SPECIAL SEMINAR - Dr Solène Florence KAMMENER -JACQUET Médecin pathologiste PUPH - CHU de Rennes

“Predicting Homologous Recombination Repair in prostate cancer with Artificial Intelligence”

Dr Solène Florence KAMMENER -JACQUET

Médecin pathologiste PUPH

CHU de Rennes

Anatomie et Cytologie pathologiques

Abstract

Prostate cancer ranks first among cancers in men (50,400 new cases in 2018), well ahead of lung and colorectal cancers. It represents 25% of all incident cancers in men and ranks third among deaths (8,100 deaths in 2018) from cancer in men. It occurs in approximately 30% of cases in men under the age of 65. In recent years, mortality has decreased slightly due to access to prostate cancer screening and improved treatments, particularly for advanced cancers.

HRR (Homologous Recombination Repair) genes such as BRCA1 and BRCA2 are involved in DNA repair and may be associated with hormonal resistance in prostate cancer. Mutations in these genes can increase the risk of developing prostate cancer and influence response to treatments, including an increased response to olaparib. Olaparib belongs to a class of drugs called PARP (Poly (ADP-ribose) polymerase) inhibitors. It was initially developed for the treatment of ovarian cancer associated with a BRCA mutation, but it has also been approved by the FDA for the treatment of patients with metastatic castration-resistant prostate cancer with a proven BRCA1 or BRCA2 mutation. The objective of the project is to develop an algorithm based on deep learning principles and architectures using digitized prostate cancer biopsies from a retrospective cohort. Good performance for this algorithm would help selecting patients before further assessed with genetic analyses.