PARP和ATM的协同抑制导致小儿骨肉瘤中未修复的DNA损伤和黏连蛋白介导的崩溃

Osteosarcoma (OS) is the most common malignant bone tumor in children. Currently, metastatic and relapsed patients with OS have long-term survival of less than 30% with available therapies. Due to high levels of genomic instability and tumor heterogeneity, novel therapeutic development has been challenging in OS. Recently, homologous recombination deficiency (HRD) has been observed using computational signatures in OS, which has led to investigations of efficacy of PARP inhibitors (PARPi) in this disease. PARPi are FDA approved for use in other HRD-marked cancers; however, in pre-clinical models, single agent PARPi has had mixed efficacy in OS. Thus, we aimed to investigate synergistic targets with PARPi in OS, with the hypothesis that we could increase the utility of this drug class, deepen understanding of the biology of DNA damage repair (DDR) in OS, and circumvent drug resistance which is seen with many targeted therapies, including PARPi. To explore rational combinations with PARPi in OS, we performed a genome-scale CRISPR-Cas9 screen in the presence of the PARPi olaparib in two OS cell lines. ATM emerged as a top druggable sensitizer to PARPi across both models. We first performed lentiviral CRISPR knockout (KO) of ATM in OS cell lines and treated with a range of PARPi doses and observed a signal of increased efficacy to PARPi, validating our screen. Next, we tested a small molecule ATM inhibitor (ATMi), AZD1390, combined with multiple PARPi, using a novel drug synergy platform in a panel of OS including low-passage patient-derived cell lines, observing profound synergy across OS models compared to non-OS cell lines using ATP-based assays and live cell imaging. Dual ATMi/PARPi demonstrated significantly increased DNA damage by immunofluorescence, G2/M cell cycle arrest by flow cytometry, and DNA replication stress by immunoblotting (IB), which led to increased apoptosis measured by increased caspase 3/7 and annexin V dyes. Given the known contribution of ATM to regulation of the cohesin complex, we next investigated modulation of the cohesin complex by ATMi/PARPi by IB. In combination-treated cells, we observed a marked reduction in the cohesin release factor WAPL across cell lines at early time points. Bortezomib therapy rescued dual-treated cells from WAPL depletion and partially rescued apoptosis in dual treated cells, confirming proteasomal degradation in this context. Based on our data, we propose that the combination reduces cohesin unloading, thereby changing chromatin accessibility and increasing DDR, leading to unresolved DNA damage and, ultimately, apoptosis. In vivo work has confirmed an increase in survival in mice bearing OS tumors treated with the combination compared to control. We are repeating these experiments to confirm the clinical relevance for OS patients using synergistic low doses to achieve efficacy without toxicity.