A combined biomarker panel and assessment model for evaluating chronic low-grade inflammation in polycystic ovary syndrome
By jointly detecting eight inflammation-related proteins and two metabolites, a logistic regression model was constructed to solve the problem of objective assessment of chronic low-grade inflammation in PCOS. This model enables stable and quantitative assessment of inflammatory status and subtype differentiation, and is suitable for individualized management of polycystic ovary syndrome.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-03-06
- Publication Date
- 2026-06-09
AI Technical Summary
Current technologies lack objective assessment methods for the chronic low-grade inflammatory state of polycystic ovary syndrome (PCOS). Single inflammatory factor detection is difficult to reflect the inflammatory network status, has large individual differences, lacks stability, is difficult to distinguish different metabolic subtypes, and lacks quantitative scoring models.
Eight inflammation-related proteins (FURIN, CCL3, CCL8, CD38, NOS2, NOS3, IL18R1, and HGF) and two metabolites (lithocholic acid-3-O-glucuronide and taurine) were detected in combination. An immunological and metabolomics approach was used to construct a logistic regression model for evaluation.
It achieves stable and quantitative assessment of chronic low-grade inflammation in PCOS, has good discriminative ability, is applicable to different metabolic subtypes, and has broad applicability and is easy to translate into clinical practice.
Smart Images
Figure CN122171811A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of biomarker technology, specifically to a combination of biomarkers and an evaluation model for assessing chronic low-grade inflammation in polycystic ovary syndrome. Background Technology
[0002] Polycystic ovary syndrome (PCOS) is a common reproductive endocrine and metabolic disorder. Numerous studies have shown that chronic low-grade inflammation is one of the core pathophysiological features of PCOS and is closely associated with insulin resistance, metabolic disorders, and long-term cardiovascular risk. However, current clinical diagnostic criteria for PCOS are mainly based on reproductive and endocrine phenotypes, and objective assessment methods for the chronic inflammatory state of PCOS are lacking.
[0003] The existing technology has the following defects and shortcomings: (1) Existing detection methods are mostly focused on single inflammatory factors, which are difficult to reflect the overall state of the inflammatory network in PCOS; (2) The reported inflammatory factors or metabolites often have large individual differences and insufficient stability, which limits their clinical application. (3) Existing methods are difficult to distinguish between different metabolic subtypes (such as obese type, insulin-resistant type, etc.), and cannot provide a basis for individualized chronic inflammation management; (4) There is currently a lack of scoring or models that can quantitatively reflect the level of systemic inflammation in PCOS and its synergistic changes with metabolic abnormalities.
[0004] Therefore, developing a multi-molecule combined detection method that can comprehensively reflect the chronic inflammatory burden of PCOS is of great significance for disease stratification, risk assessment, and subsequent individualized management. Summary of the Invention
[0005] To address the aforementioned technical problems, the present invention aims to provide a combination of biomarkers and an assessment model for evaluating chronic low-grade inflammation in polycystic ovary syndrome (PCOS), thereby resolving the existing lack of objective evaluation methods for the chronic inflammatory status of PCOS.
[0006] The technical solution of the present invention to solve the above-mentioned technical problems is as follows: A first aspect of the present invention provides a combination of combined biomarkers for assessing chronic low-grade inflammation in polycystic ovary syndrome, comprising inflammation-related proteins and metabolites; Inflammation-related proteins include FURIN, CCL3, CCL8, CD38, NOS2, NOS3, IL18R1, and HGF; Metabolites include lithocholic acid-3-O-glucuronide and taurine.
[0007] Furthermore, inflammation-related proteins and metabolites were positively correlated with the chronic low-grade inflammatory state in polycystic ovary syndrome.
[0008] A second aspect of the invention provides the application of the above-described combination of biomarkers in establishing a model for evaluating chronic low-grade inflammation in polycystic ovary syndrome.
[0009] A third aspect of the present invention provides a method for establishing a chronic low-grade inflammation model for evaluating polycystic ovary syndrome, comprising the following steps: S1. Collect peripheral blood samples from patients with polycystic ovary syndrome and healthy controls; S2. Immunological detection methods were used to quantitatively detect eight inflammation-related proteins, namely FURIN, CCL3, CCL8, CD38, NOS2, NOS3, IL18R1 and HGF, in peripheral blood samples collected in S1. S3. The metabolomics method was used to quantitatively detect two metabolites, lithocholic acid-3-O-glucuronide and taurine, in the peripheral blood samples collected in S1. S4. Based on the quantitative detection results of inflammation-related proteins and metabolites obtained from S2 and S3 in patients with polycystic ovary syndrome and healthy controls, input them into the statistical model to construct a joint diagnostic model.
[0010] Furthermore, the immunological detection methods in S2 include the Olink platform or ELISA detection.
[0011] Furthermore, the metabolomics detection methods in S3 include liquid chromatography-mass spectrometry.
[0012] Furthermore, the statistical models in S4 include either Logistic regression or regularized regression models.
[0013] A fourth aspect of the present invention provides a construction system for evaluating the establishment of a chronic low-grade inflammation model of polycystic ovary syndrome, comprising the following modules: The blood sample collection module is used to collect peripheral blood samples. The protein detection module is used to quantitatively detect eight inflammation-related proteins: FURIN, CCL3, CCL8, CD38, NOS2, NOS3, IL18R1, and HGF. The metabolite detection module is used for the quantitative detection of two metabolites: lithocholic acid-3-O-glucuronide and taurine. The model building module is used to process detection data and build a joint diagnostic model.
[0014] In a fifth aspect, the present invention provides a chronic low-grade inflammation model for evaluating polycystic ovary syndrome, obtained using the above-described establishment method.
[0015] In a sixth aspect, the present invention provides a kit for assessing chronic low-grade inflammation in polycystic ovary syndrome (PCOS), comprising the aforementioned model for assessing chronic low-grade inflammation in PCOS, a peripheral blood sample collection preparation, detection preparations for each of the eight inflammation-related proteins (FURIN, CCL3, CCL8, CD38, NOS2, NOS3, IL18R1, and HGF), and detection preparations for each of the two metabolites (lithocholic acid-3-O-glucuronide and taurine).
[0016] The present invention has the following beneficial effects: 1. Multi-molecule combination with high stability: This invention uses eight inflammation-related proteins and two metabolites to construct a panel for assessing chronic low-grade inflammation in PCOS, overcoming the shortcomings of large fluctuations in single indicators and improving the stability and repeatability of the assessment.
[0017] 2. Intuitive reflection of inflammatory status: This invention establishes a combined diagnostic model through multi-molecule combination, which can intuitively obtain the combined Panel score, realizing a systematic and quantitative assessment of the chronic low-grade inflammatory status of PCOS patients.
[0018] 3. Excellent differentiation and prediction capabilities: This invention assesses chronic low-grade inflammation in PCOS through a multi-molecule combined approach, which has a more accurate ability to distinguish between PCOS and healthy individuals, significantly superior to a single indicator.
[0019] 4. Wide applicability: This invention demonstrates good stability in obese, insulin-resistant, hyperlipidemic, and novel subtyped PCOS systems through a multi-molecule combined evaluation method, and has wide applicability.
[0020] 5. Easy to translate into clinical practice: The detection methods involved in the evaluation method of this invention are all mature detection methods in the field, which can be implemented under routine experimental conditions and have practical promotion value. Attached Figure Description
[0021] Figure 1 A schematic diagram of the screening process for combined markers; Figure 2 The results of Spearman rank correlation coefficient analysis of the combined biomarkers and inflammation scores in PCOS patients in Example 2; Figure 3 This refers to the ablation analysis results of combined biomarkers and inflammation scores in PCOS patients in Example 2; Figure 4The diagram shows the distribution of the combined Panel score in different PCOS subtypes, where a represents the overall PCOS population, b represents severe PCOS, c represents IR type, d represents HA type, e represents hyperlipidemia subtype, and f represents the four subtypes recently proposed by Academician Chen Zijiang. Detailed Implementation
[0022] The principles and features of the present invention are described below with reference to the accompanying drawings. The examples given are for illustrative purposes only and are not intended to limit the scope of the invention. Unless otherwise specified in the examples, conventional conditions or conditions recommended by the manufacturer should be followed. Reagents or instruments whose manufacturers are not specified are all commercially available products.
[0023] Example 1: Screening and stability verification of combined biomarkers A schematic diagram of the screening process for combined markers is shown below. Figure 1 As shown, patient samples were first obtained and Olink proteomics and non-target metabolomics were performed. The association analysis of the two omics ultimately identified 8 proteins (FURIN, CCL3, CCL8, CD38, NOS2, NOS3, IL18R1 and HGF) and 2 metabolites (lithocholic acid-3-O-glucuronide and taurine) associated with chronic inflammation of PCOS.
[0024] Based on the 10 differential substances obtained from the above screening, an inflammation score and a combined panel score were constructed. It was found that the panel constructed by the combined application of the 10 differential substances selected in this invention can reflect the overall condition of PCOS patients and has good and stable diagnostic performance.
[0025] Example 2: Diagnostic efficacy of combined biomarkers Ten biomarkers screened in Example 1 were used as combined biomarkers for Spearman rank correlation coefficient analysis and ablation analysis of inflammation scores in PCOS patients. The experimental results are as follows: Figure 2 and Figure 3 As shown.
[0026] The results showed that eight proteins associated with chronic inflammation in PCOS (FURIN, CCL3, CCL8, CD38, NOS2, NOS3, IL18R1, and HGF) and two metabolites associated with chronic inflammation in PCOS (lithocholic acid-3-O-glucuronide and taurine) were significantly positively correlated with inflammation scores, indicating that they can serve as biomarkers for PCOS-related inflammation. Ablation analysis showed that the model AUC gradually increased with the stepwise addition of the eight inflammation-related proteins, lithocholic acid-3-O-glucuronide (L3G), and taurine, with the panel model consisting of 10 biomarkers achieving the highest predictive performance.
[0027] Example 3: Detection performance of combined biomarkers in different PCOS subtypes PCOS patients are classified into different subtypes, including obese, insulin-resistant, hyperandrogenic, and other related subtypes. By comparing the distribution of combined Panel scores in each subtype, combined Panel scores can be used to assess the differences in chronic inflammatory burden among different PCOS subtypes. This is particularly suitable for PCOS patients with metabolic abnormalities, as it can better diagnose the patient's chronic inflammatory status. This embodiment uses combined biomarkers to detect Panel scores for different subtypes, and the experimental results are as follows: Figure 4 As shown.
[0028] The results showed that the Panel score was significantly elevated in the overall PCOS population, and remained consistently high in the overweight, IR, HA, and hyperlipidemia subtypes, indicating that the combined biomarkers used in this invention have good subtype differentiation ability in terms of PCOS inflammatory-metabolic phenotypes. Based on Academician Chen Zijiang's latest tetratyping, the Panel score also showed significant differences among different subtypes, with the OB subtype showing a higher Panel score and the SHBG subtype showing a relatively lower Panel score. This indicates that the combined inflammatory-metabolic biomarkers effectively reveal the molecular characteristics of different PCOS subtypes and have broad applicability.
[0029] Example 4: Translational Application of Combined Biomarkers in Clinical Practice The Panel score can be used to assess the chronic inflammatory status of patients with polycystic ovary syndrome (PCOS) and can be combined with clinical information for disease assessment and follow-up. The calculation results of the Panel score can provide objective evidence for the assessment of chronic inflammation, thereby assisting in individualized management and clinical decision-making for PCOS chronic inflammation.
[0030] This embodiment constructs a logistic regression model based on 10 indicators (FURIN, IL18R1, CCL3, CCL8, CD38, NOS2, NOS3, HGF, L3G, and Taurine) for risk prediction. The specific process is as follows: First, the raw measured values (x) of each indicator in the new sample are standardized according to the mean (μ) and standard deviation (σ) in the training queue, and the Z-score (z = (x)) is calculated. Next, the regression coefficient (β) corresponding to each indicator is multiplied by its Z-score to obtain the contribution value (β×z) of a single indicator, and the contribution values of the 10 indicators are summed; then, this sum is added to the model's shrinkage intercept term (Intercept, shrunk) to obtain the linear prediction value PanelScore (PanelScore= Intercept + Σ(β×z)); finally, the PanelScore is converted into the probability value PanelProb using the Logistic function (PanelProb = 1 / [1 + exp( PanelScore). PanelProb represents the predicted probability of a positive event (such as high risk of chronic low-grade inflammation in PCOS) as defined by the model. It is usually judged by a default threshold of 0.5: PanelProb ≥ 0.5 is judged as high risk, and < 0.5 is judged as low risk.
[0031] In this embodiment, the mean, standard deviation, and regression coefficients of all indicators used are derived from the training queue, and the new sample detection platform and data prediction processing method are consistent with those in the modeling stage to ensure the comparability and effectiveness of the model calculation results.
[0032] The following is a specific calculation and application method using sample N1 as an example:
[0033] Based on the data in the table, perform the following calculations: Intercept(shrunk) = 0.8627; ∑(β×Z)=-1.4425; PanelScore=Intercept(shrunk)+∑(β×Z)=-0.5798; PanelProb=1 / (1+exp(0.5798))=0.359.
[0034] Based on the calculation that PanelProb < 0.5, sample N1 is judged to be a low-risk patient with chronic inflammation of PCOS.
[0035] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.
Claims
1. A combination of biomarkers for assessing chronic low-grade inflammation in polycystic ovary syndrome, characterized in that, Including inflammation-related proteins and metabolites; The inflammation-related proteins include FURIN, CCL3, CCL8, CD38, NOS2, NOS3, IL18R1, and HGF; The metabolites include lithocholic acid-3-O-glucuronide and taurine.
2. The combined biomarker combination for assessing chronic low-grade inflammation in polycystic ovary syndrome according to claim 1, characterized in that, Both the inflammation-related proteins and the metabolites are positively correlated with the chronic low-grade inflammatory state of polycystic ovary syndrome.
3. The use of the combined biomarker combination as described in claim 1 or 2 in establishing a chronic low-grade inflammation model for evaluating polycystic ovary syndrome.
4. A method for establishing a chronic low-grade inflammation model for evaluating polycystic ovary syndrome, characterized in that, Includes the following steps: S1. Collect peripheral blood samples from patients with polycystic ovary syndrome and healthy controls; S2. Immunological detection methods were used to quantitatively detect eight inflammation-related proteins, namely FURIN, CCL3, CCL8, CD38, NOS2, NOS3, IL18R1 and HGF, in peripheral blood samples collected in S1. S3. The metabolomics method was used to quantitatively detect two metabolites, lithocholic acid-3-O-glucuronide and taurine, in the peripheral blood samples collected in S1. S4. Based on the quantitative detection results of inflammation-related proteins and metabolites obtained from S2 and S3 in patients with polycystic ovary syndrome and healthy controls, input them into the statistical model to construct a joint diagnostic model.
5. The method for establishing a chronic low-grade inflammation model for evaluating polycystic ovary syndrome according to claim 4, characterized in that, The immunological detection methods in S2 include Olink platform or ELISA detection.
6. The method for establishing a chronic low-grade inflammation model for evaluating polycystic ovary syndrome according to claim 4, characterized in that, The metabolomics detection method in S3 includes liquid chromatography-mass spectrometry.
7. The method for establishing a chronic low-grade inflammation model for evaluating polycystic ovary syndrome according to claim 4, characterized in that, The statistical model in S4 includes either a Logistic regression model or a regularized regression model.
8. A system for constructing a method for evaluating the establishment of a chronic low-grade inflammation model for polycystic ovary syndrome, characterized in that, Includes the following modules: The blood sample collection module is used to collect peripheral blood samples. The protein detection module is used to quantitatively detect eight inflammation-related proteins: FURIN, CCL3, CCL8, CD38, NOS2, NOS3, IL18R1, and HGF. The metabolite detection module is used for the quantitative detection of two metabolites: lithocholic acid-3-O-glucuronide and taurine. The model building module is used to process detection data and build a joint diagnostic model.
9. A model for evaluating chronic low-grade inflammation in polycystic ovary syndrome, characterized in that, It is obtained by using the method described in any one of claims 4-7.
10. A kit for assessing chronic low-grade inflammation in polycystic ovary syndrome, characterized in that, The invention includes the peripheral blood sample collection preparation for evaluating a chronic low-grade inflammation model of polycystic ovary syndrome as described in claim 9, detection preparations for each of the eight inflammation-related proteins (FURIN, CCL3, CCL8, CD38, NOS2, NOS3, IL18R1, and HGF), and detection preparations for each of the two metabolites (lithocholic acid-3-O-glucuronide and taurine).