Abstract Extensive trials of PARP inhibitors (PARPi) have shown that they improve progression-free survival in patients with various cancers, including ovarian cancers. This improvement is marked in patients carrying BRCA mutations, and in Homologous Recombination Deficiency (HRD) subgroups. To date, identifying patients eligible for PARPi has been a challenge for scientific and clinical teams using next-generation sequencing (NGS) analysis and the persistence of genomic scars in tumors after restoration of proficient HR or epigenetic changes is a limitation of NGS. Functional assays could be used to improve the profiling of HR status and faithfully identify HRD tumors. The Repair Capacity Test assesses the formation of RAD51 foci in proliferating cells after irradiation and can be used on fresh primary cancer tissues irradiated ex vivo as well as on Patient-Derived Tumor Organoids (PDTO). However, RAD51 foci scoring is often performed manually without any possibility for the standardization of techniques. By contrast, recent progress in whole slide imaging referring to scan a complete microscope slide and creating a single high-resolution digital file, could represent an opportunity for automatizing the evaluation of HR. The purpose of this translational study was to develop an automated tool for scoring RAD51-mediated homologous recombination, to use this tool on ovarian PDTO and to compare the result to the sensitivity to olaparib (PARPi), determined by direct exposure of PDTO to the drug. We show that immunofluorescence of RAD51 foci can be automatically detected and quantified in Cyclin A2 positive nuclei in all the cells of each PDTO slice thus offering a new opportunity for the routine management and standardization of HR assessment in PDTO that goes beyond the widely used manual estimation. Finally, this automated scoring will be optimized to directly assess HR on tumor slices to make the Repair Capacity Test a highly relevant functional precision oncology tool to identify patients who will benefit from PARPi. Citation Format: Lucie Thorel, Pierre-Marie Morice, Nicolas Elie, Louis-Bastien Weiswald, Romane Florent, Florence Giffard, Margaux Jacobs, Agathe Ricou, Raphaël Leman, Guillaume Babin, Jean-François Lebrun, Sandrine Martin, Mélanie Briand, Benoit Goudergues, Bernard Lambert, Cécile Blanc-Fournier, Dominique Vaur, Benoit Plancoulaine, Laurent Poulain. Automated scoring to assess RAD51-mediated homologous recombination in patient-derived tumor organoids of ovarian cancers [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 6112.
Read full abstract