Application of metabolites in the diagnosis of sepsis-related encephalopathy

Abstract

The present invention discloses the application of metabolites in diagnosing sepsis-related encephalopathy. The metabolite 3-Methoxy prostaglandin F1α provided by the present invention can provide high diagnostic efficacy for diagnosing SAE in sepsis patients alone. The present invention will provide direction for diagnosing SAE using metabolites. The present invention has the advantages of being objective and convenient in clinical application and not being affected by the evaluation of medical staff. It can more objectively and conveniently identify SAE patients at an early stage, thereby improving the prognosis of patients through early intervention.

Description

Technical Field

[0001] The invention belongs to the field of biotechnology and relates to the application of metabolites in diagnosing sepsis-related encephalopathy, and specifically relates to the metabolites of ar-Turmerone and / or 3-Methoxyprostaglandin F1α. Background Art

[0002] Sepsis is a life-threatening organ dysfunction caused by the body's dysregulated response to infection. Sepsis-associated encephalopathy (SAE) is a diffuse brain dysfunction associated with sepsis, but without direct central infection or other types of brain diseases. It can occur in 70% of sepsis patients, and the mortality rate of SAE patients can reach more than 40%. In clinical practice, the diagnosis of SAE in sepsis patients is an exclusionary diagnosis. There are great differences in the diagnosis and detection of SAE among medical institutions at home and abroad, and it is relatively complicated. Commonly used scores include Glasgow Coma Scale (GCS) and ICU Confusion Assessment Method (CAM-ICU), but the implementation of the score is time-consuming and inconvenient, and requires the patient's interactive cooperation, and is also subjective. Other monitoring includes imaging examinations, electroencephalograms, electroencephalograms and other neurological monitoring, but the diagnostic efficiency of SAE is not high. There are also reports on the detection of brain injury markers, but it is not routinely popularized in clinical practice, so the diagnosis of SAE in clinical practice is underestimated and lacks effective diagnostic markers. Summary of the invention

[0003] In order to solve the technical problems existing in the prior art, the present invention provides the following technical solutions:

[0004] The present invention provides the use of a reagent for detecting metabolites in a sample in the preparation of a product for predicting / diagnosing sepsis-related encephalopathy, wherein the metabolites are ar-Turmerone and / or 3-Methoxyprostaglandin F1α.

[0005] Furthermore, the product is a drug, a test kit, or a chip.

[0006] Furthermore, the sample is selected from tissue sample, cell supernatant, cell lysate, serum, plasma, vitreous fluid, lymph fluid, synovial fluid, follicular fluid, semen, amniotic fluid, milk, whole blood, blood-derived cells, urine, cerebrospinal fluid, saliva, sputum, tears, or sweat.

[0007] Furthermore, the sample is selected from serum, plasma, whole blood, and blood-derived cells.

[0008] Furthermore, the sample is plasma.

[0009] Furthermore, the sample is from a human or a non-human mammal.

[0010] Furthermore, the sample is from a human.

[0011] In some embodiments, ar-Turmerone can be used alone as a diagnostic / predictive indicator for predicting / diagnosing sepsis-related encephalopathy. In some embodiments, 3-Methoxy prostaglandin F1α can be used alone as a diagnostic / predictive indicator for predicting / diagnosing sepsis-related encephalopathy. In some embodiments, the ar-Turmerone and 3-Methoxy prostaglandin F1α can be used in combination as a diagnostic / predictive indicator for predicting / diagnosing sepsis-related encephalopathy. In some embodiments, the predicting / diagnosing sepsis-related encephalopathy is to screen patients with sepsis-related encephalopathy from subjects, and the subjects can be any population.

[0012] The term "ar-Turmerone" as used in the present invention refers to aromatic turmerone.

[0013] The term "3-Methoxy prostaglandin F1α" used in the present invention refers to 3-methoxy prostaglandin F1α.

[0014] The term "sample" as used in the present invention refers to a composition obtained or derived from a subject (e.g., an individual of interest) that contains cells and / or other molecular entities to be characterized and / or identified based on, for example, physical, biochemical, chemical and / or physiological characteristics.

[0015] The present invention provides the use of a reagent for detecting metabolites in a sample in the preparation of a product for predicting / diagnosing whether a sepsis patient has sepsis-related encephalopathy, wherein the metabolites are ar-Turmerone and / or 3-Methoxyprostaglandin F1α.

[0016] Furthermore, the product is a drug, a test kit, or a chip.

[0017] Furthermore, the sample is selected from tissue sample, cell supernatant, cell lysate, serum, plasma, vitreous fluid, lymph fluid, synovial fluid, follicular fluid, semen, amniotic fluid, milk, whole blood, blood-derived cells, urine, cerebrospinal fluid, saliva, sputum, tears, or sweat.

[0018] Furthermore, the sample is selected from serum, plasma, whole blood, and blood-derived cells.

[0019] Furthermore, the sample is plasma.

[0020] Furthermore, the sample is from a human or a non-human mammal.

[0021] Furthermore, the sample is from a human.

[0022] In some embodiments, ar-Turmerone can be used alone as a diagnostic / predictive indicator for predicting / diagnosing whether a sepsis patient has sepsis-related encephalopathy. In some embodiments, 3-Methoxyprostaglandin F1α can be used alone as a diagnostic / predictive indicator for predicting / diagnosing whether a sepsis patient has sepsis-related encephalopathy. In some embodiments, the ar-Turmerone and 3-Methoxy prostaglandin F1α can be used in combination as diagnostic / predictive indicators for predicting / diagnosing whether a sepsis patient has sepsis-related encephalopathy. In some embodiments, the prediction / diagnosis of whether a sepsis patient has sepsis-related encephalopathy is to screen out sepsis-related encephalopathy patients from confirmed sepsis patients, and the screened population is confirmed sepsis patients.

[0023] In the context of the present invention, the term "subject" used is equivalent to "object", "patient", and refers to any animal, including humans and non-human animals. Non-human animals include all vertebrates, for example, mammals, such as non-human primates (especially higher primates), sheep, dogs, rodents (such as mice or rats), guinea pigs, goats, pigs, cats, rabbits, cattle, and any livestock or pets; and non-mammals, such as amphibians, reptiles, etc. In a preferred embodiment, the subject is a human.

[0024] The present invention provides a product, which comprises a reagent for detecting a metabolite; the metabolite is ar-Turmerone and / or 3-Methoxyprostaglandin F1α.

[0025] Furthermore, the product is a drug, a test kit, or a chip.

[0026] Furthermore, the reagents for detecting metabolites include: nuclear magnetic resonance, liquid chromatography-mass spectrometry, gas chromatography-mass spectrometry, and reagents used in chromatography.

[0027] Furthermore, the reagents used in the chromatography include reagents used in HPLC and GC.

[0028] In some embodiments, the chromatography includes liquid chromatography, gas chromatography (GC), high performance liquid chromatography (HPLC), etc. In some embodiments, the ultra high performance liquid chromatography (UHPLC) also includes ultra high performance liquid chromatography (UHPLC).

[0029] Furthermore, the product is used to predict / diagnose sepsis-related encephalopathy or predict / diagnose whether a sepsis patient has sepsis-related encephalopathy.

[0030] The term "agent" used in the present invention should not be interpreted narrowly, but should be extended to small molecules, protein molecules such as peptides, polypeptides and proteins, and compositions comprising them, as well as genetic molecules such as RNA, DNA and their mimetics and chemical analogs and cellular agents.

[0031] The term "chromatography" used in the present invention refers to a method for separating related solutes in a mixture from other solutes in the mixture during binding and elution due to different rates of each solute in the mixture passing through a solid phase medium under the influence of a certain mobile phase. Common chromatographic matrices include ion exchange chromatographic matrices and hydrophobic interaction chromatographic matrices, and chromatographic matrices with the above-mentioned mixed mode effects, such as the MEP HyperCel filler matrix of Pall Corporation used in the present invention, are also common knowledge to those skilled in the art.

[0032] The term "kit" used in the present invention is any article (e.g., package or container) comprising at least one reagent (e.g., a drug for treating a condition or a probe for specifically detecting an analyte of the present invention). The kit is preferably promoted, distributed or sold in the form of a unit for performing the method of the present invention. Typically, the kit may further include a separated carrier device to receive one or more container devices, such as vials, tubes, etc., in a closed space. In particular, the container device may include one of the individual elements to be used in the method of the first aspect. The kit may further include one or more other containers, containing other materials, including but not limited to buffers, diluents, filters, needles, syringes, and package inserts with instructions for use. A label may be present on the container to indicate that the composition is used for a specific purpose, and may also indicate guidelines for in vivo or in vitro use.

[0033] The term "liquid chromatography" or "LC" used in the present invention refers to the process of selectively retarding one or more components in a fluid solution when the fluid uniformly penetrates through a column of a particulate material or through a capillary channel. The retardation is caused by the distribution of the components of the mixture between the one or more stationary phases and the free fluid (i.e., mobile phase) when the fluid moves relative to the one or more stationary phases. Methods in which the polarity of the stationary phase is higher than that of the mobile phase (e.g., using toluene as the mobile phase and silica gel as the stationary phase) are called normal phase liquid chromatography (NPLC), and methods in which the polarity of the stationary phase is lower than that of the mobile phase (e.g., using a water-methanol mixture as the mobile phase and C18 (octadecylsilyl) as the stationary phase) are called reverse phase liquid chromatography (RPLC).

[0034] The term "high performance liquid chromatography" or "HPLC" as used herein refers to a liquid chromatography method in which separation is increased by forcing the mobile phase under pressure through a stationary phase, typically a densely packed column. Typically, the column is filled with a stationary phase consisting of irregular or spherical particles, a porous monolithic layer, or a porous membrane. In the past, HPLC was divided into two different subclasses, NP-HPLC and RP-HPLC, depending on the polarity of the mobile and stationary phases.

[0035] As used herein, the term "ultra-high performance liquid chromatography" or "UHPLC" refers to HPLC methods that use high pressure (eg, 120 MPa) (17,405 lbf / in2) or about 1200 atmospheres.

[0036] The present invention provides the use of a reagent for detecting metabolites in a sample in preparing a computer diagnosis or prediction system, a device, and a computer-readable storage medium for determining whether sepsis is sepsis-related encephalopathy, wherein the metabolites are ar-Turmerone and / or 3-Methoxyprostaglandin F1α.

[0037] Furthermore, the sample is selected from tissue sample, cell supernatant, cell lysate, serum, plasma, vitreous fluid, lymph fluid, synovial fluid, follicular fluid, semen, amniotic fluid, milk, whole blood, blood-derived cells, urine, cerebrospinal fluid, saliva, sputum, tears, or sweat.

[0038] Furthermore, the sample is selected from serum, plasma, whole blood, and blood-derived cells.

[0039] Furthermore, the sample is plasma.

[0040] Furthermore, the sample is from a human or a non-human mammal.

[0041] Furthermore, the sample is from a human.

[0042] The present invention provides the use of a reagent for detecting metabolites in a sample in the preparation of a computer diagnosis or prediction system, a device, and a computer-readable storage medium for predicting / diagnosing sepsis-related encephalopathy, wherein the metabolites are ar-Turmerone and / or 3-Methoxyprostaglandin F1α.

[0043] Furthermore, the sample is selected from tissue sample, cell supernatant, cell lysate, serum, plasma, vitreous fluid, lymph fluid, synovial fluid, follicular fluid, semen, amniotic fluid, milk, whole blood, blood-derived cells, urine, cerebrospinal fluid, saliva, sputum, tears, or sweat.

[0044] Furthermore, the sample is selected from serum, plasma, whole blood, and blood-derived cells.

[0045] Furthermore, the sample is plasma.

[0046] Furthermore, the sample is from a human or a non-human mammal.

[0047] Furthermore, the sample is from a human.

[0048] The present invention provides a computer diagnosis or prediction system, which includes a result determination unit for analyzing the result of the level of the detected metabolite and outputting the result.

[0049] Furthermore, the metabolite is ar-Turmerone and / or 3-Methoxyprostaglandin F1α.

[0050] Furthermore, the computer diagnosis or prediction system includes a detection unit for detecting the level of metabolites.

[0051] Furthermore, the result determination unit includes an input part for inputting the level of the metabolite.

[0052] Furthermore, the metabolite is ar-Turmerone and / or 3-Methoxyprostaglandin F1α.

[0053] Furthermore, the result determination unit includes an analysis part for analyzing whether the subject suffers from sepsis-related encephalopathy according to the level of the metabolite, or analyzing whether the sepsis patient is a sepsis-related encephalopathy patient.

[0054] Furthermore, the metabolite is ar-Turmerone and / or 3-Methoxyprostaglandin F1α.

[0055] Furthermore, the result determination unit includes an output part for outputting the analysis result of the analysis part.

[0056] The present invention provides a computer diagnosis or prediction device, the device comprising:

[0057] A detection module, used for detecting the content of metabolites used for diagnosis / prediction of sepsis-related encephalopathy in a sample to be tested; the metabolites are ar-Turmerone and / or 3-Methoxyprostaglandin F1α;

[0058] The analysis module is used to compare the obtained test results with those of healthy persons or patients suffering from sepsis only, and determine whether it is sepsis-related encephalopathy or the risk of sepsis-related encephalopathy.

[0059] Furthermore, the device includes an input module for obtaining the detection result of the detection module.

[0060] Furthermore, the acquired detection result is a detection result acquired by an input module.

[0061] Furthermore, the device includes an output module for outputting analysis results.

[0062] Furthermore, the test results include significant difference results or ROC results.

[0063] In some embodiments, the significant difference result is that the level of the metabolite in the sample is lower than the level of the metabolite in patients with non-sepsis-related encephalopathy and shows significant difference, usually referring to a p value less than 0.05, which is called a significant difference result.

[0064] The present invention provides a computer-readable storage medium, which includes a computer program. The computer program is executed by a processor, and when the computer program is executed by the processor, the functions of different units in the computer diagnosis or prediction system described above are realized.

[0065] The present invention provides a method for constructing a sepsis-related encephalopathy diagnosis / prediction model. The method comprises the following steps: a computer program uses metabolites to construct a sepsis-related encephalopathy diagnosis / prediction model. The metabolites are ar-Turmerone and / or 3-Methoxyprostaglandin F1α.

[0066] Furthermore, the construction method includes: obtaining the level data of metabolites in the sample through a computer program, and the computer program constructs a ROC analysis curve according to the level data to obtain the critical value and sensitivity of diagnosis / prediction.

[0067] Furthermore, the level data includes concentration and / or ratio.

[0068] Furthermore, the construction method comprises the following steps:

[0069] The three metabolites ar-Turmerone and / or 3-Methoxy prostaglandin F1α, which had the same trend of content changes in sepsis-related encephalopathy and healthy people, were used as predictors, and a classification model was constructed using a machine learning algorithm. The prediction performance of the model was evaluated using accuracy, specificity, recall rate, positive predictive value, negative predictive value, F1 score and Kappa coefficient.

[0070] Furthermore, the machine learning algorithms include linear regression algorithm, support vector machine algorithm, nearest neighbor / k-nearest neighbor algorithm, logistic regression algorithm, decision tree algorithm, k-means algorithm, random forest algorithm, naive Bayes algorithm, dimensionality reduction algorithm, and gradient boosting algorithm.

[0071] The term "machine learning algorithm" used in the present invention refers to an algorithm that directly uses machine learning / computer learning capabilities without using fixed programming, so that the machine can calculate on its own and obtain an algorithm program / algorithm formula that can process or predict data.

[0072] Furthermore, the construction method comprises the following steps:

[0073] A classification model was constructed using a machine learning algorithm, using ar-Turmerone and / or 3-Methoxy prostaglandin F1α, three metabolites with consistent content change trends in sepsis-related encephalopathy and sepsis. The prediction performance of the model was evaluated using accuracy, specificity, recall rate, positive predictive value, negative predictive value, F1 score and Kappa coefficient.

[0074] Furthermore, the machine learning algorithms include linear regression algorithm, support vector machine algorithm, nearest neighbor / k-nearest neighbor algorithm, logistic regression algorithm, decision tree algorithm, k-means algorithm, random forest algorithm, naive Bayes algorithm, dimensionality reduction algorithm, and gradient boosting algorithm.

[0075] In some embodiments, the construction method uses a computer program or artificial intelligence to perform all operations, and when the operations are performed, the computer program or artificial intelligence controls the processor to achieve operations for different purposes.

[0076] In some embodiments, the threshold for ar-Turmerone diagnosis of SAE is an intensity ≤ 10 6 , the sensitivity is 0.7, the specificity is 0.9; the threshold for 3-Methoxy prostaglandin F1α to diagnose SAE is intensity ≤ 10 6 , at this time the sensitivity is 0.7 and the specificity is 0.9.

[0077] The present invention provides a sepsis-related encephalopathy diagnosis / prediction model, and the sepsis-related encephalopathy diagnosis / prediction model is obtained according to the aforementioned construction method.

[0078] Advantages and beneficial effects of the present invention:

[0079] The metabolites ar-Turmerone and 3-Methoxyprostaglandin F1α provided by the present invention can provide high diagnostic efficacy for diagnosing SAE in sepsis patients alone or in combination. The present invention will provide a direction for diagnosing SAE using metabolites. The present invention has the advantages of being objective and convenient in clinical applications and not being affected by the evaluation of medical staff. It can more objectively and conveniently identify SAE patients at an early stage, thereby improving the prognosis of patients through early intervention. BRIEF DESCRIPTION OF THE DRAWINGS

[0080] Figure 1 It is the box plot and ROC curve graph of the development group ar-Turmerone and 3-Methoxyprostaglandin F1α;

[0081] Figure 2 is the box plot and ROC curve of ar-Turmerone and 3-Methoxyprostaglandin F1α in the validation group;

[0082] Figure 3 is the ROC curve and average accuracy graph of the combined diagnosis of SAE by ar-Turmerone and 3-Methoxyprostaglandin F1α in the validation group;

[0083] Figure 4 It is the ROC curve and average accuracy graph of the development group's ar-Turmerone and 3-Methoxyprostaglandin F1α combined diagnosis of SAE. DETAILED DESCRIPTION

[0084] Receiver operating characteristic curve (ROC curve): It is a curve drawn based on a series of different binary classification methods (cutoff values ​​or decision thresholds), with sensitivity (true positive rate) as the ordinate and 1-specificity (false positive rate) as the abscissa. The area under the ROC curve is an important indicator of test accuracy. The larger the area under the ROC curve, the greater the diagnostic value of the test.

[0085] Example

[0086] 1. Experimental subjects

[0087] Patients with sepsis hospitalized in the emergency department of Peking Union Medical College Hospital were recruited.

[0088] Patient inclusion criteria:

[0089] (1) Adult patients (≥18 years old) who meet the diagnostic criteria for sepsis-3 within 7 days;

[0090] (2) The Glasgow Coma Scale (GCS) and Confusion Assessment Method in Intensive Care Unit (CAM-ICU) scores were used for assessment at enrollment.

[0091] Exclusion criteria:

[0092] (1) Primary brain diseases, including central nervous system infection, craniocerebral injury, stroke, intracranial tumor, epilepsy, etc.;

[0093] (2) Neuropsychiatric diseases with chronic changes in consciousness, including Alzheimer's disease, Parkinson's disease, depression, schizophrenia, etc.;

[0094] (3) Consciousness disturbance caused by primary metabolic disorders and organ dysfunction, including severe electrolyte disorders, dysglycemia, Wernicke's encephalopathy, hepatic encephalopathy, pulmonary encephalopathy, etc. For sepsis patients with clinical manifestations of acute encephalopathy (delirium, coma, epileptic seizures, focal neurological deficits and varying degrees of consciousness) or GCS < 15 points or positive CAM-ICU examination, SAE is considered after excluding other potential causes through clinical or laboratory examinations.

[0095] The sepsis patients included in this study were divided into SAE group and non-SAE group according to whether SAE occurred when the sample was collected. All subjects collected blood samples with EDTA anticoagulation tubes after admission. The samples were centrifuged at 2000g for 10 minutes within 2 hours after collection to separate plasma samples and collect clinical information of each patient. Finally, plasma from 127 non-SAE patients and plasma from 42 SAE patients were obtained. They were randomly divided into the development group (90 non-SAE patients and 25 SAE patients) and the validation group (37 non-SAE patients and 17 SAE patients).

[0096] The disease information of patients in the development group (development cohort, test set) is shown in Table 1.

[0097] Table 1

[0098]

[0099]

[0100]

[0101] The disease information of patients in the validation group (validation cohort, validation set) is shown in Table 2.

[0102] Table 2

[0103]

[0104]

[0105] 2. Experimental methods

[0106] This project is based on liquid chromatography-mass spectrometry (LC-MS) technology for non-targeted metabolomics research. Sample analysis was performed in positive and negative ionization modes. The raw files (.raw) obtained by mass spectrometry detection were imported into Compound Discoverer 3.3 software for spectral processing and database search to obtain the qualitative and quantitative results of metabolites, and then the data were quality controlled to ensure the accuracy and reliability of the data results. Compounds with relative peak area coefficient of variation (CV) values ​​greater than 30% in quality control samples were removed, and finally the identification and relative quantitative results of metabolites were obtained. We first used MetaboAnalyst (https: / / www.metaboanalyst.ca) (version 6.0) to perform statistical analysis of metabolites on the non-targeted metabolomics data of patients in the development group. The metabolite abundance data of patients in the SAE group and non-SAE group were obtained by two-sample t-test to obtain P values ​​and FDR, and orthogonal partial least squares discriminant analysis (OPLS-DA) was used to obtain the projected variable importance (VIP) value. Differential metabolites were selected according to VIP>1 and FDR<0.05.

[0107] 3. Experimental results

[0108] ROC curves were drawn for the differential metabolites to see their ability to diagnose SAE. The cutoff value was determined according to the maximum Youden index, and metabolites with an area under the ROC curve (AUC) > 0.8 for diagnosing SAE in the development group (test set) were selected, namely ar-Turmerone: AUC = 0.838 (0.746-0.917), 3-Methoxy prostaglandin F1α: AUC = 0.851 (0.732-0.954). The results are as follows Figure 1 shown.

[0109] Then the above two metabolite data of the patients in the validation group were analyzed in the same way as above, and the AUC values ​​of the ROC curve for diagnosing SAE were obtained as follows: ar-Turmerone: AUC = 0.868 (0.740-0.955), 3-Methoxyprostaglandin F1α: AUC = 0.803 (0.654-0.905). Figure 2 shown.

[0110] The diagnostic efficacy of the analogs was compared: (1) There were no statistically significant differences between the SAE and non-SAE groups for the other 15 prostaglandin F analogs detected in the data except 3-Methoxyprostaglandin F1α, as shown in Table 3.

[0111] Table 3

[0112]

[0113]

[0114] The two metabolites were further combined to diagnose SAE. The AUC and average accuracy after 100-fold cross-validation in the development cohort were 0.893 and 0.800, respectively; the AUC and average accuracy in the validation cohort were 0.909 and 0.835, respectively. Figure 3 , Figure 4 shown.

[0115] The description of the above embodiments is only used to understand the method and core idea of ​​the present invention. It should be pointed out that for ordinary technicians in this field, several improvements and modifications can be made to the present invention without departing from the principle of the present invention, and these improvements and modifications will also fall within the scope of protection of the claims of the present invention.

Claims

1. Use of a reagent for detecting metabolites in a sample in the preparation of a product for predicting / diagnosing whether a sepsis patient has sepsis-related encephalopathy, wherein the metabolite is 3-Methoxy prostaglandin F1α, and the sample is selected from serum, plasma, and whole blood.

2. The use according to claim 1, wherein the product is a drug, a kit or a chip.

3. A computer diagnosis or prediction system, comprising a result determination unit for analyzing the level of metabolites in a detected sample and outputting the result; the sample is selected from serum, plasma, and whole blood; The computer diagnosis or prediction system includes a detection unit for detecting the level of a metabolite in a sample; The result determination unit includes an input part for inputting the level of a metabolite; The result determination unit includes an analysis part, which is used to analyze whether the sepsis patient is a patient with sepsis-related encephalopathy according to the level of the metabolite; the metabolite is 3-Methoxy prostaglandin F1α; The result determination unit comprises an output part for outputting the analysis result of the analysis part.

4. A computer-readable storage medium, comprising a computer program, which is executed by a processor and implements the functions of different units in the computer diagnosis or prediction system according to claim 3 when the computer program is executed by the processor.

Images