools, clinical annotations, and local population frequency data enabled reliable reclassification of VUS in a Brazilian HBOC cohort. The CLASSIVUS platform proved to be a valuable tool for standardizing variant interpretation and supporting safe, reproducible, and transparent clinical decisions. This approach strengthens precision medicine practices in underrepresented populations and may serve as a model for other institutions seeking to enhance their genetic curation capabilities. Citation Format: B. E. Nascimento, L. P. Lacerda, P. F. Silva, R. M. Goveia, I. M. Calassa, K. M. Veiga, R. M. Rahal, W. D. Oliveira, D. C. Maia, E. P. Silveira-Lacerda. Classivus: a clinical platform for reclassifying uncertain variants in hereditary breast and ovarian cancer in a Brazilian cohort [abstract]. In: Proceedings of the San Antonio Breast Cancer Symposium 2025; 2025 Dec 9-12; San Antonio, TX. Philadelphia (PA): AACR; Clin Cancer Res 2026;32(4 Suppl):Abstract nr PS4-05-26.