A marker combination and its use in diagnosing dysplasia expressing spasmolytic polypeptide

Abstract

The application discloses a marker combination and application thereof in diagnosing dysplasia expressing spasmolytic polypeptide, and the marker combination comprises SCD1, PHB1 and p-ERK marker combination. The marker combination of the application is based on combined detection of SCD1, PHB1 and p-ERK markers, and a diagnostic model established based on the marker combination is significantly superior to a traditional single detection method in sensitivity and specificity, and is helpful to early identification of SPEM, a potential precancerous lesion.

Description

TECHNICAL FIELD

[0001] The present application belongs to the field of biomarkers, and particularly relates to a marker combination and application thereof in diagnosis of spasmolytic polypeptide expressing metaplasia. BACKGROUND

[0002] At present, as a potential precancerous lesion, the diagnosis of gastric metaplasia, especially spasmolytic polypeptide expressing metaplasia (SPEM) mainly relies on pathological detection and immunohistochemical techniques. The main diagnostic method is multiple immunofluorescence detection, which uses multiple specific antibodies to label cell subgroups in gastric mucosa, such as TFF2, MUC6, GIF, etc. for co-expression analysis to identify SPEM cell groups. Although certain progress has been made in SPEM detection, there are still the following shortcomings: marker selection limitation: a single marker is often difficult to balance sensitivity and specificity in a complex pathological background, although multiple marker joint detection has improved, but how to screen the most diagnostic combination from numerous potential markers is still a difficulty. Complicated detection method and high cost: the existing immunohistochemical detection technology and multiple staining method are complicated in operation, and the repeatability and accuracy are difficult to guarantee; at the same time, the high cost restricts the popularization and application of the technology in primary hospitals and large-scale screening. SUMMARY

[0003] The present application aims to solve the following key technical problems: how to construct a high-precision SPEM diagnostic model based on multiple markers, by systematically screening and combining multiple molecular markers closely related to SPEM, in order to achieve a diagnostic standard with high sensitivity and specificity. How to optimize the detection method to reduce the complexity and cost of operation: by improving the detection process and introducing semi-quantitative analysis method, the model is more simple, efficient and suitable for large-scale clinical screening under the premise of ensuring accuracy. How to improve the individual risk assessment and clinical practicability of the model: using statistical methods and machine learning algorithms, a diagnostic tool that can provide individual risk assessment is constructed, so that doctors can accurately formulate intervention measures according to the specific indicators of patients. How to provide guidance for subsequent treatment through the diagnostic model: the diagnostic model is not only used for early detection of SPEM, but also can monitor the progression of the lesion, and then guide the clinical use of targeted intervention strategies to achieve early prevention and reversal of the lesion process.

[0004] To solve the above technical problems, the present application provides a marker combination and its application in diagnosing dysplasia expressing spasmolytic polypeptide. The present application uses multi-marker joint detection and comprehensive statistical modeling method to achieve the following significant technical effects: improving diagnostic accuracy and early identification rate: based on the joint detection of SCD1, PHB1 and p-ERK markers, the diagnostic model established is significantly superior to the traditional single detection method in terms of sensitivity and specificity, which helps to identify SPEM, a potential precancerous lesion, at an early stage; simplifying the detection process and reducing costs: using semi-quantitative immunohistochemical technology and standardized data processing process, the operation is more simple, the detection cost is reduced, and it has high repeatability and generalizability, realizing personalized risk assessment: by constructing nomogram tool, the individualized risk of patients is intuitively evaluated, which provides strong basis for clinicians to develop precise intervention programs; guiding clinical intervention and pathological reversal: the present application is not limited to diagnosis, but also provides data support for subsequent intervention, indicating a comprehensive management mode from early detection to pathological reversal, which is expected to reduce the risk of gastric cancer.

[0005] The first aspect of the present application provides a marker combination comprising SCD1, PHB1 and p-ERK markers.

[0006] SCD1 (Stearoyl-CoA Desaturase 1) is an enzyme located in the endoplasmic reticulum, which catalyzes the synthesis of unsaturated fatty acids and is an important component of cell membrane lipids. It plays a key role in cell proliferation, survival, inflammatory response and cancer development.

[0007] PHB1 (Prohibitin 1) is a highly conserved protein mainly located in the inner membrane of mitochondria, which is an important regulator of mitochondrial function. PHB1 plays a role in various cellular processes, including cell cycle regulation, cell aging, apoptosis and stabilization of mitochondrial respiratory chain.

[0008] p-ERK (Phosphorylated Extracellular Signal-Regulated Kinase) is the phosphorylated form of ERK (Extracellular Signal-Regulated Kinase), which is a key molecule in the MAPK (Mitogen-Activated Protein Kinase) signaling pathway. ERK signaling pathway is widely involved in the regulation of cell proliferation, differentiation, growth and apoptosis.

[0009] The second aspect of the present application provides the use of the marker combination according to the first aspect of the present application in preparing a drug for predicting and / or diagnosing dysplasia expressing spasmolytic polypeptide.

[0010] In a preferred embodiment, the marker combination has a significant difference in expression level between dysplasia expressing spasmolytic polypeptide samples and non-dysplasia expressing spasmolytic polypeptide samples.

[0011] The third aspect of the present application provides use of a reagent for determining a marker combination in preparation of a kit for judging a spasmolytic polypeptide expression metaplasia sample of a subject, wherein the marker combination comprises SCD1, PHB1 and p-ERK.

[0012] In a preferred embodiment, the sample is from a mucosa sample, such as a mucosa sample of the greater curvature of corpus.

[0013] In a preferred embodiment, the reagent is a reagent for determining the level of a marker.

[0014] The fourth aspect of the present application provides a reagent for detecting a marker combination, the marker combination being as defined in the first aspect of the present application, the reagent being a reagent for determining the level of a marker, such as a reagent for immunohistochemical staining or Western blot;

[0015] Preferably, the marker is from a mucosa sample, such as a mucosa sample of the greater curvature of corpus.

[0016] The fifth aspect of the present application provides a diagnostic kit for spasmolytic polypeptide expression metaplasia, comprising the reagent of the fourth aspect of the present application and a control.

[0017] The sixth aspect of the present application provides a diagnostic system for spasmolytic polypeptide expression metaplasia, comprising a detection module and an analysis and judgment module; the detection module detects the level of a marker combination in a mucosa of a subject and transmits the level data to the analysis and judgment module; the analysis and judgment module judges the level data of the marker combination and outputs a diagnostic result that the subject is a spasmolytic polypeptide expression metaplasia or a non-spasmolytic polypeptide expression metaplasia; wherein the marker combination is as defined in the first aspect of the present application.

[0018] In a preferred embodiment, the analysis and judgment module judging the level data of the marker combination comprises determining whether there is a significant difference in the expression level of the marker combination between a spasmolytic polypeptide expression metaplasia sample and a non-spasmolytic polypeptide expression metaplasia.

[0019] The present application also provides a computer readable storage medium storing a computer program, characterized in that the computer program, when executed by a processor, can realize the function of the diagnostic system of the sixth aspect of the present application.

[0020] The present application also provides an electronic device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor is configured to execute the computer program to realize the function of the diagnostic system of the sixth aspect of the present application.

[0021] Based on a large sample of cases, the present patent simultaneously detects key indicators such as SCD1, PHB1, p-ERK, etc. in gastric mucosa specimens by semi-quantitative immunohistochemical detection method, and screens out the most sensitive and specific marker combination by Spearman correlation analysis and LASSO regression. A variety of statistical and machine learning algorithms such as logistic regression (LR) and XGBoost are used to comprehensively analyze the data, establish and optimize the diagnostic model, calculate the AUC value by ROC curve to verify the diagnostic performance of the model, and construct a nomogram containing multiple variables to realize individualized risk assessment. A one-stop solution from early diagnosis to intervention treatment is provided for clinical practice.

[0022] On the basis of common sense in the art, the above-mentioned preferred conditions can be combined arbitrarily, i.e. to obtain each preferred example of the present application.

[0023] Unless otherwise specified, the reagents and raw materials used in the present application are commercially available.

[0024] The positive progress effect of the present application is that the present application achieves the following significant technical effects through multi-marker joint detection and comprehensive statistical modeling method:

[0025] Improve the diagnostic accuracy and early recognition rate: based on the joint detection of SCD1, PHB1 and p-ERK markers, the diagnostic model established is significantly superior to the traditional single detection method in terms of sensitivity and specificity, which helps to early identify SPEM, a potential precancerous lesion;

[0026] Simplify the detection process and reduce the cost: the semi-quantitative immunohistochemical technology and standardized data processing process are adopted, which makes the operation more simple, reduces the detection cost, and has high repeatability and generalizability;

[0027] Realize individualized risk assessment: by constructing a nomogram tool, the individualized risk of patients is intuitively evaluated, which provides a strong basis for clinicians to develop precise intervention programs;

[0028] Guide clinical intervention and pathological reversal: the present application is not limited to diagnosis, but also provides data support for subsequent intervention, which indicates a comprehensive management mode from early detection to pathological reversal, and is expected to reduce the risk of gastric cancer. BRIEF DESCRIPTION OF DRAWINGS

[0029] Figure 1Figure 6. The diagnostic model of SPEM. A, Immunohistochemistry and semi-quantitative analysis of gastric tissue chip in obese patients; B, Correlation analysis of SPEM occurrence with SCD1, Trim21, PHB1 and p-ERK; C and D, LASSO analysis (C: green represents PHB1, black represents SCD1, blue represents Trim21, and red represents p-ERK); E, Construction of the diagnostic model by regression model and XGBoost; F, Lollipop plot results.

[0030] Figure 2 Figure 7. SHAP analysis of the diagnostic model of SPEM. A, SHAP summary plot showing the effect of each variable on the probability of SPEM diagnosis; B, SHAP waterfall plot for individual patients, indicating the contribution of features to the prediction of SPEM risk; C and D, Confusion matrix showing the performance of the logistic regression (C) and XGBOOST (D) models in SPEM classification on the training set. DETAILED DESCRIPTION

[0031] The present application is further illustrated by the following examples without thereby limiting the present application to the examples described. The experimental methods in the following examples, if not specified, are selected according to the conventional methods and conditions, or according to the product instructions.

[0032] Example 1 Sample collection and immunohistochemical detection

[0033] 1.1 Sample source and ethical approval

[0034] Sample selection: 42 cases of severe obese patients who underwent laparoscopic sleeve gastrectomy (LSG) in the Obesity Center of the Tenth People's Hospital of Shanghai were selected, and preoperative gastroscopy suggested no malignant lesions.

[0035] Tissue acquisition: About 1x1x1 cm of mucosal biopsy specimens were taken at the greater curvature of the body during the operation and immediately fixed in 4% paraformaldehyde solution for 24 hours. 3 Mucosal biopsy specimens were immediately fixed in 4% paraformaldehyde solution for 24 hours.

[0036] Ethical approval: This study was approved by the Ethics Committee of the Tenth People's Hospital of Shanghai, and all patients signed the informed consent form.

[0037] The following table is the inclusion and exclusion criteria for the obese group:

[0038]

[0039]

[0040] 1.2 Immunohistochemical (IHC) detection

[0041] Paraffin embedding and sectioning: After dehydration, transparency, and paraffin embedding of the fixed tissue, sectioning was performed on a microtome with a thickness of 4 pm.

[0042] Antigen retrieval: sections were microwave treated in citrate buffer (pH 6.0) for 10 minutes.

[0043] Blocking: 3% hydrogen peroxide for 10 minutes to inactivate endogenous peroxidase, 5% goat serum for 30 minutes.

[0044] Primary antibody incubation: anti-SCD1 (Abeam, ab19862, 1:200), anti-PHB1 (Proteintech, 10774-1-AP, 1:150), anti-p-ERK (Cell Signaling Technology, #4370, 1:100) were added respectively and incubated at 4°C overnight.

[0045] Secondary antibody and staining: HRP-labeled secondary antibody (Dako, K4001) was incubated at room temperature for 30 minutes, DAB was developed for 2 minutes, and hematoxylin was counterstained.

[0046] Mounting and scoring: neutral resin mounting after dehydration and transparency. Two pathologists blindly evaluated the staining intensity and the proportion of positive cells.

[0047] The results are shown in A of Figure 1 According to the number of immunohistochemical positive cells and staining intensity, they were divided into +, ++, +++ respectively according to the grade (B of Figure 1 for semi-quantitative grade labeling, which was used for subsequent statistical analysis.

[0048] Statistical analysis and model construction of Example 2 data

[0049] 2.1 Data preprocessing and grouping

[0050] Expression data: The original scores of group staining scores (1-3: low, medium, high) were imported into R language (v4.2.1).

[0051] SPEM definition: According to the immunofluorescence multicolor staining of pathological sections, it is defined as SPEM group (n=18) and non-SPEM group (n=24).

[0052] 2.2 Correlation analysis

[0053] Spearman rank correlation test was used to output r value and P value.

[0054] Results: As shown in B of Figure 1 SPEM level was significantly negatively correlated with SCD1 (r=-0.379, P=0.013), TRIM21 (r=-0.376, P=0.014); and significantly positively correlated with PHB1 (r=0.435, P=0.001), p-ERK (r=0.480, P=0.001).

[0055] 2.3 LASSO feature selection

[0056] Tools: R package glmnet (v4.1-4), 10-fold cross-validation, optimal lambda determined by the method of least mean square error.

[0057] Screening results: As shown in C and D of Figure 1 , the final remaining variables SCD1, PHB1, p-ERK.

[0058] 2.4 Model construction and performance evaluation

[0059] Logistic regression: using R package rms, a multivariate logistic regression model was constructed, and the results are shown in E of Figure 1 , AUC = 0.942.

[0060] XGBoost: Python version xgboost (v1.6.1), and the results are shown in E of Figure 1 , AUC = 0.89.

[0061] 2.5. Construction of individualized risk estimation tool - development of nomogram model

[0062] Software environment: R package rmda (v1.6) and rms (v6.2-0).

[0063] Input variables: SCD1, PHB1, p-ERK semi-quantitative score.

[0064] Output: SPEM occurrence.

[0065] Nomogram (F of Figure 1 ) annotates the corresponding score interval of each variable and the mapping relationship between the total score and the risk probability.

[0066] Example 3 Feature importance explanation - SHAP analysis

[0067] Tools: using Python package'shap' (v0.41.0) to perform interpretive analysis on the trained model.

[0068] SHAP summary plot: as shown in A of Figure 2 , p-ERK, PHB1 and SCD1 are the three variables that contribute most to the prediction of SPEM, among which high expression of p-ERK has the most significant positive effect.

[0069] SHAP waterfall plot: as shown in B of Figure 2 , it shows the composition structure of the prediction value of an individual SPEM patient, and high expression of p-ERK and PHB1 are the main risk enhancing factors.

[0070] Example 4 Model Performance Verification - Confusion Matrix Evaluation

[0071] Logistic Regression Model (LR): built using R package 'rms', performance as shown in C in Figure 2 . The model achieved an accuracy of 0.833 (95% CI: 0.6864 - 0.9303) in the training set, with a Kappa value of 0.6711, sensitivity and specificity of 0.9444 and 0.7500 respectively, and an F1 value of 0.8293, indicating a good discrimination ability of the model for SPEM.

[0072] XGBOOST Model: trained using Python version 'xgboost' (v1.6.1), performance as shown in D in Figure 2 . The model achieved an accuracy of 0.875 (95% CI: 0.7101 - 0.9649), with a Kappa value of 0.7241, sensitivity and specificity of 0.9000 and 0.8636 respectively, and an F1 value of 0.8182, showing a better overall performance and generalization ability than logistic regression.

Claims

1. The use of a combination of biomarkers in the preparation of a drug for predicting and / or diagnosing the expression of antispasmodic peptides, said biomarker combination being SCD1, PHB1 and p-ERK biomarkers.

2. The application as described in claim 1, characterized in that, The expression levels of the biomarker combination differed significantly between pharmacochemical samples expressing antispasmodic peptides and those expressing non-antispasmodic peptides.

3. The application of reagents for determining biomarker combinations in the preparation of kits for assessing the expression of antispasmodic peptides in subjects, wherein, The marker combination is SCD1, PHB1, and p-ERK.

4. The application as described in claim 3, characterized in that, The sample was taken from a mucosal sample.

5. The application as described in claim 4, characterized in that, The sample was taken from the mucosa at the greater curvature of the stomach.

6. The application as described in claim 4 or 5, characterized in that, The reagent is used to determine the level of the marker.

7. The application as described in claim 6, characterized in that, The reagents mentioned are those used for immunohistochemical staining or Western blot.

8. The application as described in claim 7, characterized in that, The kit also includes a control.

9. A diagnostic system for the expression of antispasmodic polypeptides, characterized in that, The diagnostic system includes a detection module and an analysis and judgment module; the detection module detects the level of a combination of biomarkers in the subject's mucosa and transmits the level data to the analysis and judgment module; the analysis and judgment module judges the level data of the biomarker combination and outputs a diagnostic result: the subject is a pharmacogene expressing antispasmodic peptides or a pharmacogene expressing non-antispasmodic peptides; wherein, the biomarker combination is SCD1, PHB1 and p-ERK biomarkers.

10. The diagnostic system as described in claim 9, characterized in that, The analysis and judgment module determines the level data of the biomarker combination by: determining whether there is a significant difference in the expression level of the biomarker combination in the antispasmodic peptide expression phantom sample and the non-antispasmodic peptide expression phantom sample.

11. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it can perform the functions of the diagnostic system as described in claim 9 or 10.

12. An electronic device comprising a memory and a processor, the memory storing a computer program, characterized in that, The processor is used to execute the computer program to implement the functions of the diagnostic system as described in claim 9 or 10.

Images