Skip to main content
Download PDF

Potential drug targets for ovarian cancer identified through Mendelian randomization and colocalization analysis

Journal of Ovarian Research volume 18, Article number: 32 (2025) Cite this article

Abstract

Background

The existing drugs for ovarian cancer (OC) are unsatisfactory and thus new drug targets are urgently required. We conducted proteome-wide Mendelian randomization (MR) and colocalization analysis to pinpoint potential targets for OC.

Methods

Data on protein quantitative trait loci (pQTL) for 734 plasma proteins were obtained from large genome-proteome-wide association studies. Genetic associations with OC were derived from the Ovarian Cancer Association Consortium, which included 25,509 cases and 40,941 controls. MR analysis was performed to evaluate the association between the proteins and the OC risk. Colocalization analysis was conducted to check whether the identified proteins and OC shared causal variants. In addition, the phenome-wide MR analysis was performed to clarify protein associations across the phenotype, and drug target databases were examined for target validation.

Results

Genetically predicted circulating levels of 44 proteins were associated with OC risk at Benjamini-Hochberg correction. Genetically predicted 17 proteins had evidence of the increased risk of OC (CLEC11A, MFAP2, TYMP, PDIA3, IL1R1, SPINK1, PLAU, DKK2, IL6ST, DLK1, LRRC15, CDON, ANGPTL1, SEMA4D, AKR1A1, TNFAIP6, and FCGR2B); 27 proteins decreased the risk of OC(SIGLEC9, RARRES1, SPINT3, TMEM132A, HAVCR2, CNTN2, TGFBI, GSTA1, HGFAC, TREML2, GRAMD1C, ASAH2, CPNE1, CCL25, MAPKAPK2, POFUT1, PREP, NTNG1, CA10, CACNA2D3, CA8, MAN1C1, MRC2, IL10RB, RBP4, GP5 and CALCOCO2). Bayesian colocalization demonstrated that GRAMD1C, RBP4, PLAU, PDIA3, MFAP2, POFUT1, MAN1C1 and DKK2 shared the same variant with OC. The phe-MR analyses assessed the side effects of these 44 identified proteins, and the drug target database offered information on both approved and investigational indications.

Conclusion

This study provides proof of a causal relationship between genetically predicted 44 proteins associated with OC risk, which could serve as promising drug targets for OC.

Introduction

Ovarian cancer (OC) is a highly heterogeneous gynecologic malignancy and the seventh most common cancer in women. It has more pathological subtypes and mainly occurs in postmenopausal women after the age of 50 [1]. OC remains an enormous challenge in early diagnosis and medical treatment. 70% of OC cases are diagnosed as advanced when detected and usually accompanied by metastasis [2]. At present, the main treatment for OC is debulking surgery and platinum-based chemotherapy, but the effect is not satisfactory [3]. The methods of targeted therapy and immunotherapy also provide a new therapeutic approach for ovarian cancer. However, most patients experience relapse or progression after resistance, with a 5-year survival rate ranging between 30% and 40% [4]. Complex biological etiology of OC restricts development of new drugs. Hence, identifying new therapeutic targets is crucial for enhancing patient survival and prognosis.

The circulating proteins in human plasma can be directly secreted into circulation or overflow into the bloodstream from the organs of origin, and circulating proteins secreted or leaked into the bloodstream play a complex role in the biological processes involved in the development of various tumors, including colorectal and brain cancers, and are considered major targets for drug therapy [5,6,7,8,9]. In clinical practice, circulating proteins serve dual roles: they can function as biomarkers, such as N-terminal pro-brain natriuretic peptide in congestive heart failure [10], and as targets for drug therapy, such as proprotein convertase subtilisin/ kexin type 9 serine protease (PCSK9) used in the treating hypercholesterolemia [11]. Compared to invasive examinations, blood-derived protein tumour markers could present an alternative, non-invasive and cost-effective approach for improvement of cancer screening [12]. The study suggests that a protein drug target, supported by genetic evidence of its link to the disease, has double the likelihood of obtaining market approval [13]. Observational studies have identified a variety of circulating proteins with therapeutic potential related with the OC onset, progression and recovery [14]. CA125 and HE4 are the most widely applied biomarkers in blood tests for OC [15,16,17], increasing the sensitivity to 94.8% and specificity to 75% of the ROMA score (Ovarian Malignancy Risk Algorithm) in a cohort of patients with predominantly advanced OC [18]. However, observational studies are prone to confounding factors and reverse causality, and their associations need to be replicated and confirmed. Randomized controlled studies are unable to explore the relationship between thousands of proteins and OC.

Proteomic data can provide evidence for effective drug development for various diseases including OC. Recent studies increasingly indicate that proteomics play a significant role in predicting the success of drug trials [13]. Mendelian randomization (MR) analysis is now widely utilized in drug target development and drug repurposing. MR utilizes genetic variation as the instrumental variable to evaluate the causal relationship between measured protein levels and OC. MR can effectively avoid confounders due to the stable random assortment of genes from parents to offspring. Recently, MR analysis method has been applied to identify potential therapeutic targets for diseases like heart failure and inflammatory bowel disease [19]. Few MR studies have been reported on the integration of GWAS and protein quantitative trait loci (pQTL) data for OC.

In this study, we leverage recent pQTL data from genome-proteome-wide association studies to explore potential targets for OC [20]. We conducted MR and colocalization analyses to identify the causal influence of proteins. Additionally, we have performed phenome-wide MR (phe-MR) and retrieved their druggability to explore their potential as therapeutic targets for OC.

Materials and methods

Exposure data

The summary-level statistics of genetic associations of plasma protein pQTL data were sourced from the study conducted by Zheng et al. [20], which involve the 5 studies previously published GWAS [21,22,23,24,25]. Only pQTLs that satisfy the criteria are considered: (a) was available at the genome-wide significant level (P < 5E-08); (b) were cis-pQTLs; (c) showed independent association [linkage disequili brium (LD) clumping r2 < 0.001]. Cis-pQTLs are located near the gene encoding the target protein and are generally considered more reliable proxies due to their stronger biological evidence for direct and independent effects on proteins. Trans-pQTLs are more likely to exhibit pleiotropy because their indirect effects on proteins can potentially violate MR assumptions and introduce bias into the results [26]. Finally, a total of 738 cis-acting SNPs were included for 734 proteins. Further details on the GWAS are available in the original publication.

Outcome data

Summary-level data for the OC were available in Ovarian Cancer Association Consortium (OCAC) [27], which was involved 66,450 participants including 25,509 OC cases and 40,941 controls. All individuals of OCAC were of European ancestryand recruited from 14 countries. All sunmmary data were downloaded from the IEU OpenGWAS project.

Statistical analysis

Mendelian randomization analysis

In the main analysis, two-sample MR analysis was employed using the R package “TwoSampleMR” to investigate the relationship between circulating proteins and OC risk. Harmonization of data on exposure and outcome was performed to align the effects of SNPs on both, ensuring they correspond to the same allele. For proteins with a single cis-pQTL, MR estimates were calculated using the Wald ratio. For those with two or more genetic instruments, the inverse-variance weighted (IVW) method was applied, accounting for the weak LD among the instruments by using an LD matrix [28, 29]. For the primary analysis, the Benjamini-Hochberg correction was applied to reduce the false discovery rate (the P value of Benjamini-Hochberg correction was < 0.05). The MR findings were expressed as odds ratio (OR) and 95% confidence interval (95% CI) for the risk of OC. We performed the heterogeneity and pleiotropy analysis to assess the robustness of the results. In addition, we used linkage imbalance score regression analysis (LDSC) to estimate the co-genetic structure between plasma proteins and OC.

Colocalization analysis

To further investigate the causality of the observed MR associations, we carried out a colocalization analysis with the R package “coloc” to evaluate the likelihood that plasma proteins and OC share the same causal variant, and to remove the confounding caused by LD [30, 31]. SNPS with significant genome-wide associations (P < 5E-08) in proteins and OC were selected. Duplicates and missing values were removed and SNPs with both exposure and outcome in overlap were extracted. For each locus, five hypotheses were considered about the shared presence of a single variant across two traits in the Bayesian co-localization analysis. Posterior probabilities were then calculated for each of these hypotheses: PPH0, no association with either trait; PPH1, associated with protein but not the OC trait; PPH2, linked to the OC trait but not protein; PPH3, related to both the OC trait and protein with different causal variants; PPH4, connected to both the OC trait and gene expression, sharing a causal variant [32]. The analysis was performed within a ± 100 kb window around the drug target gene. The shared genetic variant implicates that the protein itself contributes directly to the disease risk, rather than being altered by other biological processes. We defined genes as evidence that of co-localization based on the PPH4 > 50% [31].

Phenome-wide MR analysis

The Phe-MR analysis was used to discover significant associations between the identified pQTLs with other reported traits. Each of the cis-instruments from the IEU Open GWAS project (https://gwas.mrcieu.ac.uk/) was selected and the association of variants with traits were retrieved for P < 0.001. The causal impact of each protein on the identified traits was explored using the same MR method.

Drug-target validation

To evaluate the potential for drug targeting of the identified proteins, we reviewed several drug-target databases including Drugbank, Therapeuutic Target Database, Clinical trial, along with previously established lists of druggable genes. We also searched for drugs currently being developed against identified potential proteins. To evaluate the potential for drug formability, we categorized these proteins into four groups: (a) approved (one or more drugs targeting a specific protein have been approved); (b) in clinical trials (the target drug is currently undergoing clinical trial); (c) preclinical (the target drug is in the preclinical research phase); (d) druggable (the protein is not be found in drug databases, but listed as a medicinal target).

Results

Screening the proteome for OC causal proteins

The study design was illustrated in Fig. 1. The research investigated the MR association between 734 proteins, each with available index pQTLs signals, and the risk of OC outcomes. MR analysis revealed 44 protein-OC causal relationships, meeting the Benjamini-Hochberg corrected threshold (PB−H adjusted < 0.05) (Supplementary Table 1). Among the 44 proteins, 17 proteins are linked to a heightened risk of OC. The proteins with the most significant effects, in descending order, included CLEC11A, MFAP2, TYMP, PDIA3, IL1R1, SPINK1, PLAU, DKK2, IL6ST, DLK1, LRRC15, CDON, ANGPTL1, SEMA4D, AKR1A1, TNFAIP6, and FCGR2B. The other 27 risk-reducing proteins included SIGLEC9, RARRES1, SPINT3, TMEM132A, HAVCR2, CNTN2, TGFBI, GSTA1, HGFAC, TREML2, GRAMD1C, ASAH2, CPNE1, CCL25, MAPKAPK2, POFUT1, PREP, NTNG1, CA10, CACNA2D3, CA8, MAN1C1, MRC2, IL10RB, RBP4, GP5 and CALCOCO2 (Figs. 2 and 3). No heterogeneity and pleiotropy were detected in the analyzed proteins (Supplementary Table 2). No statistically significant genetic association was found on LDSC results (Supplementary Table 3).

Fig. 1
figure 1

Study design. OC: ovarian cancer; SNP: single nucleotide polymorphism

Full size image
Fig. 2
figure 2

Volcano plot of the MR results for plasma proteins and OC risk

Full size image
Fig. 3
figure 3

Association between 44 identified drug targets and risk for ovarian cancer. Association between 44 identified drug targets and risk for ovarian cancer. Forest plot showing 44 proteins with strong evidence of causality in MR analysis. Odds ratios per standard deviation (s.d.) of the protein and 95% confidence intervals (CI) are shown. CALCOCO2: calcium binding and coiled-coil domain 2; GP5: glycoprotein V platelet; RBP4: retinol binding protein 4; IL10RB: interleukin 10 receptor subunit beta; MRC2: mannose receptor C type 2; MAN1C1: mannosidase alpha class 1 C member 1; CA8: carbonic anhydrase 8; CACNA2D3: calcium voltage-gated channel auxiliary subunit alpha2delta 3; CA10: carbonic anhydrase 10; NTNG1:netrin G1; PREP: prolyl endopeptidase; MAPKAPK2: MAPK activated protein kinase 2; POFUT1: protein O-fucosyltransferase 1; CCL25: C-C motif chemokine ligand 25; CPNE1: copine 1; GSTA1: glutathione S-transferase alpha 1; CNTN2: contactin 2; ASAH2;ASAH2B: N-acylsphingosine amidohydrolase 2; HGFAC: HGF activator; TGFBI: transforming growth factor beta induced; TREML2: triggering receptor expressed on myeloid cells like 2; GRAMD1C: GRAM domain containing 1 C; TMEM132A: transmembrane protein 132 A; HAVCR2: hepatitis A virus cellular receptor 2; SPINT3: serine peptidase inhibitor, Kunitz type 3; RARRES1: retinoic acid receptor responder 1; SIGLEC9: sialic acid binding Ig like lectin 9;FCGR2B: Fc gamma receptor IIb; AKR1A1: aldo-keto reductase family 1 member A1; TNFAIP6: TNF alpha induced protein 6; SEMA4D: semaphorin 4D; ANGPTL1:angiopoietin like 1; CDON: cell adhesion associated, oncogene regulated; LRRC15: leucine rich repeat containing 15; DLK1: delta like non-canonical Notch ligand 1; IL6ST: interleukin 6 cytokine family signal transducer; DKK2: dickkopf WNT signaling pathway inhibitor 2; PLAU: plasminogen activator, urokinase; SPINK1: serine peptidase inhibitor Kazal type 1; IL1R1: interleukin 1 receptor type 1; PDIA3: protein disulfide isomerase family A member 3; TYMP: thymidine phosphorylase; MFAP2: microfibril associated protein 2; CLEC11A: C-type lectin domain containing 11 A

Full size image

Colocalization analysis of cis-pQTLs

MR analysis revealed a causal association between 44 proteins and OC. Steiger filtering further ensures directionality. Colocalization analyses were performed to distinguish between causal relationships from linkage disequilibrium. The PPH4 for 10 of the 44 cis-pQTLs with MR evidence exceeded 0.5 (SPINK1, IL1R1, CCL25, MAPKAPK2, TYMP, IL6ST, CDON, AKR1A1, PREP and RARRES1), in 4 proteins were > 0.75(GP5, CA10, CLEC11A and CACNA2D3), and in 8 proteins were > 0.95(GRAMD1C, RBP4, PLAU, PDIA3, MFAP2, POFUT1, MAN1C1 and DKK2) (Fig. 4; Supplementary Table 4).

Fig. 4
figure 4

Proteins with MR and colocalization evidence

Full size image

Drug-target validation and repurposing

Phenome-wide MR

To delve deeper into the comprehensive indications and side effects of the 44 proteins, phe-MR studies were conducted, revealing important safety and efficacy findings that could provide the treatment of OC (Supplementary Table 5). A significant number of identified proteins influenced risk factors for OC, such as anthropometric traits (weight, height, body mass index, hip or waist circumference, and fat mass), metabolic traits (triglycerides, cholesterol, urate, and calcium), and female reproductive traits (menstrual cycle, sex hormones). Beyond the aforementioned well-known factors, certain proteins also impact a variety of other traits, including basal metabolic rate, breast cancer, birth weight, and blood cell count. Notably, both breast cancer and basal metabolic rate have been linked to the risk of OC [33,34,35].

Additionally, phe-MR has identified several proteins as promising drug targets for different indications that align with the same therapeutic direction as OC. Proteins implicated in OC also contribute to elevated risks for various disorders affecting both the female reproductive and non-reproductive systems. Genetically determined MAPKAPK2, MFAP2, and PLAU increased the risk of rheumatoid arthritis in non-female reproductive and breast diseases. Genetic determined CALCOCO2 and DLK1 increased the Type 2 diabetes risk. As for cancer, CPNE1 and TGFBI increased the risk of breast cancer, while MAN1C1 and PDIA3 reduced it. GRAMD1C reduces the risk of pharyngeal cancer, CA8 decreased the skin cancer risk, and CALCOCO2 increases the risk of carcinoma in situ of cervix uteri. When considering proteins as targets for OC, it is crucial to also evaluate their safety. For example, CA8 reduced OC but increased the risk of varicose veins. MAPKAPK2 decreased the risk of OC but increased inflammatory bowel disease and ulcerative colitis risk. CPNE1 increased the risk of asthma. These adverse side effects should be taken into consideration when evaluating its preventive effect on OC.

Druggability of identified proteins

44 proteins identified as possible drug targets in MR analysis were searched in the drug databases to uncover records of past or ongoing clinical drug development programs associated with these proteins. Among these 44 proteins, 9 (20.45%) proteins were identified as targets under investigation in clinical trials. These 9 proteins include SEMA4D, HAVCR2, CACNA2D3, MAPKAPK2, PREP, FCGR2B, TYMP, PLAU and DLK1. Their indications include huntington disease (SEMA4D), alzheimer disease (SEMA4D), myelodysplastic syndrome (HAVCR2), Generalized anxiety disorder (CACNA2D3), solid tumor/cancer (SEMA4D, HAVCR2), mesothelioma (MAPKAPK2), airway inflammation (MAPKAPK2), coeliac disease (PREP), influenza virus infection (PREP), Autoimmune diabetes (FCGR2B), Chronic lymphocytic leukaemia (FCGR2B), inborn errors of metabolism (TYMP), ovarian cancer (PLAU), and Amyotrophic lateral sclerosis (DLK1).

There are 9 proteins that are already approved drug targets. Amlodipine and nilvadipine, both inhibitors of CACNA2D3, are used to treat hypertension and angina; Tretinoin, an agonist of RARRES1, is employed for the treatment of acne vulgaris and promyelocytic leukemia; Anakinra, an antagonist of IL1R1, is used to treat the rheumatoid arthritis and coronavirus disease 2019 (COVID-19); Ritlecitinib, substrate of GSTA1, is utilized for the treatment of alopecia areata.

None of them were found to be drug targets for OC. No information was available for SIGLEC9, CLEC11A, DKK2, MRC2, ASAH2; ASAH2B, MAN1C1, POFUT1, LRRC15, GRAMD1C, TREML2, NTNG1, CDON, TMEM132A, caloco2, SPINT3, GP5, PDIA3, AKR1A1, RBP4, SPINK1, CCL25, HGFAC, TGFBI, IL6ST, CNTN2, and ANGPTL1 in searched databases. Although only one protein for OC target was identified in the databases, they may provide new and promising targets for OC (Table 1).

Table 1 Potential repurposing opportunities of approved drugs and novel drugs under development
Full size table

Discussion

To the best of our knowledge, this study represents the inaugural attempt to leverage plasma proteomic data for identifying the causal proteins associated with OC through the utilization of two-sample MR and Bayesian colocalization methodologies. In this study, 738 cis-acting SNPs across 734 plasma proteins were analyzed for causal associations with OC through two-sample MR. Ultimately, we pinpointed 44 proteins as potential drug targets for OC. Among them, 27 proteins were negatively correlated with the risk of OC, while 17 of which increased the risk. To constrain the bias of pleiotropy, we exclusively employed cis-pQTLs as instruments, given their direct involvement in the transcription and/or translation of relevant genes. The results did not reveal heterogeneity or pleiotropy. We also calculated the LDSC as the complementary analysis to explore the genetic relationship between plasma protein and OC, although the LDSC results were not statistically significant. LDSC evaluates the shared genetic architecture between traits, identifying genetic correlations that provide insights into potential biological pathways. However, it does not infer causal relationships [36]. On the other hand, MR complements LDSC by using genetic variants as instrumental variables to infer causal effects, thereby providing directional evidence [37, 38]. Our findings contributed to the determination of circulating protein biomarkers that hold promise for early-stage detection and diagnosis of OC in clinical settings. These associations encompassed previously reported incident OC connections, such as CNTN2 and HAVCR2 [39,40,41], proteins like SEMA4D, which have documented links to prognosis, invasion, and metastasis [42, 43], and proteins such as DLK1 promoting tumorigenesis and epithelial-mesenchymal transition [44].

MR analysis plays an important role in drug target development. Multi-omics research has demonstrated the value of pQTLs in reusing existing targets for other indications and prioritizing new drug targets [20]. Compared to trans-acting pQTLs that may act through indirect mechanisms, cis-acting pQTLs are generally accorded greater biological significance and commonly utilized in the screening of drug targets [20, 45]. MR employs genetic variation as an instrumental variable for probing the causal impacts of exposure on outcomes, and the target indication that links target genes to related phenotypes is a potential cost-effective method for prioritizing the development of drug targets. Compared with observational studies, MR circumvents the impact of confounding variables. It has streamlined the discovery of prospective therapeutic targets across various conditions, including atrial fibrillation and Alzheimer’s disease [31, 46].

Several studies have reported HAVCR2 as a biomarker for OC [47]. HAVCR2, best known as TIM3, is described co-inhibitory molecule and expresses in cellular membranes. HAVCR2, through its interaction with galectin-9, facilitates T cell apoptosis and consequently promotes immunosuppression. Additionally, by recognizing nucleic acids released from apoptotic tumor cells via Toll-like receptors (TLRs), it further diminishes chemotherapeutic efficacy [48, 49]. PLAU, also called urokinase-type plasminogen activator (uPA, Urokinase), is a serine protease involved in the regulation of numerous cellular signaling pathways and the induction of diverse responses. There is now clear evidence that PLAU is expressed in primary and metastatic OC. Its overexpression correlates with the invasive metastatic potential of ovarian cancer and predicts a worse prognosis [50,51,52].

In this study, we found that GRAMD1C, RBP4, PLAU, PDIA3, MFAP2, POFUT1, MAN1C1 and DKK2 exhibited the most compelling evidence of MR and colocalization (PPH4 > 0.95). PLAU, PDIA3, MFAP2, and DKK2 were found to increase the risk of OC. The protein disulphide isomerase (PDI) family comprises multifunctional endoplasmic reticulum (ER) enzymes that are often elevated in multiple cancer types [53]. The oncogenic effects of PDI are facilitated through the UPR signaling pathway and the regulation of apoptosis [54, 55], which may be responsible for the development and progression of OC. MFAP2, the inaugural member of the MFAP family subfamily, manifests in both membrane and soluble forms. In OC, MFAP2 modulates the FOXM1/β-catenin signaling axis to enhance cell proliferation and glycolysis [56]. DKK2, a secreted proteins, acts as a Wnt signaling antagonist by binding to LDL receptor-related protein 5/6 (LRP5/6), thereby inhibiting its interaction with the Wnt-Frizzled complex [57].

In a subsequent analysis, the 44 OC drug targets underwent evaluation in databases, with exploration of their associated side effects. For instance, TNFAIP6, derived from TNF, has been linked to elevated susceptibility to various cancers, such as OC, as well as several autoimmune diseases. TNFAIP6 can promote cell migration and have anti-inflammatory property by binding to the chemokine CXCL8 [58]. TNFAIP6 has already a target for pain, inflammation, and Alzheimer’s Disease [59, 60]. FCGR2B had the function of inhibiting the overactivation of immune cells and associated with a variety of autoimmune diseases [61]. MGD010, a developed drug targeting autoimmune diseases, is currently undergoing phase II clinical trials to assess its efficacy. Our pQTL-MR study identified that PLAU (rs2227551) significantly increased the risk of OC, which consistent with previous study reports. The clinical trial of PLAU has been initiated for OC treatment in the United States (NCT00939809), which is a multicenter phase II study to access the efficacy and tolerability of urokinase-derived peptide (A6) in managing persistent or recurrent epithelial ovarian, fallopian tube, or primary peritoneal cancer. This study showed that A6 exhibited good tolerability but showed limited efficacy in patients with persistent or recurrent OC [62]. Whether PLAU merits consideration as a drug target for OC warrants investigation in forthcoming animal studies and clinical trials.

There were several strengths in this analysis. Firstly, Using MR analysis to assess the correlation between 734 proteins and OC reduced the biases caused by confounders and reverse causality. Colocalization analysis explored how specific loci exert pleiotropic effects across multiple traits. The innovative combination of proteomic data with MR analysis for OC presents a pioneering approach to pinpoint potential therapeutic targets. This advancement enhances our comprehension of ovarian cancer biology and propels future research and clinical applications. Secondly, large samples of GWAS have sufficient statistical validity to present a relationship between proteins and ovarian cancer. Thirdly, as additional analyses, the phe-MR and drug target databases improved the completeness and reliability of our findings. Fourthly, our analysis was restricted to Europeans, which minimized population stratification bias. However, there are still limitations. First, only European populations were studied, limiting the generalization of this study to other populations. Second, plasma proteins were selected instead of direct tissue samples for numerous human disorders. Interventions involving plasma proteins may not directly affect particular tissues. Thirdly, MR analysis was not a complete substitute for clinical trials. Our analysis suggested a biological relationship between the proteins and OC, offering evidence solely for the initial stage of drug development. Fourthly, the drug targets assessed in this study represented by a limited range of instruments, implying that the inferred causal effects hinge on a small set of genetic instruments. Thus, these associations indicate causality and do not confirm it definitively. Additionally, in MR analysis addressing LD is achieved by employing the generalized IVW method to gauge the effect sizes, potentially enhancing the power of MR results.

Conclusions

In summary, this study identified 44 proteins that may be attractive drug targets for OC via MR and colocalization analysis using population-based proteomic data. Additional research is warranted to confirm our discoveries and delve into the functions of these protein candidates in OC.

Data availability

No datasets were generated or analysed during the current study.

Abbreviations

IVs:

Instrumental variables

IVW:

Inverse variance weighted

LD:

Linkage disequilibrium

LDSC:

Linkage imbalance score regression analysis

MR:

Mendelian randomization

OC:

Ovarian Cancer

OCAC:

Ovarian Cancer Association Consortium

OR:

Odds ratio

SNP:

Single nucleotide polymorphism

References

  1. Sung H, Ferlay J, Siegel RL, Laversanne M, Soerjomataram I, Jemal A, Bray F. Global Cancer statistics 2020: GLOBOCAN estimates of incidence and Mortality Worldwide for 36 cancers in 185 countries. CA Cancer J Clin. 2021;71(3):209–49.

    Article  PubMed  Google Scholar 

  2. Jayson GC, Kohn EC, Kitchener HC, Ledermann JA. Ovarian cancer. Lancet. 2014;384(9951):1376–88.

    Article  PubMed  Google Scholar 

  3. Lheureux S, Braunstein M, Oza AM. Epithelial ovarian cancer: evolution of management in the era of precision medicine. CA Cancer J Clin. 2019;69(4):280–304.

    Article  PubMed  Google Scholar 

  4. Nag S, Aggarwal S, Rauthan A, Warrier N. Maintenance therapy for newly diagnosed epithelial ovarian cancer- a review. J Ovarian Res. 2022;15(1):88.

    Article  PubMed  PubMed Central  Google Scholar 

  5. Finan C, Gaulton A, Kruger FA, Lumbers RT, Shah T, Engmann J, Galver L, Kelley R, Karlsson A, Santos R et al. The druggable genome and support for target identification and validation in drug development. Sci Transl Med 2017, 9(383).

  6. Jelski W, Mroczko B. Molecular and circulating biomarkers of brain tumors. Int J Mol Sci 2021, 22(13).

  7. Liang Y, Li J, Li Q, Tang L, Chen L, Mao Y, He Q, Yang X, Lei Y, Hong X, et al. Plasma protein-based signature predicts distant metastasis and induction chemotherapy benefit in nasopharyngeal carcinoma. Theranostics. 2020;10(21):9767–78.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  8. Urata S, Iida T, Yamamoto M, Mizushima Y, Fujimoto C, Matsumoto Y, Yamasoba T, Okabe S. Cellular cartography of the organ of Corti based on optical tissue clearing and machine learning. Elife 2019, 8.

  9. Enroth S, Berggrund M, Lycke M, Broberg J, Lundberg M, Assarsson E, Olovsson M, Stalberg K, Sundfeldt K, Gyllensten U. High throughput proteomics identifies a high-accuracy 11 plasma protein biomarker signature for ovarian cancer. Commun Biol. 2019;2:221.

    Article  PubMed  PubMed Central  Google Scholar 

  10. Iwanaga Y, Nishi I, Furuichi S, Noguchi T, Sase K, Kihara Y, Goto Y, Nonogi H. B-type natriuretic peptide strongly reflects diastolic wall stress in patients with chronic heart failure: comparison between systolic and diastolic heart failure. J Am Coll Cardiol. 2006;47(4):742–8.

    Article  CAS  PubMed  Google Scholar 

  11. Rosenson RS, Hegele RA, Fazio S, Cannon CP. The evolving future of PCSK9 inhibitors. J Am Coll Cardiol. 2018;72(3):314–29.

    Article  CAS  PubMed  Google Scholar 

  12. Loke SY, Lee ASG. The future of blood-based biomarkers for the early detection of breast cancer. Eur J Cancer. 2018;92:54–68.

    Article  CAS  PubMed  Google Scholar 

  13. Nelson MR, Tipney H, Painter JL, Shen J, Nicoletti P, Shen Y, Floratos A, Sham PC, Li MJ, Wang J, et al. The support of human genetic evidence for approved drug indications. Nat Genet. 2015;47(8):856–60.

    Article  CAS  PubMed  Google Scholar 

  14. Yarmolinsky J, Bull CJ, Vincent EE, Robinson J, Walther A, Smith GD, Lewis SJ, Relton CL, Martin RM. Association between genetically proxied inhibition of HMG-CoA reductase and epithelial ovarian Cancer. JAMA. 2020;323(7):646–55.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  15. Dochez V, Caillon H, Vaucel E, Dimet J, Winer N, Ducarme G. Biomarkers and algorithms for diagnosis of ovarian cancer: CA125, HE4, RMI and ROMA, a review. J Ovarian Res. 2019;12(1):28.

    Article  PubMed  PubMed Central  Google Scholar 

  16. Felder M, Kapur A, Gonzalez-Bosquet J, Horibata S, Heintz J, Albrecht R, Fass L, Kaur J, Hu K, Shojaei H, et al. MUC16 (CA125): tumor biomarker to cancer therapy, a work in progress. Mol Cancer. 2014;13:129.

    Article  PubMed  PubMed Central  Google Scholar 

  17. Zhang M, Cheng S, Jin Y, Zhao Y, Wang Y. Roles of CA125 in diagnosis, prediction, and oncogenesis of ovarian cancer. Biochim Biophys Acta Rev Cancer. 2021;1875(2):188503.

    Article  CAS  PubMed  Google Scholar 

  18. Karlsen MA, Sandhu N, Hogdall C, Christensen IJ, Nedergaard L, Lundvall L, Engelholm SA, Pedersen AT, Hartwell D, Lydolph M, et al. Evaluation of HE4, CA125, risk of ovarian malignancy algorithm (ROMA) and risk of malignancy index (RMI) as diagnostic tools of epithelial ovarian cancer in patients with a pelvic mass. Gynecol Oncol. 2012;127(2):379–83.

    Article  PubMed  Google Scholar 

  19. Henry A, Gordillo-Maranon M, Finan C, Schmidt AF, Ferreira JP, Karra R, Sundstrom J, Lind L, Arnlov J, Zannad F, et al. Therapeutic targets for heart failure identified using proteomics and mendelian randomization. Circulation. 2022;145(16):1205–17.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  20. Zheng J, Haberland V, Baird D, Walker V, Haycock PC, Hurle MR, Gutteridge A, Erola P, Liu Y, Luo S, et al. Phenome-wide mendelian randomization mapping the influence of the plasma proteome on complex diseases. Nat Genet. 2020;52(10):1122–31.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  21. Emilsson V, Ilkov M, Lamb JR, Finkel N, Gudmundsson EF, Pitts R, Hoover H, Gudmundsdottir V, Horman SR, Aspelund T, et al. Co-regulatory networks of human serum proteins link genetics to disease. Science. 2018;361(6404):769–73.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  22. Folkersen L, Fauman E, Sabater-Lleal M, Strawbridge RJ, Franberg M, Sennblad B, Baldassarre D, Veglia F, Humphries SE, Rauramaa R, et al. Mapping of 79 loci for 83 plasma protein biomarkers in cardiovascular disease. PLoS Genet. 2017;13(4):e1006706.

    Article  PubMed  PubMed Central  Google Scholar 

  23. Suhre K, Arnold M, Bhagwat AM, Cotton RJ, Engelke R, Raffler J, Sarwath H, Thareja G, Wahl A, DeLisle RK, et al. Connecting genetic risk to disease end points through the human blood plasma proteome. Nat Commun. 2017;8:14357.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  24. Sun BB, Maranville JC, Peters JE, Stacey D, Staley JR, Blackshaw J, Burgess S, Jiang T, Paige E, Surendran P, et al. Genomic atlas of the human plasma proteome. Nature. 2018;558(7708):73–9.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  25. Yao C, Chen G, Song C, Keefe J, Mendelson M, Huan T, Sun BB, Laser A, Maranville JC, Wu H, et al. Genome-wide mapping of plasma protein QTLs identifies putatively causal genes and pathways for cardiovascular disease. Nat Commun. 2018;9(1):3268.

    Article  PubMed  PubMed Central  Google Scholar 

  26. Swerdlow DI, Kuchenbaecker KB, Shah S, Sofat R, Holmes MV, White J, Mindell JS, Kivimaki M, Brunner EJ, Whittaker JC, et al. Selecting instruments for mendelian randomization in the wake of genome-wide association studies. Int J Epidemiol. 2016;45(5):1600–16.

    Article  PubMed  PubMed Central  Google Scholar 

  27. Phelan CM, Kuchenbaecker KB, Tyrer JP, Kar SP, Lawrenson K, Winham SJ, Dennis J, Pirie A, Riggan MJ, Chornokur G, et al. Identification of 12 new susceptibility loci for different histotypes of epithelial ovarian cancer. Nat Genet. 2017;49(5):680–91.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  28. Deng YT, Ou YN, Wu BS, Yang YX, Jiang Y, Huang YY, Liu Y, Tan L, Dong Q, Suckling J, et al. Identifying causal genes for depression via integration of the proteome and transcriptome from brain and blood. Mol Psychiatry. 2022;27(6):2849–57.

    Article  CAS  PubMed  Google Scholar 

  29. Burgess S, Zuber V, Valdes-Marquez E, Sun BB, Hopewell JC. Mendelian randomization with fine-mapped genetic data: choosing from large numbers of correlated instrumental variables. Genet Epidemiol. 2017;41(8):714–25.

    Article  PubMed  PubMed Central  Google Scholar 

  30. Giambartolomei C, Vukcevic D, Schadt EE, Franke L, Hingorani AD, Wallace C, Plagnol V. Bayesian test for colocalisation between pairs of genetic association studies using summary statistics. PLoS Genet. 2014;10(5):e1004383.

    Article  PubMed  PubMed Central  Google Scholar 

  31. Ning Z, Huang Y, Lu H, Zhou Y, Tu T, Ouyang F, Liu Y, Liu Q. Novel drug targets for Atrial Fibrillation identified through mendelian randomization analysis of the blood proteome. Cardiovasc Drugs Ther 2023.

  32. Foley CN, Staley JR, Breen PG, Sun BB, Kirk PDW, Burgess S, Howson JMM. A fast and efficient colocalization algorithm for identifying shared genetic risk factors across multiple traits. Nat Commun. 2021;12(1):764.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  33. Zhang H, Qiu J, Meng F, Shu X. Insight into the causality between basal metabolic rate and endometrial and ovarian cancers: analysis utilizing systematic mendelian randomization and genetic association data from over 331,000 UK biobank participants. Eur J Clin Invest. 2023;53(6):e13971.

    Article  PubMed  Google Scholar 

  34. Bergfeldt K, Rydh B, Granath F, Gronberg H, Thalib L, Adami HO, Hall P. Risk of ovarian cancer in breast-cancer patients with a family history of breast or ovarian cancer: a population-based cohort study. Lancet. 2002;360(9337):891–4.

    Article  PubMed  Google Scholar 

  35. Muggia F, Tommasi S, Lynch H, Paradiso A. Hereditary breast and ovarian cancer: lessening the burden. Ann Oncol. 2013;24(Suppl 8):viii5–6.

    Article  PubMed  Google Scholar 

  36. Bulik-Sullivan BK, Loh PR, Finucane HK, Ripke S, Yang J, Schizophrenia Working Group of the Psychiatric, Genomics C, Patterson N, Daly MJ, Price AL, Neale BM. LD score regression distinguishes confounding from polygenicity in genome-wide association studies. Nat Genet. 2015;47(3):291–5.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  37. Evans DM, Davey Smith G. Mendelian randomization: New Applications in the coming age of hypothesis-free causality. Annu Rev Genomics Hum Genet. 2015;16:327–50.

    Article  CAS  PubMed  Google Scholar 

  38. Zhu Z, Zheng Z, Zhang F, Wu Y, Trzaskowski M, Maier R, Robinson MR, McGrath JJ, Visscher PM, Wray NR, et al. Causal associations between risk factors and common diseases inferred from GWAS summary data. Nat Commun. 2018;9(1):224.

    Article  PubMed  PubMed Central  Google Scholar 

  39. Adair SJ, Carr TM, Fink MJ, Slingluff CL Jr., Hogan KT. The TAG family of cancer/testis antigens is widely expressed in a variety of malignancies and gives rise to HLA-A2-restricted epitopes. J Immunother. 2008;31(1):7–17.

    Article  CAS  PubMed  Google Scholar 

  40. Wu JL, Zhao J, Zhang HB, Zuo WW, Li Y, Kang S. Genetic variants and expression of the TIM-3 gene are associated with clinical prognosis in patients with epithelial ovarian cancer. Gynecol Oncol. 2020;159(1):270–6.

    Article  CAS  PubMed  Google Scholar 

  41. Fucikova J, Rakova J, Hensler M, Kasikova L, Belicova L, Hladikova K, Truxova I, Skapa P, Laco J, Pecen L, et al. TIM-3 dictates functional orientation of the Immune Infiltrate in Ovarian Cancer. Clin Cancer Res. 2019;25(15):4820–31.

    Article  CAS  PubMed  Google Scholar 

  42. Chen Y, Zhang L, Pan Y, Ren X, Hao Q. Over-expression of semaphorin4D, hypoxia-inducible factor-1alpha and vascular endothelial growth factor is related to poor prognosis in ovarian epithelial cancer. Int J Mol Sci. 2012;13(10):13264–74.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  43. Chen Y, Zhang L, Liu WX, Wang K. VEGF and SEMA4D have synergistic effects on the promotion of angiogenesis in epithelial ovarian cancer. Cell Mol Biol Lett. 2018;23:2.

    Article  PubMed  PubMed Central  Google Scholar 

  44. Huang CC, Cheng SH, Wu CH, Li WY, Wang JS, Kung ML, Chu TH, Huang ST, Feng CT, Huang SC, et al. Delta-like 1 homologue promotes tumorigenesis and epithelial-mesenchymal transition of ovarian high-grade serous carcinoma through activation of notch signaling. Oncogene. 2019;38(17):3201–15.

    Article  CAS  PubMed  Google Scholar 

  45. Millwood IY, Bennett DA, Holmes MV, Boxall R, Guo Y, Bian Z, Yang L, Sansome S, Chen Y, Du H, et al. Association of CETP Gene variants with Risk for Vascular and Nonvascular diseases among Chinese adults. JAMA Cardiol. 2018;3(1):34–43.

    Article  PubMed  Google Scholar 

  46. Wingo AP, Liu Y, Gerasimov ES, Gockley J, Logsdon BA, Duong DM, Dammer EB, Robins C, Beach TG, Reiman EM, et al. Integrating human brain proteomes with genome-wide association data implicates new proteins in Alzheimer’s disease pathogenesis. Nat Genet. 2021;53(2):143–6.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  47. Kozlowski M, Borzyszkowska D, Cymbaluk-Ploska A. The role of TIM-3 and LAG-3 in the Microenvironment and Immunotherapy of Ovarian Cancer. Biomedicines 2022, 10(11).

  48. Patel J, Bozeman EN, Selvaraj P. Taming dendritic cells with TIM-3: another immunosuppressive strategy used by tumors. Immunotherapy. 2012;4(12):1795–8.

    Article  CAS  PubMed  Google Scholar 

  49. Anderson AC. Tim-3, a negative regulator of anti-tumor immunity. Curr Opin Immunol. 2012;24(2):213–6.

    Article  CAS  PubMed  Google Scholar 

  50. Wang L, Madigan MC, Chen H, Liu F, Patterson KI, Beretov J, O’Brien PM, Li Y. Expression of urokinase plasminogen activator and its receptor in advanced epithelial ovarian cancer patients. Gynecol Oncol. 2009;114(2):265–72.

    Article  CAS  PubMed  Google Scholar 

  51. Ghasemi A, Saeidi J, Mohtashami M, Hashemy SI. Estrogen-independent role of ERalpha in ovarian cancer progression induced by leptin/Ob-Rb axis. Mol Cell Biochem. 2019;458(1–2):207–17.

    Article  CAS  PubMed  Google Scholar 

  52. Chen H, Hao J, Wang L, Li Y. Coexpression of invasive markers (uPA, CD44) and multiple drug-resistance proteins (MDR1, MRP2) is correlated with epithelial ovarian cancer progression. Br J Cancer. 2009;101(3):432–40.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  53. Powell LE, Foster PA. Protein disulphide isomerase inhibition as a potential cancer therapeutic strategy. Cancer Med. 2021;10(8):2812–25.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  54. Xu S, Butkevich AN, Yamada R, Zhou Y, Debnath B, Duncan R, Zandi E, Petasis NA, Neamati N. Discovery of an orally active small-molecule irreversible inhibitor of protein disulfide isomerase for ovarian cancer treatment. Proc Natl Acad Sci U S A. 2012;109(40):16348–53.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  55. Kranz P, Neumann F, Wolf A, Classen F, Pompsch M, Ocklenburg T, Baumann J, Janke K, Baumann M, Goepelt K, et al. PDI is an essential redox-sensitive activator of PERK during the unfolded protein response (UPR). Cell Death Dis. 2017;8(8):e2986.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  56. Zhao LQ, Sun W, Zhang P, Gao W, Fang CY, Zheng AW. MFAP2 aggravates tumor progression through activating FOXM1/beta-catenin-mediated glycolysis in ovarian cancer. Kaohsiung J Med Sci. 2022;38(8):772–80.

    Article  CAS  PubMed  Google Scholar 

  57. Zhu J, Zhang S, Gu L, Di W. Epigenetic silencing of DKK2 and wnt signal pathway components in human ovarian carcinoma. Carcinogenesis. 2012;33(12):2334–43.

    Article  CAS  PubMed  Google Scholar 

  58. Dyer DP, Thomson JM, Hermant A, Jowitt TA, Handel TM, Proudfoot AE, Day AJ, Milner CM. TSG-6 inhibits neutrophil migration via direct interaction with the chemokine CXCL8. J Immunol. 2014;192(5):2177–85.

    Article  CAS  PubMed  Google Scholar 

  59. Reed MJ, Damodarasamy M, Pathan JL, Chan CK, Spiekerman C, Wight TN, Banks WA, Day AJ, Vernon RB, Keene CD. Increased Hyaluronan and TSG-6 in Association with neuropathologic changes of Alzheimer’s Disease. J Alzheimers Dis. 2019;67(1):91–102.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  60. Dyer DP, Salanga CL, Johns SC, Valdambrini E, Fuster MM, Milner CM, Day AJ, Handel TM. The anti-inflammatory protein TSG-6 regulates chemokine function by inhibiting Chemokine/Glycosaminoglycan interactions. J Biol Chem. 2016;291(24):12627–40.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  61. Wang J, Li Z, Xu L, Yang H, Liu W. Transmembrane domain dependent inhibitory function of FcgammaRIIB. Protein Cell. 2018;9(12):1004–12.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  62. Gold MA, Brady WE, Lankes HA, Rose PG, Kelley JL, De Geest K, Crispens MA, Resnick KE, Howell SB. A phase II study of a urokinase-derived peptide (A6) in the treatment of persistent or recurrent epithelial ovarian, fallopian tube, or primary peritoneal carcinoma: a Gynecologic Oncology Group study. Gynecol Oncol. 2012;125(3):635–9.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

Download references

Acknowledgements

Not applicable.

Funding

This study was supported by National Natural Science Foundation of China (No. 82372126, 82301900, 82072078), Zhongda Hospital Affiliated to Southeast University, Jiangsu Province High-Level Hospital (No. 2023GSPKY11, GSP-LCYJFH01, zdyyxy35), China Postdoctoral Science Foundation (No. 2024M750460) and Nanjing Postdoctoral Research Foundation Project (No. FTJ-bh-2).

Author information

Author notes

  1. Sicong Liu and Hao Lin contributed equally to this work.

Authors and Affiliations

  1. Department of Obstetrics and Gynecology, School of Medicine, Zhongda Hospital, Southeast University, Nanjing, 210003, China

    Sicong Liu, Ke Zhang, Quan Zhou & Yang Shen

  2. Department of Clinical Science and Research, School of Medicine, Zhongda Hospital, Southeast University, Nanjing, China

    Hao Lin

Authors
  1. Sicong Liu
    View author publications

    You can also search for this author inPubMed Google Scholar

  2. Hao Lin
    View author publications

    You can also search for this author inPubMed Google Scholar

  3. Ke Zhang
    View author publications

    You can also search for this author inPubMed Google Scholar

  4. Quan Zhou
    View author publications

    You can also search for this author inPubMed Google Scholar

  5. Yang Shen
    View author publications

    You can also search for this author inPubMed Google Scholar

Contributions

LSC, ZQ and ZK conceived the idea for the study and performed the data analyses and interpreted the results of the data analyses. LSC and LH obtained the genetic data. LSC, ZK, LH, ZQ and SY wrote the manuscript. LSC, ZQ and LH revised, polished, and verified the results of the manuscript. All authors read and approved the final manuscript.

Corresponding authors

Correspondence to Quan Zhou or Yang Shen.

Ethics declarations

Ethics approval and consent to participate

The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Clinical trial number

Not applicable.

Additional information

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Supplementary Material 1

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Liu, S., Lin, H., Zhang, K. et al. Potential drug targets for ovarian cancer identified through Mendelian randomization and colocalization analysis. J Ovarian Res 18, 32 (2025). https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s13048-025-01620-7

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s13048-025-01620-7

Keywords

Journal of Ovarian Research

ISSN: 1757-2215

Contact us