Genomic landscape of hereditary cancer syndromes in the largest cohort in Colombia: a retrospective study.

Background: Considerable fraction of global cancer cases stem from hereditary cancer predisposition syndromes (HCSs). The identification of genetic variants linked to HCSs is crucial for prompt treatment of both patients and their families. This study aimed to assess the diagnostic performance of multigene panels in detecting variants linked to hereditary cancer predisposition in a cohort of 8165 individuals from Colombia. Methods: We analyzed 8165 individuals in Colombia (2018-2024), with and without personal or family cancer history, using NGS hereditary cancer panels. Variant interpretation (P, LP, VUS) was performed with SOPHiA DDM and Varsome Clinical, following ACMG guidelines and ClinGen Hereditary Cancer Group criteria. Findings: 61.8% (n = 5049) of patients were referred from Bogotá and 38.10% (n = 3116) from other cities in Colombia. It was not possible to distinguish between ethnic groups. The age range of patients was 0-97 (mean = 55 years; SD = 12.1), 86% (n = 7024) were female and 14% (n = 1142) male. 409 P/LP variants were identified in high-penetrance genes such as BRCA1/2, in 946 individuals, resulting in an overall diagnostic yield of 9.82%. Among the most important findings were increased diagnostic yields in ovarian and colorectal cancer, as well as in unaffected individuals with a family history of cancer. Finally, 38 novel variants and recurrent alterations in 27 HCS-related genes reinforce the need to prioritize these biomarkers in diagnostic evaluations. Interpretation: This study provides insights into the performance of genetic panels for detecting HCS-associated variants in the largest Latin American cohort evaluated to date. These findings demonstrate that robust panel-based testing strategies enable the systematic detection of clinically relevant signs that would not be captured through phenotype-driven approaches alone. BRCA1 and BRCA2 were the most frequently altered high-penetrance genes, with higher diagnostic yields in breast, ovarian, and colorectal cancers, as well as in unaffected individuals with a family history. Funding: Fundación Universitaria Sanitas.