Primary membranous nephropathy prognosis biomarker and application thereof

Abstract

The invention belongs to the technical field of biomedicine, and discloses a primary membranous nephropathy prognosis biomarker and application. The prognostic marker for the primary membranous nephropathy comprises paraoxonase 1, beta-actin-like protein 2, root protein and tripeptidyl peptidase 1, and the expression level of the prognostic marker is remarkably related to clinical remission of PMN patients. Based on the prognostic marker, a combined prognostic risk model integrating urine protein concentration and clinical characteristics is established. Patients with primary membranous nephropathy can be divided into a low-risk group and a high-risk group according to the risk score, and the high-risk group is proved to have poorer prognosis. The invention also provides a prognosis evaluation product and a prognosis evaluation model containing the biomarker. The method provides a reliable basis for individualized treatment decision of PMN patients, and has a wide clinical application prospect.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of biomedical technology, and particularly relates to a primary membranous nephropathy prognosis biomarker and application. BACKGROUND

[0002] Primary membranous nephropathy (PMN) is a common cause of adult idiopathic nephrotic syndrome (massive proteinuria, hypoalbuminemia, edema, hyperlipidemia) and is an autoimmune disease characterized by diffuse and non-uniform immune complex deposition under the epithelium of the glomerular basement membrane (GBM) with diffuse thickening of the basement membrane. These deposits are mainly formed by pathogenic autoantibodies binding to target antigens. The main manifestations are massive proteinuria, edema, and hypoalbuminemia, and most patients need to be diagnosed by kidney biopsy.

[0003] The treatment of primary membranous nephropathy is mainly individualized supportive treatment, and some need to be combined with immunosuppressive agents. The main treatment goals are to reduce proteinuria, protect renal function, and prevent complications. Reliable prognostic evaluation tools for primary membranous nephropathy are still limited. Although traditional clinical indicators and blood markers have been widely used, such as proteinuria duration and level, renal function trend, and anti-PLA2R antibody titer, they still cannot fully reflect the individualized treatment response of patients, and identifying multidimensional risk factors can further guide treatment. SUMMARY

[0004] The purpose of the present application is to overcome the shortcomings of the prior art and provide a primary membranous nephropathy prognosis biomarker and application.

[0005] To achieve the above-mentioned purpose, the technical solutions adopted by the present application are as follows: In a first aspect, the present application provides a primary membranous nephropathy prognosis biomarker, which comprises paraoxonase 1, beta-actin like protein 2, radixin, and tripeptidyl peptidase 1.

[0006] The present application provides four protein expression levels as primary membranous nephropathy prognosis biomarkers for the prognosis evaluation of the disease, including PON1 (paraoxonase 1), ACTBL2 (beta-actin like protein 2), RDX (radixin), and TPP1 (tripeptidyl peptidase 1).

[0007] As a preferred embodiment of the prognostic marker of the present application, the prognostic marker is from urine. Preferably, the urine is from the midstream morning urine of the subject.

[0008] As a filtrate of blood, urine can reflect the degree and functional status of kidney injury earlier and more sensitively, while avoiding the invasive operation of blood collection. In addition, using nanoHPLC-MS / MS technology can effectively reduce the protein interference from blood, significantly improve the enrichment efficiency of kidney-derived proteins, and accurately capture the extremely low abundance of proteins in urine. It can better predict the prognosis of PMN patients, thereby helping to adjust the treatment plan and contributing to individualized precision treatment.

[0009] As a preferred embodiment of the prognostic marker of the present application, it also includes serum anti-PLA2R antibody titer, age and eGFR.

[0010] In a second aspect, the present application applies the prognostic marker of the first aspect to the preparation of a product for evaluating or predicting the prognosis of primary membranous nephropathy.

[0011] In a third aspect, the present application provides a product for evaluating or predicting the prognosis of primary membranous nephropathy, comprising a preparation for detecting the expression level of the prognostic marker of the first aspect.

[0012] As a preferred embodiment of the product of the present application, it includes a diagnostic agent, a kit, a chip, a test paper or a well plate.

[0013] In a fourth aspect, the present application applies the prognostic marker of the first aspect to the construction of a system for evaluating or predicting the prognosis of primary membranous nephropathy.

[0014] In a fifth aspect, the present application provides a system for evaluating or predicting the prognosis of primary membranous nephropathy, comprising a sample prognostic marker detection module, a data processing module, a multi-dimensional data integration module and a result output module; the sample prognostic marker detection module is used to detect the expression level of the prognostic marker of the first aspect.

[0015] Among them, the data processing module is configured for the pretreatment of midstream morning urine, including centrifugal slag removal, protein enrichment and buffer stabilization units; the prognostic marker detection module is used to quantitatively detect the expression level of the prognostic marker of the first aspect; the multi-dimensional data integration module is an internal clinical feature input interface, which is used to calculate the joint risk score in combination with the prognostic marker and clinical features of the first aspect; the result output module is used to provide a visual risk prognosis grading report on the online website.

[0016] In a sixth aspect, the present application provides a primary membranous nephropathy prognosis prediction evaluation model, characterized in that the prediction model is: Risk score = -0.462 x [PON1] + 0.415 x [ACTBL2] + 0.195 x [TPP1] - 0.444 x [RDX] + 0.0001 x PLA2R antibody titers (RU / mL) - 0.016 x Age (years) - 0.007 x eGFR (ml / min / 1.73 m 2 ); Wherein, the [PON1] is the concentration value of paraoxonase 1 in the middle morning urine of the to-be-tested person; the [ACTBL2] is the concentration value of beta-actinin 2 in the middle morning urine of the to-be-tested person; the [RDX] is the concentration value of rootin in the middle morning urine of the to-be-tested person; the [TPP1] is the concentration value of tripeptidyl peptidase 1 in the middle morning urine of the to-be-tested person; the PLA2R antibody titers is the serum anti-PLA2R antibody titer of the to-be-tested person; the Age is the age of the to-be-tested person; and the eGFR is the estimated glomerular filtration rate of the to-be-tested person.

[0017] If the risk score is greater than a reference value, it indicates a poor prognosis and belongs to a high risk; and if the risk score is less than the reference value, it indicates a good prognosis and belongs to a low risk.

[0018] As a preferred implementation form of the primary membranous nephropathy prognosis prediction and evaluation model, the concentration value is in ng / mL; the serum anti-PLA2R antibody titer is in RU / mL; the age is in years; the estimated glomerular filtration rate is in mL / min / 1.73 m²; and the reference value is -1.2205.

[0019] Compared with the prior art, the present application has the following beneficial effects: 1. The present application uses PON1, ACTBL2, RDX and TPP1 as biomarkers for evaluating the prognosis of primary membranous nephropathy, can generate a combined risk score by detecting the concentration levels of the four proteins in the urine of a patient and combining clinical characteristics, and helps to evaluate the prognosis of PMN patients, provides a reliable basis for individualized treatment decisions for PMN patients, and helps to guide the development of treatment plans by clinicians.

[0020] 2. The sample for PMN prognosis biomarker detection involved in the present application is middle morning urine, and the sample is collected in a non-invasive and convenient manner, thereby improving the compliance of patients and helping to promote the wide application in clinical practice.

[0021] 3, The product and model for evaluating or predicting the prognosis of primary membranous nephropathy can be used for evaluating the prognosis of patients with primary membranous nephropathy. The medical device product has the clinical conversion advantages of multi-element adaptation, including diagnostic agents, kits, chips, test papers and well plates, and can meet the comprehensive needs from primary screening to precise diagnosis. The final report of the model can be directly obtained on an online website, and the model has clinical practicability. BRIEF DESCRIPTION OF DRAWINGS

[0022] Figure 1 It is a schematic diagram of the overall technical route of the application.

[0023] Figure 2 It is a differential protein analysis result graph of the remission group and the non-remission group in the PMN patient of the application; wherein: A is the PCA analysis of the total protein group data of the remission group and the non-remission group; B is the volcano plot of the differential proteins identified in the remission group and the non-remission group; C is the differential expression protein clustering heat map identified in the remission group and the non-remission group; D and E are GO and KEGG analysis.

[0024] Figure 3 It is a construction and verification result graph of the prognosis risk model of the application; wherein: A is the forest plot of the single factor COX regression analysis result of urine protein; B is the importance protein ranking of the random survival forest method; C is the forest plot of the multi-factor COX regression analysis result of urine protein; D is the Kaplan-Meier curve of the patients in the low-risk group and the high-risk group, and the number of patients in different risk groups; E is the receiver operating characteristic curve of the 1, 2 and 3 year prognosis value of the urine protein model; F is the calibration curve: evaluating the calibration performance of the urine protein prognosis model on the 1 year clinical remission probability; G is the Sankey graph: showing the overlapping PMN patients between the urine protein prognosis model and the remission state.

[0025] Figure 4The correlation between the expression levels of three urinary proteins and the remission rate and clinical characteristics of PMN patients was investigated. Specifically: 4A-H represent: A, Kaplan-Meier curves of urinary protein PON1 expression level and remission rate in PMN patients (P=0.00016) and the number of patients in different groups; B, Kaplan-Meier curves of urinary protein ACTBL2 expression level and remission rate in PMN patients (P=0.07) and the number of patients in different groups; C, Kaplan-Meier curves of urinary protein RDX expression level and remission rate in PMN patients (P=0.097) and the number of patients in different groups; D, Kaplan-Meier curves of urinary protein TPP1 expression level and remission rate in PMN patients (P=0.0081) and the number of patients in different groups; E, correlation between urinary protein PON1 expression level and patient clinical characteristics; F, correlation between urinary protein ACTBL2 expression level and patient clinical characteristics; G, correlation between urinary protein RDX expression level and patient clinical characteristics; and H, correlation between urinary protein TPP1 expression level and patient clinical characteristics.

[0026] Figure 5 A heatmap showing the correlation between four types of urinary protein, risk scores, and different clinical characteristics.

[0027] Figure 6 The graph shows the correlation analysis results of the prognostic model risk score and different clinical features of the present invention; where: A and B are the correlation between risk score and clinical features; C is a forest plot of the univariate Cox regression analysis results of risk score and clinical features; D is a forest plot of the multivariate Cox regression analysis results of risk score and clinical features; E is a comparison of the C-index of three models for predicting the prognosis of PMN patients, including the 4-PM model, the single clinical model (serum anti-PLA2R antibody, age and eGFR) and the combined model (4-PM + clinical features); F is a decision curve analysis graph: showing the net benefits of the three models under different threshold probabilities.

[0028] Figure 7 This is a schematic diagram of the structure of the medical device for detecting the prognosis of primary membranous nephropathy according to the present invention. Detailed Implementation

[0029] To better illustrate the objectives, technical solutions, and advantages of this invention, the invention will be further described below with reference to specific embodiments. Those skilled in the art should understand that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0030] Unless otherwise specified, the experimental methods used in the examples are conventional methods; the materials and reagents used are commercially available unless otherwise specified.

[0031] Example 1: Construction and validation of a prognostic model for primary membranous nephropathy 1. Materials and methods (1) Acquisition of clinical samples The clinical and pathological information and mid-morning urine samples of 71 patients with PMN who were treated and followed up in the Third Affiliated Hospital of Sun Yat-sen University from 2017 to 2021 were collected. The clinical data of the patients were collected by checking the electronic medical records or by telephone inquiry. The inclusion criteria were: 1) diagnosed as primary membranous nephropathy by pathological biopsy in our hospital; 2) aged 18-80 years old, and not treated with immunosuppressive agents within 6 months before kidney biopsy; 3) received treatment in our hospital after biopsy, and had complete clinical data for at least 6 months of follow-up; 4) 24-hour urine protein > 3.5 g / d or PCR > 3500 mg / g at baseline. Patients with secondary factors leading to membranous nephropathy were excluded, including the following complications or predisposing factors: infectious diseases (hepatitis B, HIV and syphilis, etc.), autoimmune diseases, malignancies, allogeneic hematopoietic stem cell transplantation, exposure to toxic agents or use of certain drugs.

[0032] (2) Sample processing and mass spectrometry analysis The first mid-morning urine samples of 71 patients with PMN were collected and stored at -80°C. One milliliter of urine sample was centrifuged at 176,000 x g for 70 minutes to collect the precipitate. The precipitate was digested with 1 microgram of trypsin for 4 hours at 37°C according to the adjusted simplified procedure. The digested peptides were separated by self-packed C18 reverse-phase capillary chromatography column and analyzed by high-resolution mass spectrometer coupled with nanoliter liquid chromatography, with a mass spectrometer resolution not less than 60,000.

[0033] (3) Data analysis and quality control The original LC-MS / MS data were processed by a bioinformatics analysis platform, and the peptide segments were identified using a database search engine (peptide FDR <1%). Protein quantification was performed by intensity-based absolute quantification (iBAQ) algorithm. To standardize the differences in sample size, the iBAQ values were further converted to iFOT (fraction of total number), and the specific calculation method was to divide the iBAQ value of each protein by the total iBAQ of the sample, and then multiply by 10 ^ 5 for easy visualization. All missing values were replaced by zero. The trypsin digest of human HEK293 T cells was used as a quality control sample and was routinely evaluated by LC-MS / MS to ensure the reproducibility of the instrument.

[0034] (4) Statistical analysis of data Group comparisons for continuous variables were performed using Student's t-test or Mann-Whitney U test. Group comparisons for categorical variables were performed using the Chi-square test. Kaplan-Meier survival analysis was used to demonstrate the time distribution of the primary outcome. The statistical significance level was set at p < 0.05. The correlation between each predictor and the time to clinical remission was assessed using the Cox regression model. The best urinary protein markers associated with the prognosis of PMN patients were screened by random forest survival analysis and stepwise COX regression, and then a risk prediction model was constructed, which was internally validated using the Bootstrap method. To evaluate the predictive performance of the model, patients were divided into high-risk and low-risk groups according to the median risk score, and Kaplan-Meier survival analysis was performed using the R package "survival". The time-dependent receiver operating characteristic curve was constructed based on the R package "timeROC" to evaluate the sensitivity and specificity of the model. In addition, to assess whether the prognostic prediction ability based on the four urinary proteins can be independent of other clinical variables, further multivariate Cox regression analysis was performed.

[0035] 2. Experimental results (1) Basic characteristics of patient cohort As shown in Table 1: Among the 71 patients with primary membranous nephropathy diagnosed by kidney biopsy, 18 received supportive treatment (without the use of immunosuppressants), 47 received glucocorticoid + calcineurin inhibitor treatment, 4 received glucocorticoid + cyclophosphamide treatment, and 2 received calcineurin inhibitor + rituximab treatment. During the median follow-up period of 8 months, 54 patients were determined to have achieved CR or PR (remission group), and 17 patients were determined not to have achieved remission (non-remission group). The median remission time of PMN patients was 11.2 months (95% CI: 6.2-18.0 months). The clinical characteristics of the remission group (R = 54) and the non-remission group (NR = 17) at baseline are shown in Table 1.

[0036] Table 1. Baseline characteristics of PMN patients (2) Construction and validation of prognostic model 1) Differential protein analysis In the cohort of 71 patients with primary membranous nephropathy, a total of 1909 urinary proteins were identified using label-free quantitative methods. To exclude the influence of different treatment regimens on prognosis, 47 PMN patients who achieved remission and non-remission using the "glucocorticoid + calcineurin inhibitor" treatment regimen were selected for differential analysis. Principal component analysis results showed that there were certain differences in urinary proteins between PMN patients with different prognoses (Fig. 1A). The results of the Wilcoxon rank-sum test showed that 4 urinary proteins were significantly different between the remission and non-remission groups (Fig. 1B). Figure 2A). A total of 77 differentially expressed proteins were identified (p-value < 0.05, FC≥1.5), including 60 up-regulated proteins and 17 down-regulated proteins (Fig. 1A). Figure 2 B,C). GO and KEGG enrichment analysis showed that these differentially expressed proteins were significantly enriched in complement and coagulation cascade pathways and lysosomal pathways (Fig. 1B,C). Figure 2 D,E).

[0037] 2) Prognostic model construction A total of 24 urinary proteins were significantly associated with clinical remission rate in PMN patients using univariate COX regression analysis (Fig. 2A). Figure 3 A). A total of 101 machine learning model combinations were constructed using the R package "Mime1", and the results showed that the C-index of the data set was best in the random survival forest model.

[0038] Using random survival forest method and bidirectional stepwise COX regression analysis, 4 key urinary proteins (PON1, ACTBL2, RDX and TPP1) biomarkers were finally screened, and a 4-urinary protein prognostic model (4-PM model) was constructed (Fig. 3A). Figure 3 B,C). The risk score formula of the model is: risk score = -0.438×[PON1] + 0.364×[ACTBL2] + 0.225×[TPP1] - 0.403×[RDX].

[0039] 3) Model validation Internal validation of the model: using random division method, 49 cases in the data set were randomly divided into training set, and 22 cases were divided into validation set, and finally the model C-index was calculated as 0.729.

[0040] Risk stratification: according to the median of risk score, PMN patients were divided into high-risk group and low-risk group. Kaplan-Meier analysis showed that the clinical remission rate of low-risk group was significantly higher than that of high-risk group (Fig. 4D, P<0.001). Figure 3

[0041] Time-dependent ROC curve: the area under the curve (AUC) of total clinical remission rate at 1 year, 2 years and 3 years was 0.730, 0.774 and 0.805, respectively (Fig. 5E). Figure 3

[0042] Calibration consistency: the calibration curve showed high consistency with the ideal reference line in the prediction evaluation of 1-year total remission (Fig. 6F). Figure 3

[0043] ​​​Prediction stability: The integrated Brier score (IBS) based on 1000 Bootstrap resamples was 0.089 (95% CI: 0.057-0.111), further validating the reliability of the model.

[0044] (3) Prognostic risk score was associated with clinical outcomes 1) Correlation analysis To further validate the model, we first analyzed the correlation between the expression level of each of the four urinary proteins and the clinical remission rate of PMN patients. Kaplan-Meier curves showed that high expression of urinary protein PON1 was significantly associated with faster clinical remission in PMN patients (P<0.05, Figure 4 A), while high expression of urinary protein TPP1 was significantly associated with slower clinical remission in PMN patients (P<0.05, Figure 4 D). In addition, high expression of urinary proteins ACTBL2 and RDX may indicate poor prognosis in PMN patients, but this was not statistically significant (P>0.05, Figure 4 B,C).

[0045] Secondly, we analyzed the correlation between the expression level of each of the four urinary proteins and the clinical characteristics of PMN patients. The results showed that the expression of urinary protein PON1 was significantly correlated with hypertension, PLA2R antibody titer, serum creatinine and eGFR (estimated glomerular filtration rate, calculated according to CKD-EPI creatinine formula), while the expression of urinary protein ACTBL2 was significantly correlated with serum albumin, serum creatinine and eGFR (P<0.05, Figure 4 E,F, Figure 5 ). At the same time, the expression of urinary protein RDX was significantly correlated with age and eGFR, while the expression of urinary protein TPP1 was significantly correlated with age, urine PCR and eGFR (P<0.05, Figure 4 G,H, Figure 5 ).

[0046] 2) Risk score is an independent prognostic factor To analyze the correlation between risk score and clinical characteristics, we compared the differences in risk score according to different clinical characteristics. The results showed that risk score was correlated with hypertension, but not significantly correlated with gender, age, diabetes, smoking and PLA2R antibody (P>0.05, Figure 6 A,B).

[0047] Multivariate COX regression analysis showed that riskScore was an independent prognostic factor for PMN patients (HR=4.504, P<0.001), while the differences in clinical characteristics such as age, gender, treatment regimen, eGFR and serum PLA2R antibody were not statistically significant (HR=0.999, P=0.999, HR=0.999, P=0.999, HR=0.999, P=0.999, HR=0.999, P=0.999, respectively) Figure 6 C,D).

[0048] (4) The combined model is significantly better than the predictive ability of single clinical characteristics model Univariate Cox regression analysis showed that only age and eGFR were significantly associated with prognosis (HR=0.999, P=0.999, HR=0.999, P=0.999, respectively) Figure 6 C). In view of the existing research confirmed the association between serum anti-PLA2R antibody and the prognosis of membranous nephropathy (antibody positive group remission rate is lower than the negative group), further comparison of 4-PM urine protein model (PON1, ACTBL2, RDX and TPP1) and single clinical characteristics (serum anti-PLA2R antibody, age, eGFR) of the predictive efficiency.

[0049] 1) Comparison of C-index of two basic models Using multivariate COX regression analysis, two models were established, and the specific risk score formula is as follows: single clinical characteristics model = -0.001 x PLA2R antibody titers (RU / mL) -0.018 x Age (years) +0.007 x eGFR (ml / min / 1.73 m 2 ); 4-PM model = -0.438 x [PON1] +0.364 x [ACTBL2] +0.225 x [TPP1] -0.403 x [RDX].

[0050] 4-PM model C-index = 0.729.

[0051] Single clinical characteristics model C-index = 0.636.

[0052] The difference value of C-index of the two models is: 0.094 (95% CI: -0.200~0.009) Figure 6 E).

[0053] 2) Optimization verification of combined model After establishing the combined model, the predictive performance of the three models was compared. Considering the impact of clinical indicators on prognostic prediction, three clinical features were further incorporated into the 4-PM model, ultimately establishing the combined model: 4-PM & clinical features = -0.462×[PON1] + 0.415×[ACTBL2] + 0.195×[TPP1] - 0.444×[RDX] + 0.0001×PLA2R antibody titers (RU / mL) - 0.016×Age (years) - 0.007×eGFR (ml / min / 1.73 m 2 The combined model's C-index improved to 0.744, an improvement of 0.108 compared to the single clinical feature model (95% CI: 0.005–0.213, P<0.001).

[0054] Meanwhile, the combined model demonstrated clear clinical value in terms of improvement in risk stratification. As shown in Table 2, compared with the individual clinical models, the combined model showed significant improvement in reclassification [NRI=0.393(0.025, 0.800), IDI=0.199(0.088, 0.340), P<0.05].

[0055] Table 2 Comparison of C-index, NRI, and IDI for different prognostic models Decision curve analysis: The joint model provided the highest net benefit over a wide range of threshold probabilities, demonstrating its significant value in clinical decision-making. Figure 6 F).

[0056] In summary, this embodiment integrates machine learning methods to construct and validate a prognostic model for primary membranous nephropathy based on urinary protein PON1, ACTBL2, RDX, TPP1, and three clinical features. The combined model significantly outperforms single clinical feature models in predicting the prognosis of PMN patients.

[0057] Example 2: Medical Device for Prognostic Detection Based on a Joint Model of Urinary Protein Biomarkers and Clinical Features By simultaneously detecting the expression levels of urinary protein biomarkers (PON1, ACTBL2, RDX, TPP1) and combining this with three patient clinical characteristics (serum anti-PLA2R antibody, age, and eGFR), the data is input into an embedded analysis system to generate a combined risk score for prognostic stratification. Prognostic testing devices include, but are not limited to: multi-indicator diagnostic kits, protein detection chips, lateral flow chromatography test strips, and microplate detection systems.

[0058] The detection medical instrument adopts a modular design, comprising: a sample processing module: configured for pretreatment of a mid-morning urine, including a centrifugal residue removal unit, a protein enrichment unit, and a buffer stabilization unit; a prognosis marker detection module: based on an immunological analysis principle to quantitatively detect expression levels of four target proteins (PON1, ACTBL2, RDX, and TPP1); a multi-dimensional data integration module: an internal clinical characteristic input interface is provided, and a combined risk score can be calculated by combining urine proteins and clinical characteristics; and a result output module: a visual risk prognosis grading report can be provided on an online website.

[0059] (1) The specific embodiment is as follows: 1) Diagnostic kit ① Structural composition Quadruple detection microplate: 96-well plates are divided into four antibody-coated areas (A area: anti-PON1 monoclonal antibody, B area: anti-ACTBL2 polyclonal antibody, C area: anti-RDX monoclonal antibody, and D area: anti-TPP1 monoclonal antibody); Dynamic calibration module: containing gradient concentration recombinant protein calibrant and matrix correction buffer (such as: PBS-Tween 20+ 1% BSA buffer, casein-containing Tris buffer, etc.); Data integration card: loaded with a risk score algorithm two-dimensional code, which can be scanned to enter the website to input clinical characteristic data.

[0060] ② Prognosis detection method The morning urine sample is centrifuged to obtain the supernatant, mixed with buffer 1:4, 100 μL of the mixed solution is added to the microplate, and incubated at 37°C for 45 minutes; after washing the plate, HRP-secondary antibody mixed solution is added, and reacted at 37°C for 30 minutes; TMB is developed, and the absorbance value is measured at 450 nm; the protein concentration is converted according to the standard curve, and the risk report is automatically generated by scanning and inputting the clinical parameters.

[0061] 2) Protein detection chip ① Chip architecture A 4x4 array type silicon-based chip is used, comprising: a micro-reaction pool: a surface modified nanobody capture layer; and a photoelectric sensing unit: an integrated SPR sensor for real-time monitoring of binding signals.

[0062] ② Chip reader automatically completes Conversion of the concentration of the four target proteins in the urine sample; and the final data is transmitted to an online website to automatically calculate the risk score.

[0063] 3) Lateral flow chromatography test paper ① Chromatography system A lateral flow chromatography technology is used, and four independent detection lines are arranged on a nitrocellulose membrane: T1 line: PON1 detection line (anti-PON1 antibody); T2 line: ACTBL2 detection line (anti-ACTBL2 antibody); T3 line: RDX detection line (anti-RDX antibody); T4 line: TPP1 detection line (anti-TPP1 antibody); C line: quality control line (anti-rabbit IgG).

[0064] ②Quantitative interpretation The signal intensity of the detection line is collected by a portable fluorescence reader, and the signal value is converted into protein concentration by the built-in algorithm, and transmitted to the online website for automatic calculation of risk score.

[0065] 4) Microplate detection system ①Automatic platform It includes: 96 deep hole sample pretreatment module; four-channel high-speed pipetting system; multi-spectral detection head.

[0066] ②Workflow Automatically complete urine sample centrifugation, dilution, sample addition; parallel detection of 4 target protein concentrations; obtain clinical feature data through the hospital internal system; finally transmit the data to the online website for automatic calculation of risk score.

[0067] (2) Clinical data integration module Clinical data integration can directly obtain the specific values of serum anti-PLA2R antibody titer (RU / mL), age (years), and eGFR (mL / min / 1.73m²) through the hospital electronic medical record system or laboratory system.

[0068] (3) Risk calculation and output report system A Web-based prognosis prediction system is constructed, which mainly realizes prognosis evaluation through three modules (platform access website: https: / / rsf0427models.shinyapps.io / PMN_Clinical_Predictor / ): 1) Data input module An interactive input panel is provided, which includes 7 prediction variables: Urine biomarkers: PON1, ACTBL2, RDX, TPP1 concentration values (ng / mL); Clinical features: serum anti-PLA2R antibody titer (RU / mL), age (years), eGFR (mL / min / 1.73m²).

[0069] 2) Core calculation engine Embedded algorithms are used for real-time execution: Combined risk score = -0.462 x [PON1] + 0.415 x [ACTBL2] + 0.195 x [TPP1] - 0.444 x [RDX] + 0.0001 x PLA2R antibody titers (RU / mL) - 0.016 x Age (years) - 0.007 x eGFR (ml / min / 1.73 m 2 ).

[0070] Based on the score, automatically divide the risk level into low-risk group and high-risk group.

[0071] 3) Visualization report Generate an interactive clinical decision report, including: Dynamic risk indicator: red area represents high risk, green area represents low risk; risk score greater than reference value indicates poor prognosis, belonging to high risk; risk score less than reference value indicates good prognosis, belonging to low risk; reference value is -1.2205.

[0072] 1-3 years of clinical remission probability curve; Personalized treatment recommendations: high risk - recommend intensive follow-up (review urine protein every 3 months, consider immunosuppressive therapy); low risk - maintain regular follow-up (check urine protein every 6 months).

[0073] Finally, it should be explained that the above examples are only used to illustrate the technical solutions of the present application, not to limit the protection scope of the present application. Although the present application has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present application can be modified or replaced by equivalents without departing from the essence and scope of the technical solutions of the present application.

Claims

1. A prognostic biomarker for primary membranous nephropathy, characterized in that, This includes paraoxyphosphokinase 1, β-actin-like protein 2, root protein, and tripeptidyl peptidase 1.

2. The prognostic biomarker according to claim 1, characterized in that, The prognostic markers were derived from urine.

3. The prognostic biomarker according to claim 1, characterized in that, It also includes serum anti-PLA2R antibody titer, age, and eGFR.

4. The use of the prognostic biomarker according to any one of claims 1-3 in the preparation of products for assessing or predicting the prognosis of primary membranous nephropathy.

5. A product for assessing or predicting the prognosis of primary membranous nephropathy, characterized in that, The formulation includes a preparation for detecting the expression level of the prognostic biomarker as described in any one of claims 1-3.

6. The product according to claim 5, characterized in that, This includes diagnostic reagents, kits, chips, test strips, or well plates.

7. The use of the prognostic biomarker according to any one of claims 1-3 in constructing a system for assessing or predicting the prognosis of primary membranous nephropathy.

8. A system for assessing or predicting the prognosis of primary membranous nephropathy, characterized in that, It includes a prognostic biomarker detection module, a data processing module, a multi-dimensional data integration module, and a result output module; the prognostic biomarker detection module is used to detect the expression level of the prognostic biomarker as described in any one of claims 1-3.

9. A prognostic prediction and assessment model for primary membranous nephropathy, characterized in that, The prediction model is as follows: Risk score = -0.462×[PON1] + 0.415×[ACTBL2] + 0.195×[TPP1] - 0.444×[RDX] + 0.0001×PLA2R antibody titers(RU / mL)-0.016×Age(years)-0.007×eGFR(ml / min / 1.73m 2 ); Wherein, [PON1] is the concentration of paraoxygenase 1 in the midstream morning urine of the test subject; [ACTBL2] is the concentration of β-actin-like protein 2 in the midstream morning urine of the test subject; [RDX] is the concentration of root protein in the midstream morning urine of the test subject; [TPP1] is the concentration of tripeptidyl peptidase 1 in the midstream morning urine of the test subject; PLA2R antibody titers are the serum anti-PLA2R antibody titers of the test subject; Age is the age of the test subject; and eGFR is the estimated glomerular filtration rate of the test subject. If the risk score is greater than the reference value, it indicates a poor prognosis and is classified as high risk; if the risk score is less than the reference value, it indicates a good prognosis and is classified as low risk.

10. The prognostic prediction and assessment model for primary membranous nephropathy according to claim 9, characterized in that, The concentration values ​​are in ng / mL; the serum anti-PLA2R antibody titer is in RU / mL; the age is in years; and the estimated glomerular filtration rate is in mL / min / 1.73m2. The reference value is -1.2205.

Images