Biomarker Combinations and Diagnostic Model for Diabetic Kidney Disease Based on Model Analysis

Abstract

This invention, entitled "A Biomarker Combination Based on Model Analysis and a Diagnostic Model for Diabetic Nephropathy," belongs to the field of disease diagnosis technology. The technical problem it aims to solve is that existing technologies lack specific biomarker combinations for diabetes and diabetic nephropathy, have poor correlations between biomarkers and etiologies, and cannot accurately distinguish between the two and define the disease course. The key technical solution is a biomarker combination based on model analysis, consisting of P07148_FABP1 and P02144_MB. This invention significantly improves the differentiation between type 2 diabetes and early diabetic nephropathy, effectively avoiding misdiagnosis and missed diagnosis, accurately defining the disease stage, and providing strong evidence for precision clinical diagnosis and treatment.

Description

Technical Field

[0001] This invention belongs to the field of disease diagnosis technology, specifically relating to biomarker combinations based on model analysis and a diagnostic model for diabetic nephropathy. Background Technology

[0002] Diabetic kidney disease (DKD) is a chronic complication of diabetes mellitus (DM). Its pathogenesis is complex, involving the interaction of multiple processes such as metabolic disorders, oxidative stress, inflammatory responses, and renal interstitial fibrosis, ultimately leading to glomerular sclerosis and renal tubular damage, gradually progressing to end-stage renal disease. With the expanding diabetic population, the health crisis and medical burden caused by DKD are becoming increasingly prominent. Due to limited diagnostic technology, the rates of missed and misdiagnosed cases of DKD are high, hindering the standardized treatment of the disease.

[0003] Current clinical diagnostic systems for diabetic kidney disease (DKD) have significant shortcomings. Traditional indicators are not only diagnostically lagging but also inadequate for accurately assessing disease progression and prognosis. Microalbuminuria, a traditional hallmark of early DKD damage, typically shows abnormal elevations after kidney inflammation, fibrosis, and glomerular hyperfiltration, failing to detect early subclinical kidney damage and leading to missed diagnoses of non-proteinuric DKD patients. While serum creatinine and blood urea nitrogen are widely used for kidney function assessment, their diagnostic specificity is low due to interference from physiological state and comorbidities. In patients with diabetes, hypertension, or cardiovascular disease, it is difficult to distinguish the specific cause of kidney damage, easily leading to misdiagnosis.

[0004] In recent years, although some studies have identified novel biomarkers associated with kidney injury molecule-1 (KIM-1) and neutrophil gelatinase-associated lipocalin (NGAL) as being related to kidney disease (DKD), these biomarkers primarily focus on renal tubular injury or systemic inflammatory responses, lacking DKD specificity. They can also be abnormally elevated in other kidney diseases such as acute kidney injury and chronic nephritis, and therefore cannot be used as sole diagnostic criteria for DKD. Furthermore, the detection methods for these novel biomarkers are not standardized, and there is no unified consensus on diagnostic thresholds, hindering their clinical translation. In addition, DKD patients exhibit significant individual differences, with different pathological subtypes showing marked variations in disease progression rates and treatment responses. Existing biomarkers cannot achieve disease subtyping and prognostic prediction. Therefore, there is an urgent need to discover novel diagnostic biomarkers specific to DKD and establish a precise diagnostic and prognostic assessment system to support individualized clinical treatment, which has significant clinical value and application prospects.

[0005] Relevant patent documents retrieved: The publication, published in China (CA2926269C) on December 13, 2022, discloses biomarkers for kidney disease. These biomarkers include complement C3a dearginylation product (C3adesArg), macrophage inflammatory protein 1α (MIP-1α), C-reactive protein (CRP), cystatin C (CYSC), endothelial growth factor (EGF), liver fatty acid-binding protein 1 (FABP1), interleukin-8 (IL-8), soluble tumor necrosis factor receptor 1 (sTNFR1), soluble tumor necrosis factor receptor 2 (sTNFR2), D-dimer, and neutrophil gelatinase-associated lipotransferase (NGAL).

[0006] Relevant non-patent literature retrieved: The journal title is *Clinical Medical Research and Practice*, the article title is "Diagnostic Value of Pyruvate Kinase M1 and M2 in Patients with Diabetic Nephropathy", volume number 2025, 10(28), publication date 2025.10.16. This article discloses the detection of serum PKM1 and PKM2 expression levels using real-time quantitative polymerase chain reaction (RT-qPCR). Pearson correlation analysis was used to explore the association between PKM1, PKM2 and estimated glomerular filtration rate (eGFR) and urine albumin / creatinine ratio (UACR). Receiver operating characteristic (ROC) curves were used to evaluate the diagnostic efficacy of PKM1 and PKM2 for diabetic nephropathy. Results showed that the levels of PKM1 and PKM2 in the diabetic nephropathy group were lower than those in the healthy group and the diabetic group (P<0.05).

[0007] The prior art represented by the aforementioned documents has at least the following unresolved technical problems or defects: Current technologies have not constructed a specific combination of biomarkers for diabetes and diabetic nephropathy, and the etiological correlation of the biomarkers is insufficient, making it difficult to achieve accurate identification and disease course definition between the two. Summary of the Invention

[0008] The purpose of this invention is to provide: A combination of biomarkers and a diagnostic model for diabetic nephropathy based on model analysis, and related technologies, to solve the technical problems in existing technologies, such as the lack of specific biomarker combinations for diabetes and diabetic nephropathy, poor correlation between biomarkers and etiology, and the inability to accurately distinguish between the two and define the course of the disease, or a combination thereof.

[0009] Terminology Explanation: Unless otherwise defined, all technical terms used herein have the same meanings as commonly understood by one of ordinary skill in the art to which this subject matter pertains. Unless otherwise stated, all patents, patent inventions, and disclosures cited throughout this document are incorporated herein by reference in their entirety. Where multiple definitions exist for terms herein, the definitions provided in this chapter shall prevail.

[0010] It should be understood that the above brief description and the following detailed description are exemplary and for illustrative purposes only, and do not limit the subject matter of the invention in any way. In this invention, the singular is used in conjunction with the plural unless otherwise specifically stated. It should also be noted that, unless otherwise stated, the use of “or” or “or” means “and / or”. Furthermore, the use of the term “comprising” and other forms such as “including,” “containing,” and “contains” are not limiting.

[0011] Definitions of standard terms can be found in the reference books "Internal Medicine (10th Edition), People's Medical Publishing House, authors: Ge Junbo, Wang Chen, Wang Jian'an, 2024.06" and "Clinical Proteomics, Science Press, authors: Qiu Zongyin, Yin Yibing, 2008.04".

[0012] Unless otherwise stated, conventional methods within the scope of this art, such as ultracentrifugation, shall be used. Unless specifically defined, the use of all commercially available products used herein shall employ standard techniques. For example, they may be performed using the manufacturer's instructions for use with the kit, or in accordance with methods known in the art or the description of this invention. The techniques and methods described herein are generally performed according to conventional methods well known in the art, based on the descriptions in the various general and more specific documents cited and discussed in this specification.

[0013] The terms "optional / arbitrary" or "optionally / arbitrarily" mean that the event or situation subsequently described may or may not occur, including both the occurrence and non-occurrence of the event or situation. For example, according to the definition below: the diagnostic product includes any one or more of reagent kits, test strips, chips, devices, and detection systems. This indicates that the diagnostic product can be a reagent kit, or the product can be a test strip, or the product can be a chip, or the product can be a reagent kit, test strip, chip, device, and detection system.

[0014] The term "model-based biomarker combination" used in this article refers to a combined detection index consisting of two synergistic protein biomarkers. Specifically, it refers to the combination of P07148_FABP1 and P02144_MB proteins used to differentiate between type 2 diabetes and diabetic nephropathy and to enable early prediction of diabetic nephropathy. Diagnostic models can be constructed by detecting their expression levels, providing quantitative evidence for the prediction of diabetic nephropathy.

[0015] The term “P07148_FABP1” used in this article refers to Fatty acid-binding protein 1, FABP1, corresponding to UniProt accession number P07148. It is a human liver-type fatty acid-binding protein that participates in intracellular fatty acid transport and metabolic regulation and is a candidate biomarker related to liver injury and lipid metabolism disorders.

[0016] The term “P02144_MB” used in this article refers to Myoglobin, MB, corresponding to UniProt accession number P02144. Myoglobin is mainly found in cardiac and skeletal muscle and is responsible for binding and storing oxygen. It is a classic clinical biomarker for the early and rapid diagnosis of acute myocardial injury.

[0017] The term "diabetic nephropathy" used in this article refers to Diabetic Kidney Disease (DKD), a microvascular complication of the kidneys caused by long-term hyperglycemia. It is characterized by glomerular sclerosis and tubulointerstitial damage, and clinically manifests as proteinuria and progressive decline in renal function. It is one of the main causes of end-stage renal disease.

[0018] The term "type 2 diabetes" used in this article refers to Type 2 Diabetes Mellitus (T2DM), a metabolic disease with insulin resistance and relative insulin insufficiency as its core pathogenesis. It mainly has an onset in adulthood, is characterized by persistent hyperglycemia, and can induce chronic complications in multiple systems.

[0019] The term "ROC curve" used in this article refers to the Receiver Operating Characteristic Curve (ROC Curve), an evaluation curve plotted with the true positive rate (sensitivity) of the classification model on the vertical axis and the false positive rate (1-specificity) on the horizontal axis. The area under the curve (AUC) is the core quantitative indicator of the model's discriminative efficacy.

[0020] The term “data-independent acquisition” used in this article refers to Data Independent Acquisition (DIA), a mass spectrometry quantitative acquisition mode that divides the mass-to-charge ratio scanning range into a continuous window and performs uniform fragmentation detection on all ions within the window, achieving full coverage and reproducible quantification of peptides / proteins.

[0021] The term "parallel reaction monitoring" used in this article refers to Parallel Reaction Monitoring (PRM), a high-resolution targeted mass spectrometry quantitative technology that targets and fragments precursor ions, simultaneously detecting all fragment ions to achieve highly specific and accurate quantification of target proteins / peptides.

[0022] The term "logistic regression" used in this article refers to Logistic Regression (LR), a generalized linear statistical model that maps linear prediction results to a probability interval of 0-1 using a log-odds function. It is mainly used for binary and multi-class classification, risk factor analysis, and predictive modeling.

[0023] The term "Youden Index" used in this article refers to the Youden Index (YI), a comprehensive evaluation index for diagnostic tests. It is calculated as sensitivity + specificity - 1, and its value ranges from 0 to 1. A higher value indicates higher diagnostic accuracy and is used to determine the optimal diagnostic threshold.

[0024] The term "liquid chromatography-mass spectrometry" used in this article refers to Liquid Chromatography-Mass Spectrometry (LC-MS), which combines the separation capabilities of high-performance liquid chromatography with the qualitative and quantitative capabilities of mass spectrometry for the separation and detection of proteins, peptides, and metabolites in complex biological samples.

[0025] The term “mobility peak detection threshold” used in this article refers to Mobility Peak Detection Threshold (MPDT), which is the minimum response intensity threshold used in ion mobility separation-mass spectrometry to determine whether a mobility dimension signal peak is a valid detection peak, and is used to distinguish the real signal from background noise.

[0026] In a first aspect, the present invention provides: a combination of markers based on model analysis.

[0027] This includes: a combination of biomarkers based on model analysis.

[0028] Among them, the biomarker combination based on model analysis consists of P07148_FABP1 and P02144_MB.

[0029] Based on further solutions to the technical problems of the present invention, or simultaneous solutions to multiple technical problems, the preferred solution in the technical solution provided in the first aspect of the present invention includes: The first preferred option: The biomarker combination based on model analysis consists of P07148_FABP1 and P02144_MB. This technical solution, in addition to solving the technical problems of "the lack of specific biomarker combinations for diabetes and diabetic nephropathy in the existing technology, poor correlation between biomarkers and etiology, and inability to accurately distinguish between the two and define the course of the disease", further solves the technical problem of "providing a specific biomarker combination for diagnosing diabetic nephropathy".

[0030] Secondly, the present invention provides a diagnostic model for diabetic nephropathy.

[0031] This includes: diagnostic models.

[0032] The diagnostic model is constructed using computer algorithms with the expression levels of the above-mentioned biomarker combinations as input parameters.

[0033] Based on further solutions to the technical problems of the present invention, or simultaneous solutions to multiple technical problems, the preferred solution in the technical solution provided in the second aspect of the present invention includes: The first preferred solution is a diagnostic model for diabetic nephropathy, which is constructed using a computer algorithm with the expression levels of the aforementioned biomarker combination as input parameters. This solution addresses the technical problems of "the lack of specific biomarker combinations for diabetes and diabetic nephropathy in existing technologies, poor correlation between biomarkers and etiology, and inability to accurately distinguish between the two and define the disease course," and further solves the technical problem of "providing a diagnostic model for diabetic nephropathy."

[0034] Thirdly, the present invention provides a method for constructing a diagnostic model.

[0035] This includes: construction methods.

[0036] The construction method includes the following steps: S1. Collect samples from the diabetes group, diabetic nephropathy group and control group, and extract proteins; S2. Detect the expression levels of P07148_FABP1 and P02144_MB in the proteins obtained in step S1, and obtain quantitative data; S3. Using the quantitative data obtained in step S2 as the training set, a computer algorithm is used to train and obtain a diagnostic model for diabetic nephropathy.

[0037] Specifically, the samples mentioned in step S1 include any one or more of the following: urine, blood, blood spots, oral cells, semen, sperm spots, bones, hair, saliva, saliva spots, sweat, and amniotic fluid containing fetal cells.

[0038] Preferably, the sample mentioned in step S1 is urine.

[0039] Specifically, the computer algorithms described in step S3 include any one or more of the following: logistic regression, support vector machine, random forest, decision tree, gradient boosting tree, neural network, K-nearest neighbors, and Naive Bayes.

[0040] Preferably, the computer algorithm described in step S3 is a logistic regression algorithm.

[0041] Based on further solutions to the technical problems of the present invention, or simultaneous solutions to multiple technical problems, the preferred solution in the technical solution provided in the third aspect of the present invention includes: The first preferred embodiment is a method for constructing a diagnostic model for diabetic nephropathy, comprising the following steps: S1, collecting samples from a diabetic group, a diabetic nephropathy group, and a control group, and extracting proteins; S2, detecting the expression levels of P07148_FABP1 and P02144_MB in the proteins obtained in step S1, and obtaining quantitative data; S3, using the quantitative data obtained in step S2 as a training set, training the model using a computer algorithm to obtain a diagnostic model for diabetic nephropathy. This technical solution, while addressing the existing technical problems of "no specific biomarker combination for diabetes and diabetic nephropathy, poor correlation between biomarkers and etiology, and inability to accurately distinguish between the two and define the disease course," further solves the technical problem of "providing a method for constructing a diagnostic model for diabetic nephropathy."

[0042] Fourthly, the present invention provides the application of the above-mentioned combination of biomarkers in the preparation of diagnostic products for diabetic nephropathy.

[0043] This includes: Diagnostic products for diabetic nephropathy.

[0044] Among them, the diagnostic product for diabetic nephropathy is used to differentiate between diabetic nephropathy and type 2 diabetes.

[0045] Based on further solutions to the technical problems of the present invention, or simultaneous solutions to multiple technical problems, the preferred solution in the technical solution provided in the fourth aspect of the present invention includes: The first preferred solution is the application of the above-mentioned biomarker combination in the preparation of diagnostic products for diabetic nephropathy. This technical solution, based on solving the technical problems of "the lack of specific biomarker combinations for diabetes and diabetic nephropathy in the existing technology, poor correlation between biomarkers and etiology, and inability to accurately distinguish between the two and define the course of the disease", further solves the technical problem of "providing specific applications of the above-mentioned biomarker combination".

[0046] Fifthly, the present invention provides: a diagnostic product for diabetic nephropathy.

[0047] This includes: diagnostic products.

[0048] The diagnostic products include any one or more of reagent kits, test strips, chips, devices, and detection systems.

[0049] Specifically, the diagnostic product also includes reagents for detecting the expression levels of P07148_FABP1 and P02144_MB.

[0050] Based on further solutions to the technical problems of the present invention, or simultaneous solutions to multiple technical problems, the preferred solution in the technical solution provided in the fifth aspect of the present invention includes: The first preferred embodiment is a diagnostic product for diabetic nephropathy, comprising the aforementioned biomarker combination. This technical solution, while addressing the existing technical problems of "the lack of specific biomarker combinations for diabetes and diabetic nephropathy, poor correlation between biomarkers and etiology, and inability to accurately distinguish between the two and define the disease course," further addresses the technical problem of "providing a specific product for diagnosing diabetic nephropathy."

[0051] In a sixth aspect, the present invention provides a method for detecting the expression level of the above-mentioned combination of biomarkers.

[0052] This includes: methods.

[0053] The method includes any one or more of the following: mass spectrometry, chromatography, surface-enhanced Raman spectroscopy, Western blotting, flow cytometry, protein array, immunoprecipitation, and immunoadsorption.

[0054] Preferably, the method is any one or more methods selected from mass spectrometry and chromatography.

[0055] More preferably, the method is a liquid chromatography-mass spectrometry (LC-MS) analytical method.

[0056] Based on further solutions to the technical problems of the present invention, or simultaneous solutions to multiple technical problems, the preferred solution in the technical solution provided in the fifth aspect of the present invention includes: The first preferred embodiment is a method for detecting the expression level of the above-mentioned biomarker combination, wherein the method includes any one or more of the following methods: mass spectrometry, chromatography, surface-enhanced Raman spectroscopy, Western blotting, flow cytometry, protein array, immunoprecipitation, and immunoadsorption. This technical solution, based on solving the technical problems of "the lack of specific biomarker combinations for diabetes and diabetic nephropathy in the prior art, poor correlation between biomarkers and etiology, and inability to accurately distinguish between the two and define the disease course," further solves the technical problem of "providing a specific method for detecting and diagnosing biomarker combinations for diabetic nephropathy."

[0057] In this invention, embodiments 1-4 at least support the protection scope of "model-based marker combination".

[0058] The term "model-based biomarker combination" is derived from the foregoing explanation and / or the corresponding "P07148_FABP1 and P02144_MB" in Examples 1-4. Therefore, those skilled in the art can reasonably infer that "model-based biomarker combination," its subordinate concepts, its substantially equivalent technical means, and technical means that can replace it within the scope of conventional technical means and common knowledge based on the existing level of technology should all fall within the protection scope of "model-based biomarker combination." Replacing "model-based biomarker combination" with "model-based protein molecule combination," etc., still falls within the protection scope of this invention.

[0059] Examples 1-4 of this invention at least support the protection scope of "diabetic nephropathy diagnostic model".

[0060] The term "diagnostic model for diabetic nephropathy" is derived from the foregoing explanation and / or the corresponding statements in Examples 1-4, such as "the diagnostic model constructed based on the above-mentioned biomarker combination and logistic regression algorithm has high diagnostic efficacy." Therefore, those skilled in the art can reasonably infer that the "diagnostic model for diabetic nephropathy," its subordinate concepts, its essentially equivalent technical means, and technical means that can replace it within the scope of conventional technical means and common knowledge based on the existing technical level, should all fall within the protection scope of the "diagnostic model for diabetic nephropathy." Replacing the "diagnostic model for diabetic nephropathy" with the "diagnostic model for DKD," etc., still falls within the protection scope of this invention.

[0061] Examples 1-4 of this invention at least support the protection scope of "diagnostic products for diabetic nephropathy".

[0062] The term "diagnostic product for diabetic nephropathy" is derived from the foregoing explanation and / or the corresponding "reagent kit," "device," and "test strip" in Examples 1-4. Therefore, those skilled in the art can reasonably infer that "diagnostic product for diabetic nephropathy," its subordinate concepts, its essentially equivalent technical means, and technical means that can replace it within the scope of conventional and common knowledge based on the existing level of technology should all fall within the protection scope of "diagnostic product for diabetic nephropathy." Replacing "diagnostic product for diabetic nephropathy" with "diagnostic reagent kit for diabetic nephropathy," "diagnostic test strip for diabetic nephropathy," etc., still falls within the protection scope of this invention.

[0063] Examples 1-4 of this invention at least support the protection scope of "method for detecting the expression level of a combination of biomarkers".

[0064] The term "method for detecting the expression level of a combination of biomarkers" is derived from the foregoing explanation and / or the corresponding "chromatographic analysis" and "mass spectrometry analysis" in Examples 1-4. Therefore, those skilled in the art can reasonably infer that the "method for detecting the expression level of a combination of biomarkers," its subordinate concepts, its essentially equivalent technical means, and technical means that can replace it within the scope of conventional technical means and common knowledge based on the existing technical level should all fall within the protection scope of the "method for detecting the expression level of a combination of biomarkers." Replacing the "method for detecting the expression level of a combination of biomarkers" with "method for detecting the expression level of a combination of proteins predicting diabetic complications," etc., still falls within the protection scope of this invention.

[0065] The present invention has at least the following beneficial effects: Compared to existing technologies, this invention significantly improves the differentiation between type 2 diabetes mellitus (DM) and early-stage diabetic kidney disease (DKD). Experimental tests show that its AUC value, sensitivity, and specificity are excellent, effectively avoiding misdiagnosis and missed diagnosis, accurately defining the disease stage, and providing strong evidence for precision clinical diagnosis and treatment.

[0066] Furthermore, based on the present invention: Based on the comparison of Example 4 and Comparative Examples 1-3, the present invention uses technical means to combine P07148_FABP1 and P02144_MB to construct a biomarker combination based on model analysis, achieving new technical effects: better differentiation between type 2 diabetes mellitus (DM) and diabetic nephropathy (DKD).

[0067] Considering the possibility of this invention entering other countries, this invention also provides the following technical solutions: A method for diagnosing diabetic nephropathy, the method comprising using a combination of biomarkers, diagnostic models, or diagnostic products of the present invention.

[0068] Specifically, the method includes obtaining the expression levels of P07148_FABP1 and P02144_MB in the sample to be tested for the diagnosis of diabetic nephropathy.

[0069] Preferably, the method is used to differentiate between diabetes and diabetic nephropathy. Attached Figure Description

[0070] Figure 1 This is the ROC curve for Example 4.

[0071] Figure 2 This is the ROC curve for Comparative Example 1.

[0072] Figure 3 This is the ROC curve for Comparative Example 2.

[0073] Figure 4 This is the ROC curve for Comparative Example 3. Detailed Implementation

[0074] The following non-limiting embodiments are intended to enable those skilled in the art to gain a more comprehensive understanding of the present invention, but do not limit the invention in any way. The following content is merely an exemplary description of the scope of protection claimed by the present invention, and those skilled in the art can make various changes and modifications to the present invention based on the disclosed content, and such changes should also fall within the scope of protection claimed by the present invention.

[0075] The present invention will be further described below by way of specific embodiments. Unless otherwise specified, all instruments, devices, equipment, reagents, products, etc., used in the embodiments of the present invention are obtained through conventional commercial means.

[0076] Basic Example 1: Sample Collection and Allocation 1. Inclusion criteria for the diabetes group, diabetic nephropathy group, and control group 1.1 Inclusion criteria for the diabetes group: The recommendations in the "Guidelines for the Prevention and Treatment of Type 2 Diabetes in China (2020 Edition)" are shown in Table 1: Table 1

[0077] Note: OGTT in the table refers to the oral glucose tolerance test; HbA1c refers to glycated hemoglobin. Typical symptoms of diabetes include polydipsia, polyuria, polyphagia, and unexplained weight loss; random blood glucose refers to blood glucose at any time of day, regardless of the time of the last meal, and cannot be used to diagnose impaired fasting glucose or impaired glucose tolerance; fasting status refers to at least 8 hours without caloric intake. It is recommended that glycated hemoglobin be measured in medical institutions that use standardized testing methods and have strict quality control (US National Glycated Hemoglobin Standardization Program, China Glycated Hemoglobin Consistency Study Program).

[0078] 1.2 Inclusion criteria for the diabetic nephropathy group: Clinical diagnostic criteria based on the "Guidelines for the Prevention and Treatment of Diabetic Nephropathy in China (2021 Edition)" issued by the Chinese Diabetes Society and the "Chinese Guidelines for Clinical Diagnosis and Treatment of Diabetic Nephropathy" issued by the Chinese Nephrology Society in 2021: Diabetic nephropathy can be diagnosed if at least one of the following criteria is met, provided that diabetes is clearly the cause of kidney damage and other causes of chronic kidney disease have been ruled out: (1) Excluding interfering factors, at least two out of three tests within 3-6 months have a urine albumin / creatinine ratio (UACR) ≥230mg / g or a 24h urine albumin excretion rate (UAER) ≥30mg / 24h (20μg / min).

[0079] (2) Estimated glomerular filtration rate (eGFR, using the CKD-EPI formula) < 60 ml·min -1 · (1.73m) 2 ) -1 Lasting for more than 3 months.

[0080] (3) The renal biopsy is consistent with the pathological changes of diabetic nephropathy.

[0081] Case inclusion criteria: (1) Age 30-70 years, gender matched; (2) eGFR≥45ml min -1 (1.73m2 ) -1 ; (3) Meets the clinical diagnostic criteria for diabetic nephropathy and is classified accordingly; (4) The kidney biopsy results are consistent with the diagnosis of diabetic nephropathy and are classified accordingly; (5) The patient has good compliance and agrees to and signs the informed consent form.

[0082] 1.3 Inclusion criteria for the control group: (1) Age 18-75 years old, gender not limited, no underlying diseases; physical examination, routine biochemical tests, electrocardiogram, CT, ultrasound and other auxiliary examinations showed no obvious abnormalities; (2) Blood pressure: 90-140 / 60-90 mmHg; (3) Fasting blood glucose: 3.9-6.1 mmol / L; (4) Body Mass Index: 18.5-23.9 (BMI); (5) Heart rate: 60-100 beats / minute at rest; (6) Body temperature: 36-37 degrees Celsius.

[0083] 2. Sample collection Sampling criteria: Collect the first midstream urine of the study subjects in the morning: discard the first and last urination segments, collect only the midstream urine in a sterile container, immediately label and aliquot it into 50ml tubes, and collect a total of 100ml per sample. Aliquot the sample into two tubes and store them in a -80℃ freezer for later use, avoiding repeated freeze-thaw cycles.

[0084] 3. Sample allocation The training and validation sets included a total of 120 samples, with 50 samples in the diabetes group, 50 samples in the diabetic nephropathy group, and 20 samples in the control group.

[0085] Eighty samples were selected from the sample as the training set for proteomics experiments, including 30 samples in the diabetes group, 30 samples in the diabetic nephropathy group, and 20 samples in the control group.

[0086] Forty samples were selected from the sample as the validation set for the PRM experiment, including 20 samples from the diabetes group and 20 samples from the diabetic nephropathy group.

[0087] Example 1: Proteomics Experimental Procedure 1. Preparation of reagents used in this invention (1) Lysis buffer: containing 6M urea and 2M thiourea, prepared by dissolving urea and thiourea in 0.1M ABB.

[0088] (2) 0.2M TCEP solution: Dissolve tris(2-carboxyethyl)phosphine hydrochloride in 0.1M ABB (ammonium bicarbonate) to obtain 0.2M TCEP solution.

[0089] (3) 0.8M IAA solution: Dissolve iodoacetamide in 0.1M ABB to obtain 0.8M IAA solution.

[0090] (4) Trypsin enzyme solution: 100 μg of trypsin protease was dissolved in 200 μL of 0.1 mM ABB to obtain a trypsin enzyme solution with a concentration of 0.5 μg / μL.

[0091] (5) rLys-C enzyme solution: 50 μg rLys-C protease was dissolved in 200 μL 0.1 mM ABB to obtain an rLys-C enzyme solution with a concentration of 0.5 μg / μL.

[0092] 2. Sample pretreatment (1) The urine samples of the training set stored at -80℃ in the basic example 1 were ultracentrifuged at 100,000g for 70 min, and the supernatant was collected to obtain exosome samples. 200 μL of exosome sample was added to a 3KD ultrafiltration tube, 200 μL of 0.1 MABB to remove glycerol was added, and the sample was centrifuged at 12,000g for 30 min until the remaining liquid in the ultrafiltration tube was 50 μL. Then, 200 μL of lysis buffer was added to replace the liquid, and the sample was transferred to a 1.5 mL EP tube to obtain the replacement lysis buffer.

[0093] (2) In each PCT tube, add 30 μL of the above-mentioned replacement lysis buffer and mix well; add 5 μL of 0.2 MTCEP solution and 2.5 μL of 0.8 M IAA solution to each tube and mix well; add 75 μL of 0.1 M ABB, 10 μL of trypsin enzyme solution and 2.5 μL of rLys-C enzyme solution to each tube, and make up to 150 μL with 0.1 M ABB, and adjust the pH of the system to 8.0. Perform the enzymatic digestion operation at 20 kpsi, 50 seconds of high pressure, 10 seconds of normal pressure, 120 pressure cycles, and 30 °C.

[0094] (3) After enzymatic hydrolysis, the sample was transferred to a 1.5 mL EP tube and 15 μL of 10% TFA solution was added to each tube. The enzymatic hydrolysis was terminated when the final TFA concentration was 1%, and the hydrolysate was obtained.

[0095] 3. Desalination treatment The pH of the above enzymatic hydrolysate was adjusted to ensure it was between 2 and 3. Desalting was performed according to the user guide provided by the manufacturer of SOLAμ solid-phase extraction SPE plates (Thermo Fisher Scientific™, San Jose, USA). The specific steps were as follows: activation of the desalting column with 200 μL MeOH × 2 times; equilibration of the desalting column with 200 μL 80% ACN and 0.1% TFA × 2 times; washing the desalting column with 200 μL 2% ACN and 0.1% TFA × 2 times; desalting after sample loading with 200 μL 2% ACN and 0.1% TFA × 10 times; collecting the sample with 100 μL 40% ACN and 0.1% TFA × 2 times, centrifuging and concentrating at 40°C and below 10 mBar until the sample was dry, reconstituted, and peptide content was measured at A280 wavelength.

[0096] 4. Fractionation treatment Take the peptide solutions from each sample whose content has been determined in step 3, mix them in equal amounts to prepare a mixed peptide sample with a total amount of 100 μg, which will be used to construct a spectral library.

[0097] The above-mentioned mixed peptide sample was injected into a DIONEX UltiMate™ 3000 liquid chromatography system and separated using an XBridge Peptide BEH C18 column (300 Å, 5 μm × 4.6 mm × 250 mm, Waters, Milford, MA, USA) with the following parameters set: Mobile phase: Mobile phase A was a 10 mM ammonium hydroxide aqueous solution (pH=10), and mobile phase B was a mixture of 98% CAN and 10 mM ammonium hydroxide (pH=10); Flow rate: 0.5 mL / min; Gradient elution program: A gradient of 5% to 35% ACN was used over 60 minutes; Collection and pooling: Starting from the gradient elution, eluted fractions were collected every minute for a total of 60 fractions. Adjacent fractions were then pooled in elution order to obtain 30 pooled fractions. After drying, the 30 pooled fractions were resuspended in 2% ACN and 0.1% formic acid, respectively, for subsequent analysis.

[0098] 5. Mass spectrometry analysis Liquid chromatography-mass spectrometry (LC-MS) analysis was performed using a UHPLC (Bruker Daltonics, Germany) system and a timsTOF Pro mass spectrometer (Bruker Daltonics, Germany). Data acquisition was independent (DIA); library construction was performed using data-dependent acquisition (DDA). Mobile phase A: 100% water, 0.1% formic acid; Mobile phase B: 100% acetonitrile, 0.1% formic acid. All reagents were mass spectrometry grade.

[0099] During DDA collection for library construction, peptides were first loaded onto a pre-column (5 mm) at a pressure of 217.5 bar. 300 µmi.d.), then injected into the analytical column (1.9 µm, 120 Å, 150 mm) at a flow rate of 300 nL / min. Analysis was performed using a 60-minute liquid chromatography gradient (0-50 min, 5%-27% mobile phase B; 50-60 min, 27%-40% mobile phase B) at 75 µm id. Mass spectrometry parameters were as follows: PASEF MS and MS / MS mass scan range 100-1700 m / z, 1 / k0 scan range 0.6-1.6, and mobility peak detection threshold 5000; PASEF MS / MS scan number 10, charge range 0-5, and peak detection threshold 2500 cts / s.

[0100] During DIA collection of all samples, the peptides were first loaded onto a pre-column (5 mm) at a pressure of 217.5 bar. 300 µmi.d.), then injected into the analytical column (1.9 µm, 120 Å, 150 mm) at a flow rate of 300 nL / min. Analysis was performed using a 60-minute liquid chromatography gradient (0-50 min, 5%-27% mobile phase B; 50-60 min, 27%-40% mobile phase B) at 75 µm id. Mass spectrometry parameters were as follows: PASEF MS mass scan range 100-1700 m / z, 1 / k0 scan range 0.7-1.3, resolution 60000, mobility peak detection threshold 5000; PASEF MS / MS scan number 10, charge range 0-5, peak detection threshold 2,500 cts / s, window number 56.

[0101] 6. Data Analysis Mass spectrometry data were searched using DIA-NN software (version 1.8.1) with the "match-between-run" (MBR) function enabled. The DDA spectral library constructed from the uniprot fasta file (2023-07-25-reviewed-contam-UP000005640_human_pd.fasta) was used for the search analysis. Cysteine ​​carbamidomethylation was set as a fixed modification, and methionine oxidation was set as a variable modification. The analysis results provide qualitative and quantitative data, and were screened using a strict false discovery rate (FDR) criterion of less than 0.01.

[0102] Example 2 PRM Experiment 1. Sample pretreatment, desalting, and fractionation The pretreatment, desalting, and fractionation of the PRM experimental samples were performed in accordance with the methods described in Example 2, with the steps being completely identical. The only difference was that the samples used were urine samples from the validation set in Example 1.

[0103] 2. LC-MS / MS detection LC-MS / MS analysis was performed using an UltiMate 3000 RSLCnano liquid chromatography system coupled with an OrbitrapExploris™ 480 mass spectrometer (Thermo Scientific™, San Jose, USA), equipped with FAIMS Pro™ (Thermo Scientific™, San Jose, USA).

[0104] FAIMS compensation voltage (CV) was set to -65V and -45V. Buffer A: 2% acetonitrile, 98% water, containing 0.1% fatty acid; Buffer B: 80% acetonitrile aqueous solution (containing 0.1% fatty acid). All reagents were mass spectrometry grade.

[0105] At each acquisition, the peptide was loaded into a pre-column (3µm, 100Å, 20mm × 75µm inner diameter) at a flow rate of 6μL / min, followed by injection at a flow rate of 300nL / min using a 90-minute liquid chromatography gradient (buffer B from 8% to 35%) (analytical column, 1.9µm, 120Å, 150mm × 75µm inner diameter). MS1 had an m / z range of 350–1010, a resolution of 60,000, a normalized AGC target of 300%, and a maximum ion implantation time (maximum IT) of 100 ms. MS / MS experiments had a resolution of 30,000, a normalized AGC target of 1000%, and a maximum IT of 100 ms.

[0106] 3. Data Analysis Import the PRM mass spectrometry data acquired by LC-MS / MS detection in step 2 into Skyline software (version 22.2.0.527), export the peptide matrix, and extract key information columns: protein name, peptide sequence, and total area fragmentation, for subsequent biomarker validation.

[0107] Example 3: Establishment and Validation of Marker Combinations and Their Models Based on the DIA proteome quantitative data obtained in Example 1, 40 key candidate proteins were screened, as shown in Table 2. PRM validation was performed in Example 2, and 33 of these proteins were successfully identified. Seven proteins—O15127_SCAMP2, 5A3E0_POTEF, P18754_RCC1, P32926_DSG3, A0A075B6H7_IGKV3-7, O95568_METTL18, and Q92900_UPF1—were not identified by the PRM method and were excluded from subsequent model construction.

[0108] Table 2

[0109] A threshold of 0.5 was uniformly selected: when the diagnostic probability value of the combination of biomarkers is ≥0.5, it is determined to be DKD; when the diagnostic probability value of the combination of biomarkers is <0.5, it is determined to be DM.

[0110] Example 4: A combination of biomarkers based on model analysis This invention provides a biomarker combination based on model analysis constructed from the above candidate proteins, consisting of P07148_FABP1 and P02144_MB.

[0111] The DIA proteomics quantitative data (30 cases in the diabetes group and 30 cases in the diabetic nephropathy group) from Example 1 were used as the training set, and the Logistic Regression algorithm was used for model training. The software version was Python 3.9.13 and sklearn 1.1.2. The model parameters were set as follows: {'C': 0.5, 'max_iter': 100, 'penalty': 'l1', 'solver': 'liblinear'.

[0112] The trained model was validated using its own training set data and the PRM validation cohort data (20 cases in the diabetes group and 20 cases in the diabetic nephropathy group) from Example 2. The results showed that the diagnostic model constructed based on the above biomarker combination and logistic regression algorithm has high diagnostic efficacy and exhibits good diagnostic stability in the PRM validation cohort. The ROC curve is shown in Figure 1. Figure 1 As shown.

[0113] Comparative Example 1 A biomarker combination based on model analysis, consisting of P02144_MB and O95999_BCL10, was used. The DIA proteomics quantitative data (30 cases in the diabetes group and 30 cases in the diabetic nephropathy group) from Example 1 were used as the training set, and the Logistic Regression algorithm was employed for model training. Software versions: Python 3.9.13, sklearn 1.1.2; model parameters were set as follows: 'C': 4.4, 'max_iter': 100, 'penalty': 'l1', 'solver': 'liblinear'.

[0114] The trained model was validated using its own training set data and the PRM validation cohort data (20 cases in the diabetes group and 20 cases in the diabetic nephropathy group) from Example 2. The results showed that the diagnostic model constructed based on the above biomarker combination and logistic regression algorithm has high diagnostic efficacy and exhibits good diagnostic stability in the PRM validation cohort. The ROC curve is shown in Figure 1. Figure 2 As shown.

[0115] Comparative Example 2 A biomarker combination based on model analysis, consisting of P07148_FABP1 and O95999_BCL10, was used as the training set. The model was trained using the quantitative DIA proteomics data (30 cases in the diabetes group and 30 cases in the diabetic nephropathy group) from Example 1. The software version was Python 3.9.13, sklearn 1.1.2, and the model parameters were set as follows: 'C': 0.6, 'max_iter': 100, 'penalty': 'l2', 'solver': 'liblinear'.

[0116] The trained model was validated using its own training set data and the PRM validation cohort data (20 cases in the diabetes group and 20 cases in the diabetic nephropathy group) from Example 2. The results showed that the diagnostic model constructed based on the above biomarker combination and logistic regression algorithm has high diagnostic efficacy and exhibits good diagnostic stability in the PRM validation cohort. The ROC curve is shown in Figure 1. Figure 3 As shown.

[0117] Comparative Example 3 A biomarker combination based on model analysis, consisting of P00450_CP and O95999_BCL10, was used as the training set. The model was trained using the quantitative DIA proteomics data (30 cases in the diabetes group and 30 cases in the diabetic nephropathy group) from Example 1. The software versions were Python 3.9.13 and sklearn 1.1.2. The model parameters were set as follows: 'C': 0.1, 'max_iter': 100, 'penalty': 'l2', 'solver': 'liblinear'.

[0118] The trained model was validated using its own training set data and the PRM validation cohort data (20 cases in the diabetes group and 20 cases in the diabetic nephropathy group) from Example 2. The results showed that the diagnostic model constructed based on the above biomarker combination and logistic regression algorithm has high diagnostic efficacy and exhibits good diagnostic stability in the PRM validation cohort. The ROC curve is shown in Figure 1. Figure 4 As shown.

[0119] Example 4, the area under the curve (AUC value) of Comparative Examples 1-3 are shown in Table 3: Table 3

[0120] The closer the AUC value is to 1, the stronger the ability of the biomarker combination to distinguish between type 2 diabetes mellitus (DM) and diabetic nephropathy (DKD). The results showed that the biomarker combination in Comparative Example 3, P00450_CP and O95999_BCL10, had an AUC value of 0.580. In Comparative Example 2, replacing P00450_CP in the biomarker combination of Comparative Example 3 with P07148_FABP1 increased the AUC value by 4.31% compared to Comparative Example 3. In Comparative Example 1, replacing P00450_CP in the biomarker combination of Comparative Example 3 with P02144_MB increased the AUC value by 13.79% compared to Comparative Example 3. In Example 4 of this invention, replacing P00450_CP and O95999_BCL10 in the biomarker combination of Comparative Example 3 with P07148_FABP1 and P02144_MB of this invention increased the AUC value by 26.21% compared to Comparative Example 3. It is evident that the combination of P07148_FABP1 and P02144_MB in this invention has a synergistic effect.

[0121] A validation set of 20 diabetic samples and 20 diabetic nephropathy samples was used to calculate the diagnostic probability of DKD for each clinical sample and determine the test results. The results are shown in Table 4 below: Table 4

[0122] Finally, it should be noted that the above content is only used to illustrate the technical solution of the present invention, and is not intended to limit the scope of protection of the present invention. Simple modifications or equivalent substitutions made by those skilled in the art to the technical solution of the present invention do not depart from the essence and scope of the technical solution of the present invention.

Claims

1. A combination of biomarkers based on model analysis, characterized in that, The marker combination consists of P07148_FABP1 and P02144_MB.

2. A diagnostic model for diabetic nephropathy, characterized in that, The diagnostic model is constructed using a computer algorithm with the expression levels of the biomarker combination described in claim 1 as input parameters.

3. The method for constructing the diagnostic model according to claim 2, characterized in that, Includes the following steps: S1. Collect samples from the diabetes group, diabetic nephropathy group and control group, and extract proteins; S2. Detect the expression levels of P07148_FABP1 and P02144_MB in the proteins obtained in step S1, and obtain quantitative data; S3. Using the quantitative data obtained in step S2 as the training set, a computer algorithm is used to train and obtain a diagnostic model for diabetic nephropathy.

4. The construction method according to claim 3, characterized in that, The samples mentioned in step S1 include any one or more of the following: urine, blood, blood spots, oral cells, semen, sperm spots, bones, hair, saliva, saliva spots, sweat, and amniotic fluid containing fetal cells.

5. The construction method according to claim 3, characterized in that, The computer algorithms mentioned in step S3 include any one or more of the following: logistic regression, support vector machine, random forest, decision tree, gradient boosting tree, neural network, K-nearest neighbors, and Naive Bayes.

6. The use of the biomarker combination of claim 1 in the preparation of diagnostic products for diabetic nephropathy.

7. A diagnostic product for diabetic nephropathy, characterized in that, The diagnostic product includes the biomarker combination as described in claim 1.

8. The diagnostic product according to claim 7, characterized in that, The diagnostic products include any one or more of the following: reagent kits, test strips, chips, devices, and detection systems.

9. The diagnostic product according to claim 7, characterized in that, The diagnostic product also includes reagents for detecting the expression levels of P07148_FABP1 and P02144_MB.

10. A method for detecting the expression level of the biomarker combination of claim 1, characterized in that, The method includes any one or more of the following: mass spectrometry, chromatography, surface-enhanced Raman spectroscopy, Western blotting, flow cytometry, protein array, immunoprecipitation, and immunoadsorption.

Images