Neutrophil polarization characteristic-based tumor immunotherapy curative effect prediction system

Abstract

The invention provides a tumor immunotherapy curative effect prediction system based on polarization characteristics of neutrophil, and relates to the technical field of biology, the system comprises a neutrophil detection module used for determining neutrophil information; the patient information module is used for acquiring tumor information and immune information of a patient; the immunotherapy scheme information module is used for acquiring immunotherapy scheme information of the patient; the historical data acquisition module is used for acquiring historical patient information; the immune state coefficient module is used for determining the immune state coefficient of the patient; the prediction model module is used for obtaining the predicted tumor size through the trained tumor size prediction model; the immune state prediction module is used for determining a predicted immune state coefficient; and the curative effect prediction module is used for determining a predicted tumor immunotherapy curative effect score. According to the invention, the predicted tumor immunotherapy curative effect can be evaluated according to two aspects of predicting the tumor size and predicting the immune state coefficient.

Description

Technical Field

[0001] The present invention relates to the field of biotechnology, and in particular to a tumor immunotherapy efficacy prediction system based on neutrophil polarization characteristics. Background Art

[0002] In the related technology, CN118033136A discloses the use of CD74highCD63low neutrophils as molecular markers in the preparation of products for predicting the therapeutic efficacy of non-small cell lung cancer. CD74highCD63low neutrophils have the potential to present antigens and activate T cells, and their proportion is significantly increased in the peripheral blood of patients with non-small cell lung cancer who have sustained responses after ICB-based neoadjuvant therapy, and is positively correlated with the good prognosis of patients with non-small cell lung cancer. It can be used as a biomarker or therapeutic target for predicting and treating non-small cell lung cancer.

[0003] CN115078727A discloses a marker, a kit and an application for predicting the effect of cancer immunotherapy. A marker for predicting the effect of cancer immunotherapy, wherein the marker is CypB. The CypB is combined with any one or more of PD-L1, tumor mutation load TMB, EGFR mutation, tumor infiltrating lymphocyte TIL, DDR gene (DNA damage response gene), circulating tumor cell CTC, circulating tumor DNA (ctDNA), and neutrophil / lymphocyte ratio NLR as a marker for predicting the effect of cancer immunotherapy. The advantages of this scheme are: (1) high sensitivity and specificity. (2) high accuracy of efficacy prediction. (3) simple operation and short time consumption. (4) strong applicability and easy to promote.

[0004] Based on the above-mentioned related technologies, the accuracy of predicting the effect of cancer immunotherapy can be improved. However, the related technologies do not take into account the impact of the patient's immune system status and immunotherapy regimen on the prediction of the efficacy of tumor immunotherapy, that is, it is impossible to make a comprehensive prediction of the efficacy of tumor immunotherapy based on the polarization characteristics of neutrophils, tumor information, the patient's immune system status and immunotherapy regimen.

[0005] The information disclosed in the background technology section of this application is only intended to deepen the understanding of the general background technology of this application, and should not be regarded as an admission or any form of suggestion that the information constitutes the prior art already known to those skilled in the art. Summary of the invention

[0006] The present invention provides a tumor immunotherapy efficacy prediction system based on neutrophil polarization characteristics, which can solve the technical problem that related technologies cannot comprehensively predict the efficacy of tumor immunotherapy based on neutrophil polarization characteristics, tumor information, the patient's immune system status and immunotherapy regimen.

[0007] According to a first aspect of the present invention, a tumor immunotherapy efficacy prediction system based on neutrophil polarization characteristics is provided, comprising:

[0008] A neutrophil detection module is used to detect the polarization characteristic type of neutrophils in the tumor microenvironment and determine the neutrophil information, wherein the neutrophil information includes: the number of anti-tumor neutrophils and pro-tumor neutrophils;

[0009] A patient information module, used to obtain the patient's tumor information and immune information, wherein the tumor information includes: tumor differentiation degree and initial tumor size, and the immune information includes: immunoglobulin level, lymphocyte count, white blood cell count and cytokine level;

[0010] An immunotherapy regimen information module is used to obtain the patient's immunotherapy regimen information, wherein the immunotherapy regimen information includes: the dosage and frequency of drug use;

[0011] A historical data acquisition module is used to obtain historical patient information;

[0012] An immune status coefficient module, used to determine the patient's immune status coefficient based on the immune information;

[0013] A prediction model module, used to process the neutrophil information, the tumor information, the immunotherapy regimen information and the immune status coefficient through a trained tumor size prediction model to obtain a predicted tumor size;

[0014] An immune status prediction module, used to determine a predicted immune status coefficient based on the historical patient information;

[0015] The efficacy prediction module is used to determine the predicted tumor immunotherapy efficacy score based on the predicted tumor size and the predicted immune status coefficient.

[0016] Technical effect: According to the present invention, the polarization characteristics of neutrophils and the state of the patient's immune system can be accurately analyzed, and based on the polarization characteristics of neutrophils, the state of the patient's immune system, tumor information and immunotherapy information, the tumor size and the state of the patient's immune system after completing immunotherapy can be predicted. Further, the efficacy of tumor immunotherapy can be evaluated based on the predicted tumor size and the predicted immune status coefficient. When determining the patient's immune status coefficient, the patient's immune status coefficient can be determined based on the immune information, the first immunoglobulin level threshold, the second immunoglobulin level threshold, the first lymphocyte count threshold, the second lymphocyte count threshold, the first white blood cell count threshold, the second white blood cell count threshold, the first cytokine level threshold and the second cytokine level threshold. During the calculation process, the patient's immune system status can be evaluated based on the patient's immunoglobulin level status, lymphocyte count status, white blood cell count status and cytokine level status, thereby improving the comprehensiveness and accuracy of the immune status coefficient. When determining the training loss function of the tumor size prediction model, the initial tumor size, the ratio of the number of anti-tumor neutrophils to the number of pro-tumor neutrophils, the degree of tumor differentiation, the immune status coefficient, the drug dosage, and the frequency of drug use on the tumor size after immunotherapy are used to determine the influence of the above data on the error of predicting tumor size, so as to set the training loss function based on the influence and the relative error of predicting tumor size, so that the training loss function of the tumor size prediction model is reduced during the training process, and the accuracy of the tumor size prediction model is improved more targeted. When determining the first relationship function, the first relationship function can be determined based on the historical sample immune status coefficient, the historical neutrophil information, the historical tumor information, the historical immunotherapy information and the historical immune status coefficient, which accurately describes the relationship between the immune system state after immunotherapy and the immune system state before immunotherapy, the polarization characteristics of neutrophils, the degree of tumor differentiation, the dosage and the frequency of medication, and improves the accuracy and objectivity of the first relationship function.

[0017] It should be understood that the above general description and the following detailed description are exemplary and explanatory only and do not limit the present invention. Other features and aspects of the present invention will become more apparent from the following detailed description of exemplary embodiments with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other embodiments can be obtained based on these drawings without creative work.

[0019] Figure 1 A block diagram of a tumor immunotherapy efficacy prediction system based on neutrophil polarization characteristics according to an embodiment of the present invention is exemplarily shown. DETAILED DESCRIPTION

[0020] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0021] The technical solution of the present invention is described in detail with specific embodiments below. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described in detail in some embodiments.

[0022] Figure 1 A block diagram of a tumor immunotherapy efficacy prediction system based on neutrophil polarization characteristics according to an embodiment of the present invention is exemplarily shown, and the system includes:

[0023] A neutrophil detection module is used to detect the polarization characteristic type of neutrophils in the tumor microenvironment and determine the neutrophil information, wherein the neutrophil information includes: the number of anti-tumor neutrophils and pro-tumor neutrophils;

[0024] A patient information module, used to obtain the patient's tumor information and immune information, wherein the tumor information includes: tumor differentiation degree and initial tumor size, and the immune information includes: immunoglobulin level, lymphocyte count, white blood cell count and cytokine level;

[0025] An immunotherapy regimen information module is used to obtain the patient's immunotherapy regimen information, wherein the immunotherapy regimen information includes: the dosage and frequency of drug use;

[0026] A historical data acquisition module is used to obtain historical patient information;

[0027] An immune status coefficient module, used to determine the patient's immune status coefficient based on the immune information;

[0028] A prediction model module, used to process the neutrophil information, the tumor information, the immunotherapy regimen information and the immune status coefficient through a trained tumor size prediction model to obtain a predicted tumor size;

[0029] An immune status prediction module, used to determine a predicted immune status coefficient based on the historical patient information;

[0030] The efficacy prediction module is used to determine the predicted tumor immunotherapy efficacy score based on the predicted tumor size and the predicted immune status coefficient.

[0031] According to the embodiment of the present invention, the tumor immunotherapy efficacy prediction system based on neutrophil polarization characteristics can accurately analyze the neutrophil polarization characteristics and the patient's immune system status, and based on the neutrophil polarization characteristics, the patient's immune system status, tumor information and immunotherapy regimen information, predict the tumor size and the patient's immune system status after completing immunotherapy; further, the predicted tumor immunotherapy efficacy is evaluated based on the predicted tumor size and the predicted immune status coefficient.

[0032] According to one embodiment of the present invention, in a neutrophil detection module, in a tumor microenvironment, the polarization characteristic type of neutrophils is detected to determine neutrophil information, wherein the neutrophil information includes: the number of anti-tumor neutrophils and pro-tumor neutrophils.

[0033] For example, through flow cytometry and immunohistochemistry, the polarization state of neutrophils in the patient's body is detected, their molecular markers are obtained, and the neutrophils are classified into subtypes, such as N1 (anti-tumor type) and N2 (pro-tumor type), and their numbers and proportions are recorded respectively.

[0034] According to one embodiment of the present invention, in the patient information module, the patient's tumor information and immune information are obtained, wherein the tumor information includes: tumor differentiation degree and initial tumor size, and the immune information includes: immunoglobulin level, lymphocyte count, white blood cell count and cytokine level.

[0035] For example, before the start of immunotherapy, the patient's tumor and immune information is obtained, that is, the degree of tumor differentiation is detected by immunohistochemical staining. The degree of tumor differentiation refers to the degree of similarity between the tumor tissue and its source tissue in morphology and function; the location of the tumor and the initial tumor size (the size of the tumor before the patient receives immunotherapy) are detected by imaging examination; the lymphocyte count and white blood cell count are obtained by routine blood analysis; the patient's cytokine level is detected by flow cytometry; and the patient's immunoglobulin level is detected by fluorescent immunoassay.

[0036] According to an embodiment of the present invention, in the immunotherapy regimen information module, the immunotherapy regimen information of the patient is obtained, wherein the immunotherapy regimen information includes: the dosage of the drug used and the frequency of the drug used.

[0037] According to one embodiment of the present invention, in the historical data acquisition module, historical patient information is acquired.

[0038] For example, historical tumor information, historical immune information, and historical immunotherapy information of other patients tested on historical dates are recorded and stored, and the above data can be obtained in the historical data acquisition module.

[0039] According to an embodiment of the present invention, in the immune status coefficient module, the immune status coefficient of the patient is determined according to the immune information.

[0040] According to an embodiment of the present invention, determining the patient's immune status coefficient according to the immune information includes:

[0041] Obtain normal ranges for immunoglobulin levels, normal ranges for lymphocyte counts, normal ranges for white blood cell counts, and normal ranges for cytokine levels;

[0042] determining a first immunoglobulin level threshold and a second immunoglobulin level threshold according to the normal immunoglobulin level range;

[0043] Determining a first lymphocyte count threshold and a second lymphocyte count threshold according to the normal lymphocyte count range;

[0044] Determining a first white blood cell count threshold and a second white blood cell count threshold according to the normal white blood cell count range;

[0045] Determining a first cytokine level threshold and a second cytokine level threshold according to the normal cytokine level range;

[0046] The patient's immune status coefficient is determined based on the immune information, the first immunoglobulin level threshold, the second immunoglobulin level threshold, the first lymphocyte count threshold, the second lymphocyte count threshold, the first white blood cell count threshold, the second white blood cell count threshold, the first cytokine level threshold and the second cytokine level threshold.

[0047] For example, immunoglobulins, lymphocytes, leukocytes and cytokines are important components of the immune system. The ranges of immunoglobulins, lymphocytes, leukocytes and cytokines of normal adults are obtained, and the first immunoglobulin level threshold, the second immunoglobulin level threshold, the first lymphocyte count threshold, the second lymphocyte count threshold, the first leukocyte count threshold, the second leukocyte count threshold, the first cytokine level threshold and the second cytokine level threshold are determined according to the lower and upper limits of the ranges of immunoglobulins, lymphocytes, leukocytes and cytokines of normal adults, respectively; based on the immune information, the first immunoglobulin level threshold, the second immunoglobulin level threshold, the first lymphocyte count threshold, the second lymphocyte count threshold, the first leukocyte count threshold, the second leukocyte count threshold, the first cytokine level threshold and the second cytokine level threshold, the status of the patient's immune system is evaluated to determine the patient's immune status coefficient.

[0048] According to one embodiment of the present invention, determining the patient's immune status coefficient according to the immune information, the first immunoglobulin level threshold, the second immunoglobulin level threshold, the first lymphocyte count threshold, the second lymphocyte count threshold, the first white blood cell count threshold, the second white blood cell count threshold, the first cytokine level threshold and the second cytokine level threshold comprises: determining the patient's immune status coefficient Isc according to formula (1),

[0049]

[0050] Among them, α1, α2, α3 and α4 are preset weights, if is the conditional function, il is the patient's immunoglobulin level, il T1 is the first immunoglobulin level threshold, il T2 is the second immunoglobulin level threshold, lc is the patient's lymphocyte count, lc T1 is the first lymphocyte count threshold, lc T2 is the second lymphocyte count threshold, wc is the patient's white blood cell count, and wc T1 is the first white blood cell count threshold, wc T2 is the second white blood cell count threshold, cl is the patient's cytokine level, cl T1 is the first cytokine level threshold, cl T2 is the second cytokine level threshold.

[0051] According to one embodiment of the present invention, in formula (1), the conditional function The value of includes the following two cases, when il∈[il T1 ,il T2], it means that the patient's immunoglobulin level is within the normal range, and the value of the conditional function is 1. T1 ,il T2 ], it means that the patient's immunoglobulin level is not within the normal range, and the value of the conditional function is It is the relative difference between the patient's immunoglobulin level and the average of the first immunoglobulin level threshold and the second immunoglobulin level threshold. The larger the ratio is, the greater the difference between the patient's immunoglobulin level and the normal immunoglobulin level range, and the worse the patient's immune system status is.

[0052] According to one embodiment of the present invention, in formula (1), the conditional function The value of includes the following two cases, when lc∈[lc T1 , lc T2 ], it means that the patient's lymphocyte count is within the normal range, and the value of the conditional function is 1. When lc∈[lc T1 , lc T2 ], it means that the patient's lymphocyte count is not within the normal range, and the value of the conditional function is It is the relative difference between the patient's lymphocyte count and the average of the first lymphocyte count threshold and the lymphocyte count threshold. The larger the ratio is, the greater the difference between the patient's lymphocyte count and the normal lymphocyte count range is, and the worse the patient's immune system status is.

[0053] According to one embodiment of the present invention, in formula (1), the conditional function The value of includes the following two cases, when wc∈[wc T1 , wc T2 ], it means that the patient’s white blood cell count is within the normal range, and the value of the conditional function is 1. When wc∈[wc T1 , wc T2 ], it means that the patient's white blood cell count is not within the normal range, and the value of the conditional function is It is the relative difference between the patient's white blood cell count and the average of the first white blood cell count threshold and the second white blood cell count threshold. The larger the ratio is, the greater the difference between the patient's white blood cell count threshold and the normal white blood cell count threshold range is, and the worse the patient's immune system status is.

[0054] According to one embodiment of the present invention, in formula (1), the conditional function The value of includes the following two cases, when cl∈[cl T1 ,cl T2], it means that the patient's cytokine level is within the normal range, and the value of the conditional function is 1. When cl∈[cl T1 ,cl T2 ], it means that the patient's cytokine level is not within the normal range, and the value of the conditional function is It is the relative difference between the patient's cytokine level and the average of the first cytokine level threshold and the second cytokine level threshold. The larger the ratio is, the greater the difference between the patient's cytokine level and the normal cytokine level range is, and the worse the patient's immune system status is.

[0055] According to one embodiment of the present invention,

[0056]

[0057] It means that the patient's immune system status is evaluated and the immune status coefficient is determined based on the weighted sum of the patient's immunoglobulin level, lymphocyte count, white blood cell count and cytokine level.

[0058] In this way, the patient's immune status coefficient can be determined based on the immune information, the first immunoglobulin level threshold, the second immunoglobulin level threshold, the first lymphocyte count threshold, the second lymphocyte count threshold, the first white blood cell count threshold, the second white blood cell count threshold, the first cytokine level threshold and the second cytokine level threshold. During the calculation process, the patient's immune system status can be evaluated based on the patient's immunoglobulin level status, lymphocyte count status, white blood cell count status and cytokine level status, thereby improving the comprehensiveness and accuracy of the immune status coefficient.

[0059] According to one embodiment of the present invention, in the prediction model module, the neutrophil information, the tumor information, the immunotherapy regimen information and the immune status coefficient are processed by a trained tumor size prediction model to obtain a predicted tumor size.

[0060] For example, the tumor size prediction model is a deep learning neural network model. By inputting neutrophil information, tumor information, immunotherapy regimen information and immune status coefficient into the trained tumor size prediction model, the predicted tumor size can be obtained.

[0061] According to one embodiment of the present invention, the training step of the tumor size prediction model includes:

[0062] Determining historical neutrophil information, historical tumor information, historical immune information, and historical immunotherapy information of the historical patient based on the historical patient information;

[0063] Determining a historical immune status coefficient of a historical patient based on the historical immune information;

[0064] The historical neutrophil information, the historical tumor information, the historical immunotherapy regimen information and the historical immune status coefficient are processed by a tumor size prediction model to obtain a historical predicted tumor size;

[0065] Obtain historical sample tumor size of historical patients after completion of immunotherapy;

[0066] Determining a training loss function of a tumor size prediction model according to the historical neutrophil information, the historical tumor information, the historical immunotherapy regimen information, the historical immune status coefficient, the historical predicted tumor size, and the historical sample tumor size;

[0067] The tumor size prediction model is trained according to the training loss function to obtain a trained tumor size prediction model.

[0068] For example, based on historical immune information, determine the historical immune status coefficient of historical patients. The calculation method of the historical immune status coefficient is similar to formula (1) and will not be repeated here. The historical immune status coefficient represents the immune system status of historical patients before receiving immunotherapy; process historical neutrophil information, historical tumor information, historical immunotherapy regimen information and historical immune status coefficient through a tumor size prediction model to obtain a historical predicted tumor size; obtain the historical sample tumor size of historical patients after they actually received immunotherapy; determine the training loss function of the tumor size prediction model based on the error between the historical sample tumor size and the historical predicted tumor size, as well as the historical neutrophil information, historical tumor information, historical immunotherapy regimen information and historical immune status coefficient; train the tumor size prediction model based on the training loss function of the tumor size prediction model, improve the accuracy of the tumor size prediction model in predicting tumor size, and obtain a trained tumor size prediction model.

[0069] According to one embodiment of the present invention, the training loss function of the tumor size prediction model is determined based on the historical neutrophil information, the historical tumor information, the historical immunotherapy regimen information, the historical immune status coefficient, the historical predicted tumor size and the historical sample tumor size, including: determining the training loss function LOSS of the tumor size prediction model according to formula (2): T ,

[0070]

[0071] Among them, Sts k is the historical sample tumor size of the kth historical patient, Pts kThe historical predicted tumor size for the kth historical patient, N1 k is the number of anti-tumor neutrophils of the kth historical patient, N2 k is the number of tumor-promoting neutrophils in the kth historical patient, Dotd k is the tumor differentiation degree of the kth historical patient, Ots k is the initial tumor size of the kth historical patient, Isc h，k is the historical immune status coefficient of the kth historical patient, D k is the drug dosage used by the kth historical patient, f k is the frequency of drug use of the kth historical patient, K is the number of historical patients, k≤K, k and K are both positive integers.

[0072] According to one embodiment of the present invention, It is the ratio of the initial tumor size of the kth historical patient to the average initial tumor size of K historical patients. The larger the ratio is, the larger the initial tumor size of the kth historical patient is, the greater the impact on the training process, and the greater the impact on the error of predicting tumor size. is the ratio of the number of anti-tumor neutrophils to the number of pro-tumor neutrophils in the kth historical patient, that is, the N1 cell ratio of the kth historical patient, It is the ratio of the N1 cell ratio of the k-th historical patient to the average N1 cell ratio of K historical patients. The larger the ratio is, the larger the N1 cell ratio of the k-th historical patient is. is the ratio of the tumor differentiation degree of the kth historical patient to the average tumor differentiation degree of K historical patients. The larger the ratio is, the greater the tumor differentiation degree of the kth historical patient is; is the ratio of the historical immune status coefficient of the kth historical patient to the average immune status coefficient of K historical patients. The larger the ratio is, the larger the historical immune status coefficient of the kth historical patient is. It represents the ratio of the drug dosage used by the kth historical patient to the average drug dosage used by K historical patients; It represents the ratio of the drug usage frequency of the kth historical patient to the average drug usage frequency of K historical patients. The larger the ratio is, the higher the drug usage frequency of the kth historical patient is. It indicates that the initial tumor size is positively correlated with the predicted tumor size, and the ratio of the number of anti-tumor neutrophils to the number of pro-tumor neutrophils, the degree of tumor differentiation, the immune status coefficient, the dosage of drugs used, the frequency of drug use and the predicted tumor size are negatively correlated. For example, the larger the initial tumor size, the larger the tumor may mean more tumor load, which may lead to relatively poor effect of immunotherapy, then the larger the predicted tumor size, the higher the degree of differentiation of the tumor is usually more sensitive to immunotherapy, the better the immunotherapy effect, then the smaller the predicted tumor size, the larger the ratio of the number of anti-tumor neutrophils to the number of pro-tumor neutrophils, the more the number of anti-tumor neutrophils, the better the immunotherapy effect, then the smaller the predicted tumor size, the higher the immune status coefficient, the more active the patient's immune system, the better the immunotherapy effect, then the smaller the predicted tumor size, the more drug dosage used, can ensure that the immune system is fully activated, so as to more effectively identify and eliminate tumor cells, the better the immunotherapy effect, then the smaller the predicted tumor size, the more drug use frequency, can maintain the immune system in a high activation state, so as to more effectively identify and eliminate tumor cells, the better the immunotherapy effect, then the smaller the predicted tumor size. Therefore, placing the term related to the initial tumor size in the numerator indicates that the initial tumor size is relatively larger, that is, The larger the value is, the larger the predicted tumor size is, and the greater the impact on the error of predicting the tumor size is. The ratio of the number of anti-tumor neutrophils to the number of pro-tumor neutrophils, the degree of tumor differentiation, the immune status coefficient, the dosage of the drug used, and the frequency of drug use are placed in the denominator to represent The smaller the value of , the larger the predicted tumor size is, and the greater the impact on the error in predicting the tumor size is. is the relative error between the historical sample tumor size and the historical predicted tumor size of the kth historical patient.

[0073] According to one embodiment of the present invention, using

[0074] The relative errors of the predicted tumor size of each historical patient are weighted averaged to obtain the training loss function. During the training process, the above training loss function is reduced, so that the error between the predicted tumor size and the sample tumor size is reduced, and the accuracy of the tumor size prediction model for predicting tumor size is improved, thereby improving the accuracy of the tumor size prediction model.

[0075] In this way, the influence of the initial tumor size, the ratio of the number of anti-tumor neutrophils to the number of pro-tumor neutrophils, the degree of tumor differentiation, the immune status coefficient, the drug dosage used, and the frequency of drug use on the tumor size after immunotherapy can be used to determine the influence of the above data on the error in predicting tumor size. Based on this influence and the relative error in predicting tumor size, the training loss function can be set to reduce the training loss function of the tumor size prediction model during the training process, thereby improving the accuracy of the tumor size prediction model in a more targeted manner.

[0076] According to one embodiment of the present invention, in the immune status prediction module, a predicted immune status coefficient is determined based on the historical patient information.

[0077] According to one embodiment of the present invention, determining a predicted immune status coefficient based on the historical patient information includes:

[0078] Determining the immune status coefficient of the historical samples of the historical patients after completing immunotherapy according to the historical patient information;

[0079] Determining a first relationship function according to the historical sample immune status coefficient, the historical neutrophil information, the historical tumor information, the historical immunotherapy information and the historical immune status coefficient;

[0080] A predicted immune status coefficient is determined according to the first relationship function, the immune status coefficient, the neutrophil information, the tumor information and the immunotherapy information.

[0081] For example, based on historical patient information, the historical immune information of historical patients after completing immunotherapy is determined, and based on the historical immune information, the historical sample immune status coefficient is determined; the neutrophil information, historical tumor information, historical immunotherapy information and the historical immune status coefficient before receiving immunotherapy are all related to the historical sample immune status coefficient after completing immunotherapy to a certain extent. For example, the larger the historical immune status coefficient of the patient before receiving immunotherapy, the more active the patient's immune system, and the larger the historical sample immune status coefficient of the patient after completing immunotherapy. Based on the correlation of the above data, the first relationship function can be determined; the immune status coefficient, neutrophil information, tumor information and immunotherapy information are substituted into the first relationship function to determine the predicted immune status coefficient.

[0082] According to one embodiment of the present invention, determining a first relationship function according to the historical sample immune status coefficient, the historical neutrophil information, the historical tumor information, the historical immunotherapy information and the historical immune status coefficient includes: determining the equation to be fitted of the first relationship function according to formula (3),

[0083]

[0084] Among them, β1, β2, β3, β4, β5, β6 and β7 are the coefficients to be fitted, SIsc h，k is the historical sample immune status coefficient of the kth historical patient, Isc h，k is the historical immune status coefficient of the kth historical patient, N1 k is the number of anti-tumor neutrophils of the kth historical patient, N2 k The number of tumor-promoting neutrophils, Dotd k is the tumor differentiation degree of the kth historical patient, D k is the drug dosage used by the kth historical patient, f k is the frequency of drug use of the kth historical patient;

[0085] Solving the coefficient to be fitted according to the historical sample immune status coefficient, the historical neutrophil information, the historical tumor information, the historical immunotherapy information and the historical immune status coefficient to obtain a solution value of the coefficient to be fitted;

[0086] The first relationship function is determined according to the solved values ​​of the coefficients to be fitted and the equation to be fitted of the first relationship function.

[0087] According to one embodiment of the present invention, SIsc h，k is the historical sample immune status coefficient, which indicates the immune system status of historical patients after receiving immunotherapy, (β1Isc h，k +β2) indicates that the historical sample immune status coefficient of the kth historical patient is positively correlated with the historical immune status coefficient of the kth historical patient. The better the patient's immune system status at the beginning of immunotherapy, the better the patient's immunotherapy effect, and the larger the historical sample immune status coefficient. It is the ratio of the number of anti-tumor neutrophils to the number of pro-tumor neutrophils. The larger the ratio is, the better the immunotherapy effect of the patient is. The larger the historical sample immune status coefficient is, the higher the historical sample immune status coefficient is. There is a positive correlation between this ratio and D k is the drug dosage used by the kth historical patient. The larger the value, the better the immunotherapy effect of the patient. The larger the historical sample immune status coefficient, the greater the historical sample immune status coefficient has a positive correlation with this value. k is the frequency of drug use of the kth historical patient. The larger the value, the better the immunotherapy effect of the patient. The larger the historical sample immune status coefficient, the greater the historical sample immune status coefficient has a positive correlation with this value. kis the tumor differentiation degree of the kth historical patient. The larger the value, the lower the malignancy of the tumor, the slower the growth rate, the better the immunotherapy effect of the patient, and the larger the historical sample immune status coefficient. The historical sample immune status coefficient has a positive correlation with this value. Based on the above correlation, the equation to be fitted for the first relationship function can be obtained.

[0088] According to one embodiment of the present invention, fitting can be performed based on multiple parameters involved in the above equation to be fitted, that is, fitting can be performed based on the historical sample immune status coefficient, the historical neutrophil information, the historical tumor information, the historical immunotherapy information and the historical immune status coefficient to solve the above multiple coefficients to be fitted. There are 7 coefficients to be fitted, namely, β1, β2, β3, β4, β5, β6 and β7. The above 7 coefficients to be fitted are solved to obtain the solution values ​​of the above 7 coefficients to be fitted, and the solution values ​​of the above 7 coefficients to be fitted are substituted into the equation to be fitted to determine the first relationship function.

[0089] In this way, the first relationship function can be determined based on the historical sample immune status coefficient, the historical neutrophil information, the historical tumor information, the historical immunotherapy information and the historical immune status coefficient, which accurately describes the relationship between the immune system status after immunotherapy and the immune system status before immunotherapy, the polarization characteristics of neutrophils, the degree of tumor differentiation, the dosage and frequency of medication, thereby improving the accuracy and objectivity of the first relationship function.

[0090] According to one embodiment of the present invention, in the efficacy prediction module, a predicted tumor immunotherapy efficacy score is determined based on the predicted tumor size and the predicted immune status coefficient.

[0091] For example, the expected tumor size and expected immune status coefficient that are expected to be achieved through immunotherapy are obtained, and the tumor size recognition result is determined based on the predicted tumor size and the expected tumor size. If the predicted tumor size is larger than the expected tumor size, it indicates that the treatment effect is not good, and the tumor size recognition result is 0. If the predicted tumor size is smaller than the expected tumor size, it indicates that the treatment effect is good, and the tumor size recognition result is 1. Based on the predicted immune status coefficient and the expected immune status coefficient, the immune status coefficient recognition result is determined. If the predicted immune status coefficient is smaller than the expected immune status coefficient, it indicates that the treatment effect is not good, and the immune status coefficient recognition result is 0. If the predicted immune status coefficient is greater than the expected immune status coefficient, it indicates that the treatment effect is good, and the immune status coefficient recognition result is 1. Weighted summation is performed based on the tumor size recognition result and the immune status coefficient recognition result to determine the predicted tumor immunotherapy efficacy score.

[0092] According to the tumor immunotherapy efficacy prediction system based on neutrophil polarization characteristics of the embodiment of the present invention, the neutrophil polarization characteristics and the patient's immune system status can be accurately analyzed, and the tumor size and the patient's immune system status after completing the immunotherapy are predicted based on the neutrophil polarization characteristics, the patient's immune system status, tumor information and immunotherapy scheme information. Further, the tumor immunotherapy efficacy is evaluated based on the predicted tumor size and the predicted immune status coefficient. When determining the patient's immune status coefficient, the patient's immune status coefficient can be determined based on the immune information, the first immunoglobulin level threshold, the second immunoglobulin level threshold, the first lymphocyte count threshold, the second lymphocyte count threshold, the first white blood cell count threshold, the second white blood cell count threshold, the first cytokine level threshold and the second cytokine level threshold. In the calculation process, the patient's immune system status can be evaluated based on the patient's immunoglobulin level status, lymphocyte count status, white blood cell count status and cytokine level status, thereby improving the comprehensiveness and accuracy of the immune status coefficient. When determining the training loss function of the tumor size prediction model, the initial tumor size, the ratio of the number of anti-tumor neutrophils to the number of pro-tumor neutrophils, the degree of tumor differentiation, the immune status coefficient, the drug dosage, and the frequency of drug use on the tumor size after immunotherapy are used to determine the influence of the above data on the error of predicting tumor size, so as to set the training loss function based on the influence and the relative error of predicting tumor size, so that the training loss function of the tumor size prediction model is reduced during the training process, and the accuracy of the tumor size prediction model is improved more targeted. When determining the first relationship function, the first relationship function can be determined based on the historical sample immune status coefficient, the historical neutrophil information, the historical tumor information, the historical immunotherapy information and the historical immune status coefficient, which accurately describes the relationship between the immune system state after immunotherapy and the immune system state before immunotherapy, the polarization characteristics of neutrophils, the degree of tumor differentiation, the dosage and the frequency of medication, and improves the accuracy and objectivity of the first relationship function.

[0093] The present invention may be a method, an apparatus, a system and / or a computer program product. The computer program product may include a computer-readable storage medium carrying computer-readable program instructions for executing various aspects of the present invention.

[0094] It should be understood by those skilled in the art that the embodiments of the present invention described above and shown in the accompanying drawings are only examples and do not limit the present invention. The purpose of the present invention has been fully and effectively achieved. The functional and structural principles of the present invention have been demonstrated and explained in the embodiments, and the embodiments of the present invention may be deformed or modified in any way without departing from the principles.

Claims

1. A tumor immunotherapy efficacy prediction system based on neutrophil polarization characteristics, characterized in that: include: A neutrophil detection module is used to detect the polarization characteristic type of neutrophils in the tumor microenvironment and determine the neutrophil information, wherein the neutrophil information includes: the number of anti-tumor neutrophils and pro-tumor neutrophils; A patient information module, used to obtain the patient's tumor information and immune information, wherein the tumor information includes: tumor differentiation degree and initial tumor size, and the immune information includes: immunoglobulin level, lymphocyte count, white blood cell count and cytokine level; An immunotherapy regimen information module is used to obtain the patient's immunotherapy regimen information, wherein the immunotherapy regimen information includes: the dosage and frequency of drug use; A historical data acquisition module is used to obtain historical patient information; An immune status coefficient module, used to determine the patient's immune status coefficient based on the immune information; A prediction model module, used to process the neutrophil information, the tumor information, the immunotherapy regimen information and the immune status coefficient through a trained tumor size prediction model to obtain a predicted tumor size; An immune status prediction module, used to determine a predicted immune status coefficient based on the historical patient information; The efficacy prediction module is used to determine the predicted tumor immunotherapy efficacy score based on the predicted tumor size and the predicted immune status coefficient.

2. The tumor immunotherapy efficacy prediction system based on neutrophil polarization characteristics according to claim 1, characterized in that: Determining the patient's immune status coefficient according to the immune information includes: Obtain normal ranges for immunoglobulin levels, normal ranges for lymphocyte counts, normal ranges for white blood cell counts, and normal ranges for cytokine levels; determining a first immunoglobulin level threshold and a second immunoglobulin level threshold according to the normal immunoglobulin level range; Determining a first lymphocyte count threshold and a second lymphocyte count threshold according to the normal lymphocyte count range; Determining a first white blood cell count threshold and a second white blood cell count threshold according to the normal white blood cell count range; Determining a first cytokine level threshold and a second cytokine level threshold according to the normal cytokine level range; The patient's immune status coefficient is determined based on the immune information, the first immunoglobulin level threshold, the second immunoglobulin level threshold, the first lymphocyte count threshold, the second lymphocyte count threshold, the first white blood cell count threshold, the second white blood cell count threshold, the first cytokine level threshold and the second cytokine level threshold.

3. The tumor immunotherapy efficacy prediction system based on neutrophil polarization characteristics according to claim 2, characterized in that: Determining the patient's immune status coefficient according to the immune information, the first immunoglobulin level threshold, the second immunoglobulin level threshold, the first lymphocyte count threshold, the second lymphocyte count threshold, the first white blood cell count threshold, the second white blood cell count threshold, the first cytokine level threshold, and the second cytokine level threshold, comprising: According to the formula Determine the patient's immune status coefficient Isc, where α1, α2, α3 and α4 are preset weights, if is a conditional function, il is the patient's immunoglobulin level, il T1 is the first immunoglobulin level threshold, il T2 is the second immunoglobulin level threshold, lc is the patient's lymphocyte count, lc T1 is the first lymphocyte count threshold, lc T2 is the second lymphocyte count threshold, wc is the patient's white blood cell count, and wc T1 is the first white blood cell count threshold, wc T2 is the second white blood cell count threshold, cl is the patient's cytokine level, cl T1 is the first cytokine level threshold, cl T2 is the second cytokine level threshold.

4. The tumor immunotherapy efficacy prediction system based on neutrophil polarization characteristics according to claim 1, characterized in that: The training steps of the tumor size prediction model include: Determining historical neutrophil information, historical tumor information, historical immune information, and historical immunotherapy information of the historical patient based on the historical patient information; Determining a historical immune status coefficient of a historical patient based on the historical immune information; The historical neutrophil information, the historical tumor information, the historical immunotherapy regimen information and the historical immune status coefficient are processed by a tumor size prediction model to obtain a historical predicted tumor size; Obtain historical sample tumor size of historical patients after completion of immunotherapy; Determining a training loss function of a tumor size prediction model according to the historical neutrophil information, the historical tumor information, the historical immunotherapy regimen information, the historical immune status coefficient, the historical predicted tumor size, and the historical sample tumor size; The tumor size prediction model is trained according to the training loss function to obtain a trained tumor size prediction model.

5. The tumor immunotherapy efficacy prediction system based on neutrophil polarization characteristics according to claim 4, characterized in that: Determining a training loss function of a tumor size prediction model according to the historical neutrophil information, the historical tumor information, the historical immunotherapy regimen information, the historical immune status coefficient, the historical predicted tumor size, and the historical sample tumor size includes: According to the formula Determine the training loss function LOSS of the tumor size prediction model T , among which, Sts k is the historical sample tumor size of the kth historical patient, Pts k The historical predicted tumor size for the kth historical patient, N1 k is the number of anti-tumor neutrophils of the kth historical patient, N2 k is the number of tumor-promoting neutrophils in the kth historical patient, Dotd k is the tumor differentiation degree of the kth historical patient, OTs k is the initial tumor size of the kth historical patient, Isc h，k is the historical immune status coefficient of the kth historical patient, D k is the drug dosage used by the kth historical patient, f k is the frequency of drug use of the kth historical patient, K is the number of historical patients, k≤K, k and K are both positive integers.

6. The tumor immunotherapy efficacy prediction system based on neutrophil polarization characteristics according to claim 4, characterized in that: Determine the predicted immune status coefficient based on the historical patient information, including: Determining the immune status coefficient of the historical samples of the historical patients after completing immunotherapy according to the historical patient information; Determining a first relationship function according to the historical sample immune status coefficient, the historical neutrophil information, the historical tumor information, the historical immunotherapy information and the historical immune status coefficient; A predicted immune status coefficient is determined according to the first relationship function, the immune status coefficient, the neutrophil information, the tumor information and the immunotherapy information.

7. The tumor immunotherapy efficacy prediction system based on neutrophil polarization characteristics according to claim 6, characterized in that: Determining a first relationship function according to the historical sample immune status coefficient, the historical neutrophil information, the historical tumor information, the historical immunotherapy information and the historical immune status coefficient includes: According to the formula Determine the equation to be fitted of the first relationship function, where β1, β2, β3, β4, β5, β6 and β7 are coefficients to be fitted, SIsc h，k is the historical sample immune status coefficient of the kth historical patient, isc h，k is the historical immune status coefficient of the kth historical patient, N1 k is the number of anti-tumor neutrophils of the kth historical patient, N2 k The number of tumor-promoting neutrophils, Dotd k is the tumor differentiation degree of the kth historical patient, D k is the drug dosage used by the kth historical patient, f k is the frequency of drug use of the kth historical patient; Solving the coefficient to be fitted according to the historical sample immune status coefficient, the historical neutrophil information, the historical tumor information, the historical immunotherapy information and the historical immune status coefficient to obtain a solution value of the coefficient to be fitted; The first relationship function is determined according to the solved values ​​of the coefficients to be fitted and the equation to be fitted of the first relationship function.

Images