CN117330760A - Plasma exosome markers and applications for early diagnosis of pancreatic cancer - Google Patents

Abstract

The invention provides a plasma exosome marker for early diagnosis of pancreatic cancer and application thereof, wherein a series of biomarkers capable of early predicting pancreatic cancer occurrence risk are screened out by analyzing metabolites and proteins with significant differences in plasma separated exosomes of pancreatic cancer patients, pancreatitis patients and normal people by using methods of metabonomics and proteomics, and a diagnosis model of pancreatic cancer is constructed by further screening out a group of biomarkers.

Description

Plasma exosome markers and applications for early diagnosis of pancreatic cancer

Technical field

The present invention relates to the medical field. Specifically, it relates to the use of metabolomics to screen biomarkers of pancreatic cancer and its use in the diagnosis of pancreatic cancer. In particular, it relates to a method for predicting the occurrence of pancreatic cancer by detecting the abundance of metabolites contained in exosomes in plasma samples. Risk prediction systems and their applications.

Background technique

Metabolomics is a discipline that conducts qualitative and quantitative analysis of small molecule metabolites with a relative molecular weight of less than 1000 in the body. Metabolomic analysis can reflect the physiological and pathological conditions of the body, and can also distinguish differences between different individuals. With the development of mass spectrometry technology, liquid chromatography coupled with mass spectrometry (LC-MS) has become the most important research tool in metabolomics research. At present, metabolomics has been widely used in the field of clinical diagnosis, mainly discovering metabolic markers related to disease diagnosis and treatment.

At present, the clinical diagnosis methods of pancreatic cancer mainly include imaging diagnosis, histopathological diagnosis and blood immunobiochemical diagnosis. However, due to the low specificity of imaging examination and the difficulty of biopsy of the lesion in histopathological diagnosis, the expression level of tumor markers in the blood has become the main detection indicators. Since its discovery in 1979, the carbohydrate antigen CA19-9 is still the most commonly used clinical tumor marker and is the only biomarker approved by the FDA for the diagnosis of pancreatic cancer. However, about 25% of pancreatic cancer patients have no abnormal CA19-9 levels. The development of new tumor markers with both high sensitivity and specificity for early diagnosis of pancreatic cancer is urgent.

Exosomes are a type of extracellular vesicles with a size of 50 to 150 nm. They contain DNA, microRNA, proteins or other signaling molecules and play an important role in the exchange of information between cells. Analysis of abnormal plasma exosomes and the molecules they encapsulate has implications for tumor occurrence and progression. Exosomes are receiving increasing attention as a source of monitoring disease progression and discovering biomarkers. For example, the abundance of glypican-1 (GPC-1) in the serum of patients with pancreatic cancer is significantly higher than that in the normal population, and it can accurately and sensitively diagnose early pancreatic cancer; it contains 5 types of The results of the combined model of exosome-based protein markers (EGRF, EPCAM, MUC1, GPC1 and WNT2) showed higher sensitivity and specificity than the existing serum marker CA19-9. These studies confirm the superiority and possibility of using serum-derived exosomes as early diagnostic markers for pancreatic cancer, and also illustrate the need for further in-depth research and verification of potential markers.

Therefore, there is an urgent need to find a biomarker that can be easily and quickly sampled and can early predict whether an individual has a risk of pancreatic cancer, so that a more efficient assessment of pancreatic cancer risk can be achieved.

Contents of the invention

In view of the problems existing in the prior art, the present invention provides a biomarker for pancreatic cancer detection, which uses the method of metabolomics to analyze significant differences in exosomes in the blood of pancreatic cancer patients and normal people. metabolites, screen out a series of biomarkers that can early predict the risk of pancreatic cancer, and further screen out a set of biomarkers to construct a diagnostic model for pancreatic cancer, which can be used to conveniently and efficiently predict whether an individual will develop pancreatic cancer. Meet clinical needs.

In one aspect, the present invention provides the use of a biomarker in preparing a reagent for predicting whether an individual has pancreatic cancer, the biomarker being selected from one or more combinations of the following: 3-amino-2-piperidone , trans-urocanic acid, 4-cholesten-3-one, adenosine, adenine, serine, N-acetylserine, 1-stearoyl-2-arachidonoylglycerophosphatidylserine, 1-stearoylserine Fatty acyl-2-arachidonoylglycerophosphoethanolamine, N-acetylneuraminic acid.

Through non-targeted metabolomics research, the present invention uses UPLC-MS/MS high-performance liquid chromatography-tandem mass spectrometry method to analyze exosome samples from the plasma of three groups: healthy group, chronic pancreatitis patients, and pancreatic cancer patient group. Metabolites with significant differences between pancreatic cancer samples and non-pancreatic cancer control samples (healthy) were screened through four statistical methods: randomforest, sPLS-DA, difference test and SVM, and then quantified and verified through targeted metabolomics , and finally obtained 10 plasma exosome metabolites, which can be used as biomarkers to efficiently predict whether an individual has pancreatic cancer.

In some ways, the biomarkers that can be used to predict whether an individual is a pancreatic cancer reagent can be used to prepare detection reagents for the detection target, such as sample pretreatment reagents, antigens or antibodies, etc., which are suitable for the detection of the biomarkers. Biological reagents and kits; it can also be developed into standardized reagents or kits suitable for LC-UV or LC-MS detection of the biomarkers.

In some ways, the biomarkers of the present invention are obtained through plasma sample screening, and are particularly suitable for development into plasma detection reagents or kits for pancreatic cancer prediction.

Further, the detection of biomarkers in plasma samples is to detect the abundance or concentration of biomarkers of exosomes in plasma samples of individuals.

Further, the biomarker is selected from one or more of the following: 3-amino-2-piperidone, trans-urocanic acid, 4-cholesten-3-one, adenosine, adenine, Serine, N-acetylserine, 1-stearic acid-2-arachidonoyl diacylglycerol phosphatidylserine, 1-stearic acid-2-arachidonoyl diacylglycerol phosphatidylserine, N-acetylneuramine acid.

By examining the concentration differences of biomarkers in plasma exosomes of pancreatic cancer patients and normal people, and based on the verification results of targeted metabolomics, pancreatic cancer patients and pancreatitis patients or patients with pancreatic cancer or pancreatitis were further selected from 10 biomarkers. The 3 biomarkers that change stably between normal controls can be used to more effectively distinguish or predict the risk of pancreatic cancer, or be used to build a diagnostic model for pancreatic cancer.

Further, the reagent is used to detect biomarkers in plasma exosomes.

The present invention screens biomarkers of pancreatic cancer from plasma exosomes. These biomarkers have significant differences in the plasma exosomes of pancreatic cancer patients and non-pancreatic cancer patients (healthy and pancreatitis patients). By collecting Plasma exosome samples can be used to predict or assist in diagnosing whether the individual has pancreatic cancer or the possibility of having pancreatic cancer by detecting these biomarkers in an individual's plasma exosomes, or can detect a certain group of plasma exosomes These biomarkers were used to classify the group into pancreatic cancer group or non-pancreatic cancer group.

On the other hand, the present invention provides a kit or chip for predicting whether an individual has pancreatic cancer, and the kit or chip includes a detection reagent for the biomarker as described above.

Further, the reagent is used to detect biomarkers in plasma exosomes.

In another aspect, the present invention provides a biomarker combination for predicting whether an individual has pancreatic cancer. The biomarker combination includes the following biomarker combination: 3-amino-2-piperidone, trans Urocanic acid, 4-cholesten-3-one, adenosine, adenine, serine, N-acetylserine, 1-stearoyl-2-arachidonoylglycerophosphatidylserine, 1-stearoyl- 2-arachidonoylglycerol phosphoethanolamine, N-acetylneuraminic acid.

Further, the biomarker combination includes the following biomarker combination: adenosine, adenine, and N-acetylneuraminic acid.

In another aspect, the present invention provides a system for predicting whether an individual has pancreatic cancer. The system includes a data analysis module; the data analysis module is used to analyze the detection value of a biomarker, and the biomarker is selected from One or more of the following: 3-amino-2-piperidone, trans-urocanic acid, 4-cholesten-3-one, adenosine, adenine, serine, N-acetylserine, 1-hard Fatty acyl-2-arachidonoylglycerophosphatidylserine, 1-stearoyl-2-arachidonoylglycerophosphoethanolamine, N-acetylneuraminic acid.

Further, the biomarker is selected from one or more of the following: 3-amino-2-piperidone, trans-urocanic acid, 4-cholesten-3-one, adenosine, adenine, N-acetylneuraminic acid.

Further, the biomarker is selected from one or more of the following: adenosine, adenine, and N-acetylneuraminic acid.

Further, the detection value of the biomarker is the detection value of the biomarker in plasma exosomes.

Further, the detection value of the biomarker is the abundance or concentration of the biomarker in the plasma exosome sample of the detected individual.

Further, the data analysis module uses random forests or logistic regression equations to build models for analysis.

Further, the data analysis module calculates the predictive value of predicting whether the individual has pancreatic cancer by substituting the detection values of the biomarkers into the logistic regression equation, thereby evaluating whether the individual has pancreatic cancer.

Further, the logistic regression equation is:

Z=45.4514*adenosine-71.4211*adenine-0.2959*N-acetylneuraminic acid-3.1162;

Among them, the biomarker name represents the concentration (ng/mL) of the corresponding biomarker in the plasma exosome sample.

Furthermore, when Z is greater than -0.697, the probability of predicting the individual to have pancreatic cancer is high; when Z is less than -0.697, the probability of predicting the individual to be pancreatic cancer is low.

In yet another aspect, the present invention provides the use of the system as described above for constructing a detection model of a probability value for predicting whether an individual has pancreatic cancer.

The beneficial effects of the present invention are:

1. Screened out 10 new plasma exosome biomarkers that can early predict the risk of pancreatic cancer (PC);

2. Screen out random forest diagnostic models for pancreatic cancer using 1, 3, 6, and 10 biomarkers, and find that using 3 biomarkers to build a model for pancreatic cancer is the best;

3. Comparing the generalized linear regression model and the random forest model constructed using three biomarkers, it was found that the generalized linear regression model can further improve the detection accuracy and can be used to predict whether an individual has pancreatic cancer more efficiently, with an AUC value of 0.968;

4. It only needs to collect samples from plasma for testing. The method is convenient and has great advantages and prospects in clinical practice.

Description of drawings

Figure 1 is a flow chart for screening biomarkers in plasma exosomes through metabolomics in Example 1;

Figure 2 is the structural formula of 3-amino-2-piperidone in Example 1;

Figure 3 is the structural formula of trans-urocanic acid in Example 1;

Figure 4 is the structural formula of 4-cholesten-3-one in Example 1;

Figure 5 is the structural formula of adenosine in Example 1;

Figure 6 is the structural formula of adenine in Example 1;

Figure 7 is the structural formula of N-acetylneuraminic acid in Example 1;

Figure 8 is the structural formula of 1-stearoyl-2-arachidonoylglycerolphosphatidylserine in Example 1

Figure 9 is the structural formula of 1-stearoyl-2-arachidonoylglycerolphosphoethanolamine in Example 1

Figure 10 shows the 10 metabolites with the greatest impact in the predictions constructed in Example 2;

Figure 11 is the ROC curve of the model constructed by adenosine to predict whether pancreatic cancer is present in Example 2;

Figure 12 is the ROC curve of the model constructed with adenine to predict whether pancreatic cancer is present in Example 2;

Figure 13 is the ROC curve of the pancreatic cancer model predicted by N-acetylneuraminic acid in Example 2;

Figure 14 is the ROC curve of the model for predicting whether adenosine, adenine, and N-acetylneuraminic acid jointly predict pancreatic cancer in Example 2.

Detailed ways

The present invention will be described in further detail below with reference to the accompanying drawings and examples. It should be noted that the following examples are intended to facilitate the understanding of the present invention and do not limit it in any way. The reagents used in this example are all known products and were obtained by purchasing commercially available products.

Example 1 Screening biomarkers of pancreatic cancer in plasma exosomes using metabolomics

This example first conducts non-targeted metabolomics research and uses the UPLC-MS/MS ultra-high performance liquid chromatography-tandem mass spectrometry method to analyze three groups of plasma exosome samples from the healthy group, pancreatitis patients, and pancreatic cancer patient groups. . Secondly, build a model through random forest or logistic regression equation to screen metabolites that are significantly different between pancreatic cancer samples and control samples, conduct targeted metabolomics quantification and verification, and select the screened significantly different metabolites. , and finally obtained three plasma exosome metabolites as biomarkers, and verified the role of these biomarkers in the diagnosis or differentiation of pancreatic cancer (see Figure 1 for the flow chart).

Specific steps are as follows:

1. Experimental methods

①Sample collection

Patients with pancreatic cancer, patients with chronic pancreatitis, and healthy controls were recruited, including age-matched normal individuals or individuals without pancreatic disease (e.g., patients with indirect inguinal hernia). Collect 8-12 mL of blood samples from these three groups of patients, centrifuge at 1600 × g for 15 minutes at 4°C, and then 3000 × g for 15 minutes to obtain 5-7 mL of plasma samples, which are stored at -80°C for processing.

②Sample processing

Exosomes were isolated from plasma samples using the classic ultracentrifugation method. Cells and debris in plasma samples from Xiamen Lixin Technology Co., Ltd. were centrifuged at 2000×g for 30 minutes and 10000×g for 45 minutes at 4°C. The supernatant was filtered with a 0.45 μm filter membrane. Exosomes in plasma were then ultracentrifuged using a TI70 rotor at 100,000 × g for 70 min at 4°C. After discarding the supernatant, resuspend the exosomes in pre-cooled 1x PBS and ultracentrifuge again at 100,000 × g and 4°C for 70 min. The exosomes were resuspended in an appropriate amount of PBS and sent to determine the protein concentration and characterize the exosomes. The remaining exosomes were stored at -80°C.

③LC-MS/MS detection and data processing

Extract m/z ions from the original mass spectrum data obtained by LC-MS/MS detection, search the database to retrieve and identify metabolites, check the integration of the metabolite chromatographic peaks to obtain the peak area, and perform data normalization and missing value filling to obtain the data matrix. Follow-up bioinformatics analysis was performed, including sPLS-DA (sparse partial least squares discriminant analysis), volcano (volcano plot), generalized linear regression model, random forest and other statistical methods to screen the differences between pancreatic cancer samples and control samples. Ranked list of differential metabolites most effective in grouping samples.

2. Experimental results

Through sPLS-DA, difference test and volcano plot, 10 metabolites with the greatest differences between groups were screened out, that is, 10 biomarkers, as shown in Table 1.

Table 1. 10 pancreatic cancer biomarkers

Example 2: Pancreatic cancer prediction model

This embodiment uses a single biomarker or a combination of multiple biomarkers selected in Example 1 to establish a prediction or diagnosis model for pancreatic cancer. These models are used to distinguish pancreatic cancer from non-pancreatic cancer, or to screen pancreatic cancer patients from the population, or to predict whether an individual is a pancreatic cancer patient or the likelihood of an individual getting colorectal cancer. The specific models are as follows.

1. Single biomarker

Use R language software to process data. According to the grouping of pancreatic cancer patients and non-pancreatic cancer people, determine the concentration changes of metabolites in plasma exosome samples of pancreatic cancer patients and non-pancreatic cancer people, and screen out the 10 metabolites that have the greatest impact on the grouping in sPLS-DA. (Figure 8), combined with practical clinical applications, the calibration curve and ROC curve methods were further used to evaluate the regression model performance of six of the metabolites.

The analysis results proved that the six biomarkers were significantly correlated with the occurrence of pancreatic cancer. The analysis results are shown in Tables 2 and 3.

Table 2. Single biomarker ROC analysis results

The correlation between concentration changes of the six biomarkers and the occurrence of pancreatic cancer can be distinguished by the AUC values in Table 2. The higher the AUC value, the more accurately the biomarker can differentiate between people with pancreatic cancer and people without pancreatic cancer.

As can be seen from Table 2, if the concentration change of any one of the six biomarkers is used alone to distinguish the pancreatic cancer population from the non-pancreatic cancer population, the AUC value can reach more than 0.78, which is highly accurate. Among them, adenosine, 3-amino-2-piperidone, and trans-urocanic acid have the highest AUC value, with the AUC value reaching 0.911.

The 6 biomarkers provided in Table 2 were further quantified and verified through targeted metabolomics, and it was determined that adenosine, adenine, and N-acetylneuraminic acid have a clear and stable correlation with patients with pancreatic cancer. After optimizing the regression model, adenosine The AUC value further improved to 0.952 (Figure 9).

2. Combination of multiple biomarkers

Although a single biomarker can distinguish pancreatic cancer from non-pancreatic cancer plasma exosome samples or predict pancreatic cancer, the accuracy of its differentiation or prediction and the stability between individuals may be low. Therefore, multi-factor regression analysis was performed on the final three metabolites adenosine, adenine, and N-acetylneuraminic acid to establish a generalized linear regression evaluation model for predicting whether an individual has pancreatic cancer:

Z=45.4514*adenosine-71.4211*adenine-0.2959*N-acetylneuraminic acid-3.1162;

Among them, the biomarker name represents the concentration (ng/mL) of the corresponding biomarker in the plasma exosome sample.

Furthermore, when Z is greater than -0.697, the probability of predicting the individual to have pancreatic cancer is high; when Z is less than -0.697, the probability of predicting the individual to be pancreatic cancer is low.

The ROC curve of the logistic regression model provided in this embodiment for predicting whether an individual has pancreatic cancer is shown in Figure 12. The AUC value reaches 0.957, which is significantly improved compared to the separate regression model of the three biomarkers.

This generalized linear regression model is used to predict whether an individual has pancreatic cancer, and 20 clinically known pancreatic cancer patients and 31 non-pancreatic cancer patients (including 14 cases of chronic pancreatitis and 17 healthy controls) are used as the total data set. Carry out analysis and the analysis results are shown in Table 3.

Table 3. Analysis results of the model for predicting whether an individual has pancreatic cancer

As can be seen from Table 3, using the three biomarkers individually and combined to construct a generalized linear regression evaluation model to predict whether an individual has pancreatic cancer, 19 of the 20 pancreatic cancer patients were detected, and the sensitivity reached 95 %; among 31 non-pancreatic cancer patients, 1 pancreatitis patient was classified as a pancreatic cancer patient area, with a specificity of more than 96%; compared with CA19-9, the most widely used clinical pancreatic cancer marker (accuracy 70 %) Its predictive performance has been effectively improved.

Although the present invention is disclosed as above, the present invention is not limited thereto. Any person skilled in the art can make various changes and modifications without departing from the spirit and scope of the present invention. Therefore, the protection scope of the present invention should be subject to the scope defined by the claims.

Claims (11)

1. The use of biomarkers in preparing reagents for diagnosing whether an individual has pancreatic cancer. The biomarkers are selected from one or several combinations of the following: 3-amino-2-piperidone, trans-urine Acid, 4-cholesten-3-one, adenosine, adenine, serine, N-acetylserine, 1-stearoyl-2-arachidonoylglycerophosphatidylserine, 1-stearoyl-2- Arachidonoylglycerol phosphoethanolamine, N-acetylneuraminic acid. 2. The use according to claim 1, the biomarker is selected from one or more of the following: 3-amino-2-piperidone, trans-urocanic acid, 4-cholestene-3 -Ketones, adenosine, adenine, N-acetylneuraminic acid. 3. The use according to claim 1, wherein the biomarker is selected from one or more of the following: adenosine, adenine, and N-acetylneuraminic acid. 4. The use according to claim 3, wherein the combination of markers includes a combination of the following markers: adenosine, adenine, and N-acetylneuraminic acid. 5. A system for predicting whether an individual has pancreatic cancer, characterized in that the system includes a data analysis module; the data analysis module is used to analyze the detection values of biomarkers, and the biomarkers are selected from the following One or more: 3-amino-2-piperidone, trans-urocanic acid, 4-cholesten-3-one, adenosine, adenine, serine, N-acetylserine, 1-stearoyl -2-arachidonoylglycerophosphatidylserine, 1-stearoyl-2-arachidonoylglycerophosphoethanolamine, N-acetylneuraminic acid. 6. The system of claim 5, wherein the biomarker is selected from one or more of the following: 3-amino-2-piperidone, trans-urocanic acid, 4-cholester En-3-one, adenosine, adenine, N-acetylneuraminic acid. 7. The system of claim 6, wherein the biomarker is selected from one or more of the following: adenosine, adenine, and N-acetylneuraminic acid. 8. The system according to claim 7, wherein the detection value of the biomarker is a detection value of the biomarker for detecting exosomes in plasma. 9. The system of claim 5, wherein the data analysis module uses random forest or logistic regression equations to build a model for analysis. 10. The system according to claim 69, wherein the data analysis module calculates the predictive value of predicting whether the individual has pancreatic cancer by substituting the detection value of the biomarker into the logistic regression equation, thereby evaluating whether the individual has pancreatic cancer. Cancer, among others. The logistic regression equation is: Z=45.4514*adenosine-71.4211*adenine-0.2959*N-acetylneuraminic acid-3.1162. 11. The system of claim 10, wherein further, when Z is greater than -0.697, the probability of predicting the individual to have pancreatic cancer is high; when Z is less than -0.697, the probability of predicting the individual to be pancreatic cancer is low.