Subscribe to RSS
DOI: 10.1055/s-0045-1808055
Cancer risk calculator for screening hereditary cancer syndromes in the general population using artificial intelligence
Authors
Approximately 10-30% of cancers have a hereditary basis, influenced by germline genetic mutations that cause specific familial patterns of neoplasms. There are over 200 hereditary cancer predisposition syndromes associated with more than 400 genes, most of which are autosomal dominant in inheritance. The identification of hereditary cancers is based on pedigree analysis and genetic testing, with family history playing a crucial role in distinguishing between sporadic and hereditary cases. Various mathematical models, such as Gail, BOADICEA, BRCAPRO, and PREMM, have been developed to predict cancer risk based on family history. Mendelian models and artificial intelligence (AI) tools have also been applied to calculate genetic risks. This study proposes to develop an oncogenetic screening tool using AI to distinguish sporadic from hereditary cases in a cohort of Brazilian patients. The Department of Oncogenetics of the A.C. Camargo Cancer Center, with a digital database containing family histories and genetic tests of over 3,000 patients, in collaboration with the AI expertise of INSPER's Engineering department, has been working on developing this tool, named the Oncogenetic Risk Calculator. The goal is to create a highly sensitive and specific Brazilian risk calculator to differentiate hereditary cancers from sporadic ones, which can be used for high-risk population screening. The initial prototype (training set) used the clinical data of 3,340 patients and their first-degree relatives, with a Machine Learning algorithm that showed moderate performance. The next step will be to incorporate oncological data from second-degree relatives and then validate the algorithm in a real-world setting (testing set) across different populations.
Corresponding author: Jose Claudio Casali da Rocha (e-mail: casali.rocha@accamargo.org.br).
No conflict of interest has been declared by the author(s).
Publication History
Article published online:
06 May 2025
© 2025. The Author(s). This is an open access article published by Thieme under the terms of the Creative Commons Attribution 4.0 International License, permitting copying and reproduction so long as the original work is given appropriate credit (https://creativecommons.org/licenses/by/4.0/)
Thieme Revinter Publicações Ltda.
Rua Rego Freitas, 175, loja 1, República, São Paulo, SP, CEP 01220-010, Brazil
Jose Claudio Casali-da-Rocha, Jéssica Piro Barragan, Gabriel de Araujo Alves, Sarah Azevedo Pimenta da Costa, Keiya Nishio, Ricardo Mourão Rodrigues Filho, Fabio Jose Ayres. Cancer risk calculator for screening hereditary cancer syndromes in the general population using artificial intelligence. Brazilian Journal of Oncology 2025; 21.
DOI: 10.1055/s-0045-1808055