Cumulative defects in DNA repair pathways drive the PARP inhibitor response in high-grade serous epithelial ovarian cancer cell lines

// Hubert Fleury 1, 2 , Euridice Carmona 1, 2 , Vincent G. Morin 1, 2 , Liliane Meunier 1, 2 , Jean-Yves Masson 3, 4 , Patricia N. Tonin 5, 6, 7 , Diane Provencher 1, 2, 8 and Anne-Marie Mes-Masson 1, 2, 9 1 Centre de Recherche du Centre Hospitalier de l’Universite de Montreal (CRCHUM), Montreal, Canada 2 Institut du cancer de Montreal, Montreal, Canada 3 Genome Stability Laboratory, CHU Research Center, Quebec City, Canada 4 Department of Molecular Biology, Medical Biochemistry and Pathology, Laval University Cancer Research Center, Quebec City, Canada 5 Cancer Research Program (CRP), The Research Institute of the McGill University Health Centre, Montreal, Canada 6 Department of Human Genetics, McGill University, Montreal, Canada 7 Department of Medicine, McGill University, Montreal, Canada 8 Division of Gynecologic Oncology, Universite de Montreal, Montreal, Canada 9 Department of Medicine, Universite de Montreal, Montreal, Canada Correspondence to: Anne-Marie Mes-Masson, email: anne-marie.mes-masson@umontreal.ca Keywords: olaparib, high-grade serous epithelial ovarian cancer, DNA repair pathways, NER, MMR Received: January 27, 2016     Accepted: June 09, 2016     Published: June 27, 2016 ABSTRACT PARP inhibitors (PARPi), such as Olaparib, have shown promising results in high-grade serous (HGS) epithelial ovarian cancer (EOC) treatment. PARPi sensitivity has been mainly associated with homologous recombination (HR) deficiency, but clinical trials have shown that predicting actual patient response is complex. Here, we investigated gene expression microarray, HR functionality and Olaparib sensitivity of 18 different HGS EOC cell lines and demonstrate that PARPi sensitivity is not only associated with HR defects. Gene target validation show that down regulation of genes in the nucleotide excision repair (NER) and mismatch repair (MMR) pathways ( ERCC8 and MLH1 , respectively) increases PARPi response. The highest sensitivity was observed when genes in both the HR and either NER or MMR pathways were concomitantly down regulated. Using clinical samples, patients with these concurrent down regulations could be identified. Based on these results, a novel model to predict PARPi sensitivity is herein proposed. This model implies that the extreme responders identified in clinical trials have deficiencies in HR and either NER or MMR.