Ubiquitin-specific proteases in ovarian cancer: molecular mechanisms and therapeutic implications

Abstract

Ovarian cancer (OC) remains one of the most lethal gynecological malignancies, largely due to late-stage diagnosis, frequent recurrence, and the development of resistance to standard therapies. These clinical challenges highlight the urgent need for a deeper understanding of the biological processes driving OC and the identification of novel therapeutic targets. The ubiquitin-proteasome system is a central mechanism for selective protein degradation. Accumulating evidence indicates that deubiquitinases, particularly ubiquitin-specific proteases (USPs), are key regulators of oncogenic signaling and therapy response in OC. This review provides a comprehensive overview of how dysregulated USPs shape major cancer hallmarks in OC, including cell-cycle control, DNA damage repair and genomic stability, apoptosis, metastasis, metabolic reprogramming, and immune evasion. We further summarize advances in developing USP-directed pharmacologic inhibitors, discuss emerging translational strategies, and evaluate the rationale for combining USP inhibition with established treatments such as platinum-based chemotherapy and poly-ADP ribose polymerase inhibitors (PARPis) to overcome resistance. Collectively, understanding USP-dependent networks may facilitate the identification of novel biomarkers and therapeutic targets, offering new strategies for precision treatment in OC.

Introduction

Ovarian cancer is the fifth leading cause of cancer-related death in women [1]. Due to the lack of early warning symptoms and effective screening strategies, most cases are diagnosed at an advanced stage. Despite the use of cytoreductive surgery and platinum-based chemotherapy, many patients experience recurrence and develop resistance to treatment, particularly those with high-grade serous ovarian cancer (HGSOC) [2]. Current maintenance therapies, including anti-angiogenic agents, PARPis, and immune checkpoint inhibitors, show limited efficacy due to the development of drug resistance and the immune-desert phenotype in OC [3, 4]. Therefore, there is an urgent need for the development of innovative, precision-targeted therapies to overcome these limitations.

Proteins undergo various post-translational modifications (PTMs), which critically influence their stability, localization, and functional activity within diverse cellular signaling and regulatory networks. Among PTMs, ubiquitination stands out as a prevalent and critical process. Ubiquitination involves a highly orchestrated cascade of enzymatic reactions, catalyzed by E1 ubiquitin-activating enzymes, E2 ubiquitin-conjugating enzymes, and E3 ubiquitin ligases, which together transfer ubiquitin molecules to substrate proteins. This modification can mark proteins for degradation via the proteasome or alter their function and localization [5, 6]. Reversibly, deubiquitinases (DUBs) can cleave ubiquitin chains from substrate proteins, edit ubiquitin chains, and process ubiquitin precursors [7]. An overview of the ubiquitin–proteasome system and the deubiquitination process is shown in Fig. 1.

Fig. 1.

Ubiquitination and deubiquitination in the ubiquitin–proteasome system. Legend: Ubiquitination is mediated through the sequential actions of E1 ubiquitin-activating enzymes, E2 ubiquitin-conjugating enzymes, and E3 ubiquitin ligases, which transfer ubiquitin to substrate proteins. Ubiquitinated substrates may undergo proteasomal degradation or functional modulation, depending on the ubiquitin linkage context. Deubiquitinases reverse or remodel ubiquitin modifications by removing ubiquitin chains from substrate proteins, thereby contributing to substrate stabilization and functional regulation

Accumulating evidence indicates that deubiquitinases (DUBs) are involved in multiple hallmarks of cancer [8]. These findings have positioned DUBs as promising therapeutic targets in cancer. However, the biological functions, substrates, and regulatory mechanisms of many DUBs remain incompletely understood, particularly in specific cancer. According to integrated annotations for Ubiquitin and Ubiquitin-like Conjugation Database (iUUCD), DUBs have been divided into mainly seven broad classes. Among them, USPs are the most diversified and extensively studied class, making them a representative class for understanding the oncogenic roles of DUBs. Importantly, a growing subset of USPs, including USP1, USP7, and USP13, has been specifically linked to OC pathogenesis [9, 10]. In parallel, increasing efforts have been devoted to the development of USP-targeted inhibitors, highlighting their therapeutic potential in OC.

In this review, we summarize the biological functions of USPs in OC, focusing on the molecular mechanisms through which they promote malignant progression and therapeutic resistance. We also review recent advances in the preclinical and clinical development of USP-targeted inhibitors, and discuss their translational potential as emerging strategies for precision therapy in OC. By integrating current evidence, this review aims to identify therapeutic vulnerabilities and inform future efforts to translate USP-directed approaches into clinical application in OC.

The role of USPs in OC progression

To provide an overview of the mechanisms summarized in this section, Fig. 2 illustrates the major hallmarks of OC regulated by representative USPs discussed in this review, whereas Table 1 summarizes reported USP–substrate interactions in OC, validated cell models, and associated functional consequences.

Fig. 2.

Representative USPs involved in major hallmarks of ovarian cancer. Legend: This figure summarizes the major hallmarks of ovarian cancer regulated by representative ubiquitin-specific proteases discussed in this review, including cell proliferation, genome stability, apoptotic signaling, invasion and metastasis, cellular metabolism, immune evasion, and cancer stemness. The outer ring lists representative USP family members associated with each hallmark on the basis of the current evidence summarized in Sect. USPs involved in tumor proliferation–USPs sustain cancer stemness and tumor repopulation. This figure is intended as a conceptual overview rather than a comprehensive mechanistic map

Table 1.

Reported USP–substrate interactions in OC models, together with their functional consequences and direct molecular mechanisms

Family member	Cell model(s) for mechanistic validation	Functional consequence	Direct substrate and molecular mechanism	Refs.
USP5	OVCAR3 and CAOV3	Promote cell proliferation	Deubiquitinates and stabilizes HDAC2, leading to repression of p21 and p27	[11]
USP15	TYK-Nu	Promote cell proliferation	Deubiquitinates and stabilizes p53-R175H	[12]
USP7	A2780 and SKOV3	Promote cell proliferation, migration and invasion	Deubiquitinates and stabilizes LARS1, thereby enhancing mTORC1 signaling	[9]
USP13	EFO-27	Promotes cisplatin and PARPi resistance by enhancing DDR	Deubiquitinates and stabilizes RAP80 and promotes RAP80 recruitment to DNA damage sites	[10]
USP1	COV-318 and OVCAR-8	Promotes PARPi tolerance by limiting PARP1 trapping	Deubiquitinates PARP1 and promote its auto-PARylation	[13]
USP36	OVCAR8	Supports replication stress tolerance and resistance to cisplatin/olaparib	Deubiquitinates and stabilizes PrimPol following replication stress	[14]
USP28	SKOV3 and UWB1.289	Promotes PARPis resistance by enhancing DDR	Deubiquitinates and stabilizes SOX9, thereby enhancing DDR gene expression	[15]
USP13	HEY	Supports survival under stress conditions	Deubiquitinates and stabilizes MCL1	[16]
DUB3(USP17)	OVCAR3, OVCA433 or SKOV3	Promotes anti-apoptotic signaling and platinum resistance	Deubiquitinates and stabilizes MCL1	[17]
USP14	A2780CP	Promotes cisplatin resistance by suppressing apoptosis	Deubiquitinates and stabilizes BCL6	[18]
USP11	OVCAR-3	Promotes EMT-associated migration and invasion	Deubiquitinates and stabilizes Snail	[19]
USP1	MDAH-2774 and OVCAR-8	Promotes migration, invasion, and platinum-resistant dissemination	Deubiquitinates and stabilizes Snail following ATM/ATR-dependent phosphorylation	[20]
USP45	SKOV3 and OVCAR3	Promotes migration, invasion, and cisplatin resistance	Deubiquitinates and stabilizes Snail	[21]
USP22	A2780 and SKOV3	Promotes EMT-associated migration and invasion	Deubiquitinates and stabilizes ZEB1	[22].
USP14	TOV21G and RMG-1	Promotes glycogen-dependent metabolic adaptation and platinum resistance	Participates in the p53/RNF144A/GYS1 regulatory circuit through deubiquitinating and stabilizing p53	[23]
USP35	ID8	Restrains STING-dependent antitumor immune signaling	Deubiquitinates and inactivates STING, thereby suppressing STING–TBK1–IRF3 signaling	[24]
USP9X	CAOV3 and ID8	Sustains cancer stemness and chemoresistance	Deubiquitinates and stabilizes HIF-2α	[25]

USPs involved in tumor proliferation

Multiple USPs have been implicated in promoting OC cell proliferation. In this section, we distinguish those acting through relatively direct control of proliferation-associated proteins from those that support growth indirectly by reinforcing oncogenic signaling pathways.

Some USPs promote proliferation through relatively direct regulation of proteins involved in cell-cycle progression or sustained proliferation. Downregulation of USP5 significantly suppresses OC cell proliferation by inducing G0/G1 phase arrest [11]. Mechanistically, USP5 deubiquitinates HDAC2, thereby repressing p27 and p21 expression and favoring cell-cycle progression. USP15 has also been linked to proliferative maintenance through stabilization of mutant p53 proteins. This is mechanistically important because common gain-of-function p53 mutants, such as p53-R175H, p53-R248Q, and p53-R273H, are highly stable and prone to forming oncogenic aggregates [26]. Accordingly, depletion of USP15 promotes selective clearance of mutant p53 through a ubiquitin-dependent lysosomal pathway and decreases cell viability in p53-R175H-expressing cells [12].

Other USPs appear to support proliferation indirectly through oncogenic signaling networks. USP7 promotes pro-growth signaling by deubiquitinating LARS1 through direct interaction with the 256–605 amino acid region of LARS1 [9]. LARS1 is upregulated in OC and exerts oncogenic effects by activating mTORC1 signaling. USP18 has also been reported as a positive regulator of OC proliferation and aggressiveness [27]. STAT3 overexpression enhances USP18 promoter activity, whereas silencing USP18 decreased the phosphorylated STAT3 levels, suggesting a potential positive feedback loop between USP18 and the Jak-STAT3 signaling in OC [27]. In addition, USP18 promotes OC progression by regulating the transcription of FBXO6, an oncogenic protein that drives degradation of the tumor suppressor RNASET2 [27, 28].

USP22 and USP39 have been associated with proliferative phenotypes and cell-cycle alterations in OC, although direct substrate-level evidence linking their deubiquitinating activity to cell-cycle control remains less clearly defined in this setting. Inhibition of USP22 induces G0/G1 arrest and suppresses tumor growth in OC xenograft models, accompanied by reduced TGF-β1, cyclin D2, CDK4, and CDK2 expression and increased p27 levels [29], whereas USP39 knockdown results in G2/M arrest together with upregulation of p53 and p21 expression [30].

USPs regulate genome stability

USPs have emerged as key modulators of DNA damage response (DDR) and replication stress responses in OC, thereby shaping sensitivity to platinum chemotherapy and PARPis. Current evidence suggests that these effects can be broadly organized into three mechanistic categories.

One major mechanism involves direct regulation of homologous recombination repair (HRR). USP13 promotes DDR and resistance to cisplatin and PARPis [10, 31]. Upon DNA damage, ATM phosphorylates USP13, enabling it to remove K63-linked ubiquitin from RAP80 and facilitate RAP80 recruitment to damage sites [10], which in turn supports BRCA1 assembly at DNA double-strand breaks (DSBs) through the scaffold protein CCDC98, ultimately enhancing repair capacity [32]. A related mechanism has been described for USP15, which enhances HRR by deubiquitinating the BRCT domain of BARD1 and stabilizing the BRCA1/BARD1 complex at repair foci after ATM-dependent phosphorylation and MDC1-mediated recruitment to DSBs [33]. However, unlike USP13, this mechanism has been established primarily in non-ovarian models, including HEK293T, U2OS, and MCF7 cells, and therefore remains extrapolative in the context of OC.

A second mechanism involves regulation of DNA lesion processing and replication stress responses. Beyond BRCA1/homologous recombination(HR)-centered pathways, USP1 regulates PARP1 signaling at DNA lesions by deubiquitinating PARP1 and removing K63-linked polyubiquitin chains [34].Notably, this modification does not primarily affect PARP1 protein stability, but rather modulates its function. Inhibition of USP1 suppresses PARP1 auto-PARylation, thereby enhancing PARP1 trapping at DNA lesions and increasing sensitivity to PARP inhibitors [34]. In parallel, pharmacologic inhibition of USP1 has been shown to induce accumulation of mono- and polyubiquitinated PCNA together with reduced total PCNA, thereby impairing DNA replication and repair [13]. Importantly, this PCNA dysregulation creates a synthetic lethal vulnerability in BRCA1/2-deficient tumors, as ectopic PCNA expression reverses USP1 inhibitor sensitivity [13]. Although USP1 has been reported to regulate monoubiquitinated PCNA [35], this study primarily focused on the consequences of USP1 inhibition rather than further dissecting the direct mechanism of USP1-mediated PCNA deubiquitination in OC. USP36 also promotes replication stress recovery by deubiquitinating PrimPol (removing K29-linked ubiquitin), thereby preserving PrimPol-dependent replication fork restart. Conversely, loss of USP36 shortens replication tracts and increases vulnerability to cisplatin and olaparib in OVCAR8 and PEO1 cell lines [14].

A third mechanism involves indirect enhancement of DNA repair through transcriptional upregulation of DDR-related genes. USP28 has been linked to olaparib resistance by stabilizing SOX9, which transcriptionally upregulates key DDR genes, including SMARCA4, UIMC1, and SLX4 [15].

USPs modulate apoptotic signaling

USPs influence apoptosis of OC cells largely by stabilizing anti-apoptotic factors or attenuating stress-induced death signaling, thereby shaping treatment response. Among the currently described mechanisms, the most coherent axis centers on MCL1-dependent survival, whereas additional USPs affect apoptosis through other anti-apoptotic proteins or through broader signaling pathways.

Several USPs directly regulate MCL1 protein stability. USP13 deubiquitinates and stabilizes MCL1 in OC cells [16]. Notably, although USP13 depletion reduces MCL1 protein levels, it does not significantly affect cell proliferation or migration under normal growth conditions. Instead, USP13 loss markedly increases apoptosis under hypoxic or oxidative stress, suggesting that USP13-dependent MCL1 stabilization becomes particularly important for survival under stress conditions, rather than initiating the apoptotic process. DUB3 also stabilizes MCL1 and promotes chemoresistance [17]. Upstream of this axis, MGMT transcriptionally activates DUB3, and inhibition of MGMT with PaTrin-2 suppresses cell growth in OC models with elevated MGMT, DUB3, and MCL1 expression. Histone deacetylase inhibitors (HDACis) further induce MGMT expression and sensitize OC cells to PaTrin-2, such that combined treatment with HDACis and PaTrin-2 produces enhanced antitumor effects in multiple OC models. USP9X has been identified as a deubiquitinase that stabilizes MCL1 by removing K48-linked polyubiquitin chains [36]. However, this direct mechanism was established in 293T cells rather than in ovarian cancer-specific models in the original study. In ES2 cells, subsequent work showed that BAG3 interacts with USP9X and positively regulates its protein expression, while USP9X silencing reduces MCL1 levels, thereby supporting BAG3–USP9X–MCL1 axis associated with resistance to apoptosis in OC [37].

Beyond MCL1, USPs also modulate apoptosis through other apoptosis-related proteins. USP14 promotes cisplatin resistance and suppresses apoptosis by deubiquitinating and stabilizing the anti-apoptotic protein BCL6 [18]. USP14 has also been linked to maintenance of Bcl-xL, suggesting that it may inhibit apoptosis through multiple anti-apoptotic effectors [38].

Additional USPs appear to regulate apoptosis more indirectly through broader signaling programs rather than through direct stabilization of a single apoptosis regulator. USP39 has been implicated in carboplatin resistance by sustaining EGFR-, AKT-, and ERK-associated survival signaling. Consistent with this, USP39 knockdown enhances carboplatin-induced apoptosis in SKOV3 cells, accompanied by activation of PARP and caspase-3 [39]. A related pattern is observed for USP8 in cisplatin-resistant OC cells [40]. Elevated USP8 expression is associated with the resistant state, whereas USP8 silencing decreases EGFR and AKT activation, reduces FLIPL expression, and increases cisplatin-induced caspase-3/7 activation, thereby sensitizing OC cells to apoptosis.

USPs regulate invasion and metastasis

Accumulating evidence indicates that several USPs have been implicated in OC invasion and metastasis by reinforcing epithelial–mesenchymal transition (EMT) at multiple regulatory levels.

At the level of core EMT transcription factors, USP11 promotes EMT by deubiquitinating and stabilizing Snail [19]. Similar Snail-centered mechanisms have also been implicated for USP1 and USP45 [20, 21]. Following cisplatin treatment, USP1 is directly phosphorylated by ATM and ATR on S42 and S331, leading to its interaction with Snail and preventing Snail degradation [20]. Likewise, MYH9 recruits USP45 to deubiquitinate Snail, thereby enhancing migration/invasion and cisplatin resistance in serous OC models [21, 41]. In addition to Snail, another core EMT transcription factor, ZEB1, can also be stabilized by USP22-mediated deubiquitination, thereby promoting EMT and malignant progression in OC. Moreover, SRSF9 binds to and stabilizes USP22 mRNA, while ZEB1 directly binds to the SRSF9 promoter and enhances its transcription, thereby forming a SRSF9/USP22/ZEB1 positive feedback loop that sustains EMT signaling in OC [42].

At the level of indirect pro-metastatic signaling, USP39 enhances EGFR-dependent PI3K/AKT and MAPK/ERK signaling, and elevated USP39 is associated with increased peritoneal dissemination, a hallmark of OC progression [39]. Similarly, USP13 overexpression promotes peritoneal metastasis and hemorrhagic ascites in vivo, whereas USP13 inhibition suppresses OC progression and increases susceptibility to AKT inhibitor, suggesting that USP13 and PI3K/AKT may share a downstream pathway [31]. In addition, USP7 and MARCH7 expression is associated with poor prognosis and lymph node metastasis in OC, and functional studies in SKOV3 cells showed that USP7 regulates E-cadherin and β-catenin expression through the MARCH7 pathway [22]. Notably, USP7-mediated stabilization of MARCH7 was originally demonstrated in HEK293T cells [43], while immunoprecipitation in SKOV3 cells further supports their interaction in OC [22]. Together, these findings suggest that USP7 may contribute to OC cell migration and invasion through ubiquitin-dependent regulation of MARCH7, although the direct mechanism remains to be fully established in OC models.

USPs deregulate cellular metabolism

USPs have also been implicated in metabolic reprogramming in OC. Amplification of USP13 has been shown to promote OC progression by rewiring cellular metabolism [44]. Mechanistically, USP13 was shown in HEK293T cells to deubiquitinate and stabilize ATP citrate lyase (ACLY) and oxoglutarate dehydrogenase (OGDH), two key enzymes that coordinate glutaminolysis, mitochondrial respiration, and fatty acid synthesis [45–47]. In OC cells, this mechanism is supported functionally by the observation that USP13 amplification promotes glutamine anaplerosis to replenish tricarboxylic acid (TCA) cycle intermediates and sustain ATP production, while simultaneously supporting acetyl-CoA supply for lipid biosynthesis [44]. Conversely, USP13 depletion disrupts OGDH- and ACLY-dependent metabolism, leading to defective glutamine anaplerosis, suppressed lipogenesis, mitochondrial dysfunction, and ultimately metabolic stress-induced cancer cell death [44].

In addition, USP14 has been implicated in metabolic adaptation during platinum resistance in ovarian clear cell carcinoma (OCCC). In this setting, p53 transcriptionally upregulates RNF144A, which promotes GYS1 ubiquitination, whereas GYS1 facilitates proteasomal degradation of p53 by competitively binding to USP14, thereby forming a p53/RNF144A/GYS1 feedback circuit [23]. Under platinum stress, this circuit facilitates mobilization of accumulated glycogen, which fuels energetic NADPH production, resulting in resistance to disulfidptosis and increased platinum resistance in OCCC. These observations suggest that USP-associated metabolic programs may represent actionable vulnerabilities in OC.

USPs involved in immune evasion

Emerging evidence implicates USPs in modulating antitumor immunity in OC. Integrative analyses of OC datasets indicate that USP35 overexpression correlates with reduced immune infiltration [24]. Mechanistically, USP35 binds the cytoplasmic C-terminal region of STING and deubiquitinates STING via its C-terminal catalytic domains [24]. In SKOV3 cells, USP35 was found to inactivate STING, while STING activation facilitated its binding to USP35 in a STING phosphorylation-dependent manner. Functionally, USP35 silencing enhances the expression of type I interferon genes and downstream chemokines through the STING–TBK1–IRF3 axis in response to cytosolic DNA or cGAMP, and increases cisplatin-induced interferon signaling in OC xenografts, highlighting its potential role in modulating the tumor immune microenvironment and chemotherapeutic responsiveness.

USPs sustain cancer stemness and tumor repopulation

USPs can maintain cancer stem cell (CSC) programs that enable tumor initiation, chemoresistance, and post-treatment repopulation in HGSOC. A representative example is USP9X, which links hypoxia and TGF-β signaling to CSC maintenance [25]. Hypoxic tumor microenvironment is an initiator of TGF-β signaling, which upregulates USP9X via Smad2/3-mediated transcriptional regulation, and USP9X subsequently deubiquitinates and stabilizes HIF-2α, thus promoting stemness reprogramming [25]. Clinically, USP9X expression correlates with TGF-β signatures, CSCs characteristics, EMT behaviors, and chemotherapy responsiveness, along with HIF-2ɑ [25].

Translational research on USP inhibitors in OC therapy

Several USP-directed agents have entered early-phase clinical trials in advanced solid tumors, and a subset of these studies explicitly include OC among eligible tumor types, generating initial safety and pharmacokinetic/pharmacodynamic (PK/PD) data. However, no USP inhibitor has yet been developed in an OC-specific clinical program (Table 2). The two main translational strategies discussed in this section are summarized in Fig. 3.

Table 2.

Early-phase clinical trials of USP inhibitors enrolling patients with advanced solid tumors

Agent	Sponsor	Target	ClinicalTrials.gov ID	Current status	Phase	Population / key eligibility	Regimen
Bardoxolone methyl (RTA 402)	Biogen	USP7	NCT00529438	Completed	I	Advanced solid tumors and lymphoid malignancies	Monotherapy
TNG348	Tango Therapeutics	USP1	NCT06065059	Terminated due to safety	I/II	BRCA1/2-mutant or other HRD+ advanced/metastatic solid tumors, including OC among eligible populations	Monotherapy or with olaparib
TQB3217	Chia Tai Tianqing	USP1	NCT07291050	Not yet recruiting	I	Advanced solid tumors	Monotherapy
SIM0501	Jiangsu Simcere	USP1	NCT06331559	Terminated due to company development strategy change	I	Advanced solid tumors	Monotherapy or with olaparib
XL309 (ISM3091)	Exelixis	USP1	NCT05932862	Recruiting	I	Advanced solid tumors, including HGSOC among eligible populations	Monotherapy or with olaparib
RO7623066 (KSQ-4279)	Hoffmann-La Roche	USP1	NCT05240898	Completed	I	Advanced solid tumors	Monotherapy or with olaparib or carboplatin
HSK39775	Xizang Haisco	USP1	NCT06314373	Recruiting	I/II	Advanced solid tumors	Monotherapy

Fig. 3.

Translational strategies for USP-targeted therapy in ovarian cancer. Legend: Two main translational strategies are summarized. Strategy 1 involves selective inhibition of individual USPs to disrupt defined oncogenic dependencies, with development spanning preclinical studies in OC models and early-phase clinical evaluation in advanced solid tumors. Strategy 2 involves inhibition of 19S proteasome-associated deubiquitinases, particularly USP14 and UCHL5, to induce proteotoxic stress. The figure also highlights the importance of biomarker-guided translation and rational combination strategies in future development

Inhibitors targeting individual USPs

Preclinical development in OC models

Among individual USP-directed strategies, USP1 has the most extensive preclinical evidence in OC. Building on the established role of USP1 in DNA lesion processing, replication stress responses, and the regulation of PARP1 activity, preclinical studies have provided a strong translational rationale for targeting USP1 in OC [20, 34, 48, 49]. In OC models, combined treatment with the selective USP1 inhibitor SJB3-019 A (SJB) and niraparib not only enhanced PARP1 trapping at DNA lesions, but also prevented reversal of niraparib-induced effects after drug withdrawal, thereby maintaining PARP1 retention at DNA damage sites and sustaining replication stress and DNA damage signaling [34]. Similar results were observed with the USP1-specific inhibitor KSQ-4279 [34]. The translational relevance of USP1 inhibition was further supported in resistant disease models. In several isogenic platinum-resistant epithelial ovarian cancer (EOC) models that also displayed cross-resistance to PARP inhibitors, combined SJB and niraparib significantly reduced cell viability compared with niraparib alone, highlighting the therapeutic potential of the USP1/PARPi axis in both platinum-sensitive and platinum-resistant EOC irrespective of HR status [34]. In vivo, SJB alone or in combination with niraparib showed no overt toxicity and did not increase hematologic toxicity compared with niraparib monotherapy, supporting the feasibility of this combination strategy [34]. In resistant models, including ID8 p53-null/PTEN-null tumors and a BRCA1-mutant HGSOC PDX with acquired platinum/PARPi resistance, SJB/KSQ-niraparib combinations restored niraparib sensitivity, reduced ascites and tumor burden, and restricted metastatic dissemination [34]. In addition, USP1 inhibitors (SJB3-019 A or pimozide) were reported to enhance cisplatin sensitivity [20]. Additional preclinical findings further support clinical translation of USP1 inhibition. Simoneau et al. showed that the reversible USP1 inhibitor I-138 reduces DNA synthesis and induces S-phase DNA damage in I-138-sensitive cells [13]. Similarly, ML323 suppressed OC cell proliferation by blocking S-phase progression [49].

Other USP inhibitors remain at the preclinical stage but have also shown activity in OC models. Pharmacologic inhibition of USP13 by spautin-1 significantly decreases MCL1 abundance and can sensitize tumor cells to BCL-2 homology 3 (BH3) mimetics such as ABT-263, a selective antagonist of BCL-2 and Bcl-xL, providing a mechanistically grounded rationale for combining USP inhibition with apoptosis-targeting agents in MCL1-dependent settings [16]. The USP2 inhibitor ML364 suppresses OC cell growth and migration and induces apoptosis, accompanied by pro-ferroptotic effects through increasing ROS and PARP cleavage [50]. Similarly, the USP28 specific inhibitor AZ1 destabilizes SOX9 and increases the sensitivity of OC cells to olaparib, consistent with the role of USP28 in SOX9-dependent DDR [15]. In addition, USP9X-targeting DUB inhibitor WP1130 suppresses CSC features and improves chemosensitivity across patient-derived models [25].

For USP14 inhibition, however, published results are conflicting: IU1 was reported to sensitize A2780CP cells to cisplatin [18], whereas another study found that IU1 or USP14 knockdown reduced cisplatin cytotoxicity in A2780 cells, potentially through indirect modulation of Cx32 and gap junction intercellular communication [51]. These discrepancies indicate that the chemosensitizing effect may require specific biological conditions. Thus, before considering clinical translation, it is essential to elucidate the underlying mechanism and define the settings in which benefit can be expected.

Clinical-stage development

Based on the mature preclinical rationale described above, several USP1 inhibitors have entered early-phase evaluation in advanced solid tumors. KSQ-4279 (NCT05240898) and TNG348 (NCT06065059) have completed phase I clinical trials. KSQ-4279 reported an acceptable safety profile during a phase I dose escalation study, with anemia being the most common side effect, and demonstrated robust PK, PD, and clinical activity. However, TNG348 monotherapy was terminated due to grade 3/4 liver function abnormalities. This hepatotoxicity is believed to be specific to TNG348 rather than a class-wide adverse event, as similar liver function abnormalities were not observed in NCT05240898. Other USP1 programs currently in early-phase development include XL309 (NCT05932862) and HSK39775 (NCT06314373), whereas SIM0501 (NCT06331559) was terminated because of company development strategy change.

Bardoxolone methyl (CDDO-Me), a synthetic triterpenoid reported to inhibit USP7, also represents a clinical-stage individual USP-targeting strategy. In preclinical OC models, including drug-resistant settings, CDDO-Me showed anti-tumor activity [52]. In paclitaxel-resistant OVCAR8TR and cisplatin-resistant A2780cp70 cells, CDDO-Me reduced IL-6 secretion, suppressed Src, Jak2, and STAT3 phosphorylation, and promoted apoptosis by modulating STAT3-regulated genes [53]. Additional studies reported that CDDO-Me induced ROS generation and inhibited the expression of p-AKT, p-mTOR, NF-κBp65 and NF-κB-regulated antiapoptotic proteins [54, 55]. In a phase I clinical trial enrolling patients with advanced solid tumors and lymphoma, CDDO-Me was reported to be tolerated up to 900 mg/day, and objective tumor responses were observed [56]. NQO1 mRNA levels, indicative of Nrf2 activation, were increased in peripheral blood mononuclear cells, and NF-κB and cyclin D1 levels were decreased in tumor biopsy samples. These data show that CDDO-Me modulates antioxidant and anti-inflammatory targets [56].

Inhibitors targeting 19 S proteasome-associated DUBs

In addition to selective inhibition of individual USPs, inhibition of 19 S proteasome-associated DUBs, notably USP14 and UCHL5, offers a class-based strategy to induce proteotoxic stress. Platinum pyrithione (PtPT) was reported to selectively inhibit USP14/UCHL5, leading to accumulation of ubiquitinated proteins, activation of the unfolded protein response (UPR), and apoptosis and G2/M cell cycle arrest in OC cells [57, 58]. Similarly, RA-9, another inhibitor of 19 S proteasomal DUBs, induced UPR in OC and inhibited cell viability in vitro while prolonging survival in mouse models through G2/M arrest and caspase-3-mediated apoptosis [59].

Despite these promising preclinical data, this class has already encountered translational limitations. The 19 S DUB inhibitor VLX1570 showed anti-myeloma activity in a phase I trial (NCT02372240), but development was terminated due to pulmonary toxicity [60].

Biomarker-guided translational strategies

An important next step in USP-directed therapy is to move beyond broad target nomination toward biomarker-guided translational strategies, with USP1 currently providing the most mature example. Da Costa et al. showed that BRCA1-mutant HGSOC organoids were more sensitive to the USP1 inhibitor TNG6132 than BRCA1-wild-type counterparts [48]. This sensitivity correlated with single-stranded DNA (ssDNA) gap accumulation after USP1 inhibition, which is a direct consequence of dysregulated PCNA ubiquitination as described by Simoneau et al. [13]. In the same study, USP1 inhibition overcame PARP inhibitor resistance in a BRCA1-mutant xenograft model and was synergistic with PARP and POLQ inhibition in BRCA1-mutant cells, with all of these responses accompanied by ssDNA gap accumulation [48]. Consistently, in patient-derived OC organoids, sensitivity to USP1 inhibition alone or in combination correlated with ssDNA gap accumulation, supporting ssDNA gap assessment in organoid models as a rapid functional approach to anticipate response to USP1 inhibition in ongoing clinical studies [48]. Moreover, elevated USP1 mRNA expression was also associated with TNG6132 sensitivity, which may serve as a candidate predictive biomarker [48].

Other biomarker-linked combination strategies are also suggested by available preclinical data. Selective inhibition of USP8 by caffeic acid phenethyl ester (CAPE) enhanced cisplatin efficacy in TOV112D cells, the cisplatin-resistant IGROV-1/Pt1 variant, and PEO1 cells, all of which express relatively high USP8 levels, but not in PEO4 cells with reduced USP8 expression, suggesting a possible biomarker-linked therapeutic avenue [61]. In addition, USP5 amplification was shown to increase proapoptotic effect of HDACi PXD101, compared to cells without USP5 amplification, indicating that USP5 amplification may serve as a candidate biomarker for HDAC-targeted therapy selection [11]. By contrast, biomarker development for proteasome-associated DUB inhibition remains limited.

Conclusions and future directions

USPs are increasingly recognized as central regulators of OC biology. Through direct deubiquitination of oncogenic or tumor-suppressive substrates, as well as broader control of signaling networks, USPs influence multiple hallmarks of OC, including proliferation, genome maintenance, apoptosis, metastasis, metabolic adaptation, immune evasion, and cancer stemness. Collectively, the reported USP–substrate relationships in OC, as summarized in Table 1, provide a mechanistic framework linking dysregulated proteostasis to malignant phenotypes and therapeutic resistance. Despite the growing body of evidence, the role of USPs in OC warrants further investigation. Some family members are still supported mainly by correlative clinical observations rather than direct functional and mechanistic validation. For example, USP10 expression is significantly lower in OC and has been associated with poor prognosis, potentially due to CpG island hypermethylation [62]. USP19 has also been proposed as a candidate prognostic marker in HGSOC [63].

From a translational perspective, current evidence supports two main therapeutic directions. One focuses on selectively targeting individual USPs to disrupt defined oncogenic dependencies, whereas the other involves inhibition of 19 S proteasome-associated deubiquitinases, such as USP14 and UCHL5, to induce proteotoxic stress. Given the risk of off-target effects among DUB inhibitors, selectivity and target engagement should be stringently validated in OC models using chemical–genetic concordance and rescue experiments. Although no USP inhibitor has yet been developed in an OC-specific clinical program, early-phase trials in advanced solid tumors have already generated initial safety and PK/PD data, particularly for USP1 inhibitors. At the same time, emerging safety signals, including pulmonary toxicity with VLX1570 and hepatotoxicity observed during development of TNG348, highlight the need for careful attention to therapeutic window and toxicity management during future clinical translation.

To strengthen clinical translation, future studies should first define the OC subgroups most likely to benefit from USP inhibition, rather than assuming uniform activity across histotypes and treatment settings. This will require pairing each USP target with trial-feasible predictive biomarkers and PD readouts that directly reflect mechanism and target engagement, such as ssDNA gap accumulation for USP1 inhibitors. Combination strategies also warrant further investigation. Given the central role of several USPs in maintaining MCL1-dependent survival, including USP13, DUB3, and USP9X, BH3 profiling, a functional mitochondrial assay that measures apoptotic priming and anti-apoptotic protein dependence, may offer a useful translational tool. Functionally, it may serve as a predictive biomarker to identify patients with MCL1-dependent OC who are more likely to benefit from combinations of USP inhibitors and BH3 mimetics. In addition, it may provide a screening platform for identifying novel synergistic combinations with USP inhibitors, facilitating the rational design of precision combination regimens.

Emerging artificial intelligence (AI)- and machine learning (ML)-based approaches may facilitate the development of USP-targeted therapies in OC. Structure-based modeling and virtual screening could improve the discovery and optimization of selective USP inhibitors [64], while ML-assisted integration of multi-omics datasets may enable the identification of predictive biomarkers, including USP expression- or mutation-related signatures [65]. These advances may facilitate biomarker-guided patient stratification and support more precise clinical development of USP-directed therapies in OC.

Abbreviations

BH3

BCL-2 homology 3

CDDO-Me

Bardoxolone methyl

CSC

Cancer stem cell

DDR

DNA damage response

DSB

DNA double-strand break

DUB

Deubiquitinase

EMT

Epithelial–mesenchymal transition

EOC

Epithelial ovarian cancer

HDACis

Histone deacetylase inhibitors

HGSOC

High-grade serous ovarian cancer

HR

Homologous recombination

HRR

Homologous recombination repair

iUUCD

Integrated annotations for Ubiquitin and Ubiquitin-like Conjugation Database

ML

Machine learning

OC

Ovarian cancer

OCCC

Ovarian clear cell carcinoma

PARPi

Poly(ADP-ribose) polymerase inhibitor

PCNA

Proliferating cell nuclear antigen

PD

Pharmacodynamic

PDX

Patient-derived xenograft

PK

Pharmacokinetic

PTMs

Post-translational modifications

ssDNA

Single-stranded DNA

TCA

Tricarboxylic acid

UPR

Unfolded protein response

USP

Ubiquitin-specific protease

Author contributions

LD, JY were responsible for conceiving the ideas. LD, YZJ wrote the initial manuscript. YZJ and DNW contributed to editing. JY and LYW were responsible for revising the manuscript. All authors read and approved the final manuscript.

Funding

The publication of this article was funded by the CAMS Innovation Fund for Medical Sciences (CIFMS) (Grant No. 2025-I2M-KJ-004).

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

The manuscript does not report on or involve any animals, humans, human data, human tissue or plants. All authors of this manuscript consent to participate.

Consent for publication

All authors of this manuscript consent to publication.

Competing interests

The authors declare no competing interests.