Combinatorial Efficacy of Olaparib with Radiation and ATR Inhibitor Requires PARP1 Protein in Homologous Recombination-Proficient Pancreatic Cancer.

PARP inhibitor monotherapy (olaparib) was recently FDA approved for the treatment of BRCA1/2-mutant, homologous recombination (HR) repair-deficient pancreatic cancer. Most pancreatic cancers, however, are HR proficient and thus resistant to PARP inhibitor monotherapy. We tested the hypothesis that combined therapy with radiation and ataxia telangiectasia and Rad3-related (ATR) inhibitor (AZD6738) would extend the therapeutic indication of olaparib to HR-proficient pancreatic cancers. We show that olaparib combined with AZD6738 significantly reduced radiation survival relative to either agent alone, regardless of HR status. Whereas catalytic inhibition of PARP with low concentrations of olaparib radiosensitized HR-deficient models, maximal sensitization in HR-proficient models required concentrations of olaparib that induce formation of PARP1-DNA complexes. Furthermore, CRISPR-Cas9-mediated PARP1 deletion failed to recapitulate the effects of olaparib on radiosensitivity and negated the combinatorial efficacy of olaparib and AZD6738 on radiosensitization, suggesting that PARP1-DNA complexes, rather than PARP catalytic inhibition, were responsible for radiosensitization. Mechanistically, therapeutic concentrations of olaparib in combination with radiation and AZD6738 increased DNA double-strand breaks. DNA fiber combing revealed that high concentrations of olaparib did not stall replication forks but instead accelerated replication fork progression in association with an ATR-mediated replication stress response that was antagonized by AZD6738. Finally, in HR-proficient tumor xenografts, the combination of olaparib, radiation, and AZD6738 significantly delayed tumor growth compared with all other treatments. These findings suggest that PARP1-DNA complexes are required for the therapeutic activity of olaparib combined with radiation and ATR inhibitor in HR-proficient pancreatic cancer and support the clinical development of this combination for tumors intrinsically resistant to PARP inhibitors.