Application of plasma circulating protein ADM2 in predicting occurrence and prognosis of ovarian cancer

Abstract

Researches show that the plasma circulating protein ADM2 plays an important role in occurrence and development of the ovarian cancer and can be used as a biomarker for early diagnosis and treatment monitoring of the ovarian cancer. The identified high-level blood plasma circulation ADM2 protein level is remarkably related to the ovarian cancer risk increase, and the potential of the protein in ovarian cancer prognosis evaluation is prompted. Further co-localization analysis provides evidence of high support between the plasma circulating protein ADM2 and the ovarian cancer, and full phenotype group association study (PheWAS) does not find that the plasma circulating protein ADM2 is obviously associated with other diseases or phenotypes (no obvious side reaction), so that the feasibility of the plasma circulating protein ADM2 as a potential drug target is defined. Queue research based on the real world further verifies the potential application value of the plasma circulating protein ADM2 in ovarian cancer specific diagnosis.

Description

Technical Field

[0001] The present invention belongs to tumor diagnostic markers, and specifically relates to a combined diagnostic marker for early screening of ovarian cancer risk in a female population and predicting the prognosis of ovarian cancer patients. Background Art

[0002] Ovarian cancer is one of the common malignant tumors of the female reproductive system, and its global incidence is on the rise. According to the World Health Organization, ovarian cancer ranks eighth among female tumors, and its mortality rate is relatively high, partly because its early symptoms are hidden and atypical, mainly manifested as abdominal distension, abdominal pain, loss of appetite and abnormal urination, etc., so it is often mistaken for other diseases of the digestive system or urinary system, causing patients to delay medical treatment. The existing mainstream treatment methods mainly include surgery, chemotherapy and targeted therapy. Although the treatment methods for ovarian cancer have gradually diversified, due to the high heterogeneity of ovarian cancer and its resistance to chemotherapy drugs, and many patients are already in the late stage of the disease when diagnosed, the overall prognosis of patients is not ideal. Therefore, the development and verification of accurate biomarkers is of great significance for the early screening of ovarian cancer and improving patient prognosis.

[0003] The application of tumor biomarkers has become an important tool for early diagnosis, disease prognosis assessment, and treatment monitoring. Biomarkers can not only reflect the presence and nature of tumors, but also provide information about the patient's response to treatment. With the advancement of biotechnology, a variety of biomarkers associated with different cancers have been identified. For example, prostate-specific antigen (PSA) is used for the detection of prostate cancer, alpha-fetoprotein (AFP) is widely used for the screening of liver cancer, and carcinoembryonic antigen (CEA) is often used for the monitoring of colon cancer. In the field of ovarian cancer, CA-125 is currently the most commonly used biomarker, and its increased level is closely related to the presence and prognosis of ovarian cancer. However, the specificity of CA-125 is limited and it is often elevated in non-neoplastic diseases. Therefore, it is urgent to find new biomarkers to improve the sensitivity and specificity of early diagnosis.

[0004] Plasma circulating proteins as biomarkers show broad application prospects in monitoring tumor occurrence and development. These proteins exist in the blood and are easy to obtain. Efficient diagnosis and monitoring can be achieved through simple blood tests. Compared with traditional tissue biopsies, the detection of plasma circulating proteins has the advantages of less trauma, good patient tolerance and strong repeatability. They can not only reflect the presence of tumors, but also provide important information about the stage of tumor development, response to treatment and risk of recurrence. This makes them play a key role in early screening, prognosis assessment and personalized treatment of tumors.

[0005] Based on this, the study of biomarkers for ovarian cancer will provide new ideas and methods for clinical practice. Through in-depth research on plasma circulating proteins, we can better understand the biological characteristics of ovarian cancer, strengthen early identification of the disease and improve the overall prognosis of patients. Therefore, finding accurate biomarkers for ovarian cancer not only has important scientific value, but will also bring significant clinical benefits to the treatment and management of patients.

[0006] It is still unclear whether the protein level of ADM2 in plasma has a causal relationship with the risk and prognosis of ovarian cancer, and further research is needed to explore and clarify it. Summary of the invention

[0007] The purpose of the present invention is to provide a novel biomarker for diagnosing the onset of ovarian cancer and predicting the prognosis of patients. This signature has accurate predictive efficacy for early screening and prognostic assessment of ovarian cancer. Using large-scale protein quantitative trait loci (pQTL) data based on 35,559 Icelanders and related data from genome-wide association studies (GWAS), the causal association between plasma circulating proteins and three common gynecological tumors (cervical cancer, endometrial cancer and ovarian cancer) was systematically analyzed. We identified circulating proteins that were significantly associated with the risk of three tumors and evaluated the effects of these proteins by two-sample Mendelian randomization (MR) analysis. In the MR analysis of ovarian cancer, we paid special attention to the circulating protein adrenomedullin 2 (ADM2) in plasma and found that its high level was significantly associated with an increased risk of ovarian cancer. This suggests that the plasma circulating protein ADM2 may serve as a potential biomarker and target for ovarian cancer. Further co-localization analysis revealed highly supportive evidence between the plasma circulating protein ADM2 and ovarian cancer, suggesting its potential for application in treatment. In addition, in order to explore the feasibility of circulating proteins as potential drug targets, we analyzed the drugability of these proteins and evaluated their potential side effects in clinical applications. Through MR-PheWAS, we found that drugs targeting plasma circulating protein ADM2 have no significant side effects, which provides important clinical information for the future application of plasma circulating protein ADM2 in the treatment of ovarian cancer. The clinical efficacy of serum protein ADM2 levels in screening ovarian cancer was further verified based on a real-world clinical cohort, providing support for plasma circulating protein ADM2 to become a new biomarker for ovarian cancer.

[0008] The technical solution of the present invention is: The present invention provides the use of a reagent for detecting the expression level of plasma circulating protein ADM2 in preparing an early screening reagent or a detection kit for the risk of ovarian cancer in a population.

[0009] The present invention also provides the use of a reagent for detecting the expression level of plasma circulating protein ADM2 in preparing a prognosis prediction reagent or a prognosis detection kit for ovarian cancer patients.

[0010] Preferably, the reagent for detecting the expression level of plasma circulating protein ADM2 is a human blood ADM2 enzyme-linked immunosorbent assay reagent.

[0011] The higher the serum ADM2 protein level, the higher the individual's likelihood of developing ovarian cancer; for ovarian cancer patients, the higher the serum ADM2 protein level, the worse the patient's prognosis.

[0012] The present invention will be further explained and described below: (1) We first investigated the causal association between 1347 proteins with available indexed pQTL signals and the risk of three common gynecological tumors in a proteome-wide Mendelian randomization (MR) analysis. The overall landscape of the MR analysis results is shown in Figure 2 As shown. We identified 38 significant protein-cervical cancer associations (P < 0.05). After excluding the HEIDI trial and performing multiple test correction (FDR), the associations that did not survive showed that increased levels of three circulating proteins, PDE5A (cGMP-specific 3',5'-cyclic phosphatase), VEGFA (vascular endothelial growth factor A), and CD14 (monocyte differentiation antigen CD14), were significantly associated with cervical cancer risk ( Figure 2 , OR < 1). In the SMR analysis, the results of PDE5A, VEGFA, and CD14 remained stable. However, in the study of Sakaue S et al., the association of PDE5A, VEGFA, and CD14 with the risk of cervical cancer could not be replicated.

[0013] Similarly, we found 95 circulating proteins that were causally associated with endometrial cancer (P < 0.05). However, after FDR correction and excluding the HEIDI experiment, only the GLCE-endometrial cancer association survived, and increased serum levels of GLCE may lead to an increased risk of endometrial cancer ( Figure 2 , OR>1). However, the results of GLCE were negative in SMR analysis and could not be replicated in the study of Sakaue S et al.

[0014] In the MR analysis based on ovarian cancer, we screened 76 circulating proteins with genome-wide significant association (P < 0.05). After FDR and HEIDI screening, we found that high levels of circulating ADM2 (adrenomedullin 2) were associated with an increased risk of ovarian cancer ( Figure 2). For every 1 standard deviation increase in the genetically predicted ADM2 protein level, the risk of ovarian cancer increased by 4.262 times (95%CI: 2.159-8.412). In the multi-SNPs-SMR analysis, the association between ADM2 and ovarian cancer still exists. Sakaue S's study repeated the relationship between ADM2 and ovarian cancer risk and obtained a consistent causal direction. The specific analysis results are shown in Table 1 below: Table 1 Mendelian randomization analysis and colocalization of circulating proteins in ovarian cancer

[0015] (2) In order to explore whether the circulating proteins screened in the MR analysis have the potential to become drug targets, we performed co-localization analysis on the pQTL signals of circulating proteins and the GWAS signals of cervical cancer, endometrial cancer and ovarian cancer. The results of the co-localization analysis showed that in cervical cancer, APOM had high co-localization evidence with PIANP, and GSKIP and PDE5A had moderate co-localization evidence. Ovarian cancer had high support for co-localization with ADM2 and RRM2B, and moderate support for co-localization with NT5M. For endometrial cancer, there was moderate to high support for co-localization with 14 circulating proteins, including CXCL9, CYR61, IFNAR1, LECT1, OIT3, ST3GAL2, STMN4, BIRC2, C5orf38, ENG, IGF2R, IL3RA, ISLR2 and U2AF2 ( Figure 3 ).

[0016] Combined with the results of MR analysis, we found that among the five circulating proteins screened, evidence highly supporting co-localization was observed between ADM2 and ovarian cancer (P H4 =0.897), and was identified as a first-level drug target. Moderate evidence supporting colocalization was observed between PDE5A and cervical cancer (P H4 =0.630), and was identified as a Level 2 drug target. The remaining three circulating proteins were identified as Level 3 drug targets (P H4 <0.5).

[0017] (3) Next, we explored the drugability of the identified proteins. We searched the drug database for the five circulating proteins identified as possible drug targets in the MR analysis. None of them were found to be drug targets for cervical cancer, endometrial cancer, and ovarian cancer. Drugs targeting VEGFA and PDE5A are mainly used to treat other cancers. In addition, we found several drugs targeting CD14, one of which has been approved for the treatment of sepsis and pneumonia. No drug information was found for GLCE and ADM2. We also retrieved drug information for proteins with highly supporting evidence of co-localization. Several drugs targeting IFNAR1, named pegylated interferon α-2b, interferon α-2b, interferon β-1b, and pegylated interferon β-1a, have been approved for the treatment of various malignancies including melanoma. Drugs targeting STMN4, CYR61, APOM, and RRM2B were found to be potential cancer targets. No relevant information about ST3GAL2, OIT3, CXCL9, LECT1, PIANP and other sites was retrieved from the drug database.

[0018] (4) Furthermore, we performed MR-PheWAS to investigate whether there are significant side effects of targeted therapy against the ovarian cancer-related core protein ADM2. A total of 700 traits or diseases in UKBB were used for MR-PheWAS screening, and the Manhattan plot showed that ADM2 had no significant association with any disease phenotype ( Figure 4 ).

[0019] (5) Finally, we collected serum samples from 71 ovarian cancer patients and 19 healthy women in a real cohort based on the Third Xiangya Hospital and detected the protein concentration of ADM2 in the samples by enzyme-linked immunosorbent assay (ELISA). Figure 5 A shows the ELISA normalized regression curve between protein concentration and OD450nm. We found that the serum ADM2 concentration of ovarian cancer patients was significantly higher than that of the control group ( Figure 5 B), and as a predictor of ovarian cancer risk, the ROC curve of serum ADM2 level showed an AUC value of 0.919 ( Figure 5 C, 95% CI: 0.844-0.963), indicating that it has a high predictive efficacy. Multivariate logistic regression analysis showed that serum ADM2 level was a risk factor independent of age and hypertension (Figure 5D). KM survival analysis of the ovarian cancer cohort showed that the progression-free survival (PFS) of patients with high ADM2 expression was significantly lower than that of patients with low ADM2 expression ( Figure 5E). The results of clinical subgroup analysis showed that the serum ADM2 protein concentration of patients with progressive ovarian cancer was significantly increased, and the proportion of ovarian cancer patients with high ADM2 expression and disease progression was significantly higher than that of patients with low ADM2 expression ( Figure 5 F).

[0020] Compared with the prior art, the advantages of the present invention are: The advantage of the present invention is that it systematically evaluates the potential role of circulating plasma protein ADM2 in ovarian cancer and related diseases, and provides strong support for its feasibility as a drug target. Through Mendelian randomization analysis, it was revealed that increased levels of circulating ADM2 protein were significantly associated with increased risk of ovarian cancer (OR = 4.262). In addition, we verified it by constructing a real clinical cohort. We collected serum samples from 71 ovarian cancer patients and 19 healthy women, and detected ADM2 levels by ELISA. It was found that the serum ADM2 concentration of ovarian cancer patients was significantly higher than that of the control group, and the AUC value of the ROC curve was 0.919, indicating its good efficacy in predicting the risk of ovarian cancer. This result suggests that ADM2 has the potential to become a new biomarker in the early screening and monitoring of ovarian cancer. At the same time, combined with KM survival analysis, it was shown that patients with high expression of ADM2 had a significantly lower progression-free survival than patients with low expression, suggesting that ADM2 may be associated with prognosis and is of great significance in clinical applications. In addition to evaluating the diagnostic efficacy of ADM2 for ovarian cancer, this study further identified ADM2 as a drug target with great potential (level 1 evidence) through co-localization analysis, and revealed through pheGWAS analysis that ADM2 as a drug target has no significant side effects and has high feasibility. Therefore, circulating plasma protein ADM2 can be used as a target site in ovarian cancer screening reagents.

[0021] The detailed structure of the present invention is further described below in conjunction with the accompanying drawings and specific implementation methods. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 Roadmap for the design of this study. FDR, false discovery rate; HEIDI, heterogeneity of causal instruments; MR, Mendelian randomization.

[0023] Figure 2Manhattan plots showing the associations of 1347 circulating protein levels predicted by genes with three gynecological tumors in MR analysis. (A) Association of circulating protein levels predicted by genes with cervical cancer. (B) Association of circulating protein levels predicted by genes with endometrial cancer. (C) Association of circulating protein levels predicted by genes with ovarian cancer. The labeled and colored genes refer to MR findings with a P value < 0.05 after FDR correction (two-sample MR analysis). Red genes indicate a positive effect of circulating proteins on the results; blue genes indicate a negative effect of circulating proteins on the results. Results are plotted according to the starting position of the gene.

[0024] Figure 3 Figure 2. Highly supported co-localization evidence between circulating proteins and results in the discovery and replication datasets. The size of the circle represents the P value of the H4 hypothesis in the co-localization analysis, and the color of the circle represents the classification of the evidence.

[0025] Figure 4 Figure 1 shows the results of MR-pheWAS analysis of targeted therapy for serum ADM2 in ovarian cancer. The Manhattan plot shows that no significant association was found between ADM2 and any disease phenotype.

[0026] Figure 5 Figure 1 is an exploration of the clinical significance of serum ADM2 in ovarian cancer based on a real cohort. (A) Regression curve of ADM2 protein expression level in serum and OD450nm obtained by fitting the ELISA standard data. (B) Difference in ADM2 serum protein expression level between ovarian cancer patients and the control group based on ELISA. (C) ROC curve was used to verify the predictive efficacy of serum ADM2 level on the risk of ovarian cancer. (D) Multivariate logistic regression analysis showed that serum ADM2 level was a risk factor independent of age and hypertension and may be associated with the occurrence of ovarian cancer. (E) KM survival analysis of the effect of serum ADM2 protein expression level on progression-free survival (PFS) of ovarian cancer patients. (F) Difference in serum ADM2 expression level between patients with disease progression and non-progression during follow-up. ns, no statistically significant difference; * p < 0.05; ** p < 0.01; ** p < 0.001. DETAILED DESCRIPTION

[0027] 1. Study design and GWAS data sources This study was designed as Figure 1Summary-level statistics (https: / / www.decode.com / summarydata / ) of genetic associations with levels of 4907 circulating proteins were extracted from a large-scale protein quantitative trait loci (pQTL) study in 35,559 Icelanders as proteomic data for this study. A multiplexed, modified aptamer binding assay (SOMAscan V4) was used for proteomic analysis. In genome-wide association analysis under the BOLT-LMM linear mixed model, residuals were normalized using an inverse normal transformation and the normalized values ​​were used as phenotypes. pQTL data were used as exposure factors in a two-sample Mendelian randomization (MR) analysis to screen for proteins with genome-wide significant associations (p < 5e-8) with outcome events. All proteins with summary-level data were included in the colocalization analysis. Three common gynecological tumors were considered outcome events for the MR analysis in this study. GWAS data for cervical, endometrial, and ovarian cancers were obtained from the IEU openGWAS project, the Ovarian Cancer Association Consortium, and the study by O'Mara et al., including 1604 cervical cancers, 12906 endometrial cancers, and 1954 ovarian cancers, respectively. The study by Sakaue S et al. was used for replication, including 909 cervical cancers, 2188 endometrial cancers, and 1588 ovarian cancers, respectively. All participants were of European ancestry, and there was no overlap between data sources. All included studies were approved by the respective ethical review committees.

[0028] 2. Mendelian randomization (MR) analysis based on pQTL We performed two-sample MR analysis based on indexed single nucleotide polymorphisms (SNPs) on circulating proteins to obtain causal associations between them and three common gynecological tumors. The "TwoSampleMR" R package was used to perform the MR analysis, and SNPs with F statistics greater than 10 were screened for the next analysis to exclude weak instrumental variables. The odds ratios (ORs) and corresponding p-values ​​of the associations between circulating proteins and three common gynecological tumors were calculated using the "Inversevariance weighted" and "Wald ratio" methods, respectively. The false discovery rate (FDR) method was used for multiple testing at α = 0.05. To strengthen the evidence of causal associations between circulating proteins and three common gynecological tumors, the Mendelian randomization (SMR) method based on pooled data was used to test the significance level using multiple SNPs for each circulating protein. The heterogeneity test for dependent instrumental variables (HEIDI) was used to test heterogeneity and pleiotropy, and P > 0.05 was considered to indicate that the results did not have heterogeneity and pleiotropy. The SMR software tool (v1.0.3) was used to perform the SMR analysis.

[0029] 3. Colocalization Analysis The “COLOC” R package was used to perform colocalization analysis to test whether the associations between the identified proteins and cervical, endometrial, and ovarian cancers were driven by linkage disequilibrium. For each locus, a Bayesian approach evaluated five different hypothesis tests (H0, H1, H2, H3, and H4), corresponding to the following five exclusive hypotheses: 1) no association with either trait; 2) association with trait 1 only; 3) association with trait 2 only; 4) association with both traits, but different causal variants for each trait; 5) association with both traits, and the causal variants for both pathways are the same. The prior probability that a SNP is associated with trait 1 (P1) and the prior probability that a SNP is associated with trait 2 (P2) were both set to 1e-4. The prior probability that a SNP is associated with both traits 1 and 2 (P12) was set to 1e-5. When the posterior probability of sharing a causal variant (P H4 ) ≥ 0.8, the GWAS signal and the pQTL signal were considered to have a high degree of co-localization evidence. <P H4 When <0.8, the two signals were considered to have moderate evidence of colocalization.

[0030] Subsequently, the circulating proteins with FDR-corrected P values ​​< 0.05 in MR analysis and P values ​​> 0.05 in Heidi test were divided into three groups according to the results of colocalization analysis. H4 Proteins with colocalization evidence (≥0.8) were considered primary targets; proteins with moderate support for colocalization evidence (0.5 <P H4 <0.8) were considered secondary targets; the remaining proteins were considered tertiary targets.

[0031] 4. Identification of druggable proteins To evaluate the drugability of the circulating proteins obtained from the screening, we searched in DrugBank, Dependency Map, Connectivity Map, ChEMBL databases, and lists of druggable genes from previous studies. For proteins identified in drug databases, information about the drug name and drug development process was recorded. To evaluate potential drug efficacy, we divided these proteins into four categories: 1) approved (one or more drugs targeting a specific protein have been approved); 2) in clinical trials (targeted drugs are currently undergoing clinical trials); 3) preclinical (targeted drugs are undergoing preclinical studies); 4) druggable (proteins cannot be identified in drug databases but are listed as druggable targets).

[0032] 5. Phenotype-Wide Association Study (PheWAS) To further explore the potential side effects of drugs targeting the ovarian cancer-related core protein ADM2, we performed MR-PheWAS analysis based on the “ieugwasr” and “TwoSampleMR” R packages. The cis-pQTL of ADM2 was used as the exposure factor, and the disease summary statistics of the UKBB cohort with a sample size of up to 408,961 people were used as the outcome factor. In the process of conducting GWAS, the researchers used a scalable and accurate implementation of the generalized mixed model (SAIGE V.0.29) method to adjust the unbalanced case-control ratio. To ensure sufficient statistical power, we selected 700 traits or diseases with more than 500 cases from 1403 traits (diseases) as outcomes. These 700 traits or diseases were defined according to the International Classification of Diseases and Related Health Problems (ICD-9 / -10). We performed MR analysis based on the Wald ratio method, and the causal association was considered statistically significant if the Bonferrni method-corrected p value was <0.05 / 700.

[0033] 6. Real World Queues We collected serum samples from 71 ovarian cancer cases and 19 healthy female controls from May to October 2024. All serum samples were collected from the inpatient wards and physical examination center of the Third Xiangya Hospital. The pathological diagnosis of ovarian cancer was confirmed by the pathologist and the attending physician. Blood samples were collected from the patient's vein using aseptic techniques and collected into vacuum blood collection tubes. The collected whole blood samples were placed at room temperature for 30 minutes. Subsequently, the serum was separated by centrifugation at 3000 rpm for 15 minutes in a centrifuge, and the supernatant was transferred to a clean cryotube and stored at -80°C for further analysis. This study was approved by the Ethics Committee of the Third Xiangya Hospital. All patients signed informed consent before blood collection.

[0034] 7. Enzyme-linked immunosorbent assay (ELISA) The protein expression levels of all sera were determined by ELISA. A double antibody one-step sandwich ELISA kit was used to detect the levels of human ADM2 in serum samples. Specific capture antibodies were pre-coated in the microplate wells, and then samples, standards, and HRP-labeled detection antibodies were added in sequence, followed by incubation and thorough washing. The color reaction was initiated by adding TMB substrate, and the color changed to blue in the presence of peroxidase and to yellow under acidic conditions. The depth of the color was positively correlated with the level of ADM2 in the sample. The absorbance (OD value) was measured at 450nm using an ELISA reader to calculate the sample concentration. Based on the ELISA results, we verified the association between the expression of ADM2 in patient serum and the clinical characteristics and prognosis of ovarian cancer, and explored its clinical significance.

[0035] 8. Statistical analysis All bioinformatics analyses in this study were performed using R (v4.2.1) and Perl, with Perl being used primarily for data organization and cleaning. Unless otherwise specified, in the bioinformatics section, the “limma” R package was used to perform differential analysis in this study, the Wilcoxon test was used to compare differences between two groups, and the Kruskal–Wallis test was used to compare differences between three or more groups. The Spearman correlation coefficient was used to evaluate correlation analysis in this study. The “survival” and “surviminer” R packages were used for survival analysis in this study, with the default methods being the KM method and the log-rank test. In the experimental section, Student's t test was used for comparisons between two samples. All statistical tests were two-sided, with p < 0.05 considered statistically significant.

Claims

1. The use of a reagent for detecting the expression level of plasma circulating protein ADM2 in the preparation of an early screening reagent or detection kit for the risk of ovarian cancer in the population.

2. The use of a reagent for detecting the expression level of plasma circulating protein ADM2 in the preparation of a prognosis prediction reagent or a prognosis detection kit for ovarian cancer patients.

3. The use according to claim 1 or 2, characterized in that: The reagent for detecting the expression level of plasma circulating protein ADM2 is a human blood ADM2 enzyme-linked immunosorbent assay reagent.

Images