PARP inhibitors in the treatment of ovarian cancer: a narrative review

Ovarian cancer remains a life-threatening malignancy that exerts significant impacts on women’s health across the world. Characterized by its complex pathogenesis, diverse histological subtypes, and highly variable clinical outcomes, ovarian cancer represents a major focus in gynecologic oncology. According to 2020 global cancer statistics, there were approximately 314,000 new cases of ovarian cancer worldwide, accounting for about 3.7% of all new cancer diagnoses in women. In the same year, the disease caused an estimated 207,000 deaths, accounting for 4.7% of all female cancer-related mortality ( 1 ). Histopathologically, ovarian cancers are broadly classified into four main categories: epithelial carcinoma, malignant germ cell tumors, sex cord stromal tumors, and metastatic tumors. Epithelial ovarian carcinoma (EOC) constitutes the vast majority (approximately 80–90%) of cases ( 2 ). Among patients with EOC, the 5-year overall survival rate ranges between 40% and 50%. EOC itself can be further subdivided into five principal histological subtypes: high-grade serous carcinoma (HGSOC), low-grade serous carcinoma (LGSOC), endometrioid carcinoma (ENOC), clear cell carcinoma (CCC), and mucinous carcinoma (MOC). Among these, HGSOC is the most prevalent, representing approximately 75% of all EOC cases ( 3 ). Despite the considerable advances achieved in therapeutic strategies, approximately 70% of patients with advanced-stage ovarian cancer (predominantly of epithelial origin) experience relapse within 2–3 years of initial treatment, with the median progression-free survival (PFS) being about 16 months. Consequently, the overall 5-year survival rate for ovarian cancer has remained stubbornly low over the past few decades. Significant challenges persist, particularly in the areas of early detection, development of individualized treatment approaches, and the overcoming of therapy resistance ( 4 ). It is within this challenging clinical context that the administration of poly(ADP-ribose) polymerase inhibitors (PARPis) has emerged as a transformative therapeutic strategy for this disease. Recent comprehensive reviews have contextualized PARPis within modern treatment algorithms, emphasizing their integration into multidisciplinary management strategies that encompass surgery, chemotherapy, and targeted maintenance approaches ( 5 ). This narrative review was conducted to synthesize the literature on PARPis and provide a comprehensive critical appraisal; examine the mechanistic basis, pivotal clinical trials, and evolution of its application; address the pressing challenges of resistance; and outline the future trajectory of PARPi-based therapy in epithelial ovarian cancer. We present this article in accordance with the Narrative Review reporting checklist (available at https://tcr.amegroups.com/article/view/10.21037/tcr-2026-1-0258/rc ).

Ovarian cancer is often diagnosed at an advanced stage due to the absence of specific early symptoms. Understanding risk factors is crucial for prevention and screening in high-risk populations. Hereditary predisposition is the most significant risk factor, with germline mutations in BRCA1 and BRCA2 conferring lifetime risks of approximately 44% and 17%, respectively ( 6 ). Other homologous recombination repair (HRR) genes (e.g., RAD51C , RAD51D , BRIP1 , PALB2 ) and Lynch syndrome (mismatch repair gene mutations) also increase risk ( 7 ). Reproductive and hormonal factors play a key role: nulliparity, early menarche, and late menopause increase risk, while oral contraceptive use and multiparity are protective ( 8 - 10 ). Endometriosis is a well-established precursor for clear cell and endometrioid ovarian cancers, driven by chronic inflammation, local estrogen production, and somatic ARID1A mutations ( 11 - 14 ). Other factors, including polycystic ovary syndrome and pelvic inflammatory disease, have been suggested but lack consistent evidence ( 15 - 18 ). Recognizing these risk factors, particularly hereditary syndromes, guides prevention, early detection, and directly influences therapeutic decisions, including the use of PARPis.

The cornerstone of ovarian cancer management is a multimodal approach that integrates cytoreductive surgery and platinum-based chemotherapy. The primary surgical goal is maximal tumor cytoreduction. The choice between primary cytoreductive surgery (PCS) and neoadjuvant chemotherapy followed by interval debulking surgery (NACT-IDS) is guided by disease stage (International Federation of Gynecology and Obstetrics classification), tumor distribution, and patient performance status. This treatment paradigm is strongly endorsed by international guidelines. Joint recommendations from the International Gynecologic Cancer Society (IGCS) and the European Society of Gynecological Oncology (ESGO) emphasize that complete resection (R0), defined as no macroscopic residual disease, is the most significant prognostic factor for survival in patients with advanced-stage disease. The guidelines stipulate that PCS should be pursued when R0/R1 resection is feasible. Conversely, NACT-IDS is recommended for patients with a high tumor burden for whom complete cytoreduction is unlikely or for those with severe comorbidities ( 19 ). The landmark phase III EORTC 55971 trial provided pivotal evidence supporting neoadjuvant chemotherapy, establishing the NACT-IDS strategy as noninferior to PCS in terms of overall survival [hazard ratio (HR) =0.98; 90% confidence interval (CI): 0.84–1.13]. Importantly, it achieved a significantly higher rate of complete resection (80.6% vs. 41.6%) and reduced perioperative morbidity and mortality, thereby solidifying NACT-IDS as a standard option ( 20 ). Following optimal cytoreduction, platinum-based combination chemotherapy remains the cornerstone of first-line systemic treatment. The current preferred standard is the carboplatin and paclitaxel doublet. This treatment paradigm was first established by the pivotal GOG-111 study, which demonstrated the superiority of paclitaxel plus a platinum agent (cisplatin) ( 21 ) and later reinforced by trials such as AGO-OVAR 3, which confirmed carboplatin’s comparable efficacy to cisplatin with a more favorable toxicity profile ( 22 ). Despite high initial response rates, a major challenge persists: approximately 70–80% of patients with advanced disease develop platinum-resistant recurrence within 5 years ( 3 ). The emergence of platinum resistance typically heralds rapid disease progression, limited subsequent treatment options, and a poor prognosis. This pressing clinical reality has been the primary impetus for the development of novel therapeutic strategies, most notably targeted agents such as PARPis, which are discussed in the following sections.

As outlined previously, the management of advanced ovarian cancer is fundamentally challenged by the high rate of relapse after platinum therapy. PARPis address this unmet need, representing a paradigm shift in treatment. The cornerstone of their efficacy is the property of synthetic lethality, which involves the concurrent disruption of two independent DNA repair pathways—PARP-mediated base excision repair (BER) for single-strand breaks and BRCA1/2 -mediated HRR for double-strand breaks—which selectively kills cancer cells while sparing normal ones ( 23 , 24 ). PARPis exert cytotoxicity through a dual mechanism. Beyond the catalytic inhibition of PARP, the more potent mechanism involves “PARP trapping”, in which the inhibitor sequesters PARP on DNA. This results in cytotoxic complexes that impede replication fork progression, leading to fork stalling and collapse. Consequently, single-strand breaks are converted into lethal double-strand breaks. In tumor cells with pre-existing homologous recombination deficiency (HRD), the accurate repair of these breaks is compromised, forcing reliance on error-prone backup pathways and culminating in genomic instability and apoptosis ( 25 - 27 ). Notably, PARPis also exhibit activity in a subset of homologous recombination-proficient tumors. The mechanisms underlying this observation are under active investigation and may involve effects on immune activation, ribosome biogenesis, and transcriptional regulation ( 28 - 30 ). Bolstered by this robust preclinical rationale, the clinical application of PARPi has evolved from later-line therapy to a cornerstone of first-line maintenance treatment following platinum-based chemotherapy. This strategy exploits the DNA damage inflicted by platinum agents and the HRD context of the tumor. By sustaining the suppression of DNA repair, PARPis target minimal residual disease, thereby profoundly delaying recurrence and significantly improving PFS, an approach which marks a breakthrough in the long-term management of advanced ovarian cancer.

The compelling mechanistic rationale for PARP inhibition, as detailed above, has been validated by multiple phase III trials. These studies have established several PARPis as cornerstone maintenance therapies, each with a distinct clinical profile and indication. To facilitate comparison, the key efficacy and safety data for approved PARPis are summarized in Table 2 . 1L, first-line; AML, acute myeloid leukemia; gBRCAmut, germline BRCA mutation; HR, hazard ratio; HRD+, homologous recombination deficiency-positive; ITT, intention-to-treat; MDS, myelodysplastic syndrome; OS, overall survival; PARPi, poly(ADP-ribose) polymerase inhibitor; PFS, progression-free survival; PSROC, platinum-sensitive recurrent ovarian cancer. In the SOLO-1 trial, maintenance olaparib significantly improved PFS compared with placebo in patients with newly diagnosed advanced ovarian cancer and BRCA mutation (HR =0.30; 95% CI 0.23–0.41) ( 31 ). The final overall survival analysis confirmed a 45% reduction in the risk of death (HR =0.55), with a 7-year overall survival rate of 67.0% with olaparib vs. 46.5% with placebo ( 32 ). This landmark outcome solidified olaparib as the standard of care in this population. Building on the success of olaparib monotherapy in patients with BRCA -mutated ovarian cancer, subsequent research sought to extend the benefits of PARP inhibition to a broader patient population through rational combinations. To this end, the PAOLA-1 trial investigated the combination of olaparib with bevacizumab as first-line maintenance. In the overall population, the addition of olaparib to bevacizumab improved PFS (HR =0.59; 95% CI: 0.49–0.72), with a more pronounced benefit in the HRD-positive subgroup (HR =0.33; 95% CI: 0.25–0.45) ( 33 ). This study successfully expanded the indication of PARPi-based therapy from BRCA -mutated patients to the broader HRD-positive population. The safety profile of olaparib is characterized primarily by anemia and gastrointestinal disturbances, which are generally manageable with dose adjustments and supportive care. Long-term use requires monitoring for rare but serious adverse events such as myelodysplastic syndrome (MDS) and acute myeloid leukemia (AML), though the incidence remains low. The PRIMA study established niraparib as a first-line maintenance treatment for a biomarker-unselected, all-comer population. In the overall cohort, niraparib significantly improved median PFS compared to placebo (HR =0.62; 95% CI: 0.50–0.76; median, 13.8 vs. 8.2 months). The benefit was most pronounced in the HRD-positive subgroup, where niraparib more than doubled the median PFS (HR =0.43; 95% CI: 0.31–0.59; median, 21.9 vs. 10.4 months) ( 34 ). PRIMA established niraparib as the first PARPi approved for first-line maintenance without biomarker selection, providing a critical option for patients with unknown or HRD-negative status. Niraparib is associated with higher rates of thrombocytopenia and hypertension; individualized starting doses based on baseline platelet count and body weight reduce grade ≥3 thrombocytopenia without compromising efficacy ( 34 ). Rucaparib has demonstrated efficacy in both platinum-sensitive recurrent and first-line maintenance settings. The ARIEL3 trial showed significant PFS improvement in patients with platinum-sensitive recurrent ovarian cancer: median PFS was 16.6 vs. 5.4 months (HR =0.23; 95% CI: 0.16–0.34) in those with BRCA mutations, and 13.6 vs. 5.4 months (HR =0.32; 95% CI: 0.24–0.42) in the broader HRD population ( 35 ). In the first-line setting, the ATHENA-MONO trial reported improved PFS with rucaparib monotherapy compared with placebo (HR =0.52; 95% CI: 0.40–0.68) in an all-comer population, with greatest benefit in HRD-positive patients (median PFS, 28.7 vs. 11.3 months; HR =0.47; 95% CI: 0.31–0.72) ( 36 ). Rucaparib has been linked to elevations in liver enzymes and creatinine, as well as fatigue, requiring regular laboratory monitoring. These effects are typically reversible with dose interruption or reduction. As the first PARPi independently developed in China, fuzuloparib has emerged as an important agent for platinum-sensitive relapsed disease. In the pivotal phase III FZOCUS-2 trial, fuzuloparib maintenance therapy significantly extended median PFS to 12.9 months, as compared to 5.5 months with placebo (HR =0.25; 95% CI: 0.17–0.36) ( 37 ). Based on these results, China’s National Medical Products Administration (NMPA) approved fuzuloparib in 2020 for the maintenance treatment of patients with germline BRCA -mutated, platinum-sensitive relapsed ovarian cancer, providing a novel therapeutic option in the Chinese clinical landscape. Fuzuloparib’s safety profile appears similar to other PARPis, with hematologic toxicity being most frequent. Long-term safety data are still accumulating. The SOLO-1 trial established biomarker-driven maintenance therapy in newly diagnosed advanced ovarian cancer with BRCA mutations, demonstrating a marked reduction in the risk of disease progression (HR =0.30). Long-term follow-up confirmed durable benefit and an overall survival advantage at 7 years (HR =0.55), solidifying BRCA mutation status as a robust predictive biomarker ( 31 , 32 ). PAOLA-1 subsequently expanded this paradigm by combining olaparib with bevacizumab. In HRD-positive tumors, the combination significantly reduced the risk of progression (HR =0.33) and improved overall survival (HR =0.62), whereas no benefit was observed in HRD-negative disease ( 33 ). Notably, PAOLA-1 randomized patients to olaparib + bevacizumab versus placebo + bevacizumab after response to platinum-based chemotherapy; benefit was concentrated in HRD-positive tumors, with limited/no benefit seen in HRD-negative patients. In contrast, the PRIMA trial adopted a biomarker-unselected design and demonstrated a PFS benefit in the overall population (HR =0.62), with greater magnitude in HRD-positive tumors (HR =0.43) but retained activity in HRD-negative patients (HR =0.68) ( 34 ). Collectively, these trials illustrate complementary strategies: precision selection in biomarker-defined populations versus cautious expansion into broader risk groups. PARPis are generally well tolerated, but each agent exhibits a distinct toxicity profile that requires proactive monitoring and management. Common class effects include hematologic toxicities (anemia, thrombocytopenia, neutropenia) and gastrointestinal symptoms (nausea, vomiting, fatigue). These are typically grade 1–2 and manageable with dose adjustments and supportive care. Niraparib is associated with a higher incidence of hypertension and thrombocytopenia. Individualized starting doses based on baseline platelet count and body weight significantly reduce grade ≥3 thrombocytopenia without compromising efficacy, as demonstrated in subsequent dose-optimization analyses and reflected in the approved prescribing information ( 34 ). Regular blood pressure monitoring and antihypertensive management are recommended. Rucaparib frequently causes transient elevations in alanine aminotransferase (ALT), aspartate aminotransferase (AST), and creatinine. These abnormalities are generally reversible and not associated with progressive liver injury; they typically stabilize or resolve with continued treatment or dose interruption ( 35 , 36 ). Olaparib’s most common adverse events are anemia and gastrointestinal disturbances; long-term use requires vigilance for rare events such as MDS and AML, which occur in approximately 1–2% of patients in long-term follow-up analyses ( 31 , 32 ). Fuzuloparib’s safety profile appears similar to other PARPis, with hematologic toxicity being most frequent; long-term safety data are still accumulating ( 37 ). Overall, patient education, regular laboratory monitoring, and individualized dose modification are essential to maintain treatment adherence and quality of life during long-term maintenance therapy.

Despite the marked clinical success of PARPis, the emergence of therapy resistance poses a critical challenge that ultimately limits long-term patient benefit. A comprehensive understanding of the underlying resistance mechanisms is fundamental to devising strategies for overcoming this limitation and improving outcomes. Accumulating mechanistic and translational studies have begun to elucidate the biological underpinnings of PARPi resistance. Current evidence supports three principal resistance pathways: restoration of HRR function, replication fork stabilization, and intracellular adaptations, including drug efflux and tumor microenvironment (TME) remodeling. Table 3 summarizes these mechanisms and highlights their corresponding therapeutic strategies and degree of clinical actionability. ctDNA, circulating tumor DNA; MDSC, myeloid-derived suppressor cell; PARPi, poly(ADP-ribose) polymerase inhibitor; TAM, tumor-associated macrophage; TME, tumor microenvironment. Under physiological conditions, stalled replication forks are subjected to nucleolytic degradation by enzymes such as MRE11 and MUS81 , a process essential for initiating DNA repair ( 38 ). BRCA1/2 proteins protect the fork, while factors such as EZH2 , PTIP , and RADX facilitate nuclease recruitment ( 39 , 40 ). In BRCA-deficient cells, however, downregulation of EZH2 and PTIP impairs this recruitment, leading to fork stabilization. This fork protection allows tumor cells to tolerate replication stress without relying on homologous recombination, thereby diminishing the synthetic lethal effect of PARPis and conferring resistance ( 24 , 39 ). The overexpression of ATP-binding cassette (ABC) transporters, notably ABCB1 (P-glycoprotein), constitutes a common multidrug resistance mechanism. ABCB1 mediates the efflux of PARPis from the cell in an ATP-dependent manner, thereby diminishing intracellular drug accumulation ( 41 ). Upregulation of ABCB1 expression is observed in ovarian cancer tissues following multiple chemotherapy lines, which can similarly confer resistance to PARPis ( 42 ). By lowering intracellular PARPi concentrations, this mechanism promotes tumor cell survival despite HRD, effectively attenuating the synthetic lethal interaction ( 43 ). Tumor cells can evade PARPi toxicity by restoring proficient HRR through several molecular strategies: ❖ Genetic reversion: secondary mutations in BRCA1 / 2 or other HRR genes can restore the open reading frame and produce functional protein, directly reversing the initial homologous recombination defect ( 44 ). Lord and Ashworth systematically elucidated this mechanism in a seminal review, demonstrating how reversion mutations restore BRCA1/2 function and confer resistance to both platinum chemotherapy and PARPis ( 45 ). Clinical evidence from ovarian cancer further supports this mechanism: approximately 46% of patients with platinum-resistant BRCA -mutant tumors harbor tumor-specific secondary mutations that restore the BRCA1/2 open reading frame ( 46 ). ❖ Exon skipping: alternative splicing that bypasses mutant exons can generate truncated but partially functional protein isoforms, circumventing the deleterious effects of the original mutation ( 47 ). ❖ Dysregulated end resection: in BRCA1 -deficient cells, overexpression of CtBP-interacting protein (CtIP), a key regulator of DNA end resection, promotes excessive DNA end resection, facilitating RAD51 loading and restoration of a functional homologous recombination pathway. This process can be augmented by Syk kinase-mediated phosphorylation of CtIP, which stabilizes CtIP at damage sites and enhances resection efficiency, thereby driving resistance ( 48 , 49 ). Emerging evidence suggests that the TME is associated with PARPi resistance. Long-term PARPi exposure can activate inflammatory signaling pathways within TME. For example, PARP inhibition can induce STAT3 activation, thereby promoting the formation of immunosuppressive states and providing favorable conditions for the survival of drug-resistant clones. This process enhances the recruitment and infiltration of myeloid-derived suppressor cells (MDSCs) and tumor-associated macrophages (TAMs), further dampening antitumor immune responses ( 50 ). In addition, hypoxia within the TME can downregulate HRR-related gene expression, potentially altering PARPi sensitivity while simultaneously promoting angiogenesis and metabolic reprogramming. These dynamic TME interactions suggest that strategies targeting tumor cells alone may be insufficient to overcome resistance durably. Instead, combination approaches addressing both tumor-intrinsic pathways and microenvironment-mediated mechanisms may provide greater therapeutic benefit. Some of these resistance mechanisms are already clinically feasible. The detection of BRCA1/2 reversion mutations via liquid biopsy [circulating tumor DNA (ctDNA)] is a classic example. As Lord and Ashworth noted in the seminal review, tumor-acquired resistance to PARPi and platinum-based drugs is closely associated with BRCA1/2 secondary mutations ( 45 ). If such mutations are detected in a patient progressing on a PARPi, retreatment with PARPis or platinum should be avoided, and enrollment in clinical trials of ATR or other novel inhibitors should be considered. Currently, early-phase clinical trials with ATR inhibitors (e.g., ceralasertib and berzosertib) targeting replication fork protection are ongoing. In contrast, mechanisms such as drug efflux pump overexpression lack effective clinical inhibitors; therefore, strategies to modulate the TME to overcome resistance remain largely preclinical or in early clinical development. Among these, BRCA reversion mutations are currently the most clinically relevant, guiding therapy switches via ctDNA monitoring, while replication fork protection is being targeted by ATR inhibitors in trials, and drug efflux or TME-mediated resistance remains investigational.

Given the well-elucidated mechanisms of PARPi resistance, the development of effective countermeasures has emerged as a critical research focus. Among these, the combination of PARPis with antiangiogenic agents represents a particularly promising strategy, leveraging multitargeted intervention to circumvent resistance. The synergistic potential of this combination is underpinned by several key mechanisms. The first is normalization of the TME, in which agents, such as bevacizumab, inhibit vascular endothelial growth factor (VEGF), promoting partial normalization of the disorganized tumor vasculature, which enhances intratumoral blood perfusion and drug delivery. Furthermore, by alleviating tissue hypoxia and downregulating hypoxia-inducible factor-1α (HIF-1α), antiangiogenic therapy may indirectly compromise DNA damage repair capacity, thereby sensitizing tumors to PARPi cytotoxicity ( 51 , 52 ). The second key mechanism is the dual inhibition of DNA damage repair. Evidence indicates functional crosstalk between the VEGF/VEGF receptor 2 (VEGFR2) signaling pathway and key DNA repair proteins (e.g., ATM and BRCA1 ) ( 53 ). VEGF pathway inhibition reduces phosphorylation of these repair proteins and impairs the formation of DNA repair foci, thereby acting in concert with PARPis to suppress the DNA damage response (DDR) ( 54 ). The third mechanism involves polyclonal suppression and resistance prevention: this strategy concurrently targets genomic instability (via PARPis) and the supportive TME (via antiangiogenics) by applying multifaceted selective pressure on both cancer cells and their niche. This dual targeting theoretically facilitates a more effective eradication of heterogeneous tumor cell populations and reduces the opportunity for the expansion of resistant subclones dependent on a single survival pathway ( 55 ). This robust mechanistic rationale is strongly corroborated by clinical evidence. As discussed in the “Olaparib: defining biomarker-driven efficacy and establishing survival milestones” section, the PAOLA-1 trial established a significant overall survival benefit of the olaparib-bevacizumab combination in the HRD-positive population, underscoring the long-term value of this approach ( 33 ). The application of this combination is now being extended to more challenging clinical scenarios. For instance, the phase II AVANOVA2 trial demonstrated that the combination of niraparib and bevacizumab significantly improved PFS compared to niraparib monotherapy (median PFS: 11.9 vs. 5.5 months; HR =0.35) in an all-comer population of patients with platinum-sensitive recurrent ovarian cancer ( 56 ). This breakthrough not only provides a viable option for overcoming established resistance but also supports the feasibility of an effective chemotherapy-free targeted therapy paradigm in this recurrent patient setting.

Building on the success of coupling PARPis with antiangiogenic agents, research has expanded into other rational combinations, notably with immune checkpoint inhibitors and novel DDR pathway inhibitors, to combat resistance through complementary mechanisms. PARPis can profoundly remodel the tumor immune microenvironment, creating a foundation for powerful synergy with immunotherapy. This interplay operates through multiple mechanisms: ❖ Enhanced immunogenicity: PARPi-induced DNA damage accumulation increases tumor mutational burden and neoantigen generation, thereby promoting immune recognition ( 57 ). ❖ Activation of innate immunity: cytosolic DNA fragments resulting from genomic instability activate the cGAS/STING pathway, driving the production of type I interferons and other inflammatory cytokines that recruit and activate cytotoxic T cells ( 58 , 59 ). ❖ Upregulation of immune checkpoints: PARPis concurrently upregulate PD-L1 expression on tumor cells, providing a strong mechanistic rationale for combining them with PD-1/PD-L1 blockade ( 60 ). This compelling preclinical synergy is now being translated into the clinical arena. The phase III FIRST/ENGOT-OV44 trial on patients with newly diagnosed ovarian cancer provided preliminary validation. Compared to first-line chemotherapy followed by niraparib maintenance, the addition of the PD-1 inhibitor dostarlimab yielded a modest PFS benefit (median PFS, 20.6 vs. 19.2 months; HR =0.85) ( 61 ). Despite a strong preclinical rationale, the modest clinical benefit observed to date suggests that optimal patient selection (e.g., HRD-positive, immune-infiltrated tumors) or alternative scheduling (e.g., sequential rather than concurrent) may be required to fully realize this synergy. Another strategy involves the precision cotargeting of other critical nodes within the DDR network. When tumor cells evade PARPi lethality by activating backup repair pathways or adapting cell-cycle checkpoints, inhibiting these bypass mechanisms can effectively block escape routes. The promising candidates are detailed below. ❖ ATR inhibitors: these agents can reverse the PARPi resistance associated with SLFN11 deficiency by disrupting the replication stress response, resensitizing tumors to PARPi-induced catastrophe ( 62 ). ❖ WEE1 inhibitors: by inhibiting CDK1/2 phosphorylation, WEE1 inhibitors force cells with unrepaired DNA damage to prematurely enter mitosis, synergizing with PARPi to induce mitotic catastrophe ( 63 , 64 ). ❖ POLQ inhibitors: targeting the error-prone microhomology-mediated end joining) pathway with POLQ inhibitors eliminates a key backup repair mechanism, thereby enhancing synthetic lethality in homologous repair–deficient tumors and overcoming the resistance driven by this pathway ( 65 ). Collectively, these mechanism-driven combination strategies signify a pivotal evolution in overcoming PARPi resistance—from initial broad combinations to rationally designed, precision-targeted approaches. They represent a promising therapeutic frontier for patients with difficult-to-treat diseases.

The successful implementation of strategies to overcome PARPi resistance is highly dependent on robust biomarkers for patient selection and therapy monitoring. In this context, liquid biopsy, particularly through the analysis of ctDNA, has emerged as an indispensable noninvasive tool for the dynamic surveillance of resistance mechanisms. ctDNA analysis enables the comprehensive profiling of the key molecular alterations driving PARPi resistance. This includes the detection of acquired BRCA1 / 2 reversion mutations, epigenetic changes such as RAD51C promoter methylation, and dysregulation of alternative DNA repair pathways (e.g., ATM/CHK2 abnormalities), with high concordance with traditional tissue biopsies ( 66 , 67 ). Longitudinal ctDNA studies have been instrumental in validating these mechanisms. While landmark trials such as ARIEL2 established the efficacy of rucaparib in patients with BRCA -mutant ovarian cancer ( 68 , 69 ), the limitation of both primary and acquired resistance became apparent. Subsequent ctDNA analyses have elucidated BRCA reversion mutations as a central mechanism. Notably, such research discovered that a subset of patients exhibits pre-existing BRCA reversion mutations in baseline ctDNA, which is associated with minimal clinical benefit and elucidates a fundamental mechanism of primary resistance ( 70 ). HRD testing itself is evolving. Current methods for assessing genomic instability are primarily based on integrated scoring systems that incorporate loss of heterozygosity (LOH), telomeric allelic imbalance (TAI), and large-scale state transitions (LST) ( 71 ), as validated in clinical trials of PARPis (34 and related assay validation studies). Multiple commercial platforms (e.g., myChoice CDx, FoundationOne CDx) are available; however, standardization across assays remains a challenge ( 72 ). It is critical to distinguish between static and dynamic HRD classification. Current commercial assays provide a static snapshot based on accumulated genomic scars. However, HRD status is dynamic and can evolve under treatment pressure. For example, BRCA reversion mutations or epigenetic changes can restore homologous recombination proficiency, converting an HRD tumor into an HR-proficient one ( 45 , 46 ). This underscores the need for longitudinal monitoring via liquid biopsies rather than reliance on a single baseline test. Furthermore, disparities in access to genetic testing and HRD assays across different regions create inequities in precision medicine and limit the broader implementation of PARPi in resource-limited settings ( 72 ). These findings underscore the dual clinical utility of ctDNA analysis: it not only facilitates the early detection of “molecular progression” during therapy but also holds the potential for prospective efficacy prediction through the identification of primary resistance at baseline. Consequently, dynamic genomic monitoring via ctDNA provides a critical window for pre-emptive therapeutic intervention, enabling timely treatment adjustment and facilitating the personalized management of PARPi resistance.

To identify relevant literature for this narrative review, a structured literature search strategy was employed, with a focus on capturing key developments in PARPi for treating ovarian cancer. The primary aim was to gather comprehensive information on the mechanism of action, clinical trial evidence, resistance mechanisms, and novel combination strategies for PARPis. As this is a narrative review, we prioritized data from large-scale randomized controlled trials, meta-analyses, and authoritative guidelines to ensure a high level of evidence. The detailed search strategy is summarized in Table 1 . Where applicable, we prioritized randomized controlled trials, phase III studies, high-quality meta-analyses, and guideline statements. No formal risk-of-bias assessment or quantitative evidence grading was performed, as this work was designed as a narrative (rather than systematic) review. ESGO, European Society of Gynecological Oncology; MeSH, medical subject headings; NCCN, National Comprehensive Cancer Network; PARP, poly(ADP-ribose) polymerase; PARPi, poly(ADP-ribose) polymerase inhibitor.

PARPis have irrevocably transformed the standard of care for ovarian cancer, yet their long-term benefit for many patients is curtailed by the emergence of acquired resistance and the need to manage toxicity. To propel the field forward and overcome this bottleneck, a multipronged research strategy is essential. First, a more profound, systematic dissection of resistance evolution is necessary. This entails the application of integrated multiomics analyses coupled with longitudinal liquid biopsies to map the complex evolutionary trajectories and clonal heterogeneity of resistant tumors. Second, the development and optimization of mechanism-informed combination therapies must be accelerated. The rational combinations discussed in this review—including the addition of PARPis to antiangiogenic agents, immunotherapies, or novel DDR inhibitors—should be rigorously evaluated within precisely defined molecular subgroups to maximize efficacy and minimize overlapping toxicities. Third, a paradigm shift in clinical trial design is urgently needed. Future trials must be fundamentally translational and biomarker-driven and systematically integrate predictive biomarkers (e.g., HRD status and SLFN11 expression) and dynamic resistance markers (e.g., emergent BRCA reversions) for the construction of flexible clinical decision-making pathways. The overarching goal is to transition decisively from the current model of empirical stratification to a future of authentic, dynamically guided individualized therapy. Finally, broader considerations should also be addressed: the economic cost of long-term PARPi therapy, and the ethical implications of unequal access to these drugs—linked to disparities in genetic and HRD testing—require attention from policymakers, healthcare systems, and the oncology community to ensure equitable benefits for patients. These integrated efforts may further extend the durability of PARPi-based therapy and improve long-term disease control.

The article’s supplementary files as