Simultaneous Targeting of PARP1 and RAD52 Triggers Dual Synthetic Lethality in BRCA-Deficient Leukemia Cells

Abstract

Cancer cells may be addicted to DNA repair, which is supported by more selective targeting of tumor cells by recently developed novel drugs/compounds against specific DNA repair mechanisms. The success of PARP1 inhibitor olaparib in BRCA1 and BRCA2-deficient breast and/or ovarian tumors has established a proof-of-concept for personalized cancer therapy utilizing synthetic lethality. Unfortunately, therapeutic effect is usually short-lived and tumor cells become unresponsive to PARP1 inhibitors. Recently using personalized medicine approach we discovered that cohorts of acute and chronic leukemias display deficiencies in BRCA1/2-RAD51 -mediated homologous recombination (HR) repair, but these leukemia cells could not be completely eradicated by PARP1 inhibitors (Nieborowska-Skorska et al., J.Clin.Invest., 2017). Therefore, more radical elimination of BRCA1/2-deficient leukemia cells is required to prevent time-dependent emergence of PARP1 inhibitor-resistant/refractory clones. Alternative mechanisms of DNA repair may emerge in tumor cells exhibiting genetic and/or epigenetic deficiencies in the BRCA protein network to protect cells from lethal effect of DNA double strand breaks (DSBs). It has been suggested that RAD52-RAD51 provides an alternative HR pathway in BRCA1/2-deficient cells. We detected that RAD52-RAD51 mediated HR repair of DSBs remained active in PARP1 inhibitor-treated BRCA1/2-deficient tumor cells. Therefore, we hypothesized that RAD52-RAD51 pathway may be a potential escape route from PARP1 inhibitor-mediated synthetic lethality in BRCA1/2-deficient cells and that targeting RAD52 should enhance synthetic lethal effect of PARP1 inhibitor. Here we show that simultaneous short-term targeting of PARP1 and RAD52 with dominant-negative mutants or small molecule inhibitors resulted in synergistic accumulation of DSBs and/or cell death of BRCA1/2-deficient solid tumor and leukemia cells, but not BRCA/1/2-proficient cells. Moreover, long-term simultaneous treatment with PARP1 inhibitor and RAD52 inhibitor completely eradicated BRCA1/2-deficient tumor cells in vitro. Combination of PARP1 inhibitor + RAD52 inhibitor was not toxic for normal cells and tissues. Moreover, Parp1-/-;Rad52-/- double-knockout mice are indistinguishable from their wild-type littermates. SImultaneous targeting of PARP1 and RAD52 also reduced leukemia burden in vivo. BCR-ABL1-induced BRCA1-deficient Parp1-/-;Rad52-/- leukemias displayed prolonged latency when compared to Parp1-/- or Rad52-/- counterparts. Finally, the combination of PARP1 inhibitor and RAD52 inhibitor exerted synergistic activity against BRCA1/1-deficient primary leukemia xenografts in immunodeficient mice. In conclusion, we postulate that addition of RAD52 inhibitor may dramatically improve therapeutic outcome of BRCA1/2-deficient leukemias treated with PARP1 inhibitor. DisclosuresValent:Celgene: Honoraria, Research Funding; Novartis: Honoraria.