A kit and method for evaluating the malignancy of digestive system cancer

Abstract

The embodiments of the present invention disclose a kit and method for evaluating the malignancy of digestive system cancer. A kit for evaluating the malignancy of digestive system cancer includes reagents for detecting the following indicators: sialic acid accumulation level; infiltration level of B cells, macrophages and neutrophils; and hypoxia level. The present invention obtains sialic acid accumulation level; infiltration level of B cells, macrophages and neutrophils; and hypoxia level, which are key factors affecting the malignancy of digestive system cancer. The model constructed by the above key factors has good sensitivity and specificity, can systematically and comprehensively evaluate the malignancy of digestive system cancer, and is of great significance for the early diagnosis and treatment of digestive system cancer.

Description

Technical Field

[0001] The embodiments of the present invention relate to the field of biomedical technology, and in particular to a kit and method for evaluating the malignancy of digestive system cancer. Background Art

[0002] Cancer is a major disease that seriously threatens human health worldwide, especially digestive system cancer, which has high morbidity and mortality rates. Advanced digestive system cancer is difficult to treat, patients have short survival periods, and are prone to recurrence and metastasis. Among digestive system cancers, cholangiocarcinoma is a rare and highly aggressive cancer, and its research is particularly important. Cholangiocarcinoma originates from the bile duct and has an average five-year survival rate of 9%, the lowest among the 33 major cancer types publicly released worldwide. This is mainly due to its unique tumor microenvironment and metabolic reprogramming. Cholangiocarcinoma is the most easily metastatic type among the seven types of digestive system cancer (i.e., bile duct cancer, colon adenocarcinoma, esophageal cancer, hepatocellular carcinoma, pancreatic adenocarcinoma, rectal adenocarcinoma, and gastric adenocarcinoma) due to high levels of bile acid production. The latest data released shows that there will be 988,627 cases and 847,780 deaths worldwide in 2022.

[0003] Current methods for the early diagnosis, efficacy assessment, and prognosis of digestive system cancers have certain limitations. Histopathological examinations or imaging scans are commonly used for qualitative and quantitative analysis of cancer. However, these two methods can only provide doctors with reference information and cannot fully and accurately assess a patient's condition. For example, tissue sections may not fully capture tumor samples, increasing the risk of misdiagnosis; and imaging cannot display organ function, which can affect doctors' judgment. Therefore, there is an urgent need for a more accurate, simple, economical, and reliable cancer malignancy assessment kit and method in clinical practice, which would have significant practical value in improving diagnostic accuracy, reducing costs, and accelerating medical treatment. Summary of the Invention

[0004] We conducted a comprehensive computational study, with the following key findings: (1) The biggest factor in cholangiocarcinoma's high malignancy is that its sialic acid biosynthesis and accumulation on the cancer cell surface are the highest among seven cancer types, driving cancer cell migration and invasion; (2) Bile acids play a key role in inhibiting cell proliferation and indirectly promoting sialic acid synthesis and accumulation; (3) Further extending the study to seven digestive system cancer types, we found that the five-year survival rate can be well explained by the accumulation of sialic acid on the cancer cell surface, the infiltration of B cells, macrophages, and neutrophils, and the level of hypoxia. This is the first study to explain the high malignancy of a cancer type from the perspective of cancer chemical state and host organ characteristics.

[0005] To this end, embodiments of the present invention provide a kit and method for assessing the malignancy of digestive system cancer, so as to address the defects of low accuracy and low sensitivity of existing methods for assessing the malignancy of digestive system cancer.

[0006] In order to achieve the above objectives, the embodiments of the present invention provide the following technical solutions:

[0007] According to a first aspect of an embodiment of the present invention, the present invention provides a kit for assessing the malignancy of digestive system cancer, comprising the following indicators for detecting tumor samples: sialic acid accumulation level; infiltration level of B cells, macrophages and neutrophils; and hypoxia level.

[0008] Furthermore, it is characterized in that the calculation method of the sialic acid accumulation level is as follows:

[0009] The average level of sialic acid accumulation in tumor samples at the Localized, Regional, Distant, and Combined stages was detected using the following formula (1):

[0010]

[0011] Among them, i∈{Localized, Regional, Distant, Combined}, G1={ST6GALNAC1, ST6GALNAC2, ST6GALNAC4}, G2=NEU3, Ave i (G,S) represents the average expression value of G in all samples S at stage i;

[0012] The following formula (2) is used to evaluate the level of sialic acid accumulation up to stage I:

[0013] ΔSA(I)=∑ i≤I ΔSA' i (2)

[0014] Among them, I = Localized represents the tumor sample of the primary lesion, I = Regional represents the tumor sample near the primary lesion, and I = Distant represents the metastatic tumor sample.

[0015] Furthermore, the xCell calculation tool was used to calculate enrichment scores to detect the infiltration levels of B cells, macrophages, and neutrophils in tumor samples. Specifically, the infiltration levels of B cells, macrophages, and neutrophils in a tumor sample can be assessed by comparing the enrichment scores of B cells, macrophages, and neutrophils in the tumor sample to the total enrichment scores of all immune cell types in the tumor sample.

[0016] Furthermore, the expression of HIF1A was used to assess the hypoxia level in tumor samples. HIF1A, or hypoxia-inducible factor 1α subunit, is a protein that plays a key role in the cellular response to hypoxia. Higher HIF1A expression indicates a higher level of hypoxia.

[0017] Furthermore, the digestive system cancer is bile duct cancer, colon adenocarcinoma, esophageal cancer, hepatocellular carcinoma, pancreatic adenocarcinoma, rectal adenocarcinoma or gastric adenocarcinoma.

[0018] According to a second aspect of an embodiment of the present invention, the present invention provides a method for assessing the malignancy of digestive system cancer, the method comprising:

[0019] Tumor samples were tested for sialic acid accumulation; B cell, macrophage, and neutrophil infiltration levels; and hypoxia levels.

[0020] Comprehensive evaluation was performed based on the multivariate linear regression model established with the five-year survival rate.

[0021] Furthermore, the multiple linear regression model includes:

[0022] Localized regression model model1:

[0023] SR=1.0158-0.4043ΔSA+0.2528H+0.0434B+0.004M-0.4158N (3)

[0024] Regional regression model model2:

[0025] SR=1.3789-0.5405ΔSA+0.3069H+0.1469B-0.1133M-0.8068N (4)

[0026] Distant regression model model3:

[0027] SR=0.3679-0.1086ΔSA-0.0948H+0.136B+0.0665M-0.124N (5)

[0028] Combined regression model model4:

[0029] SR=1.58-0.3908ΔSA+0.0104H+0.2268B+0.0937M-0.6073N (6)

[0030] Among them, SR represents the five-year survival rate; ΔSA represents the sialic acid accumulation level; H represents the hypoxia level; B, M, and N represent the infiltration levels of B cells, macrophages, and neutrophils, respectively.

[0031] Furthermore, 60% is set as a threshold. If the five-year survival rate is higher than the threshold, digestive system cancer is a low-malignancy tumor; if the five-year survival rate is lower than the threshold, digestive system cancer is a high-malignancy tumor.

[0032] The present invention has the following advantages:

[0033] The present invention obtains the sialic acid accumulation level; the infiltration level of B cells, macrophages and neutrophils and the hypoxia level are key factors affecting the malignancy of digestive system cancer. The model constructed by the above key factors has good sensitivity and specificity, and can systematically and comprehensively evaluate the malignancy of digestive system cancer, which is of great significance for the early diagnosis and treatment of digestive system cancer. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] To more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for the embodiments or the description of the prior art. Obviously, the drawings described below are merely exemplary, and those skilled in the art can derive other implementation drawings based on the provided drawings without inventive effort.

[0035] Figure 1 Figure 3: Hypoxia levels in localized (A), regional (B), distant (C), and combined (D) tumor samples and control samples. HIF1A expression is used to reflect hypoxia levels in tumor samples and control samples at various cancer stages across seven cancer types. The y-axis in each figure represents hypoxia levels using the logarithmic transformation of TPM (gene expression format).

[0036] Figure 2 : Evaluated infiltration levels of B cells (A), macrophages (B), and neutrophils (C) in the control (normal tissue), localized, regional, distant, and combined tissues in seven cancer types, where the y-axis represents the percentage of infiltrating cells in the tissue sample; (D) shows the relationship between the B cell infiltration level (y-axis) and the sialic acid biosynthesis and attachment level (x-axis) measured using the first principal component of the expression of marker genes corresponding to sialic acid biosynthesis and attachment in the relevant samples, with a correlation coefficient of 0.823, as shown in Table 3.

[0037] Figure 3 :(A) De novo biosynthesis of nucleotides (NT), biosynthesis of sialic acid (SA) and their use to neutralize OH produced by Fenton reaction -The x-axis represents the contribution weight of nucleotide biosynthesis, the y-axis represents the contribution weight of sialic acid biosynthesis and utilization, and the points on the diagonal represent the contribution of the two biosynthetic processes to the neutralization of OH produced by the Fenton reaction. - The contributions of the three genes are equally weighted; (B) Predicted levels of sialic acid accumulation in each cancer type; (C) Correlations between bile acid (BA) biosynthesis and de novo nucleotide biosynthesis, and between bile acid biosynthesis and sialic acid biosynthesis; (D) Gene enrichment pathways related to cell cycle progression that are positively correlated with GPBAR1 in cholangiocarcinoma; (E) The bile acid receptor GPBAR1 is negatively correlated with cell cycle marker genes, where each point represents a cholangiocarcinoma sample, its normalized GPBAR1 expression level (x-axis) and the first principal component of normalized cell cycle-related gene expression levels (y-axis); (F) Correlations between de novo nucleotide synthesis and sialic acid biosynthesis and utilization at different stages. The x-axis represents the normalized de novo nucleotide synthesis level, and the y-axis represents the normalized sialic acid synthesis level. All relevant marker genes are listed in Table 3.

[0038] Figure 4 : The five-year survival probability (y-axis) of seven cancer types and the sialic acid accumulation level, hypoxia level, infiltration level of B cells, macrophages and neutrophils were used to obtain the predicted five-year survival probability (x-axis) through multiple linear regression analysis, where (A), (B), (C) and (D) are for Localized, Regional, Distant and Combined sample sets, respectively.

[0039] Figure 5 Evaluating model accuracy. A threshold of 60% was set. Survival rates above the threshold were considered low-grade malignancy, while survival rates below the threshold were considered high-grade malignancy. Receiver-operating characteristic (ROC) curves of the SVM-based model were used to evaluate the accuracy of (A) stage I, (B) stage II, and (C) stages III-IV. DETAILED DESCRIPTION

[0040] The following describes the implementation of the present invention using specific embodiments. Those skilled in the art will readily understand the other advantages and benefits of the present invention from the disclosure herein. Obviously, the embodiments described are only a portion of the present invention, not all of it. All other embodiments derived by persons of ordinary skill in the art based on the embodiments of the present invention without inventive effort are intended to fall within the scope of protection of the present invention.

[0041] This study focused on cancers of the digestive organs, namely, the esophagus, stomach, liver, bile duct, pancreas, colon, and rectum. Based on the five-year survival reports provided by the Surveillance, Epidemiology, and End Results Program and Pathology (SEER), the five-year survival rates for these seven digestive cancers are summarized in Table 1 below.

[0042] Table 1. Five-year survival rates for seven cancers of the digestive system

[0043]

[0044] Note: The four stages in Table 1—Localized, Regional, Distant, and Combined—are arranged in a progressive order. Localized refers to the primary lesion; Regional refers to invasion beyond the primary lesion; Distant refers to metastasis; and Combined refers to a mixture of the first three. Localized corresponds to clinical stages I-II; Regional corresponds to clinical stage III; and Distant corresponds to clinical stage IV.

[0045] We compared transcriptome data from 1,835 cancer tissue samples across seven cancer types with 330 control tissue samples, all from the public database "The Cancer Genome Atlas" (TCGA). We investigated various cellular and molecular factors that statistically explain the five-year survival rates reported in the SEER reports for these seven cancer types. We then used the statistical relationships we detected to infer potential factors contributing to the high malignancy of cholangiocarcinoma.

[0046] Due to the complexity and heterogeneity of the tumor microenvironment, changes in the tumor immune microenvironment, and individual differences, the evaluation results vary greatly. In addition, the evaluation process focuses on a single factor and does not weigh multiple factors, making it difficult to fully capture the malignant biological behavior of the tumor. In this study, we focused on the following microenvironmental conditions in cancer tissues: (1) hypoxia level; (2) infiltration level of neutrophils, macrophages, and B cells; and (3) sialic acid accumulation level on the cancer cell surface. A multivariate linear regression model was established to compare the five-year survival rate with the predicted level of the factors in each tissue sample of the seven cancer types mentioned above. Finally, the contribution level of each factor was analyzed based on the Bayesian information criterion. By comparing the contribution levels, we can find out which factors of cholangiocarcinoma jointly lead to the lowest five-year survival rate of cholangiocarcinoma.

[0047] The reasons for considering these factors are as follows: (1) It has long been known that hypoxia is associated with poor prognosis in cancer. Hypoxia is usually caused by local neutrophils and macrophages producing high levels of H2O2 and (1) The presence of sialic acid in the tumor tissues is known to negatively regulate anti-tumor immunity, thereby inhibiting the anti-tumor response. Therefore, the higher the level of B cell infiltration, the worse the prognosis. (2) Since the 1960s, it has been found that cancers with higher levels of sialic acid biosynthesis are more likely to metastasize. Research on sialic acid and cancer metastasis has mainly focused on the signal transduction effect of sialic acid.

[0048] We also considered in our analysis that bile acids produced in the bile duct can activate the G protein-coupled bile acid receptor GPBAR1, which is known to inhibit cell proliferation in cancer cells. This is of great significance in our analysis, which follows the following chain of inferences: (1) All cancer tissues in TCGA have Fenton reaction (abbreviated as FR):

[0049] Fe 2+ +H2O2→Fe 3+ +OH - +·OH (7)

[0050] In the cytoplasm As a reducing molecule, Fe 3+ Reduction back to Fe 2+ , so that the Fenton reaction continues and continuously produces OH - (2) A sustained alkaline environment can affect tumor cells and bring about stress that threatens cell life, driving tumor cells to initiate a variety of acid-producing metabolic reprogramming biological processes to maintain a stable pH value; (3) The two most important acidification metabolic reprogramming biological processes utilized by all cancers in TCGA are de novo nucleotide biosynthesis and sialic acid biosynthesis. For example, the biosynthesis of purines produces 8 to 9 H + , which together have the following properties:

[0051] R(OH - ,FR)=R(H + ,NT)+R(H + ,SA)+ε (8)

[0052] Where R(OH - ,FR) indicates that Fenton reaction in cancer tissue produces OH - The rate, R(H + ,NT) and R(H + , SA) are respectively produced by nucleotide biosynthesis and sialic acid biosynthesis +The rate of H production by all other metabolic reprogramming is ε. + , and R(H + ,NT)+R(H + ,SA), its number is relatively small. From the above formula (8), it can be seen that if R(H + ,NT) cannot be as high as other cancers, so R(H + ,SA) must be high enough so that their sum is equal to OH - The production rates were largely matched to keep the pH stable.

[0053] We estimated the levels of each of the above factors based on the expression levels of each cancer tissue-associated marker gene in all seven cancer types and verified that the five-year survival rate of each cancer type could be well represented as a function of the expression levels of these factors by multiple linear regression analysis with high statistical significance.

[0054] Example 1

[0055] 1 Materials and Methods

[0056] 1.1 Data

[0057] We downloaded transcriptome data of cancer tissues and matched controls from the TCGA database for seven cancer types. Detailed information about the datasets is shown in Table 2.

[0058] Table 2. Number of samples of the seven cancer types used in this study

[0059]

[0060] For each factor analyzed, we used the expression levels of marker genes widely used in the literature (see Table 3): (1) hypoxia levels, (2) infiltration levels of B cells, macrophages, and neutrophils, and (3) sialic acid accumulation levels in seven cancer types (which also involved bile acid levels related to sialic acid, and cell cycle).

[0061] Table 3. Marker genes for tumor microenvironmental conditions

[0062]

[0063]

[0064] 1.2 Batch Effect Analysis

[0065] An empirical Bayesian approach was used to estimate batch effects by modeling the relationship between gene expression levels and batch variables. The estimated batch effects were then subtracted from the original data to ensure that individually collected transcriptome data could be directly compared with each other.

[0066] 1.3 Estimation of sialic acid accumulation on cancer cell surfaces

[0067] O-glycosylated proteins on the cell surface are involved in cancer metastasis. Therefore, the present invention evaluated the accumulation level of sialic acid on O-glycosylated proteins through sialyltransferases (i.e., ST6GALNAC1, ST6GALNAC2, ST6GALNAC3, ST6GALNAC4, and STA8SI6). We noted that only ST6GALNAC1, ST6GALNAC2, and ST6GALNAC4 were upregulated in digestive system cancers and were therefore considered in our analysis. In addition, NEU3 is primarily involved in the degradation of sialic acid. We designed a calculation method to evaluate the average level of sialic acid accumulation in samples of the Localized, Regional, Distant, and Combined stages of cancer types:

[0068]

[0069] Among them, i∈{Localized, Regional, Distant, Combined}, G1={ST6GALNAC1, ST6GALNAC2, ST6GALNAC4}, G2=NEU3, Ave i (G,S) represents the expression of G averaged over all samples S at stage i. The following is defined as the level of sialic acid accumulation assessed up to stage I:

[0070] ΔSA(I)=∑ i≤I ΔSA' i (2)

[0071] Among them, I=Localized represents tumor samples from the primary lesion, I=Regional represents tumor samples that have invaded near the primary lesion, and I=Distant represents metastatic tumor samples. The duration of the three stages is roughly the same.

[0072] 1.4xCell is used to assess the infiltration level of immune cells

[0073] xCell is a widely used computational tool for calculating enrichment scores based on the expression of marker genes for 64 immune and stromal cell types in a given cancer tissue sample. Using the calculated enrichment scores, the infiltration level of B cells, neutrophils, and macrophages in each sample was assessed by evaluating the ratio of each enrichment score to the total enrichment score of the entire tissue sample.

[0074] 1.5 Principal component analysis of gene expression analysis

[0075] Principal component analysis (PCA) was used to capture the main change directions of multiple marker gene sets in terms of overall expression changes.

[0076] 1.6 Evaluating Contribution Based on the Bayesian Information Criterion

[0077] The Bayesian Information Criterion (BIC) is used to select the best model among given options and is defined as follows:

[0078] BIC=k×ln(n)-2 ln(L) (9)

[0079] Where L is the maximum likelihood function of the model, n is the sample size, and k is the number of free parameters in the model.

[0080] Use | BIC k |-|BIC i |To evaluate the effect of the i-th free parameter on the model model k The contribution level of BIC k is to use R 2 The best model for score measurement k The quality of BIC i is the best model based on all parameters except the ith parameter. Therefore, the contribution percentage of the ith parameter can be calculated as follows:

[0081]

[0082] 1.7 Multiple Linear Regression

[0083] Multiple linear regression (MLR) is used to model the linear relationship between a single dependent variable and multiple independent variables. We used BIC-based MLR to linearly regress a range of possible contributing factors on the five-year survival rate for each sample across seven cancer types. BIC provides information about the contribution of each independent variable to the regression outcome.

[0084] 1.8SVM-ROC Evaluation Test

[0085] In the SVM model, by changing the threshold of the decision function, we can determine the true positive rate and false positive rate at the set threshold, thereby providing parameters for ROC calculation. The area under the ROC curve (AUC) is an important indicator for measuring model performance. A higher AUC value indicates a stronger model's evaluation capability.

[0086] 2 Results

[0087] 2.1 Cholangiocarcinoma is the most hypoxic of the seven cancer types

[0088] We used the expression level of HIF1A to reflect the level of hypoxia in tissue samples; the higher the HIF1A expression level, the higher the degree of hypoxia. Figure 1 The expression levels of HIF1A in localized, regional, distant, and combined cancer samples and control samples of seven cancer types are shown respectively. Figure 1 As can be seen in the figure, cholangiocarcinoma has the largest increase in HIF1A expression among cancer samples of all stages compared with control samples, indicating that cholangiocarcinoma is the most hypoxic cancer type among the seven cancer types.

[0089] 2.2 Cholangiocarcinoma has the highest level of B cell infiltration

[0090] We used xCell to assess the infiltration levels of B cells, macrophages, and neutrophils in each tissue sample across control, localized, regional, distant, and combined groups across seven cancer types. Figure 2 The calculation results are shown, from which we can see that cholangiocarcinoma has the highest level of B cell infiltration on average among all seven cancer types.

[0091] To understand why cholangiocarcinoma has the highest concentration of B cell infiltration, pathway enrichment analysis was performed to identify key genes (see Table 3) whose expression was highly correlated with genes involved in sialic acid biosynthesis and attachment. Among the signaling pathways, B cell activation was one of the most relevant activities (see Figure 2 D) Therefore, we found that higher levels of sialic acid attracted more B cells into the tumor area.

[0092] 2.3 Cholangiocarcinoma cells have the highest level of sialic acid accumulation on their cell surface

[0093] As mentioned above, cancer cells utilize de novo nucleotide biosynthesis and sialic acid biosynthesis as their major H + Producers to neutralize the OH produced by the Fenton reaction - . Figure 3 AB shows the role of sialic acid in maintaining OH produced by the Fenton reaction in seven cancer types. - Neutralization contribution weight. We noticed that cholangiocarcinoma stands out in terms of the level of contribution to sialic acid biosynthesis. To understand why sialic acid has a particularly high contribution to maintaining the stability of cancer cell solute pH, we investigated the possible role of bile acids in this process. First, we noted that: (1) bile acid levels measured using expression data of relevant marker genes (Table 3) were closely correlated with sialic acid biosynthesis levels and sialic acid accumulation levels, (2) bile acid levels were negatively correlated with five-year survival rate, as shown in Table 3. Figure 3Figure 3 (C) strongly suggests that bile acids play a role in driving sialic acid biosynthesis and attachment accumulation.

[0094] To elucidate a possible link between the two, we examined the expression levels of the bile acid receptor GPBAR1 and found that it was upregulated, consistent with the elevated bile acid levels in cholangiocarcinoma. Our literature search showed that GPBAR1 can inhibit cell proliferation by activating the cAMP signaling pathway, which is known to be involved in cancer bone metastasis. To test whether this is the case in cholangiocarcinoma, we noted that GPBAR1 expression is associated with cell cycle-related pathways ( Figure 3 D, Table 3) and cell cycle progression levels ( Figure 3 E) showed an inverse correlation, thus indicating that GPBAR1 inhibits cell proliferation in cholangiocarcinoma.

[0095] As mentioned above, de novo nucleotide biosynthesis and sialic acid biosynthesis maintain the OH produced by the Fenton reaction. - Neutralization plays a leading role. Figure 3 The data shown in Figure F explain why sialic acid biosynthesis levels in cholangiocarcinoma stand out among the seven cancer types, as bile acids inhibit cell proliferation and thus nucleotide biosynthesis. Therefore, it is ultimately bile acids that drive elevated sialic acid biosynthesis and accumulation, leading to the high motility of cholangiocarcinoma and, consequently, the lowest survival rate among the seven cancer types.

[0096] 2.4 Key factors contributing to low survival rates in cholangiocarcinoma

[0097] We established a quantitative relationship between the five-year survival rate and the above factors. We performed a multivariate linear regression analysis on the five-year survival rate and five factors: ΔSA, hypoxia level, and infiltration levels of B cells, neutrophils, and macrophages. The following shows four regression models for the Localized (Model 1), Regional (Model 2), Distant (Model 3), and Combined (Model 4) samples. The regression results are as follows: Figure 4 shown.

[0098] Regression model model1:

[0099] SR=1.0158-0.4043ΔSA+0.2528H+0.0434B+0.004M-0.4158N (3)

[0100] Regression model model2:

[0101] SR=1.3789-0.5405ΔSA+0.3069H+0.1469B-0.1133M-0.8068N (4)

[0102] Regression model model3:

[0103] SR=0.3679-0.1086ΔSA-0.0948H+0.136B+0.0665M-0.124N (5)

[0104] Regression model model4:

[0105] SR=1.58-0.3908ΔSA+0.0104H+0.2268B+0.0937M-0.6073N (6)

[0106] Where SR represents the five-year survival rate; ΔSA represents the sialic acid accumulation level; H represents the hypoxia level; B, M, and N represent the infiltration levels of B cells, macrophages, and neutrophils, respectively.

[0107] We provide the contribution level of each factor based on BIC analysis, as shown in Table 4. We conclude that sialic acid accumulation, determined by bile acid levels, plays the most important role in making cholangiocarcinoma the most lethal of the seven cancers studied, from the early cancer stage (model 1) to the final late cancer stage (model 4). B cell infiltration levels are the second-most important factor in the worst survival rate in cholangiocarcinoma, which is partly due to the high sialic acid levels discussed above.

[0108] Table 4. Performance results of regression models

[0109]

[0110] We used SVM-ROC to evaluate the accuracy of the above model. We set 60% as the threshold. If the survival rate is higher than the threshold, it is a low-malignancy tumor, and if the survival rate is lower than the threshold, it is a high-malignancy tumor. We randomly shuffled the samples of each clinical stage, put the model into a part of the samples to evaluate the malignancy of the cancer for training, and then put the model into another part of the samples to test the effect. Figure 5 It can be seen that the model evaluation effect is good in clinical phases I, II, and III+IV.

[0111] Although the present invention has been described in detail above using general descriptions and specific embodiments, it will be apparent to those skilled in the art that modifications and improvements may be made thereto. Therefore, such modifications and improvements, without departing from the spirit of the present invention, are intended to be within the scope of protection claimed herein.

Claims

1. A kit for evaluating the malignancy of digestive system cancer, characterized in that: The invention comprises reagents for detecting the following indicators of tumor samples: sialic acid accumulation level; infiltration level of B cells, macrophages and neutrophils; hypoxia level; The method for detecting the sialic acid accumulation level comprises: The average level of sialic acid accumulation in tumor samples at the Localized, Regional, Distant, and Combined stages was detected using the following formula (1): ， Among them, i∈{Localized, Regional, Distant, Combined}, G1={ ST6GALNAC1 , ST6GALNAC2 , ST6GALNAC4 }, G2= NEU3 , Ave i (G, S) represents the expression of G averaged over all samples S at stage i; The following formula (2) is used to evaluate the level of sialic acid accumulation up to stage I: ， Among them, I = Localized represents tumor samples from the primary lesion, I = Regional represents tumor samples that have invaded near the primary lesion, and I = Distant represents tumor samples that have metastasized.

2. The kit for assessing the malignancy of digestive system cancer according to claim 1, characterized in that: The xCell computational tool was used to calculate the enrichment scores to detect the infiltration levels of B cells, macrophages, and neutrophils in tumor samples.

3. The kit for assessing the malignancy of digestive system cancer according to claim 1, characterized in that: use HIF1A The expression level of TNF-α was used to detect the hypoxia level of tumor samples.

4. The kit for assessing the malignancy of digestive system cancer according to claim 1, characterized in that: The digestive system cancer is bile duct cancer, colon adenocarcinoma, esophageal cancer, hepatocellular carcinoma, pancreatic adenocarcinoma, rectal adenocarcinoma or gastric adenocarcinoma.

5. A method for evaluating the malignancy of digestive system cancer, characterized in that: The method comprises: using a kit according to any one of claims 1 to 4 to detect the sialic acid accumulation level of a tumor sample; the infiltration level of B cells, macrophages and neutrophils; and the hypoxia level, and performing a comprehensive evaluation based on a multivariate linear regression model established with the five-year survival rate.

6. The method for evaluating the malignancy of digestive system cancer according to claim 5, characterized in that: The multiple linear regression model includes: 。 7. The method for evaluating the malignancy of digestive system cancer according to claim 6, characterized in that: Set 60% as the threshold. If the five-year survival rate is higher than the threshold, digestive system cancer is a low-malignancy tumor. If the five-year survival rate is lower than the threshold, digestive system cancer is a high-malignancy tumor.

Images