Tumor prognosis prediction method and device based on tumor-immune-stromal cells

Abstract

A tumor prognosis prediction method and device based on tumor-immune-stromal cells. The method comprises: constructing an open tumor prognosis simulation model based on tumor-immune-stromal cells, to simulate the prognosis of patients with different actual situations under different treatments from two aspects, i.e., tumor-immune-stromal cell behaviors and interactions, and then to determine individual prognosis results and group prognosis results of the patients.

Description

Tumor prognosis prediction method and device based on tumor-immune-stromal cells

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] The present disclosure claims priority to Chinese patent application CN202311660879.8, filed on December 5, 2023, entitled “Tumor prognosis prediction method and device based on tumor-immune-stromal cells,” the entire contents of which are incorporated by reference into the present disclosure. Technical Field

[0003] The present disclosure relates to the field of theoretical biology technology, and in particular to a method and device for predicting tumor prognosis based on tumor-immune-stromal cells. Background Art

[0004] Research on tumor treatment options and their mechanisms of interaction between tumors, immune cells, and the stromal system is increasing. In particular, advances in sequencing technologies have provided more evidence supporting intercellular interactions. Numerous studies have found that immune and stromal cell subsets are closely associated with tumor development, progression, and prognosis. These data provide realistic support for models of tumor growth dynamics.

[0005] However, current theoretical model studies have relatively simple settings for tumor and immune cell types and are generally limited to specific tumors and treatment plans, resulting in the theoretical models rarely being able to predict patient prognosis.

[0006] Summary of the Invention

[0007] The present disclosure provides a tumor prognosis prediction method and device based on tumor-immune-stromal cells, which are used to determine the prognosis prediction results of tumor patients under different tumor treatments.

[0008] In the first aspect, the present disclosure provides a tumor prognosis prediction method based on tumor-immune-stromal cells, comprising: determining the type and / or state of the cell population affected after treatment, and all cell behaviors of the cell population affected after treatment according to the acquired information on the type of anti-tumor treatment to be analyzed; the cell population type affected after treatment includes: tumor cells, immune cells and stromal cells; determining the post-treatment related cell behavior related to the information on the type of anti-tumor treatment to be analyzed among all the cell behaviors; the cell behavior is the cell-independent behavior and cell-interaction behavior that affects the number of tumor microenvironment cell populations; the cell-independent behavior includes: proliferation, death and movement; the cell-interaction behavior includes: the clearance behavior of the tumor cells by the immune cells, the recruitment behavior of the tumor cells to the immune cells, and the transformation behavior between different cell types or states; using the tumor evolution module, the tumor cells are classified to obtain each tumor cell subpopulation, and the initial number ratio of the tumor cell subpopulation is calculated in combination with the genomic data sequenced by the patient; based on all the cell-independent behaviors and cell-interaction behaviors of the immune cells and the stromal cells, the microenvironment regulation module is used to construct the Tumor microenvironment; based on the type and / or state of the cell population in the tumor microenvironment, all the cell behaviors, and third-party cell regulation behavior, the tumor-immune-stromal cell model framework is formed; the third-party cell regulation is the regulation of the cell behavior by the tumor cells, the immune cells and the stromal cells; using the tumor treatment module, the impact of the treatment on the type and / or state of the cell population and the post-treatment related cell behavior is converted into post-treatment tumor behavior parameters and / or tumor treatment parameters in the tumor-immune-stromal cell model framework; the tumor treatment parameters can be supplemented by selecting real treatment cases from batch sequencing with treatment and prognosis information, and / or single-cell sequencing database to solve multivariate equations; through sequencing technology, the type of cell population to be tested and the proportion of the initial number to be tested are determined from the tumor microenvironment batch sequencing / single-cell sequencing data at the initial stage of the patient's treatment; based on the type of cell population to be tested, the proportion of the initial number to be tested, combined with the post-treatment tumor behavior parameters and / or the tumor treatment parameters, the long-term prognosis module is used to evolve in the tumor-immune-stromal cell model framework to obtain tumor prognosis results for multiple patients.

[0009] In a second aspect, the present disclosure provides a tumor prognosis prediction device based on tumor-immune-stromal cells, comprising: an information determination module for determining the type and / or state of the cell population affected after treatment, and all cell behaviors of the cell population affected after treatment, based on the acquired information on the type of anti-tumor treatment to be analyzed; the cell population types affected after treatment include: tumor cells, immune cells and stromal cells; a post-treatment related cell behavior determination module for determining, among all the cell behaviors, the post-treatment related cell behavior related to the information on the type of anti-tumor treatment to be analyzed; the cell behavior is the cell-independent behavior that affects the number of cell populations in the tumor microenvironment and cell interaction behaviors; the cell-independent behaviors include: proliferation, death and migration; the cell-interaction behaviors include: the clearance behavior of the tumor cells by the immune cells, the recruitment behavior of the tumor cells to the immune cells, and the transformation behavior between different cell types or states; a subpopulation and its proportion determination module is used to classify the tumor cells using the tumor evolution module to obtain each tumor cell subpopulation, and calculate the initial number proportion of the tumor cell subpopulation in combination with the genomic data sequenced by the patient; a microenvironment construction module is used to regulate the microenvironment based on all the cell-independent behaviors and cell-interaction behaviors of the immune cells and the stromal cells A module constructs the tumor microenvironment; a framework forming module is used to form the tumor-immune-stromal cell model framework based on the type and / or state of the cell population in the tumor microenvironment, all the cell behaviors, and the third-party cell regulation behavior; the third-party cell regulation is the regulation of the cell behavior by the tumor cells, the immune cells and the stromal cells; a parameter determination module is used to use the tumor treatment module to convert the effect of the treatment on the type and / or state of the cell population and the post-treatment related cell behavior into the post-treatment tumor behavior parameters and / or tumor treatment parameters in the tumor-immune-stromal cell model framework; the tumor Tumor treatment parameters can be supplemented by batch sequencing with treatment and prognosis information, and / or selecting real treatment cases from the single-cell sequencing database to solve multivariate equations; the sequencing module is used to determine the type of cell population to be tested and the proportion of the initial number to be tested from the batch sequencing / single-cell sequencing data of the tumor microenvironment at the initial stage of patient treatment through sequencing technology; the prognosis result determination module is used to obtain tumor prognosis results for multiple patients based on the type of cell population to be tested, the proportion of the initial number to be tested, combined with the tumor behavior parameters after treatment and / or the tumor treatment parameters, using the long-term prognosis module evolved in the tumor-immune-stromal cell model framework.

[0010] In a third aspect, the present disclosure provides an electronic device comprising a processor and a memory, wherein the memory stores computer-readable instructions. When the computer-readable instructions are executed by the processor, the steps in the method provided in the first aspect are executed.

[0011] In a fourth aspect, the present disclosure provides a storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, runs the steps of the method provided in the first aspect above. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] In order to more clearly illustrate the embodiments of the present disclosure or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present disclosure. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0013] Figure 1 shows one of the classic tumor dynamics models;

[0014] Figure 2 is the second classic tumor dynamics model;

[0015] Figure 3 is the third classic tumor dynamics model;

[0016] Figure 4 is the fourth classic tumor dynamics model;

[0017] Figure 5 is the fifth classic tumor dynamics model;

[0018] FIG6 is a flowchart of an embodiment of a tumor prognosis prediction method based on tumor-immune-stromal cells disclosed herein;

[0019] FIG7 is a flowchart of an embodiment of a tumor prognosis prediction method based on tumor-immune-stromal cells disclosed herein;

[0020] Figure 8 is a schematic diagram of the mechanism of drug resistance and immune escape;

[0021] Figure 9 is a schematic diagram of the tumor-immune-stromal cell model framework;

[0022] FIG10 is a schematic diagram of a curve showing changes in relative tumor volume over time in patients;

[0023] FIG11 is a structural block diagram of an embodiment of a tumor prognosis prediction device based on tumor-immune-stromal cells disclosed herein. DETAILED DESCRIPTION

[0024] The embodiments of the present disclosure provide a tumor prognosis prediction method and device based on tumor-immune-stromal cells, which are used to determine the prognosis prediction results of tumor patients under different tumor treatments.

[0025] In order to make the purpose, features, and advantages of the invention disclosed herein more obvious and easy to understand, the technical solutions 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 embodiments described below are only part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0026] The theoretical basis of tumor growth kinetic models is a system dynamics model that simulates tumor cell growth. Since then, these models have evolved, incorporating diverse modules such as immune cells and drug mechanisms of action. Several classic models are summarized in Figures 1 through 5, showing the classic tumor kinetic models. The model in Figure 1 assumes that changes in tumor size depend on the balance between tumor cell proliferation and cell death. The effect of anticancer drugs is modeled as a reduction in cell growth rate, which is influenced by drug concentration and drug sensitivity. Specifically, this model is based on the classic tumor kinetic model developed by Palmer and Adam (2017) under the supremacy of single-drug hypothesis, adapted from Steel (1967). This model introduces the concept of acquired drug resistance and simulates the progression-free survival (PFS) curves for both monotherapy and combination therapy. In 1994, the tumor-immune cell model proposed by Kuznetsov-Taylor first introduced immune cells into the theoretical model. Subsequently, this model has incorporated modules such as time lag, drug mechanisms of action, and multiple immune cell types, striving to better simulate the interaction between tumors and immune cells. Building on the above model, Bekker et al. (2022) constructed the model shown in Figure 2 based on the effects of drugs on tumors and immune cells. This model constructs a tumor-immune cell growth game model and categorizes tumor treatment options based on this model: Treatment 1 is therapies that simultaneously reduce both cancer cells and immune cells, such as cytotoxic therapy and radiotherapy; Treatment 2 is therapies that only increase the number of immune cells, such as cell therapy; and Treatment 3 is therapies that affect the development or survival of tumor or immune cells, such as targeted therapy and immune checkpoint inhibitors. This model provides a theoretical basis for evaluating drug combinations, but has not been verified or predicted using actual data. Furthermore, Lai et al. (2017) constructed a model of tumor cells and three immune cells (dendritic cells, CD4+ T cells, and CD8+ T cells) as shown in Figure 3. This model, based on the model in Figure 2, takes into account different immune cell types. For example, when immune cell 1 is a T cell and 2 is a circulating lymphocyte, 2 does not interact with the tumor, but 2 has a promoting effect on T cells. 42 ; when 1 is a cytotoxic T cell (CTL) and 2 is a resting T cell, 2 does not interact with the tumor, but 2 is converted to 1. 43 The three immune cells can be divided into dendritic cells, CD4+ T cells, and CD8+ T cells. This model is used to analyze and predict the efficacy of immune checkpoint inhibitors combined with cancer vaccines.

[0027] In recent years, models have further considered incorporating more types of tumor cell and immune cell interactions to simulate tumor adaptability and immune escape. For example, Piretto et al. (2018) proposed a two-tumor competition-immune cell model, shown in Figure 4. This model considers drug resistance and the competitive growth of two tumors and one immune cell type between tumor cells. It can predict the efficacy of immunotherapy combined with paclitaxel, and the results are highly consistent with the short-term tumor response (tumor changes within 30 days) obtained in the combination trial. Robinson et al. (2019) constructed a nonlinear mathematical model of two tumor cells and two immune cells, in which tumors are divided into immune-resistant and immune-sensitive types, and immune cells are divided into innate immune system cells and acquired T cells. Mahasa et al. (2016) considered the existence of tumor cells that can completely escape the immune system and constructed a three-tumor cell-two immune cell model, shown in Figure 5. This model simultaneously considers drug resistance and the tumor-immune cell model of different immune cells. Tumor 3 is a complete immune escape, while tumor 1 is a partial immune escape. Overall, the aforementioned models each have their own specific focus, but most are focused on theoretical framework design for equilibrium solutions or simulated solely through individual case studies based on animal experiments. They lack efficacy prediction and systematic validation based on clinical big data. Furthermore, the few studies that have used individual case-based predictions of combined effects suggest that these models primarily predict short-term tumor responses, lacking the ability to predict long-term clinical endpoints, and their predictive capabilities still need to be further improved.

[0028] The tumor microenvironment (TME) is an ecosystem composed of multiple components, including tumor cells, immune cells, stromal cells, the extracellular matrix, and growth factors. These various factors, through complex interactions and signaling networks, can influence tumor cell proliferation, invasion, and metastasis. These factors influence tumor therapeutic response and prognosis by regulating multiple biological processes, including tumor cell metabolism and survival, promoting or inhibiting the onset and progression of tumor inflammatory responses, and modulating tumor immune escape. Current theoretical models mostly consider tumor-immune cell interactions, but none incorporate all stromal cells into their mathematical models.

[0029] Stromal cells, including fibroblasts and endothelial cells, are a key component of the tumor immune microenvironment. These cells synthesize and secrete a variety of matrix proteins and extracellular matrix molecules, constructing the extracellular matrix scaffold surrounding the tumor, forming the tumor stroma. The tumor stroma provides the structural and nutritional foundation of the microenvironment, directly or indirectly influencing the function and behavior of tumor cells. It also regulates the directional migration and aggregation of immune cells, playing a crucial role in tumor development, progression, and response to therapeutics. Studying and addressing the inhibitory effects of the tumor stroma is a hot topic in cancer research. Currently, several drugs are targeted at the tumor stroma, such as anti-angiogenic (VEGF) drugs targeting tumor blood vessels and anti-fibrotic agents targeting tumor fibrosis. Furthermore, novel therapeutic strategies, such as CAR-T cell therapy and immune checkpoint inhibitors, also achieve therapeutic effects by influencing the tumor stroma and the immune microenvironment. Therefore, incorporating stromal cells into theoretical models of tumor-associated disease will improve their predictive accuracy.

[0030] Example 1, please refer to FIG6 , which is a flowchart of an embodiment of a tumor prognosis prediction method based on tumor-immune-stromal cells according to an embodiment of the present disclosure, and the method includes S101 to S108 .

[0031] Step S101, based on the acquired information on the type of anti-tumor treatment to be analyzed, determine the type and / or state of the cell population affected by the treatment, as well as all cell behaviors of the cell population affected by the treatment; the cell population types affected by the treatment include: tumor cells, immune cells and stromal cells.

[0032] Step S102, determining the post-treatment related cell behavior related to the anti-tumor treatment type information to be analyzed among all the cell behaviors; the cell behavior is the cell-independent behavior and cell-interaction behavior that affects the number of tumor microenvironment cell populations; the cell-independent behavior includes: proliferation, death and migration; the cell-interaction behavior includes: the clearance behavior of the immune cells against the tumor cells, the recruitment behavior of the tumor cells to the immune cells, and the conversion behavior between different cell types or states.

[0033] In step S103, the tumor cells are classified using a tumor evolution module to obtain tumor cell subpopulations, and the initial number ratio of the tumor cell subpopulations is calculated in combination with the genomic data sequenced by the patient.

[0034] Step S104 : constructing the tumor microenvironment using the microenvironment regulation module based on all independent cell behaviors and cell interaction behaviors of the immune cells and the stromal cells.

[0035] Step S105, forming the tumor-immune-stromal cell model framework based on the type and / or state of the cell population in the tumor microenvironment, all the cell behaviors, and third-party cell regulation behaviors; the third-party cell regulation refers to the regulation of the cell behaviors by the tumor cells, the immune cells, and the stromal cells.

[0036] Step S106: Using a tumor treatment module, the effects of the treatment on the type and / or state of the cell population and the post-treatment related cell behavior are converted into post-treatment tumor behavior parameters and / or tumor treatment parameters in the tumor-immune-stromal cell model framework; the tumor treatment parameters can be supplemented by batch sequencing with treatment and prognosis information, and / or selecting real treatment cases from a single-cell sequencing database to solve multivariate equations.

[0037] Step S107 , using sequencing technology, determines the type of cell population to be tested and the proportion of the initial number to be tested from the tumor microenvironment batch sequencing / single-cell sequencing data at the initial stage of the patient's treatment.

[0038] Step S108, based on the type of cell population to be tested, the proportion of the initial number to be tested, combined with the tumor behavior parameters after treatment and / or the tumor treatment parameters, a long-term prognosis module is evolved in the tumor-immune-stromal cell model framework to obtain tumor prognosis results for multiple patients.

[0039] In one embodiment, the tumor prognosis result includes an individual prognosis result and a group prognosis result; based on the type of cell population to be tested, the proportion of the initial number to be tested, combined with the tumor behavior parameters after treatment and / or the tumor treatment parameters, a long-term prognosis module is used to evolve in the tumor-immune-stromal cell model framework to obtain tumor prognosis results for multiple patients, including: based on the type of cell population to be tested, the proportion of the initial number to be tested, the tumor behavior parameters after treatment and / or the tumor treatment parameters, the long-term prognosis module is used to evolve in the tumor-immune-stromal cell model framework to obtain a tumor prognosis simulation model of the tumor-immune-stromal cells; the individual prognosis results of multiple patients are determined from the tumor prognosis simulation model of the tumor-immune-stromal cells; and the individual prognosis results of multiple patients are statistically analyzed to obtain the group prognosis result.

[0040] In one embodiment, the individual prognostic results include: individual tumor progression time and best response; determining multiple individual prognostic results of the patient from the tumor prognosis simulation model of the tumor-immune-stromal cells, including: determining the change information of the population number of each tumor cell subpopulation under growth interaction and drug action over time from the tumor prognosis simulation model of the tumor-immune-stromal cells; based on the change information, combined with the average volume of the cell population, calculating the change curve of the tumor relative volume over time; the tumor relative volume is the tumor volume relative to the initial tumor volume; according to the change curve, determining the individual tumor progression time and the best response of the patient.

[0041] The presently disclosed embodiments provide a tumor prognosis prediction method based on tumor-immune-stromal cells. By constructing an open tumor-immune-stromal cell tumor prognosis simulation model, the method simulates the prognosis of patients in different actual situations under different treatments from the perspectives of tumor-immune-stromal cell behavior and interaction, thereby determining the individual and group prognosis results of the patients.

[0042] For the second embodiment, please refer to FIG7 , which is a flowchart of a method for predicting tumor prognosis based on tumor-immune-stromal cells according to an embodiment of the present disclosure. The method includes S201 to S212 .

[0043] Step S201, based on the acquired information on the type of anti-tumor treatment to be analyzed, determine the type and / or state of the cell population affected by the treatment, as well as all cell behaviors of the cell population affected by the treatment; the cell population types affected by the treatment include: tumor cells, immune cells and stromal cells.

[0044] Step S202, determining the post-treatment related cell behavior related to the anti-tumor treatment type information to be analyzed among all the cell behaviors; the cell behavior is the cell-independent behavior and cell-interaction behavior that affects the number of tumor microenvironment cell populations; the cell-independent behavior includes: proliferation, death and migration; the cell-interaction behavior includes: the clearance behavior of the immune cells against the tumor cells, the recruitment behavior of the tumor cells to the immune cells, and the transformation behavior between different cell types or states.

[0045] In some cases, a tumor prognosis prediction method based on tumor-immune-stromal cells of an embodiment of the present disclosure is implemented on the basis of a tumor prognosis simulation model based on tumor-immune-stromal cells, and the tumor prognosis simulation model based on tumor-immune-stromal cells is mainly formed by the combined action of a tumor evolution module, a microenvironment regulation module, a tumor treatment module and a long-term prognosis module.

[0046] In the disclosed embodiments, after clarifying the anti-tumor type to be studied in the tumor prognosis simulation model based on tumor-immune-stromal cells, combined with the research progress of drug resistance, immune surveillance and matrix regulation under this type of treatment, the cell population type and cell interaction behavior information affected by this treatment under the current evidence system are clarified.

[0047] In step S203, the tumor cells are classified using a tumor evolution module to obtain tumor cell subpopulations, and the initial number ratio of the tumor cell subpopulations is calculated in combination with the genomic data sequenced by the patient.

[0048] In some cases, the starting point of the tumor prognosis simulation model based on tumor-immune-stromal cells is the initial number ratio of each type of tumor cell population in the tumor microenvironment before treatment and as close to the treatment moment as possible.

[0049] In the disclosed embodiment, the tumor evolution module is used to classify the tumor cell population and set the initial proportion of each tumor cell subpopulation.

[0050] Please refer to Figure 8, which is a schematic diagram of the mechanism of drug resistance and immune escape. In the disclosed embodiments, tumor cells mutate due to their own high variability or after being induced by certain drugs, thereby producing tumor cell subpopulations with different characteristics. After drug treatment, the original tumor cells and most subpopulations are killed by drugs and / or immune cells. However, tumor cell subpopulations with high adaptability (drug resistance or immune escape) are relatively unaffected. During a period of tumor evolution, highly adaptable tumor cell subpopulations are naturally selected and dominate the number of tumor cell populations.

[0051] In order to simulate the long-term growth state of tumor tissue and thus evaluate and predict the impact of combined therapy on the long-term survival benefit of patients, the disclosed embodiment incorporates drug resistance and immune escape caused by mutation and adaptability into the system dynamics model of the tumor evolution module. Its mechanism of action specifically includes the following processes: (1) Drugs (such as chemotherapy drugs) induce gene mutations in tumor cells, thereby leading to drug resistance or immune escape; (2) During the clonal evolution process, pre-existing tumor cell subpopulations with high adaptability will be naturally selected; (3) After drug treatment, the number of the original tumor cell population will significantly decrease, and the newly evolved high-adaptability subpopulation (such as a subpopulation with high drug resistance in targeted drug therapy, or a subpopulation of tumor cells with other immune escape mechanisms in immunotherapy) will gradually gain the upper hand; (4) When the number of tumor cells increases to the point where the tumor diameter increases by 20% compared to the diameter before treatment, it can be defined as progression according to the clinically used RECIST v1.1 criteria.

[0052] In some cases, due to the high variability of tumor cells, combined with research on tumor drug resistance, there is a certain probability that a highly adaptable subpopulation will already exist at the beginning of tumor treatment (this parameter can be accurately calculated using tumor cell gene sequencing technology). At the same time, as the mutation progresses during treatment, a highly adaptable variant subpopulation will be generated (under the assumption that the mutation is directionless, the mutation parameter during treatment is assumed to be 0). These highly adaptable subpopulations have a competitive advantage after tumor treatment, and their growth rate is much faster than that of subpopulations generated by random mutation during treatment (the growth rate of other tumor populations is assumed to be no statistically different from that of the original cell population). Therefore, tumor subpopulations do not need to be distinguished based on all mutations, but rather based on whether they are drug-resistant or immune-escape.

[0053] Step S204 : constructing the tumor microenvironment using the microenvironment regulation module based on all independent cell behaviors and cell interaction behaviors of the immune cells and the stromal cells.

[0054] Step S205, forming the tumor-immune-stromal cell model framework based on the type and / or state of the cell population in the tumor microenvironment, all the cell behaviors, and third-party cell regulation behaviors; the third-party cell regulation refers to the regulation of the cell behaviors by the tumor cells, the immune cells, and the stromal cells.

[0055] In order to accurately simulate the interactions between drugs, tumors, and the immune microenvironment, and thus more accurately predict the actual efficacy of combined therapy, the embodiments of the present disclosure take into account immune surveillance and matrix regulation in the tumor microenvironment and construct a tumor-immune-stromal cell model framework as shown in Figure 9.

[0056] In the disclosed embodiment, a tumor-immune-stromal cell model framework is constructed through a microenvironment regulation module, and then the types of immune cell populations and stromal cell populations and their initial number ratios, as well as the cell behaviors or cell interaction parameters in the tumor microenvironment are input. In a specific implementation, cells of the same type can be further divided into different states based on the input of the types of immune cell populations and stromal cell populations. On the other hand, the initial number ratios of the types of immune cell populations and stromal cell populations can be determined by sequencing technologies such as batch analysis or single-cell analysis methods.

[0057] In one embodiment, the microenvironment regulation module includes an immune surveillance submodule and a matrix regulation submodule. The immune surveillance submodule takes into account various immune cells, different states of immune cells, and their interactions. The matrix regulation submodule includes factors such as matrix cells and their secreted cytokines, extracellular matrix, and the spatial distribution (boundaries) of matrix cells, while considering their responses to different tumor therapies, their effects on tumor cells and immune cells, and their interactions.

[0058] The framework of the tumor-immune-stromal cell model includes: a tumor cell population on the left, stromal cells in the center, and immune cells on the right. Arrows between tumor cells, immune cells, and stromal cells represent competitive and transformative interactions between tumor cell subpopulations and stromal cell types, as well as transformation or promotion and inhibition of proliferation among immune cells. Arrows from immune cells to tumor cell populations represent immune cell effects on tumors, such as immune clearance. Arrows from tumor cell populations to immune cells represent tumor recruitment and other effects on immune cells. Arrows from stromal cells to tumor cell populations, and arrows from stromal cells to immune cells, represent various pro- and inhibitory cytokines secreted by stromal cells, affecting tumors, immune cells, and being affected by tumor immune cells.

[0059] In practice, various variant subpopulations exist within the tumor cell population, such as highly drug-resistant subpopulations, immune escape subpopulations, subpopulations with both high drug resistance and immune escape, and other characteristic subpopulations. Under the assumption that cell growth resources are limited (scarce), these subpopulations grow together with the original tumor population and compete with each other. Different immune cell types exist within the immune cell population, such as CD4+ T cells, CD8+ T cells, regulatory T cells, natural killer cells, macrophages, and dendritic cells, which can directly or indirectly (e.g., by secreting cytokines) promote or inhibit the survival and function of other immune cells. Matrix cell populations also contain different types, such as fibroblasts, pericytes, and endothelial cells, which also compete for growth and may secrete cytokines to promote or inhibit the growth or function of other stromal cells. Tumor cell populations secrete or induce chemokines, cell adhesion factors, or immunosuppressive factors, inhibiting immune cell populations. Immune cell populations, in turn, primarily eliminate tumor cell populations through immune surveillance, or immune cells with immunosuppressive properties can promote tumor growth. Stromal cells also play a crucial role in tumor growth. They can compete with tumor cells for growth and secrete cytokines and extracellular matrix to influence the growth, immune clearance, and recruitment of tumor and immune cells. Furthermore, the complex spatial structure of the extracellular matrix secreted by stromal cells can influence tumor growth and invasion, inhibit immune cell infiltration, modulate immune cell function, and promote tumor angiogenesis. Because this model considers long-term prognostic benefits and has a large temporal granularity, it can transform the cytokine action process into direct cell-cell interactions, disregarding very short-term diffusion effects.

[0060] Step S206: Using a tumor treatment module, the effects of the treatment on the type and / or state of the cell population and the post-treatment related cell behavior are converted into post-treatment tumor behavior parameters and / or tumor treatment parameters in the tumor-immune-stromal cell model framework; the tumor treatment parameters can be supplemented by batch sequencing with treatment and prognosis information, and / or selecting real treatment cases from a single-cell sequencing database to solve multivariate equations.

[0061] In the disclosed embodiment, according to the tumor treatment module, the collected drug efficacy evidence is converted into cell population behavior and / or cell population interaction in the tumor-immune-stromal cell model framework, and then the cell population behavior and / or cell population interaction in the tumor-immune-stromal cell model framework is converted into cell population behavior and / or cell population interaction in the tumor-immune-stromal cell model framework.

[0062] In addition, real treatment cases can be selected through batch sequencing and / or single-cell sequencing databases with treatment and prognosis information to solve multivariate equations and obtain the required tumor treatment parameters.

[0063] In practice, to differentiate the impact of different tumor treatments on the results generated by the tumor-immune-stroma model framework, the disclosed embodiments incorporate several basic tumor treatment types into the tumor-immune-stroma model. Based on therapeutic targets and potential therapeutic effects, current mainstream anti-tumor therapies are divided into six categories, as shown in the following table:

[0064] Among them, drug therapy generally regulates cell-independent behavior (such as growth) and cell-to-cell interactions (such as immune clearance). Under sufficient drug treatment (continuous administration), the drug effect can be set as a continuous effect, regardless of the time of drug delivery and metabolism, such as continuously inhibiting the rate of tumor proliferation; non-drug treatment generally directly affects the number of the active cell population. For example, immune cell therapy directly increases the absolute number of immune cell populations during treatment. In this model, it is set as an instantaneous effect, regardless of the duration of treatment.

[0065] The tumor treatment module in the disclosed embodiment uses a binary classification method to distinguish tumor subpopulations, that is, it is divided only into high-adaptability (drug resistance, immune escape) subpopulations and original subpopulations (assuming that the treatment is effective, so this subpopulation is dominant). For tumor cells under combination therapy, there are different high-adaptability tumor subpopulations corresponding to different treatments. In theory, the tumor cell subpopulation that is resistant to all treatments or immune escapes has the highest competitive advantage. However, when a certain treatment regimen is ineffective, the dominant population may change. When the combination therapy acts on the interactions or independent behaviors of different cells, this model directly adds independently; when it acts on the same cell interaction or independent behavior, this article refers to the highest single drug hypothesis as the most conservative estimate, which is equal to the efficacy of the drug with the greatest impact (more accurate data requires direct measurement of cell experiments of combination therapy, and the combination therapy can be regarded as a treatment regimen at this time).

[0066] Step S207 , using sequencing technology, determines the type of cell population to be tested and the proportion of the initial number to be tested from the tumor microenvironment batch sequencing / single-cell sequencing data at the initial stage of the patient's treatment.

[0067] The disclosed embodiments use sequencing technology in combination with CIBERSORTx and other related deconvolution algorithms to determine the type of cell population to be tested and the proportion of the initial number to be tested from the collected patient data.

[0068] Step S208, based on the type of cell population to be tested, the proportion of the initial number to be tested, the tumor behavior parameters after treatment and / or the tumor treatment parameters, the long-term prognosis module is evolved in the tumor-immune-stromal cell model framework to obtain a tumor prognosis simulation model of the tumor-immune-stromal cells.

[0069] Step S209 , determining information on changes in the population size of each tumor cell subpopulation over time under growth interaction and drug action from the tumor prognosis simulation model of the tumor-immune-stromal cells.

[0070] Step S210 , based on the change information and in combination with the average volume of the cell population, a curve of the change of the relative tumor volume over time is calculated; the relative tumor volume is the tumor volume relative to the initial tumor volume.

[0071] Step S211, determining the individual tumor progression time and the optimal response of the patient according to the change curve, may specifically include: taking the duration of time when the relative volume of the tumor in the change curve is less than 1.728 as the individual tumor progression time.

[0072] Determine whether there is a portion in the change curve where the tumor relative volume is less than 0.343 and lasts for more than 4 weeks; if so, the best response is complete remission or partial remission; if not, the best response is disease stabilization or disease progression.

[0073] Step S212, performing statistical analysis on the individual prognostic results of multiple patients to obtain the group prognostic results, may include: calculating the prognostic curve of the patient group based on the individual tumor progression time and the best response of the multiple patients; summarizing and analyzing the prognostic curves to determine the change in the proportion of non-progressing patients in the group under the preset treatment cycle, and determining the median tumor progression time of the group based on the median of the proportion change; based on the prognostic curves, determining the group proportion of complete remission and partial remission in the best response under the preset treatment cycle, and determining the objective response rate of the group based on the group proportion.

[0074] In the disclosed embodiment, under the action of the tumor evolution module, the microenvironment regulation module and the tumor treatment module, the individual patient prognosis results and the group prognosis results are obtained based on the conditions of each tumor cell population under the action of drugs in the tumor prognosis simulation model of tumor-immune-stromal cells evolved from the long-term prognosis module.

[0075] In the specific implementation, the model is set to have heterogeneous patients equivalent to heterogeneous tumor cells, that is, the number of various types of cell populations in the tumor microenvironment is inconsistent at the beginning of treatment. This data can be combined with single-cell sequencing, or inferred and determined by deconvolution algorithm based on batch sequencing data. Combined with drug treatment evidence and cell-based behavior parameters, the changes in the population of each cell subpopulation over time under growth interaction and drug action after tumor treatment are simulated. Then, based on the average volume of each cell population, without considering cell extrusion deformation, the curve of the tumor volume relative to the initial tumor volume (defined as the relative tumor volume) over time is calculated, as shown in Figure 10, where the curve in Figure 10 shows the case of BOR=PR or CR, and the ordinate is the relative volume of the total tumor population. When the BoR of the tumor is less than 0.343 times the original volume, it can be defined as PR or CR; when the relative volume of the tumor does not exceed 1.728, it is the tumor progression time.

[0076] In some cases, the clinical RECIST v1.1 criteria include CR: all target lesions disappear, and the short diameter of all pathological lymph nodes, including target nodules and non-target nodules, must be reduced to <10 mm; PR: the sum of the target lesion diameters is reduced by at least 30% compared with the baseline level (i.e., the volume is reduced to 0.73 times) and maintained for more than 4 weeks.

[0077] At the same time, among the individual prognostic indicators of patients considered in this model, the primary outcome indicator is the time from the start of treatment to tumor progression (diameter increase of 20%, i.e., tumor relative volume = 1.23), i.e., individual tumor progression time. The secondary outcome indicator is the best overall efficacy (BOR) of the individual tumor, i.e., the best response. According to the tumor relative volume curve, it is judged whether the relative volume meets the following conditions for at least 4 weeks: (1) when the relative volume = 0, it is a complete response (CR); (2) when the relative volume ≤ 0.73, it is a partial response (PR); (3) when 0.73 < relative volume ≤ 1.23, it is stable disease (SD); (4) the rest are progressive disease (PD).

[0078] By summarizing the outcome indicators of multiple heterogeneous patients, a prognostic curve for the patient population can be calculated. The primary prognostic indicator of the prognostic curve is the patient's tumor progression-free curve (corresponding to the difference between the progression-free survival (PFS) curve and the overall survival (OS) curve in the clinical trial), which measures the proportion of patients with no progression over time to the total population. The median of these proportions is calculated to measure the overall prognosis of the patient population. A secondary prognostic indicator is the objective response rate (ORR), which is the proportion of patients with a BOR of CR and PR to the total population.

[0079] Example 3, please refer to Figure 11, which is a structural block diagram of an embodiment of a tumor prognosis prediction device based on tumor-immune-stromal cells disclosed in the present invention, including: an information determination module 301, a post-treatment related cell behavior determination module 302, a subpopulation and its proportion determination module 303, a microenvironment construction module 304, a framework formation module 305, a parameter determination module 306, a sequencing module 307 and a prognosis result determination module 308.

[0080] The information determination module 301 is used to determine the type and / or state of the cell population affected by the treatment, as well as all cellular behaviors of the cell population affected by the treatment, based on the acquired information on the type of anti-tumor treatment to be analyzed; the cell population types affected by the treatment include: tumor cells, immune cells and stromal cells.

[0081] The post-treatment related cell behavior determination module 302 is used to determine the post-treatment related cell behavior related to the anti-tumor treatment type information to be analyzed among all the cell behaviors; the cell behavior is the cell-independent behavior and cell-interaction behavior that affects the number of cell populations in the tumor microenvironment; the cell-independent behavior includes: proliferation, death and movement; the cell-interaction behavior includes: the clearance behavior of the immune cells against the tumor cells, the recruitment behavior of the tumor cells to the immune cells, and the transformation behavior between different cell types or states.

[0082] The subpopulation and its proportion determination module 303 is used to classify the tumor cells using the tumor evolution module to obtain each tumor cell subpopulation, and calculate the initial number proportion of the tumor cell subpopulation based on the patient's sequenced genomic data.

[0083] The microenvironment construction module 304 is configured to construct the tumor microenvironment using the microenvironment regulation module based on all independent cell behaviors and cell interaction behaviors of the immune cells and the stromal cells.

[0084] The framework forming module 305 is used to form the tumor-immune-stromal cell model framework based on the type and / or state of the cell population in the tumor microenvironment, all the cell behaviors, and third-party cell regulation behavior; the third-party cell regulation is the regulation of the cell behavior by the tumor cells, the immune cells, and the stromal cells.

[0085] The parameter determination module 306 is used to utilize the tumor treatment module to convert the effect of the treatment on the type and / or state of the cell population and the post-treatment related cell behavior into post-treatment tumor behavior parameters and / or tumor treatment parameters in the tumor-immune-stromal cell model framework; the tumor treatment parameters can be supplemented by batch sequencing with treatment and prognosis information, and / or selecting real treatment cases from a single-cell sequencing database to solve multivariate equations.

[0086] The sequencing module 307 is used to determine the type of cell population to be tested and the proportion of the initial number to be tested from the tumor microenvironment batch sequencing / single-cell sequencing data at the beginning of the patient's treatment through sequencing technology.

[0087] The prognosis result determination module 308 is used to obtain tumor prognosis results for multiple patients based on the type of cell population to be tested, the proportion of the initial number to be tested, the tumor behavior parameters after treatment and / or the tumor treatment parameters, and the evolution of the long-term prognosis module in the tumor-immune-stromal cell model framework.

[0088] In an optional embodiment, the tumor prognosis result includes an individual prognosis result and a group prognosis result; the prognosis result determination module 308 includes: a simulation model determination submodule, which is used to evolve the long-term prognosis module in the tumor-immune-stromal cell model framework based on the cell population type to be tested, the proportion of the initial number to be tested, the tumor behavior parameters after treatment and / or the tumor treatment parameters, to obtain the tumor prognosis simulation model of the tumor-immune-stromal cells; an individual prognosis result determination submodule, which is used to determine the individual prognosis results of multiple patients from the tumor prognosis simulation model of the tumor-immune-stromal cells; and a group prognosis result determination submodule, which is used to perform statistical analysis on the individual prognosis results of multiple patients to obtain the group prognosis result.

[0089] In an optional embodiment, the individual prognosis result includes: individual tumor progression time and optimal response; the individual prognosis result determination submodule includes: a change information determination unit, which is used to determine the change information of the population number of each tumor cell subpopulation over time under growth interaction and drug action from the tumor prognosis simulation model of the tumor-immune-stromal cells; a change curve determination unit, which is used to calculate the change curve of the tumor relative volume over time based on the change information and combined with the average volume of the cell population; the tumor relative volume is the tumor volume relative to the initial tumor volume; the individual prognosis result determination unit is used to determine the patient's individual tumor progression time and the optimal response based on the change curve.

[0090] In an optional embodiment, the individual prognosis result determination unit includes: a progression time determination subunit, used to use the duration of time in the change curve when the tumor relative volume is less than 1.728 as the individual tumor progression time; a judgment subunit, used to judge whether there is a part in the change curve where the tumor relative volume is less than 0.343 and lasts for more than 4 weeks; if so, the best response is complete remission or partial remission; if not, the best response is disease stabilization or disease progression.

[0091] In an optional embodiment, the group prognosis results include: median tumor progression time and objective response rate; the group prognosis result determination submodule includes: a prognosis curve calculation unit, which is used to calculate the prognosis curve of the patient group based on the individual tumor progression time and the best response of multiple patients; a summary unit, which is used to summarize and analyze the prognosis curve, determine the change in the proportion of non-progressing patients in the group under the preset treatment cycle, and determine the median tumor progression time of the group based on the median of the proportion change; an objective response rate determination unit, which is used to determine the group proportion of complete remission and partial remission in the best response under the preset treatment cycle based on the prognosis curve, and determine the objective response rate of the group based on the group proportion.

[0092] Embodiment 4. The present disclosure also provides an electronic device, including a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of a tumor prognosis prediction method based on tumor-immune-stromal cells as described in any of the above embodiments.

[0093] Embodiment 5. The present disclosure also provides a computer storage medium having a computer program stored thereon. When the computer program is executed by the processor, the steps of a tumor prognosis prediction method based on tumor-immune-stromal cells as described in any of the above embodiments are implemented.

[0094] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0095] In the several embodiments provided in the present disclosure, it should be understood that the methods, devices, electronic devices and storage media disclosed in the present disclosure can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0096] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0097] In addition, the functional units in the various embodiments of the present disclosure may be integrated into a single processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0098] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a readable storage medium, including a number of instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned readable storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0099] As described above, the above embodiments are only used to illustrate the technical solutions of the present disclosure, rather than to limit the same. Although the present disclosure has been described in detail with reference to the above embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the above embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present disclosure.

Claims

1. A tumor prognosis prediction method based on tumor-immune-stromal cells, wherein, it includes: Determine the type and / or state of the cell population affected after treatment, and all cell behaviors of the cell population affected after treatment according to the obtained anti-tumor treatment type information to be analyzed; The types of the cell populations affected after treatment include: tumor cells, immune cells and stromal cells; Determine the post-treatment related cell behaviors related to the anti-tumor treatment type information to be analyzed among all the cell behaviors; the cell behaviors are the cell independent behaviors and cell interaction behaviors that affect the number of cell populations in the tumor microenvironment; the cell independent behaviors include: proliferation, death and migration; the cell interaction behaviors include: the clearance behavior of the immune cells on the tumor cells, the recruitment behavior of the tumor cells on the immune cells, and the transformation behavior between different cell types or states; Use the tumor evolution module to classify the tumor cells to obtain each tumor cell subset, and calculate the initial quantity proportion of the tumor cell subsets in combination with the genomic data of the patient's sequencing; Based on all the cell independent behaviors and cell interaction behaviors of the immune cells and the stromal cells, use the microenvironment regulation module to construct the tumor microenvironment; Based on the type and / or state of the cell population in the tumor microenvironment, all the cell behaviors, and the third-party cell regulation behaviors, form the tumor-immune-stromal cell model framework; the third-party cell regulation is the regulation of the cell behaviors by the tumor cells, the immune cells and the stromal cells; Use the tumor treatment module to convert the influence of the treatment on the type and / or state of the cell population and the post-treatment related cell behaviors into the post-treatment tumor behavior parameters and / or tumor treatment parameters in the tumor-immune-stromal cell model framework; the tumor treatment parameters can be supplemented by performing multiple equation solutions on real treatment cases selected from the batch sequencing with treatment and prognosis information, and / or the single-cell sequencing database; Determine the type of the cell population to be measured and the initial quantity proportion to be measured from the batch sequencing / single-cell sequencing data of the tumor microenvironment at the initial stage of the patient's treatment through sequencing technology; Based on the type of the cell population to be measured, the initial quantity proportion to be measured, in combination with the post-treatment tumor behavior parameters and / or the tumor treatment parameters, use the long-term prognosis module to evolve in the tumor-immune-stromal cell model framework to obtain the tumor prognosis results of multiple patients.

2. The tumor prognosis prediction method based on tumor-immune-stromal cells according to claim 1, wherein, The tumor prognosis results include individual prognosis results and population prognosis results; based on the type of the cell population to be measured, the initial quantity proportion to be measured, in combination with the post-treatment tumor behavior parameters and / or the tumor treatment parameters, use the long-term prognosis module to evolve in the tumor-immune-stromal cell model framework to obtain the tumor prognosis results of multiple patients, including: Based on the type of the cell population to be measured, the initial quantity proportion to be measured, the tumor behavior parameters after treatment, and / or the tumor treatment parameters, the long-term prognosis module is used to evolve within the tumor-immune-stromal cell model framework to obtain the tumor prognosis simulation model of the tumor-immune-stromal cells; Determine the individual prognosis results of multiple patients from the tumor prognosis simulation model of the tumor-immune-stromal cells; Perform statistical analysis on the individual prognosis results of multiple patients to obtain the population prognosis result.

3. The tumor prognosis prediction method based on tumor-immune-stromal cells according to claim 2, wherein, The individual prognosis results include: the individual tumor progression time and the best response; determining the individual prognosis results of multiple patients from the tumor prognosis simulation model of the tumor-immune-stromal cells includes: Determine the information on the change in the population quantity of each tumor cell subset over time under growth interaction and drug action from the tumor prognosis simulation model of the tumor-immune-stromal cells; Based on the change information, combined with the average volume of the cell population, calculate the change curve of the relative tumor volume over time; the relative tumor volume is the tumor volume relative to the initial tumor volume; According to the change curve, determine the individual tumor progression time and the best response of the patient.

4. The tumor prognosis prediction method based on tumor-immune-stromal cells according to claim 3, wherein, According to the change curve, determining the individual tumor progression time and the best response of the patient includes: Taking the duration when the relative tumor volume in the change curve is less than 1.728 as the individual tumor progression time; Judge whether there is a part in the change curve where the relative tumor volume is less than 0.343 and lasts for more than 4 weeks; if so, the best response is complete remission or partial remission; if not, the best response is stable disease or disease progression.

5. The tumor prognosis prediction method based on tumor-immune-stromal cells according to claim 3, wherein, The population prognosis results include: the median tumor progression time and the objective response rate; performing statistical analysis on the individual prognosis results of multiple patients to obtain the population prognosis results includes: Based on the individual tumor progression time and the best response of multiple patients, calculate the prognosis curve of the patient population; Summarize and analyze the prognosis curve, determine the change in the proportion of non-progressed patients in the population quantity under the preset treatment cycle, and determine the median tumor progression time of the population according to the median of the proportion change; Based on the prognosis curve, determine the population proportion of complete remission and partial remission in the best response under the preset treatment cycle, and determine the objective response rate of the population according to the population proportion.

6. A tumor prognosis prediction device based on tumor-immune-stromal cells, wherein, Comprises: An information determination module, configured to determine the type and / or state of the cell population affected after treatment, and all cell behaviors of the cell population affected after treatment according to the obtained anti-tumor treatment type information to be analyzed; The types of cell populations of the post-treatment effects include: tumor cells, immune cells, and stromal cells; A post-treatment related cell behavior determination module, configured to determine post-treatment related cell behaviors related to the information on the type of anti-tumor treatment to be analyzed among all the cell behaviors; the cell behaviors are cell independent behaviors and cell interaction behaviors that affect the cell population quantity in the tumor microenvironment; the cell independent behaviors include: proliferation, death, and migration; the cell interaction behaviors include: the clearance behavior of the immune cells on the tumor cells, the recruitment behavior of the tumor cells on the immune cells, and the transformation behavior between different cell types or states; A subgroup and its proportion determination module, configured to classify the tumor cells by using the tumor evolution module to obtain each tumor cell subgroup, and calculate the initial quantity proportion of the tumor cell subgroup in combination with the genomic data of the patient's sequencing; A microenvironment construction module, configured to construct the tumor microenvironment by using the microenvironment regulation module based on all the cell independent behaviors and cell interaction behaviors of the immune cells and the stromal cells; A framework formation module, configured to form the tumor-immune-stromal cell model framework based on the types and / or states of the cell populations in the tumor microenvironment, all the cell behaviors, and third-party cell regulation behaviors; the third-party cell regulation is the regulation of the cell behaviors by the tumor cells, the immune cells, and the stromal cells; A parameter determination module, configured to use the tumor treatment module to convert the influence of the treatment on the types and / or states of the cell populations and the post-treatment related cell behaviors into the post-treatment tumor behavior parameters and / or tumor treatment parameters in the tumor-immune-stromal cell model framework; the tumor treatment parameters can be supplemented by solving multiple equations by selecting real treatment cases from a batch sequencing database with treatment and prognosis information, and / or a single-cell sequencing database; A sequencing module, configured to determine the types of cell populations to be measured and the initial quantity proportion to be measured from the batch sequencing / single-cell sequencing data of the tumor microenvironment at the initial stage of the patient's treatment by using sequencing technology; A prognosis result determination module, configured to obtain the tumor prognosis results of multiple patients by evolving the tumor-immune-stromal cell model framework by using the long-term prognosis module based on the types of cell populations to be measured, the initial quantity proportion to be measured, in combination with the post-treatment tumor behavior parameters and / or the tumor treatment parameters; 7. The tumor prognosis prediction device based on tumor-immune-stromal cells according to claim 6, wherein, the tumor prognosis results include individual prognosis results and population prognosis results; the prognosis result determination module includes: a simulation model determination sub-module, configured to obtain the tumor prognosis simulation model of the tumor-immune-stromal cells by evolving the tumor-immune-stromal cell model framework by using the long-term prognosis module based on the types of cell populations to be measured, the initial quantity proportion to be measured, the post-treatment tumor behavior parameters and / or the tumor treatment parameters; An individual prognosis result determination sub-module, configured to determine the individual prognosis results of multiple patients from the tumor prognosis simulation model of tumor-immune-stromal cells; A population prognosis result determination sub-module, configured to perform statistical analysis on the individual prognosis results of multiple patients to obtain the population prognosis results.

8. The tumor prognosis prediction device based on tumor-immune-stromal cells according to claim 7, wherein, the individual prognosis results include: individual tumor progression time and best response; the individual prognosis result determination sub-module includes: A change information determination unit, configured to determine the change information of the population quantity of each of the tumor cell subsets over time under growth interaction and drug action from the tumor prognosis simulation model of tumor-immune-stromal cells; A change curve determination unit, configured to calculate the change curve of the relative tumor volume over time based on the change information and in combination with the average volume of the cell population; the relative tumor volume is the tumor volume relative to the initial tumor volume; An individual prognosis result determination unit, configured to determine the individual tumor progression time and the best response of the patient according to the change curve.

9. An electronic device, wherein, it includes a processor and a memory, and the memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, the method according to any one of claims 1-5 is run.

10. A storage medium, on which a computer program is stored, wherein, when the computer program is executed by a processor, the method according to any one of claims 1-5 is run.

Images