Discovery of AZD4956, a potent and selective inhibitor of DNA polymerase theta, a clinical candidate for treatment of HRR-deficient tumors in combination with saruparib

Deficiency in the homologous recombination repair (HRR) pathway, which is key for high-fidelity DNA double-strand break (DSB) repair, is prevalent across various cancers (including ovarian, breast, pancreatic, and prostate). It frequently arises from loss-of-function mutations in BRCA1 , BRCA2 , or other HRR-associated genes. PARP inhibitors (PARPi) have transformed the treatment landscape and survival outcomes in patients with HRR-deficient tumors. However, despite the remarkable efficacy of PARPi, clinical benefit is variable, and a significant number of patients develop resistance. Combinations of PARPi with targeted agents blocking the cellular DNA damage response have been pursued to improve efficacy of PARPi treatments, but enhanced myelosuppression has limited dosing and, consequently, efficacy of these approaches. DNA polymerase theta (Polθ; POLQ) is a key component of microhomology-mediated end joining (MMEJ) DSB repair pathway. Under HRR-deficient conditions, cellular DSB repair becomes increasingly dependent on POLQ-driven MMEJ. We report the identification and optimization of a novel chemotype of POLQ inhibitors, culminating in the discovery of the clinical candidate AZD4956. AZD4956 is a potent and selective inhibitor of the polymerase domain of POLQ (IC 50 <10 nM) in biochemical assays. AZD4956 suppresses cellular MMEJ activity with single digit nanomolar potency (IC 50 3.95 nM) and shows potent antiproliferative activities (IC 50 values between 3-10 nM) and induction of genomic instability (assessed by micronuclei quantification) in a wide range of HRR defective cell lines, with no measurable activity in HRR proficient genetic backgrounds. AZD4956 further enhances genomic instability and antiproliferative activity induced by the PARP1-selective inhibitor, saruparib, in multiple HRR defective cell lines, with no activity of the combination in HRR proficient settings. Additionally, combination of AZD4956 and saruparib does not exacerbate hematotoxicity of saruparib treatment in a 3D-bone marrow microphysiological system. Oral daily treatment of AZD4956 combined with saruparib drives sustained regression of the HRR deficient BRCA2 -/- DLD‑1 xenograft model, superior to the respective activities of the two agents as monotherapies. Increased efficacy of the combination can be pharmacodynamically tracked by increased accumulation of micronuclei in circulating red blood cells. The preclinical profile, spanning efficacy, safety, DMPK and physicochemical properties, supports clinical development of AZD4956 as a combination partner with saruparib in patients with HRR deficient cancers. Clinical evaluation of AZD4956 is ongoing in a first-in-human, open-label, multicenter, phase 1/2a study in patients with HRR deficient solid tumors (PARTHENON study) in combination with saruparib.