Ovarian cancer (OC) ranks eighth in incidence and mortality among female cancers1. The global annual incidence of new OC cases increased from 238,700 to 324,398, while deaths rose from 151,900 to 206,839 from 2012 to 2022, respectively1,2. China reported the highest global numbers of OC in 2022 with 61,100 new cases and 32,600 deaths3. OC continues to pose a significant threat to women’s health in China and worldwide.
The 5-year survival rate for early-stage OC after standardized treatment ranges from 70%–95%4. Due to the insidious onset and difficulty in early detection of OC, approximately 75% of patients are diagnosed at advanced stages. Treatment of advanced stage OC is frequently complicated by disease recurrence and the development of resistance to platinum-based chemotherapy, which together result in a 5-year survival rate of <40%4. The advent of poly(ADP-ribose) polymerase (PARP) inhibitors over the past decade has significantly expanded the therapeutic landscape of OC. Several PARP inhibitors (Table 1) are now approved for maintenance or later-line therapy for OC in China and internationally5. The series of studies involving fuzuloparib led by our team have provided high-level evidence supporting the clinical application of PARP inhibitors in China. However, acquired resistance remains a major challenge that limits therapeutic efficacy in patients. Elucidating the mechanisms underlying PARP inhibitor resistance and developing effective strategies to improve the prognosis of patients with drug resistance remain major priorities in current and future research. In this Perspective the mechanisms of action for PARP inhibitors, major clinical research progress of PARP inhibitors in China, and the mechanisms of resistance are summarized. Evidence from published cohort studies was integrated to outline the trial design strategies of PARP inhibitors to provide insights that improve efficacy and expand clinical benefit.
Clinically approved PARP inhibitors and their first applications
Mechanisms of action for PARP inhibitors
PARP inhibitors have shown remarkable clinical efficacy in OC, particularly in patients with homologous recombination deficiency (HRD), including patients harboring BRCA1/2 mutations. The therapeutic advantage of PARP inhibitors lies in the ability to exploit synthetic lethality. Mechanistically, this effect is closely linked to a hierarchical cascade involving DNA damage, activation of the DNA damage response (DDR), and failure of homologous recombination (HR)-mediated repair (Figure 1). Therefore, understanding the interplay among DNA damage, DDR pathways, and synthetic lethality is essential for elucidating the antitumor effects of PARP inhibitors.
The synthetic lethality mechanism underlying PARP inhibitors. The HR repair mechanism efficiently repairs DSBs in BRCA1/2 wild-type cells, preventing cell death. In contrast, HR repair is defective in BRCA1/2-mutant and BRCAness-like cells, which leads to the accumulation of unrepaired DSBs. In addition, PARP inhibitors block PARP activity, which prevents SSB repair and further promotes the accumulation of DSBs. This accumulation triggers synthetic lethality in BRCA1/2-mutant and BRCAness-like cells, ultimately resulting in tumor cell death. BER, base excision repair; BRCA1, breast cancer type 1 susceptibility protein; BRCA 2, breast cancer type 2 susceptibility protein; DSB, double-strand break; HR, homologous recombination; PARP, poly(ADP-ribose) polymerase; SSB, single-strand break. All elements shown in this figure were created by Procreate and Adobe Illustrator.
DNA damage
DNA damage is closely associated with the biology of OC. Approximately 50% of high-grade serous OCs exhibit genetic variations in DNA HR repair genes4. Single-strand breaks (SSBs) are the most common form of DNA damage. If not repaired, SSBs can be converted into double-strand breaks (DSBs). To prevent the cascade collapse triggered by DNA damage, OC cells frequently rely on DDR pathways.
DDR
DDR has a crucial role in repairing DNA damage and maintaining genomic stability. As a coordinated network, DDR pathways maintain genomic integrity through multiple repair pathways, including base excision repair (BER), HR, and non-homologous end joining (NHEJ). Among these repair pathways, BER is the primary pathway for repairing SSBs and is highly dependent on PARP proteins, particularly PARP1, PARP2, and PARP3. In addition to SSB repair, PARP1 also contributes to replication fork stability and chromatin regulation, underscoring a central role for PARP1 in DDR and the potential as a therapeutic target.
Synthetic lethality mechanism of PARP inhibitors
PARP1 functions as an early sensor in DDR when an SSB occurs. The zinc finger domain of PARP1 quickly recognizes and binds to the lesion, triggering conformational activation and catalyzing the transfer of ADP-ribose units from nicotinamide adenine dinucleotide (NAD)+ to PARP1 and surrounding proteins, a process known as PARylation (Figure 1). This catalysis leads to the synthesis of linear and branched poly(ADP-ribose) chains. In this way, PARP1 functions as a molecular scaffold, recruiting key downstream repair factors, such as XRCC1, DNA polymerase β, and DNA ligase III, enabling the efficient repair SSBs6.
PARP inhibitors exert antitumor effects through two primary mechanisms. First, PARP inhibitors competitively bind to the catalytic domain of PARP, inhibiting PARylation and preventing SSB repair. The accumulation of unrepaired SSBs ultimately leads to DSBs and cell death. Second, PARP inhibitors stabilize the PARP–DNA complex, a phenomenon known as “PARP trapping,” which impedes replication fork progression and promotes replication fork collapse (Figure 1).
Although PARP inhibitors can induce the formation of DSBs, cells have evolved several pathways to repair DSBs, among which HR is one of the most important. In cells with intact HR, such as BRCA1/2 wild-type cells, BRCA1 is involved in the end resection of DSBs, generating ssDNA and providing a template for HR repair. BRCA1 also regulates the recruitment of other repair factors to ensure efficient repair progression. Compared to BRCA1, BRCA2 facilitates this process by binding to RAD51, which helps stabilize its accumulation on ssDNA, promoting homologous pairing and completing the repair. The combined actions of BRCA1 and BRCA2 ensure the accuracy and efficiency of HR, thereby preventing cell death from PARP inhibitors. However, HR repair is defective in BRCA1/2 mutant cells or cells exhibiting a BRCAness phenotype. As a result, DSBs induced by PARP inhibitors cannot be repaired through the HR pathway. Instead, these cells rely on the error-prone NHEJ repair mechanism, resulting in further accumulation of DNA damage and ultimately leading to cell death. This phenomenon, known as synthetic lethality (Figure 1), accounts for the potent antitumor effects of PARP inhibitors in BRCA1/2-mutant and BRCAness-like tumors5.
Major clinical research progress involving PARP inhibitors in China
Given the role in suppressing DNA damage repair, PARP inhibitors have been developed and translated into clinical applications, providing novel options for patients with OC. To date, multiple clinical studies have investigated the role of PARP inhibitors in different scenarios, including later-line treatment for recurrent disease, maintenance therapy for recurrent platinum-sensitive OC (PSOC), and first-line maintenance therapy for newly diagnosed advanced OC. In particular, the emergence of Chinese PARP inhibitors, such as fuzuloparib and pamiparib, has further advanced the clinical management of OC in China. The major clinical research progress involving PARP inhibitors in these therapeutic settings will be briefly summarized.
Later-line treatment for recurrent OC
Initially, PARP inhibitors were used to explore the efficacy and safety as a later-line treatment for recurrent OC. In 2020 fuzuloparib received initial approval in China for the later-line treatment of recurrent PSOC in patients harboring germline BRCA1/2 mutations. This approval was based on a phase II, open-label, multicenter, single-arm study (Registration No. NCT03509636)7. Among 113 enrolled patients, fuzuloparib treatment achieved an Independent Review Committee (IRC)-assessed objective response rate (ORR) of 69.9% and a median progression-free survival (PFS) of 12.0 months. These findings supported fuzuloparib as an active later-line option for recurrent BRCA1/2-mutant OC in China.
Pamiparib is the first PARP inhibitor approved in China for the later-line treatment of recurrent PSOC and platinum-resistant OC (PROC). This approval was based on a phase II, open-label study (NCT03333915)8. Among 113 enrolled patients, the IRC-assessed ORR was 64.6% in the PSOC group and 31.6% in the PROC group. These results demonstrated the antitumor activity and manageable safety profile of pamiparib.
Maintenance therapy for recurrent PSOC
Following the application for later-line treatment, PARP inhibitors were further investigated as maintenance therapy for patients with recurrent PSOC. Several international multicenter studies have shown that PARP inhibitors significantly improve the median PFS in maintenance therapy for recurrent PSOC. FZOCUS-2 (NCT03863860) was the first phase III clinical trial conducted exclusively in a Chinese population to evaluate fuzuloparib in this setting9. Patients were randomly assigned to receive fuzuloparib or placebo with blinded IRC-assessed PFS as the primary endpoint. Fuzuloparib significantly improved the PFS compared to placebo in the overall population and showed a consistent benefit in germline BRCA1/2 mutation-positive and -negative subgroups. These findings supported the approval of fuzuloparib as maintenance therapy for recurrent PSOC in China.
Maintenance therapy for newly diagnosed advanced epithelial OC
Notably, the FZOCUS-1 trial, which was led by our team, suggested that fuzuloparib monotherapy may be sufficient to achieve substantial clinical benefit for HRD patients without the need for additional anti-angiogenic therapy, thereby reducing the treatment burden and potential toxicity. Specifically, the FZOCUS-1 trial (Registration No. NCT04229615) compared the combination of fuzuloparib with apatinib (a VEGFR-2 inhibitor) to fuzuloparib monotherapy and placebo in patients with newly diagnosed advanced OC10. A total of 674 patients were randomized into 3 groups. Both fuzuloparib plus apatinib and fuzuloparib monotherapy significantly improved the median PFS compared to placebo after a median follow-up of 40 months [26.9 and 29.9 vs. 11.1 months; hazard hatio (HR) = 0.57 and 0.58, respectively]regardless of BRCA1/2 mutation status. Although fuzuloparib alone exhibited clinical benefit in HRD patients, the addition of apatinib showed a trend toward improved PFS compared to fuzuloparib monotherapy for patients with HR proficiency [HRP] (16.6 vs. 11.0 months; HR = 0.73), indicating the necessity of drug combination in HRP patients.
In addition, although PARP inhibitors were generally well-tolerated, hematologic adverse events, particularly anemia and neutropenia, were frequently reported across the abovementioned clinical trials. Therefore, future strategies should focus on improving efficacy and reducing treatment-related toxicity. It has been reported that the specific inhibition of PARP-1, rather than PARP-2, can reduce the toxicity and side effects11. Furthermore, rational dose optimization, dynamic toxicity monitoring, and biomarker-guided patient enrollment may collectively enhance the durability and precision of PARP inhibitor therapy in OC.
In addition to the side effects, drug resistance also warrants attention. Current clinical studies of Chinese PARP inhibitors have primarily focused on efficacy and safety outcomes, whereas specific resistance mechanisms have not been systematically investigated. These unresolved issues also highlight the importance of integrating clinical outcomes with molecular analyses of resistance.
Biological mechanism underlying resistance to PARP inhibitors
Despite the substantial clinical benefit of PARP inhibitors in OC, acquired resistance has emerged as a major challenge limiting long-term efficacy. Tumor cells can evade PARP inhibition through multiple mechanisms, including restoration of HR repair, reduced PARP trapping, replication fork stabilization, and metabolic alternation (Figure 2A). The major biological mechanisms underlying PARP inhibitor resistance are discussed below.
PARP inhibitor resistance mechanisms and evolving clinical trial strategies in ovarian cancer. (A) Main mechanisms of acquired resistance to PARP inhibitors, including HR restoration, reduced PARP trapping, replication fork stabilization, and metabolic adaptation. (B) Schematic summary of PARP inhibitor trial design in ovarian cancer, including tumor type, inclusion criteria, therapeutic regimen and endpoint evaluation, and the trend of patient selection progresses from BRCA-mutant patients (past) to patients with HRD (present) and biomarker-guided populations (future). HR, homologous recombination; HRD, homologous recombination deficiency; OC, ovarian cancer. All elements shown in this figure were created by Procreate and Adobe Illustrator.
HR restoration
PARP inhibitors exert therapeutic effects by blocking PARP-mediated DNA damage repair and inducing the formation of DSBs, thereby selectively targeting HRD tumors through synthetic lethality. However, restoration of HR function can induce resistance to PARP inhibitors, particularly in patients with a recurrence who initially responded to PARP inhibitors.
Reversion mutations, which are capable of restoring function of crucial HR genes, such as BRCA1/2, represent the most prevalent mechanism of resistance. Specifically, these mutations typically involve insertions or deletions, which result in frameshift mutations that can restore the open reading frame of BRCA1/2 and potentially lead to the complete restoration of BRCA1/2 protein function. In addition, translation initiation downstream of frameshift mutations can drive the expression of truncated BRCA1 proteins. The truncated BRCA1 can be recruited to DNA damage sites, facilitating RAD51 binding and enhancing HR repair capacity, thereby contributing to PARP inhibitor resistance.
Interestingly, BRCA gene reversion is not the only mechanism that can lead to HR restoration. It has been reported that BRCA1/2 reversion mutations only account for a subset of resistant cases, suggesting that additional mechanisms, such as epigenetic reactivation of BRCA1, may also contribute to treatment failure.
Decreased PARP trapping
The primary target of PARP inhibitors is PARP1. Mutations in PARP1 can alter the DNA-binding properties and impair PARP trapping at sites of DNA damage, ultimately resulting in resistance.
The CRISPR-Cas9 “tag-mutation-enrichment” screening results indicated that mutations both within and outside the DNA-binding zinc finger domain of PARP1 contribute to PARP inhibitor resistance12. Moreover, the receptor tyrosine kinase, C-Met, has been shown to interact with PARP1, promoting phosphorylation at the Y907 site within the catalytic domain, thereby impeding the interaction of PARP inhibitors13.
In addition to PARP1 mutations, the inhibited or absent activity of poly (ADP-ribose) glycohydrolase (PARG), an enzyme responsible for removing PAR chains after DNA repair, may also contribute to resistance by attenuating the synthetic lethality.
Stable replication forks
Another key mechanism of PARP inhibitor resistance is maintenance of replication fork stability. As mentioned above, BRCA1/2 are essential for DSB repair. Factors, such as EZH2 and PTIP, promote the recruitment of nucleases, including MUS81 and MRE11, to stalled replication forks in BRCAness-like cells, leading to fork collapse and cell death. However, the inhibited or absent activity of EZH2 or PTIP can prevent nuclease recruitment, stabilize replication forks, and ultimately confer resistance to PARP inhibitors14,15. FANCD2 can prevent excessive DNA resection by nucleases, such as MRE11 and DNA2, during fork stalling and reversal in contrast to the functions of EZH2 and PTIP, thereby preserving fork stability. Studies have shown that in BRCA1/2-mutant breast cancer, FANCD2 induces PARP inhibitor resistance by maintaining fork stability16. In addition, the loss of SLFN11 expression in BRCA1/2-deficient cells reduces the accumulation of replication gaps, thereby decreasing sensitivity to PARP inhibitor treatment, which is consistent with reduced SLFN11 expression in OC patients and poor PARP inhibitor efficacy17.
Metabolic pathways contributing to PARP inhibitor resistance
Because NAD+ serves as the ADP-ribose donor for PARylation, increased NAD+ availability may reduce the inhibitory effect of PARP inhibitors and contribute to resistance. It is well-known that NAD+ can be synthesized through multiple pathways, including the de novo pathway (derived from tryptophan), the Preiss–Handler pathway (derived from nicotinic acid), and the nucleoside pathway (derived from nicotinic acid riboside or nicotinamide riboside)18. NAMPT and NAPRT are key regulatory enzymes in these pathways. In addition, metabolic reprogramming, including enhanced oxidative phosphorylation (OXPHOS) and fatty acid oxidation, may further promote resistance by supporting energy production and facilitating DNA damage repair. Together, these findings suggest that metabolic adaptation is another important mechanism underlying PARP inhibitor resistance.
Conclusions
Currently, PARP inhibitors have shown significant clinical benefits for OC patients through synthetic lethality. The results of FZOCUS-1 trial, which was conducted by our team, supported the use of fuzuloparib monotherapy in patients with HRD, whereas the addition of anti-angiogenic therapy may provide further benefit in patients with HRP, suggesting the importance of personalized treatment strategies in future OC therapy.
Existing studies on resistance mechanisms provide essential theoretical foundations for the development of targeted therapies and combination strategies with PARP inhibitors. For example, targeting FANCD2 to reduce replication fork stability or activating SLFN11 to increase replication fork gap accumulation are potential strategies for enhancing PARP inhibitor efficacy. Furthermore, preclinical studies have shown that NAMPT inhibitors, such as FK866, potentiate anti-tumor effects when combined with olaparib by reducing NAD+ synthesis. However, there are currently no clearly established Phase I/II trials specifically testing NAMPT inhibitors or FANCD2-targeted strategies in combination with fuzuloparib or pamiparib. In addition, the PDBAG1 peptide increases sensitivity to the PARP inhibitor, PJ34-HCL, treatment by inhibiting OXPHOS activity. Thus, targeting metabolic pathways and inhibiting metabolic adaptation in tumor cells represents another promising combination therapy strategy to overcome resistance. Furthermore, recent studies demonstrated that USP1 inhibition can overcome PARP inhibitor resistance in vitro and in vivo19,20. These results suggested that USP1 targeting may provide a potential translational strategy for overcoming PARP inhibitor resistance.
Another important trend in the clinical development of PARP inhibitors is the evolution of patient selection strategies (Figure 2B). Early studies mainly focused on BRCA1/2-mutant populations, in whom the therapeutic benefit of PARP inhibition were most evident. Subsequent trials were expanded to include HRD populations because HR impairment may also result from other genomic or epigenetic alterations. It can be predicted that the field is moving toward a more refined biomarker-guided framework, in which BRCA and HRD status may be integrated with additional molecular and functional markers to improve patient selection and extend the benefit of PARP inhibitor therapy. This refinement is also essential to avoid the unnecessary side effects of PARP inhibitors in patients with limited likelihood of benefit, thereby reducing potential toxicity and alleviating the treatment burden on patients.
In conclusion, future research should focus on the systematic analysis of the mechanisms underlying PARP inhibitor resistance, followed by the development of novel combination therapies. The use of biomarker-based screening and dynamic monitoring of treatment responses, such as ctDNA-based liquid biopsy, will support personalized treatment strategies. These measures are expected to overcome resistance, ultimately improving the survival rate and quality of life for OC patients.
Conflict of interest statement
No potential conflicts of interest are disclosed.
Author contributions
Conceived and designed the analysis: Lingying Wu, Wei Cui, Jing Yu and Maochen Li.
Collected the data: Jing Yu, Maochen Li, Yuhan Zhu, Shengnan Wang and Yuzhao Jiang.
Wrote the paper: Jing Yu and Maochen Li.
Illustrated the table and figures: Jing Yu and Maochen Li.
Reviewed and revised the paper: Lingying Wu and Wei Cui.
- Received February 2, 2026.
- Accepted April 21, 2026.
- Copyright: © 2026, The Authors
This work is licensed under the Creative Commons Attribution-NonCommercial 4.0 International License.
References
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.