AZD-7648

Ovarian cancer, especially the drug-resistant subtype, has a treatment response rate of less than 30 %, primarily due to enhanced DNA damage repair mechanisms that reduce the efficacy of chemotherapy.Moreover, the low level of immune cell infiltration in ovarian tumors limits the therapeutic response to immune checkpoint inhibitors (ICIs).Addressing both chemoresistance and immune activation is essential to improve outcomes.DNA-dependent protein kinase catalytic subunit (DNA-PKcs), a critical component of the DNA damage repair pathway, plays a key role in the repair of DNA double-strand breaks (DSBs).Inhibition of DNA-PKcs not only sensitizes tumors to chemotherapy but also activates the cGAS-STING innate immunity pathway.Herein, we developed a glutathione (GSH)-responsive nanoparticle (NP2), self-assembled from a GSH-sensitive doxorubicin prodrug (PHHM-SS-DOX) and a DNA-PKcs inhibitor (AZD7648).NP2 responds to elevated GSH levels in cancer cells and releases DOX and AZD7648.AZD7648 inhibits DNA-PKcs phosphorylation, suppressing the non-homologous end joining (NHEJ) pathway and exacerbating doxorubicin-induced DSBs.Then sustained accumulation of dsDNA further activates the cGAS-STING pathway.In vivo, NP2 demonstrated significant tumor growth inhibition and modulation of antitumor immunity.It activated the cGAS-STING pathway and enhanced the release of inflammatory cytokines, maturation of dendritic cells, infiltration of CD8⁺ T cells, and polarization of tumor-associated macrophages toward the pro-inflammatory M1 phenotype.These effects reprogrammed the ovarian cancer microenvironment into an "immune-hot" tumor, significantly improving the response to ICIs.This strategy provides a novel therapeutic avenue to overcome chemoresistance and enhance the efficacy of immunotherapy in ovarian cancer.

AZD7648 is a potent inhibitor of DNA-dependent protein kinase (DNA-PK), which is part of the non-homologous end-joining DNA repair pathway. When combined with the PARP inhibitor olaparib, AZD7648 shows robust combination activity in pre-clinical ATM-knockout mouse xenograft models. To understand the combination activity of AZD7648 and olaparib, we developed a semi-mechanistic pharmacokinetic/pharmacodynamic (PK-PD) model that incorporates the mechanism of action for each drug which links to proliferating, quiescent, and dying cell states with an additional Allee effect-like term to account for the non-linear growth and regression observed at low cell densities. Model parameters were fitted to training data sets that contained continuous treatment of either monotherapy or the combination. The observed interaction of AZD7648 on olaparib PK was incorporated in the PK-PD model by an effect function specific for each of the drug's MoA and was found essential to quantify drug effects at high dose levels of combination treatments. The model was able to adequately describe the observed efficacy for both monotherapy and sustained regressions in combination groups, mainly driven by maintaining a > 2:1 AUC ratio of apoptotic:proliferating cell fractions. We found that this model was suitable for forecasting intermittent dosing schedules a priori and resulted in accurate predictions when compared to xenograft efficacy data, without the need for extra, descriptive terms to describe supra-additive effects under combined dose regimes. This model provides quantitative understanding on the combination effect of AZD7648 and olaparib and allows for the exploration of the full exposure landscape without the need to experimentally test all scenarios. Furthermore, the model can be utilized to assess what exposures would be necessary in the clinic by linking it to observed or predicted human PK exposures. The model suggests 64.9 uM olaparib is sufficient to achieve tumor stasis in the absence of AZD7648, while the combination of AZD7648 and olaparib only requires plasma concentrations of 20.2 uM AZD7648 and 19.9 uM olaparib at steady-state to achieve the same effect.