Immune effector cell mediated neurotoxic syndrome risk prediction system and construction method and application thereof
By establishing a logistic multivariate analysis model based on the traceability concentration of free DNA in oligodendrocytes, B cells and megakaryocytes, the problem of early diagnosis of ICANS was solved, high sensitivity and specificity of prediction was achieved, and the misdiagnosis rate was reduced.
Patent Information
- Application Number
- CN202510953229.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-07-10
- Publication Date
- 2025-10-17
AI Technical Summary
In existing technologies, the diagnosis of immune effector cell-mediated neurotoxicity syndrome (ICANS) relies on clinical symptoms and lacks objective biomarkers, resulting in a high misdiagnosis rate, difficulty in providing early warning, and affecting treatment efficacy.
The concentrations of cell-free DNA traceability in oligodendrocytes, B cells, and megakaryocytes were used as predictive factors. A risk prediction model was established through logistic multivariate analysis. A prediction system was constructed using machine learning methods, and peripheral blood samples were used to assess ICANS risk.
It achieved early prediction of ICANS with a sensitivity of 80% and a specificity of 76%, reduced the misdiagnosis rate, provided a more accurate basis for early diagnosis, and reduced invasive examinations.
Smart Images
Figure CN120809212A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application relates to the field of biotechnology, in particular to a risk prediction system for immune effector cell-mediated neurotoxicity syndrome and application thereof. BACKGROUND
[0002] Immune effector cell-mediated neurotoxicity syndrome (ICANS) is the second most common severe toxicity reaction in CAR-T cell therapy. The main symptoms of patients with toxic encephalopathy include coma, aphasia, ataxia, delirium, epilepsy and brain edema. Cell-free DNA, abbreviated as cfDNA, is mainly short fragment DNA, which is produced by apoptosis, necrosis or active release, and exists in human body fluids. During CAR-T cell therapy, CAR-T cells proliferate and target kill tumor cells. In this process, a large amount of CAR-T cell and tumor cell DNA fragments will be released. If CRS (cytokine release syndrome), ICANS or even organ toxicity occurs, a large amount of cell DNA fragments will also be released from related immune cells and damaged organs into body fluids such as peripheral blood, pleural effusion, cerebrospinal fluid, etc. The free DNA fragments from different tissue cells contain tissue cell-specific molecular characteristics which can be mined.
[0003] In clinical practice, the occurrence of ICANS is often diagnosed and predicted based on the abnormal symptoms and signs exhibited by patients, but the symptoms of ICANS are diverse and mostly abnormal in the nervous system, such as disturbance of consciousness, epilepsy, language disorders, etc. These symptoms are often similar to other common neurological diseases, and diagnosis relies on the high level of professional knowledge and rich clinical experience of doctors. In addition, since the pathogenesis of ICANS is not fully understood, existing prediction tools and biomarkers are not completely reliable and it is difficult to provide sufficient early warning information. The existing diagnosis of ICANS relies on clinical symptoms and lacks objective biomarkers, resulting in a misdiagnosis rate as high as 30%-40%. Therefore, there is an urgent need for effective objective indicators to promote the early identification and accurate diagnosis of immune effector cell-mediated neurotoxicity syndrome, help clinicians identify potential patients early, detect changes in the disease in a timely manner, take pre-emptive measures and nursing interventions, and have important significance for improving the success rate of patient treatment. SUMMARY
[0004] The present application aims to provide a set of immune effector cell-mediated neurotoxicity syndrome risk prediction systems and their construction methods and applications, which can be used to evaluate immune effector cell-mediated neurotoxicity syndrome and possibly guide the selection of treatment options.
[0005] To achieve the above-mentioned purpose, the technical scheme adopted by the present application is as follows: In a first aspect of the present application, a risk prediction system for immune effector cell-mediated neurotoxicity syndrome is provided, which comprises: At least one of the free DNA trace concentration of oligodendrocytes, B cells and megakaryocytes is used as a predictor independent variable, and the occurrence probability of immune effector cell-mediated neurotoxicity syndrome is used as a dependent variable, binary Logistic multifactor analysis is carried out, and a prediction model is established.
[0006] Further, the free DNA trace concentration of oligodendrocytes, B cells and megakaryocytes is used as an independent variable, and the formula of the risk prediction model of immune effector cell-mediated neurotoxicity syndrome is as follows: ln[p(1-p)]=-0.2989+α×A+β×B+γ×C; In the above formula, A is the free DNA trace concentration of oligodendrocytes, B is the free DNA trace concentration of B cells, and C is the free DNA trace concentration of megakaryocytes. The coefficients α, β and γ are obtained by performing Logistic multifactor regression analysis on the free DNA trace concentrations of oligodendrocytes, B cells and megakaryocytes.
[0007] As a specific embodiment, in the coefficients, α=3.8477, β=0.7318 and γ=-2.0054.
[0008] The model further comprises: The formula of the risk prediction model of immune effector cell-mediated neurotoxicity syndrome with the free DNA trace concentration of oligodendrocytes as an independent variable is as follows: ln[p(1-p)]=-1.7073+0.8621×A, A is the free DNA trace concentration of oligodendrocytes; The formula of the risk prediction model of immune effector cell-mediated neurotoxicity syndrome with the free DNA trace concentration of B cells as an independent variable is as follows: ln[p(1-p)]=-1.8460+0.4161×B, B is the free DNA trace concentration of B cells; The formula of the risk prediction model of immune effector cell-mediated neurotoxicity syndrome with the free DNA trace concentration of megakaryocytes as an independent variable is as follows: ln[p(1-p)]=-0.8882-0.3168×C, C is the free DNA trace concentration of megakaryocytes; The formula of the risk prediction model of immune effector cell-mediated neurotoxicity syndrome with the free DNA trace concentrations of B cells and megakaryocytes as independent variables is as follows: ln[p(1-p)]=-0.3052+0.6538xB-1.6449xC, B is the free DNA trace concentration of B cells, and C is the free DNA trace concentration of megakaryocytes; Taking the free DNA trace concentrations of oligodendrocytes and megakaryocytes as independent variables, the formula of the risk prediction model of the immune effector cell-mediated neurotoxicity syndrome is as follows: ln[p(1-p)]=-0.7623+1.9095xA-0.7190xC, A is the free DNA trace concentration of oligodendrocytes, and C is the free DNA trace concentration of megakaryocytes; Taking the free DNA trace concentrations of oligodendrocytes and B cells as independent variables, the formula of the risk prediction model of the immune effector cell-mediated neurotoxicity syndrome is as follows: ln[p(1-p)]=-1.7384-1.3564xA+0.4329xB, A is the free DNA trace concentration of oligodendrocytes, and B is the free DNA trace concentration of B cells.
[0009] In the second aspect of the present application, a method for constructing the risk prediction system of the immune effector cell-mediated neurotoxicity syndrome is provided, and the method comprises the following steps: Peripheral blood samples of the patient are collected regularly from the day of CAR-T reinfusion of the patient; Tissue trace data are obtained by performing tissue trace on the free DNA of the collected blood sample; The tissue trace data are sorted according to importance to obtain an elastic network factor importance sorting graph, the top four factors in importance are selected, a neural cell related factor is added, a logistic regression model is trained, and a clinical prediction model is constructed.
[0010] Further, in the tissue trace, a tissue cell trace method based on a free DNA methylation deconvolution method is used.
[0011] As a specific embodiment, peripheral blood samples of the patient are collected regularly from the day of CAR-T reinfusion of the patient; free DNA is extracted from the plasma sample, and the methylation level of the free DNA is detected.
[0012] The tissue trace method based on the methylation density deconvolution method is used to trace the free DNA to obtain tissue trace data.
[0013] In the third aspect of the embodiment of the present application, a risk prediction system of an immune effector cell-mediated neurotoxicity syndrome is provided, and the system comprises: a processor and a memory, the memory is coupled to the processor, and the memory stores instructions which, when executed by the processor, use the following steps: inputting the numerical values of the key model indicators into the prediction model to obtain the probability of occurrence of the immune effector cell-mediated neurotoxicity syndrome.
[0014] In the technical solution, the free DNA trace concentration of oligodendrocytes, the free DNA trace concentration of B cells, and the free DNA trace concentration of megakaryocytes are brought into the risk prediction model to obtain P values through calculation. The determination rule for predicting the risk of the immune effector cell-mediated neurotoxicity syndrome according to the P values obtained through calculation is as follows: If the P value is higher than the cutoff value, it is predicted that the patient will develop ICANS. If the P value is lower than the cutoff value, it is predicted that the patient will not develop ICANS. The cutoff value is 0.1244997.
[0015] In the fourth aspect of the embodiments of the present application, a computer readable storage medium is provided, which stores a computer program. When the computer program is executed by a processor, the following steps are used: inputting the numerical values of the key model indicators into the prediction model to obtain the probability of occurrence of the immune effector cell-mediated neurotoxicity syndrome.
[0016] In the fifth aspect of the embodiments of the present application, a computer program product is provided, which includes computer programs / instructions. When the computer programs / instructions are executed by a processor, the calculation steps of the prediction model are implemented.
[0017] In the sixth aspect of the embodiments of the present application, the application of the immune effector cell-mediated neurotoxicity syndrome risk prediction system, the computer readable storage medium, or the computer program product in preparing a product for predicting the risk of the immune effector cell-mediated neurotoxicity syndrome is provided.
[0018] In the seventh aspect of the embodiments of the present application, an immune effector cell-mediated neurotoxicity syndrome risk prediction product is provided, which includes: a determination product of the key model indicators of the immune effector cell-mediated neurotoxicity syndrome risk prediction system; and the prediction system, the computer readable storage medium, or the computer program product.
[0019] The one or more technical solutions in the embodiments of the present application have at least the following technical effects or advantages: The immune effector cell-mediated neurotoxicity syndrome risk prediction system has the following advantages: 1、The application constructs an ICANS prediction model based on tissue cell tracing technology of free DNA and machine learning, which can realize early prediction of ICANS, the sensitivity is 80%, among 35 patients in the verification set, 2 of 10 ICANS patients are predicted not to occur ICANS, the specificity is 76%, 6 of 25 patients without ICANS are predicted to occur ICANS, which solves the problem of untimely diagnosis and inaccurate intervention in clinic.
[0020] 2、The application has the advantage of small invasion by collecting peripheral blood samples of patients instead of cerebrospinal fluid or tissue samples, and the blood sample collection is less harmful to patients and has higher acceptance of patients.
[0021] 3、The application analyzes plasma DNA through high-resolution methylation spectrum, and can trace to 36 different tissue types, which is a new application in technology, and the wide tissue tracing capability can more accurately identify the contribution of each tissue to plasma, and provide more comprehensive and accurate biological basis for early diagnosis of diseases. BRIEF DESCRIPTION OF DRAWINGS
[0022] In order to more clearly illustrate the technical solutions in the embodiments of the application, the drawings needed in the embodiment description will be briefly introduced as follows.
[0023] Figure 1 The flow chart of the method for constructing the immune effector cell-mediated neurotoxicity syndrome risk prediction model of the application; Figure 2 The importance ranking diagram of the related factors screened out by the elastic network model, the absolute value of the regression coefficient (|coef|) under the elastic network model is used for importance ranking; Figure 2 The corresponding Chinese name of each English is: Blood B is B cell, Megakaryocytes is megakaryocyte, Lung Ep.Bron is alveolar epithelial cell, Smooth.Musc is smooth muscle cell, Pancreas.Alpha is pancreatic alpha cell, Ovary.Ep is ovarian epithelial cell, Liver.Hep is hepatocyte, Eryth.prog is erythrocyte precursor cell, Epid.kerat is epithelial keratinocyte, Colon.Ep is colon epithelial cell, Breast.Luminal.Ep is breast luminal epithelial cell, Breast.Basal.Ep is breast basal epithelial cell, Blood.T is T cell, Blood.NK is killer cell, Blood.Mono.Macro is mononuclear macrophage, Bladder.Ep is bladder epithelial cell, and Adipocytes is adipocyte.
[0024] Figure 3 The two indexes screened out based on the elastic network machine learning method are used to establish an early prediction model performance chart using logistic regression.
[0025] Figure 4 The validation chart of the model in the validation queue is shown in the figure. The X-axis (1 - Specificity) represents the specificity, i.e., the false positive rate, 0 represents that the false positive rate is 0, the model has no any misjudgment, and 1 represents that the false positive rate is 100%, all negative samples are misjudged as positive; the Y-axis (Specificity) represents the sensitivity, i.e., the true positive rate, 0 represents that the model does not detect any positive sample, and 1 represents that the model detects all positive samples. DETAILED DESCRIPTION
[0026] The advantages and various effects of the embodiments of the present application will be more clearly presented hereinafter in combination with specific embodiments and examples. Those skilled in the art should understand that these specific embodiments and examples are used to illustrate the embodiments of the present application, rather than limit the embodiments of the present application.
[0027] Throughout the specification, unless otherwise specifically indicated, the terms used herein are understood as having the meanings commonly used in the art. Therefore, unless otherwise defined, all technical and scientific terms used herein have the same meanings as generally understood by those skilled in the art to which the embodiments of the present application belong. If there is a conflict, the present specification takes precedence.
[0028] The related terms of the present application are explained as follows: 1. Traceable concentration: refers to the absolute concentration of cfDNA released by a specific cell type in the plasma, which is quantified by free DNA (cfDNA) methylation organization tracing technology, and the unit is ng / mL of plasma.
[0029] The determination process of the traceable concentration is as follows: (1) Tissue tracing ratio: based on cell-specific methylation map (such as Loyfer or Moss database), the relative percentage contribution of cfDNA of each cell type is calculated by deconvolution algorithm.
[0030] Formula: the percentage of a certain cell (%) = the number of methylation characteristic fragments of the cell / the number of total cfDNA methylation characteristic fragments × 100%; (2) Absolute concentration conversion: convert the relative percentage into absolute concentration: Traceable concentration = total plasma cfDNA concentration (ng / mL) × cfDNA percentage of the cell type (%); Example: if the total cfDNA = 50 ng / mL, the oligodendrocyte percentage is 5% → traceable concentration = 2.5 ng / mL.
[0031] 2. cfDNA (cell-free DNA): refers to short fragment DNA in circulation, which is apoptotic, necrotic or actively released, and exists in body fluids such as plasma, cerebrospinal fluid, etc. (traceable to tissue origin (such as oligodendrocytes) through methylation characteristics, and the concentration unit is ng / mL.
[0032] 3. Methylation Deconvolution: based on tissue-specific methylation profiles (such as the Loyfer database), the contribution ratio of each cell type in plasma cfDNA is analyzed by algorithm. Key steps: detect CpG site methylation rate; match reference methylation profile; calculate tissue trace percentage; formula: cell proportion = number of methylation characteristic fragments of this cell / total cfDNA methylation characteristic fragment number x 100%.
[0033] The overall idea of the technical solution of the application is as follows: The application extracts plasma free cfDNA and performs methylation detection, and then performs tissue cell tracing of cfDNA, which can effectively reflect or predict the damaged organs in CAR-T cell treatment. Based on these data, relevant factors are screened using machine learning methods, and a prediction model is constructed using logistic regression, which can provide early prediction information of ICANS and assist clinical diagnosis and treatment decisions.
[0034] In summary, the immune effector cell-mediated neurotoxicity syndrome risk prediction system constructed in the application is convenient and fast, which facilitates the prediction of immune effector cell-mediated neurotoxicity syndrome and can more accurately and earlier screen.
[0035] Unless otherwise specified, the various raw materials, reagents, instruments and equipment used in the embodiments of the application can be purchased from the market or can be prepared by existing methods.
[0036] The immune effector cell-mediated neurotoxicity syndrome risk prediction system and its application of the application will be described in detail below in combination with examples, comparative examples and experimental data.
[0037] Example 1: Immune effector cell-mediated neurotoxicity syndrome risk prediction system based on free DNA cell tracing technology and its construction method I. Method 1. Starting from the day of CAR-T infusion, the patient's peripheral blood is collected regularly and the plasma sample is separated, and stored at -80℃ or liquid nitrogen.
[0038] The plasma cfDNA was extracted using the magnetic bead method free DNA extraction kit (Magen), the concentration of the extracted cfDNA was detected by fluorescence quantification using the Qubit dsDNA HS Assay Kit, and the total concentration of the plasma cfDNA (unit: ng / mL of plasma) data was obtained by volume conversion.
[0039] 2. The plasma cfDNA was traced to the tissue cells using the tissue tracing method based on the methylation deconvolution method.
[0040] The plasma cfDNA was subjected to high-depth whole genome methylation sequencing to obtain the methylation rate detection value of the tissue-specific methylation site. Comparative analysis was performed with the specific methylation site reference atlas of human organ tissues, and the tissue tracing of cfDNA was performed using the deconvolution method. Specifically, the human cell methylation atlas and tracing analysis method established by Loyfer et al. (see the literature: Loyfer N, Magenheim J, Peretz A, et al. A DNA methylation atlas of normal human cell types [J]. Nature, 2023, 613(7943): 355-364.) was used, and the calculation tool wgbstools (https: / / github.com / nloyfer / wgbs_tools) was used to quantify the percentage of cfDNA from different tissue cell types in the plasma, or the method of 25 tissue-specific methylation atlas constructed by Moss et al. (see the literature: Moss J, Magenheim J, Neiman D, et al. Comprehensive human cell-type methylation atlas reveals origins of circulating cell-free DNA in health and disease [J]. Nature communications, 2018, 9(1): 5068.) was used to quantify the tracing proportion of different tissue cells in the plasma. The total concentration of cfDNA (unit: ng / mL) was multiplied by the percentage of each cell tracing to obtain the cfDNA concentration (unit: ng / mL) of each cell source for subsequent analysis.
[0041] 3. The obtained tissue tracing data was processed, the elastic network model and the logistic regression model were trained, the AUC, the Youden index, the sensitivity, the specificity, the cutoff value and other indicators were calculated, and the clinical prediction model was constructed.
[0042] The cfDNA histiocytic tracing concentration data of the first eight days after CAR-T reinfusion and before the occurrence of ICANS were selected, and the maximum value of the histiocytic tracing concentration of the patient sample at different time points was screened out as the initial data for model construction using the MAXIFS function. The collected patient sample data was divided into two batches according to the sampling batch, one batch was the training set, and the other batch was the validation set. The first batch of data was imported into the elastic network model for training, and the importance of the above 36 factors (cfDNA cell tracing concentration) was sorted by using the machine learning method, and the elastic network factor importance sorting result was obtained. The top two important factors were selected, and the factors of oligodendrocyte and neuron cfDNA tracing concentration were added, and the logistic regression model was trained, the sensitivity and specificity were calculated, and the prediction performance was evaluated by drawing the ROC curve. The second batch of validation data was brought into the logistic regression model, and the ROC curve was drawn to evaluate the performance of the prediction model. All models were constructed by RStudio and R4.4.3 of R language.
[0043] II. Construction of immune effector cell-mediated neurotoxicity syndrome risk prediction model 1. The free DNA tracing concentration of oligodendrocytes was used as the independent variable to establish the formula of the risk prediction model of immune effector cell-mediated neurotoxicity syndrome as follows: ln[p(1-p)]= -1.7073+0.8621×A, A is the free DNA tracing concentration of oligodendrocytes; The free DNA tracing concentration of B cells was used as the independent variable to establish the formula of the risk prediction model of immune effector cell-mediated neurotoxicity syndrome as follows: ln[p(1-p)]= -1.8460+0.4161×B, B is the free DNA tracing concentration of B cells; 2. The free DNA tracing concentration of megakaryocytes was used as the independent variable to establish the formula of the risk prediction model of immune effector cell-mediated neurotoxicity syndrome as follows: ln[p(1-p)]= -0.8882-0.3168×C, C is the free DNA tracing concentration of megakaryocytes; 3. The free DNA tracing concentration of B cells and megakaryocytes was used as the independent variable to establish the formula of the risk prediction model of immune effector cell-mediated neurotoxicity syndrome as follows: ln[p(1-p)]=-0.3052+0.6538×B-1.6449×C, B is the free DNA tracing concentration of B cells, and C is the free DNA tracing concentration of megakaryocytes; 4. The free DNA tracing concentration of oligodendrocytes and megakaryocytes was used as the independent variable to establish the formula of the risk prediction model of immune effector cell-mediated neurotoxicity syndrome as follows: ln[p(1-p)]=-0.7623+1.9095×A-0.7190×C, A is the traceable concentration of free DNA in oligodendrocytes, C is the traceable concentration of free DNA in megakaryocytes; 5. Using the free DNA traceability concentrations of oligodendrocytes and B cells as independent variables, the risk prediction model for immune effector cell-mediated neurotoxic syndrome was established using the following formula: ln[p(1-p)]=-1.7384-1.3564×A+0.4329×B, where A is the cell-free DNA traceability concentration of oligodendrocytes, and B is the cell-free DNA traceability concentration of B cells.
[0044] 6. Using the traceable concentrations of cfDNA in oligodendrocytes, B cells, and megakaryocytes as independent variables and the probability of the neurotoxic syndrome mediated by immune effector cells as the dependent variable, a binary logistic multivariate analysis was performed to establish a prediction model: ln[p(1-p)]=-0.2989+3.8477×A+0.7318×B-2.0054×C; In the above formula, A is the cfDNA traceability concentration of oligodendrocytes, B is the cfDNA traceability concentration of B cells, and C is the cfDNA traceability concentration of megakaryocytes.
[0045] Table 3 Binary Logistic Regression Analysis
[0046] 3. Model Validation The cfDNA cell-derived concentrations of oligodendrocytes, B cells, and megakaryocytes were used as predictors, and the receiver operating characteristic (ROC) curve was established to evaluate their clinical value in diagnosing ICANS.
[0047] The results are as follows Figure 4 As shown in Figure 3, the ROC curve of the three combined to predict ICANS (AUC = 0.777, 95% CI: 0.743-0.925). The area under the ROC curve of this model is 0.777, and when the sensitivity is 75%, the specificity is 75%.
[0048] Example 2: Risk prediction system for immune effector cell-mediated neurotoxicity syndrome An embodiment of the present invention provides a risk prediction system for immune effector cell-mediated neurotoxic syndrome, the system comprising: A processor and a memory, the memory being coupled to the processor, the memory storing instructions that, when executed by the processor, perform the following steps: The numerical value of the key model index is input into the prediction model to obtain the probability of occurrence of the immune effector cell-mediated neurotoxicity syndrome.
[0049] Embodiment 3, computer-readable storage medium The embodiment of the present application provides a computer-readable storage medium, which stores a computer program. The computer program is executed by a processor to implement the steps of the method in the embodiment 1 and / or the method in the embodiment 2.
[0050] Of course, the storage medium provided by the embodiment of the present application includes computer executable instructions, and the computer executable instructions are not limited to the method operations described above, but can also perform related operations in the method provided by any embodiment of the present application.
[0051] Through the above description of the embodiments, those skilled in the art can clearly understand that the present application can be realized by software and necessary general hardware, and of course, it can also be realized by hardware, but in many cases, the former is a better embodiment. Based on this understanding, the technical solutions of the present application can be embodied in the form of a software product, which can be stored in a computer readable storage medium, such as a floppy disk, a read-only memory (ROM), a random access memory (RAM), a FLASH, a hard disk or an optical disk, etc., including a plurality of instructions to make an electronic device (which can be a personal computer, a server, or a network device, etc.) execute the method described in each embodiment of the present application.
[0052] It is worth noting that in the above embodiments, each unit and module included is only divided according to the function logic, but is not limited to the above division, as long as the corresponding function can be realized; in addition, the specific name of each functional unit is only for easy distinction, and does not limit the protection scope of the present application.
[0053] Application Example 1, predicting the probability of occurrence of clinical immune effector cell-mediated neurotoxicity syndrome I. Collect the tissue trace cell trace concentration data index after the plasma free DNA methylation detection of the immune effector cell-mediated neurotoxicity syndrome (ICANS) received by the clinic, and input it into the model in the embodiment 1.
[0054] Example: As an example, a 30-year-old patient with a concentration of 0 oligodendrocytes, a concentration of 0 B cells and a concentration of 0.07234 megakaryocytes is taken. The above three indicators are brought into the logistics regression equation: ln[p(1-p)]=-0.2989+3.8477xA+0.7318xB-2.0054xC.
[0055] In an example, the probability P of occurrence of the clinical immune effector cell-mediated neurotoxicity syndrome is calculated to be 0.999247.
[0056] II. Application value evaluation The second batch of data is brought into the formula to calculate, and 8 of the 10 ICANS patients are predicted to have ICANS, with a sensitivity of 80%; 6 of the 25 patients without ICANS are predicted to have ICANS, with a specificity of 76%.
[0057] As can be seen from the above, the tissue cell tracing concentration of free DNA can be quantified, the probability of occurrence of the clinical immune effector cell-mediated neurotoxicity syndrome can be predicted, it is convenient and fast, the immune effector cell-mediated neurotoxicity syndrome can be self-checked, and the immune effector cell-mediated neurotoxicity syndrome patients can be screened out more accurately and earlier.
[0058] Finally, it should be noted that the terms "comprising", "containing" or any other variant thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or inherent to such a process, method, article or device.
[0059] Although the preferred embodiments of the embodiments of the present application have been described, those skilled in the art can make further changes and modifications to the embodiments once they know the basic inventive concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications falling within the scope of the embodiments of the present application.
[0060] Obviously, those skilled in the art can make various modifications and variations to the embodiments of the present application without departing from the spirit and scope of the embodiments of the present application. Thus, if these modifications and variations of the embodiments of the present application fall within the scope of the claims of the embodiments of the present application and their equivalent technologies, the embodiments of the present application also intend to include these modifications and variations.
Claims
1. A risk prediction model for immune effector cell-mediated neurotoxicity syndrome, characterized in that: The prediction model includes: A prediction model was established by binary logistic multivariate analysis, using at least one of the free DNA traceability concentrations of oligodendrocytes, B cells, and megakaryocytes as the predictor independent variable and the probability of occurrence of immune effector cell-mediated neurotoxic syndrome as the dependent variable.
2. The risk prediction model for immune effector cell-mediated neurotoxicity syndrome according to claim 1, characterized in that: The risk prediction model for immune effector cell-mediated neurotoxic syndrome was established using the free DNA traceability concentrations of oligodendrocytes, B cells, and megakaryocytes as independent variables. The formula is as follows: ln[p(1-p)]=-0.2989+α×A+β×B+γ×C; In the above formula, A is the cell-free DNA tracing concentration of oligodendrocytes, B is the cell-free DNA tracing concentration of B cells, and C is the cell-free DNA tracing concentration of megakaryocytes; The coefficients α, β, and γ are obtained by performing a logistic multivariate regression analysis on the free DNA traceability concentrations of oligodendrocytes, B cells, and megakaryocytes.
3. The risk prediction model for immune effector cell-mediated neurotoxicity syndrome according to claim 2, characterized in that: The model also includes: Using the free DNA traceability concentration of oligodendrocytes as the independent variable, the risk prediction model for immune effector cell-mediated neurotoxic syndrome was established using the following formula: ln[p(1-p)]= -1.7073+0.8621×A, A is the traceable concentration of free DNA in oligodendrocytes; The formula for establishing a risk prediction model for immune effector cell-mediated neurotoxic syndrome using the free DNA traceability concentration of B cells as the independent variable is as follows: ln[p(1-p)]= -1.8460+0.4161×B, B is the traceable concentration of free DNA in B cells; The formula for establishing a risk prediction model for immune effector cell-mediated neurotoxic syndrome using the free DNA traceability concentration of megakaryocytes as the independent variable is as follows: ln[p(1-p)]= -0.8882-0.3168×C, C is the free DNA traceability concentration of megakaryocytes; The formula for establishing a risk prediction model for immune effector cell-mediated neurotoxic syndrome using the cell-free DNA traceability concentrations of B cells and megakaryocytes as independent variables is as follows: ln[p(1-p)]=-0.3052+0.6538×B-1.6449×C, B is the cell-free DNA traceability concentration of B cells, and C is the cell-free DNA traceability concentration of megakaryocytes; The risk prediction model for immune effector cell-mediated neurotoxic syndrome was established using the free DNA traceability concentrations of oligodendrocytes and megakaryocytes as independent variables. The formula is as follows: ln[p(1-p)]=-0.7623+1.9095×A-0.7190×C, A is the traceable concentration of free DNA in oligodendrocytes, and C is the traceable concentration of free DNA in megakaryocytes; Using the free DNA traceability concentrations of oligodendrocytes and B cells as independent variables, the risk prediction model for immune effector cell-mediated neurotoxic syndrome was established using the following formula: ln[p(1-p)]=-1.7384-1.3564×A+0.4329×B, where A is the cell-free DNA traceability concentration of oligodendrocytes, and B is the cell-free DNA traceability concentration of B cells.
4. A method for constructing a risk prediction model for immune effector cell-mediated neurotoxicity syndrome according to claim 1, characterized in that: The method comprises: Starting from the day of CAR-T transfusion, peripheral blood samples of patients are collected regularly; Conduct tissue tracing on the cell-free DNA in the collected blood samples to obtain tissue tracing data; The tissue traceability data were ranked by importance to obtain an elastic network factor importance ranking diagram, and the top two factors were selected. Given that the occurrence of ICANS is closely related to blood-brain barrier damage, glial cell activation and dysfunction, and neuronal damage, neuronal cell-related factors, namely oligodendrocytes and neurons, were added to train a logistic regression model and construct a clinical prediction model. The model with the highest AUC in the training set was selected as the prediction model and brought into the validation set for verification.
5. A risk prediction system for immune effector cell-mediated neurotoxicity syndrome, characterized in that: The system comprises: A processor and a memory, the memory being coupled to the processor, the memory storing instructions that, when executed by the processor, perform the following steps: The numerical values of the traceable concentrations of free DNA of oligodendrocytes, B cells and megakaryocytes are input into the prediction model as described in claim 1 to obtain the probability of occurrence of immune effector cell-mediated neurotoxic syndrome.
6. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the following steps are performed: The numerical values of the traceable concentrations of free DNA of oligodendrocytes, B cells and megakaryocytes are input into the prediction model as described in claim 1 to obtain the probability of occurrence of immune effector cell-mediated neurotoxic syndrome.
7. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instructions are executed by a processor, the calculation steps of the prediction model according to claim 1 are realized.
8. Use of the risk prediction system for immune effector cell-mediated neurotoxicity syndrome according to claim 1, the prediction system according to claim 5, the computer-readable storage medium according to claim 6, or the computer program product according to claim 7 in preparing a risk prediction product for immune effector cell-mediated neurotoxicity syndrome.
9. A product for predicting the risk of neurotoxic syndrome mediated by immune effector cells, characterized in that: include: A product for measuring the traceability concentration of free DNA in oligodendrocytes, B cells, and megakaryocytes, which are key model indicators of the risk prediction system for immune effector cell-mediated neurotoxic syndrome as described in claim 1; and the prediction system of claim 5 or the computer-readable storage medium of claim 6 or the computer program product of claim 7.
10. Use of at least one of the traceable concentrations of free DNA of oligodendrocytes, B cells, and megakaryocytes as a predictive factor in constructing a risk prediction model for immune effector cell-mediated neurotoxicity syndrome or in preparing a product predicting the risk of immune effector cell-mediated neurotoxicity syndrome.