Combining PARP with ATR inhibition overcomes PARP inhibitor and platinum resistance in ovarian cancer models

High-grade serous ovarian cancer (HGSOC) remains a significant challenge, with high recurrence rates and mortality despite advances in understanding its genetics and the introduction of targeted therapies. Standard frontline care involves surgical debulking and platinum-based chemotherapy, yet over 80% of women experience recurrence. PARP inhibitors (PARPi) have shown promise, receiving FDA approval for both maintenance and recurrent settings, exploiting the synthetic lethality observed in HGSOC with homologous recombination (HR) defects (e.g., BRCA1/2 mutations). However, resistance to both chemotherapy and PARPi inevitably develops, leaving patients with limited treatment options. Approximately 50% of HGSOCs have HR defects, exhibiting initial sensitivity to platinum and PARPi, but eventually acquiring resistance. PARPi monotherapy shows limited efficacy, with complete responses being rare. The remaining 50% of HGSOCs are HR-proficient, with around 40% showing increased Cyclin E expression (CCNE1HIGH), linked to poor prognosis and platinum resistance. Overcoming this drug resistance is crucial for improving patient outcomes. Several mechanisms contribute to platinum and PARPi resistance, including altered drug accumulation, inactivation, increased DNA repair, impaired apoptosis, HR restoration (through BRCA gene restoration or other HR pathway components), altered PARP activity, upregulation of drug efflux pumps, increased replication fork stabilization, RAS pathway activation, and PI3K/AKT pathway upregulation. Combination strategies targeting alternative DNA repair pathways are a rational approach to overcome resistance. PARP inhibition disrupts single-strand DNA break repair, leading to double-strand breaks (DSBs) that are normally repaired via HR. PARP also helps stabilize replication forks. ATR, activated by replication stress, stabilizes these forks and activates cell cycle checkpoints. In cancers with high replication stress (e.g., TP53 loss or CCNE1 amplification), ATR inhibition causes replication fork collapse and loss of G2-M checkpoints, leading to mitotic catastrophe and cell death. This research hypothesizes that combining PARP and ATR inhibition (PARPi-ATRi) will increase DSBs and cell death regardless of HR status due to simultaneous targeting of different fork stabilization mechanisms.

The literature extensively documents the challenges posed by acquired resistance to platinum-based chemotherapy and PARP inhibitors in ovarian cancer. Studies have identified various mechanisms underlying this resistance, including restoration of BRCA function through secondary mutations or reversions, alterations in other HR pathway components, upregulation of drug efflux pumps, changes in PARP activity, and modifications to replication fork stability and cell survival pathways. Research into combination therapies, aiming to overcome single-agent limitations, has gained significant traction. Previous work from the authors demonstrated the potential of combining PARPi with ATRi in BRCA-deficient models, showing a decrease in ATR/CHK1 signaling, release of the G2/M checkpoint, increased DNA breaks, and subsequent tumor regression. This current study expands upon this prior work to include a more diverse range of ovarian cancer models, incorporating various genetic alterations.

This study employed a comprehensive approach using both in vitro and in vivo models to investigate the efficacy of combining PARP and ATR inhibition. **In vitro models:** PARPi and platinum-resistant cell lines were generated from BRCA1MUT and BRCA2MUT parental cell lines through prolonged drug exposure (approximately 1.5 years). De novo resistant lines and CCNE1-amplified models were also included. Whole-genome sequencing was performed to identify genetic alterations acquired during the development of resistance. Western blotting assessed ATR/CHK1 signaling. In vitro drug response was assessed using MTT assays and colony formation assays to determine IC50 values and evaluate synergy. Cell cycle analysis (flow cytometry), immunofluorescence (to detect RAD51 foci), and apoptosis assays were conducted to elucidate mechanisms of action. Replication fork speed and asymmetry were evaluated using DNA combing. **In vivo models:** Patient-derived xenografts (PDXs) from patients with acquired PARPi resistance were used to test the in vivo efficacy of the PARPi-ATRi combination. Whole-exome sequencing characterized the PDXs, including the presence of germline and newly acquired mutations. Tumors were pre-treated with a PARPi until a two-fold increase in volume was observed. Treatment groups were randomized to control, PARPi alone, ATRi alone, and the combination. Tumor growth and survival were monitored. Immunohistochemistry was performed on tumor samples to assess markers of replication stress (pCHK1) and DNA damage (γH2AX) and apoptosis (cleaved caspase-3). **Statistical analyses:** MTT and colony formation data were analyzed using one-way ANOVA with Tukey's post-hoc test. Drug synergy was assessed using the coefficient of drug interaction (CDI). For in vivo studies, longitudinal analysis of tumor growth employed linear mixed-effects modeling, and survival curves were analyzed with the Mantel-Cox log-rank test. Synergy in PDX models was determined using the Bliss independence definition of synergy.

The study found that acquired PARPi and platinum resistance in ovarian cancer models were associated with increased baseline ATR/CHK1 signaling. ATRi monotherapy showed modest effects on cell viability and colony formation in resistant cells. However, combining PARPi with ATRi (PARPi-ATRi) exhibited synergistic cytotoxicity, significantly reducing viability and colony formation across all tested resistant models including those with BRCA1/2 mutations, BRCA reversions and CCNE1 amplification. This synergy was associated with increased replication fork stalling and asymmetry, leading to an accumulation of DSBs (γH2AX), and ultimately apoptosis (increased cleaved caspase-3 levels). Mechanistically, the effects of PARPi-ATRi varied. In some models, the synergy correlated with a restoration of HR capacity indicated by RAD51 foci formation following PARPi treatment, and ATRi's ability to suppress this restoration and impede HR. In others, the mechanism appeared independent of HR, implicating other potential factors like XPC in acquired resistance. In vivo studies using PDXs confirmed the efficacy of PARPi-ATRi. Acquired PARPi resistance in BRCA2 mutant and BRCA1 mutant (with a reversion mutation) PDXs and platinum-resistant CCNE1 amplified models showed significant tumor regression and increased overall survival (OS) compared to monotherapies. These findings were consistent with the in vitro data and suggest multiple synergistic mechanisms of PARPi-ATRi action. The study also highlights that the presence of RAD51 foci may not always be a predictor of PARPi sensitivity. Importantly, long-term treatment (>50 weeks) with the PARPi-ATRi combination was well-tolerated in the PDX models. The in vivo efficacy correlated with increased γH2AX and cleaved caspase-3 expression in tumors treated with the combination.

This study demonstrates the potential of PARPi-ATRi as a therapeutic strategy to overcome PARPi and platinum resistance in ovarian cancer, independent of the specific resistance mechanism. The synergistic anti-tumor activity observed across various resistant models, including BRCA-mutant and CCNE1-amplified tumors, underscores the broad applicability of this combination. The observed mechanisms of action suggest that PARPi-ATRi creates multiple genomic vulnerabilities through its effects on replication fork stability and HR repair pathways. The in vivo data provide strong preclinical evidence supporting the clinical evaluation of this combination. The findings also highlight the complexity of acquired resistance, emphasizing that restoration of RAD51 is not universally predictive of PARPi resistance and underscores the need to further investigate alternative resistance mechanisms.

This research provides compelling preclinical evidence for the efficacy of combining PARP and ATR inhibition in overcoming resistance to PARP inhibitors and platinum-based chemotherapy in ovarian cancer. The synergistic effects observed in various resistant models, both in vitro and in vivo, support the clinical translation of this therapeutic strategy. Further research should focus on identifying biomarkers that can effectively predict the response to PARPi-ATRi and refine treatment strategies for optimal efficacy and minimal toxicity.

While this study provides robust preclinical data, limitations exist. The in vivo models are limited in their ability to fully replicate the complexity of human disease. Additional research with a larger panel of patient-derived xenografts representing a wider range of genetic backgrounds is warranted. Furthermore, the optimal dosing and scheduling of PARPi-ATRi for clinical translation need further investigation. Finally, the precise mechanistic pathways and their relative contributions to the synergistic effects remain to be fully elucidated.