A method for constructing a prediction model for distant metastasis of nasopharyngeal carcinoma

Abstract

The present application relates to a method for constructing a prediction model for distant metastasis of nasopharyngeal carcinoma, and relates to the field of biomedicine. A method for constructing a prediction model for distant metastasis of nasopharyngeal carcinoma, wherein after collecting clinical nasopharyngeal carcinoma patient information and summarizing test data to obtain multi-dimensional test indicators, five main indicators with significant differences are screened through difference analysis and random forest analysis, and further PCA dimensionality reduction analysis is performed to obtain a principal component calculation formula, thereby obtaining N （PCA） Value; then N obtained by calculation （PCA） The calculation formula of N value is numerically deduced; the optimal calculation formula of N value, namely the prediction formula of distant metastasis of nasopharyngeal carcinoma, is determined by ROC analysis and verified by Spearman analysis; the N value calculated by the prediction formula of distant metastasis of nasopharyngeal carcinoma is output and used to assess the risk of distant metastasis of nasopharyngeal carcinoma.

Description

Technical Field

[0001] The present application relates to the field of biomedicine, and specifically to a method for constructing a prediction model for distant metastasis of nasopharyngeal carcinoma. Background Art

[0002] Nasopharyngeal carcinoma (NPC) is a common head and neck malignancy, often arising from the roof and lateral walls of the nasopharynx. With the advancement and innovation of radiotherapy, surgery, chemotherapy, targeted therapies, and immunotherapy, treatment has become more precise, significantly improving local control and survival rates for NPC patients. However, the current mainstream approach to cancer treatment is to select treatment plans based on relevant guidelines without considering individual factors (such as immune status), resulting in poor prognosis for some patients.

[0003] As cancer research deepens, a growing consensus is emerging that cancer treatments must focus not only on the tumor cells themselves but also on immune status, which plays a crucial role in cancer treatment. In highly immunosuppressive nasopharyngeal carcinoma, immune monitoring plays a crucial role in efficacy evaluation and treatment optimization. Summary of the Invention

[0004] To improve the problem that current mainstream tumor treatments fail to take individual factors (such as immune status) into account, leading to poor prognosis for some patients, this application provides a method for constructing a prediction model for distant metastasis of nasopharyngeal carcinoma, using the following technical solutions:

[0005] A method for constructing a prediction model for distant metastasis of nasopharyngeal carcinoma, comprising:

[0006] (1) Selecting a sample population: The sample population includes nasopharyngeal carcinoma patients with TNM stage II or above, and the sample population is divided into a II-IVA group and an IVB group. The nasopharyngeal carcinoma patients in the II-IVA group have no distant metastasis, while the nasopharyngeal carcinoma patients in the IVB group have distant metastasis;

[0007] (2) Collecting sample data: Counting the age and BMI of the sample population, collecting peripheral blood from the II-IVA group and the IVB group, testing the collected blood samples, and obtaining data on test indicators for the II-IVA group and the IVB group; the test indicators include biochemical indicators and mitochondrial indicators;

[0008] (3) Screening and testing indicators:

[0009] Performing differential analysis on the detection indicators to screen out differential indicators with significant differences between the two groups;

[0010] Perform random forest analysis on the difference indicators to select the top five main indicators with the highest contribution rate;

[0011] (4) Analyze indicator data:

[0012] Performing min-max normalization on the data of the main indicator to obtain normalized data of the main indicator;

[0013] Performing PCA dimensionality reduction analysis on the standardized data of the main indicators to obtain a principal component expression;

[0014] The variance percentage corresponding to the principal component is used as the coefficient of each component, and the sum is N （PCA） Value calculation formula;

[0015] The N value is the distant metastasis index of nasopharyngeal carcinoma;

[0016] (5) Establish the calculation formula:

[0017] The characteristic index of average reduction of Gini coefficient>3 is obtained by random forest analysis in step (3). （PCA） The N value calculation formula obtained by the N value calculation formula determines the N value calculation formula composed of characteristic indicators and BMI;

[0018] Draw N （PCA） The ROC curve of the value calculation formula and the N value calculation formula is calculated, and the AUC value is calculated. When the AUC value of the N value calculation formula is greater than N （PCA） When the AUC value of the formula is calculated, the prediction formula for distant metastasis of nasopharyngeal carcinoma is obtained;

[0019] (6) Verify the calculation formula

[0020] The TNM staging results of the sample population are quantified in the form of integral data to obtain the quantitative staging value;

[0021] The quantitative stage values ​​of patients were compared with N by Spearman nonparametric analysis. （PCA） The correlation between the N value calculated by the value calculation formula and the N value calculation formula was used to verify the prediction formula for distant metastasis of nasopharyngeal carcinoma;

[0022] (7) Evaluate whether the patient's nasopharyngeal carcinoma cells have distant metastasis based on the calculation results of the nasopharyngeal carcinoma distant metastasis prediction formula.

[0023] By adopting the above technical solutions, through the comprehensive acquisition of age, BMI, biochemical indicators and mitochondrial indicators, a comprehensive picture of the patient's health is outlined from the individual's basic characteristics and metabolic status to the cellular microscopic level, providing sufficient data support for in-depth exploration of the factors affecting distant metastasis.

[0024] By combining differential analysis and random forest analysis with screening detection indicators, the main indicators were filtered out, greatly improving the specificity of the model. At the same time, min-max normalization processing and PCA dimensionality reduction analysis can further optimize data quality and model structure, simplify computational complexity, retain the core characteristics of the data, and enhance the generalization ability of the computational model.

[0025] Through N （PCA） The ROC curve and AUC values ​​were compared and optimized using the stage calculation formula and the N value calculation formula to ensure that the prediction formula for nasopharyngeal carcinoma distant metastasis has accurate predictive efficacy and can effectively distinguish patients with high and low risks of distant metastasis. The Spearman nonparametric analysis quantified the correlation between the stage values ​​and the predicted values, verifying the reliability of the formula from a clinical perspective. This allows the prediction formula to be effectively applied in clinical practice, identify the risk of distant metastasis in advance, and develop personalized treatment plans for patients, thereby improving the treatment effect and quality of life of nasopharyngeal carcinoma patients.

[0026] Preferably, the number of difference indicators screened in step (3) is 18, and the 18 difference indicators are:

[0027] Resting Treg: resting regulatory T cells, the percentage of Treg cells in an unactivated state;

[0028] T4Tef.MM: effector helper T cell mitochondrial mass value;

[0029] CK: creatine kinase activity;

[0030] LDH: lactate dehydrogenase activity;

[0031] FIB: fibrinogen concentration;

[0032] T8Tn.MMP: Initial CD8 ＋ Percentage of T cell mitochondria with low membrane potential;

[0033] T8Tcm.MMP: percentage of mitochondrial low membrane potential in central memory killer T cells;

[0034] Na + : sodium ion concentration value;

[0035] MONO: percentage of monocytes;

[0036] PDW: platelet distribution width;

[0037] CD8 + .MMP: percentage of low mitochondrial membrane potential in killer T cells;

[0038] RBC: red blood cell concentration;

[0039] HGB: hemoglobin concentration;

[0040] CD4 + : percentage of helper T cells;

[0041] MCHC: mean RBC hemoglobin concentration;

[0042] CD4 + .MM: percentage of mitochondrial low membrane potential in helper T cells;

[0043] RDW-CV: red blood cell distribution density CV value;

[0044] RDW-SD: red blood cell distribution density SD value.

[0045] Preferably, the main indicators screened in step (3) are: T4Tef.MM, LDH, Resting Treg, CK and FIB.

[0046] Preferably, there are 5 principal component expressions in step (4), and the 5 principal component expressions are:

[0047] in,

[0048] in, x 1 - x 5 These are the normalized values ​​of Resting Treg, T4Tef.MM, CK, LDH, and FIB;

[0049] N in step (4) （PCA） The value calculation formula is:

[0050] .

[0051] By adopting the above technical solution, 5 indicators are obtained through screening. The PCA model constructed based on these 5 main indicators naturally generates 5 principal components, which completely retains the core characteristic information of the original data. Since the number of principal components is moderate (only 5), there is no need to further screen the cumulative variance percentage to simplify the calculation, which ensures the simplicity and accuracy of the formula calculation.

[0052] Preferably, in step (5), the number of characteristic indicators for which the average reduction of the Gini coefficient is greater than 3 is 8, and the 8 characteristic indicators are: T4Tef.MM, LDH, Resting Treg, CK, FIB, RBC, T8Tn.MMP and T8Tcm.MMP.

[0053] Preferably, the number of N value calculation formulas established in step (5) is 4, and the 4 N value calculation formulas are:

[0054]

[0055] Preferably, the prediction formula for distant metastasis of nasopharyngeal carcinoma is:

[0056] Any one of the following;

[0057] in,

[0058] Resting Treg is the percentage of resting regulatory T cells, i.e., Treg cells in an unactivated state;

[0059] T4Tef.MM is the mitochondrial mass value of effector helper T cells;

[0060] CK is creatine kinase activity;

[0061] RBC is the red blood cell concentration;

[0062] LDH is lactate dehydrogenase activity;

[0063] FIB is fibrinogen concentration;

[0064] T8Tn.MMP is also T8Tn.MMP low (%) is the abbreviation of initial CD8 ＋ Percentage of T cell mitochondria with low membrane potential;

[0065] T8Tcm.MMP is also T8Tcm.MMP low (%) is the abbreviation of the low membrane potential percentage of mitochondria in central memory killer T cells;

[0066] BMI is body mass index.

[0067] By adopting the above technical solutions, the metabolism of immune cells can effectively reflect the immune status. Mitochondria are the main metabolic regulators of T cells. Mitochondrial metabolism can change the fate and function of lymphocytes. This application combines mitochondrial indicators (T4Tef.MM, T8Tn.MMP and T8Tcm.MMP) with immunological indicators (Resting Treg), blood routine indicators (RBC), biochemical indicators (CK, LDH and FIB) and BMI to construct a prediction formula for nasopharyngeal carcinoma distant metastasis, realizing the organic integration of multi-dimensional data:

[0068] Mitochondria are the core hub of T cell metabolism. T4Tef.MM, T8Tn.MMP, and T8Tcm.MMP can accurately capture the metabolic activity and functional status of T cells.

[0069] Resting Treg directly reflects the balance of the body's immune regulation from an immunological perspective;

[0070] The concentration of red blood cells (RBC) directly affects the body's oxygen-carrying capacity, and tumor growth and metastasis are highly dependent on adequate oxygen supply;

[0071] Abnormal increases in creatine kinase (CK) and lactate dehydrogenase (LDH) activity often reflect cellular metabolic disorders and tissue damage, and are closely related to the invasion process of tumors.

[0072] Changes in fibrinogen concentration (FIB) in the tumor microenvironment can promote tumor cell adhesion, migration, and distant colonization by affecting the coagulation system;

[0073] Body mass index (BMI) is an important indicator for measuring an individual's nutritional and metabolic status. Its abnormality is also potentially associated with the occurrence and development of tumors. The chronic inflammatory microenvironment of the body in an obese state may accelerate the distant metastasis of tumor cells.

[0074] The method for constructing a nasopharyngeal carcinoma distant metastasis prediction model in this application can comprehensively evaluate the immune status, body metabolic level and tumor microenvironment of nasopharyngeal carcinoma patients from a holistic level through a comprehensive analysis of multi-dimensional indicators. It breaks through the limitations of traditional single indicator prediction and significantly improves the model's prediction accuracy for nasopharyngeal carcinoma distant metastasis.

[0075] Preferably, N （PCA） The AUC value of the ROC curve is 78.11, the AUC value of the ROC curve of N1 is 91.53, the AUC value of the ROC curve of N2 is 86.68, the AUC value of the ROC curve of N3 is 91.38, and the AUC value of the ROC curve of N4 is 92.02.

[0076] Preferably, the prediction formula for distant metastasis of nasopharyngeal carcinoma is:

[0077] .

[0078] By adopting the above technical solution, the AUC value of the ROC curve of N4 is greater than that of N （PCA） The value calculation formula and the AUC value of N1-N3 indicate that N4 has the highest accuracy in predicting distant metastasis of nasopharyngeal carcinoma, so N4 is the preferred choice.

[0079] Preferably, the N value of the nasopharyngeal carcinoma patients in the II-IVA group is 0.19-0.29, and the N value of the nasopharyngeal carcinoma patients in the IVB group is 0.11-0.16.

[0080] By adopting the above technical solution, the data of the patient's relevant indicators are substituted into the formula for calculation. When the patient's N value is within the range of 0.19-0.29, it can be considered that the patient has a high probability of no distant metastasis; when the patient's N value is within the range of 0.11-0.16, it can be considered that the patient has a high probability of distant metastasis.

[0081] In summary, this application has the following beneficial effects:

[0082] 1. This application combines differential analysis and random forest analysis to screen detection indicators and filter out the main indicators, greatly improving the specificity of the model. At the same time, min-max normalization processing and PCA dimensionality reduction analysis further simplify the computational complexity while retaining the core features of the data and improving the generalization ability of the model.

[0083] 2. The proposed formula for predicting distant metastasis of nasopharyngeal carcinoma uses a comprehensive analysis of multi-dimensional indicators to comprehensively assess the immune status, metabolic level, and tumor microenvironment of nasopharyngeal carcinoma patients. This formula breaks through the limitations of traditional single-indicator prediction and significantly improves the accuracy of the formula for predicting distant metastasis of nasopharyngeal carcinoma.

[0084] 3. This application simplifies N by establishing the N value calculation formula （PCA） The calculation process of the ROC value and the comparative optimization of the ROC curve and AUC value ensure that the constructed nasopharyngeal carcinoma distant metastasis prediction model has accurate prediction effectiveness and can effectively distinguish between high-risk and low-risk patients with distant metastasis. BRIEF DESCRIPTION OF THE DRAWINGS

[0085] Figure 1 Flowchart of the method for constructing a prediction model for distant metastasis of nasopharyngeal carcinoma in this application.

[0086] Figure 2 This is the random forest analysis curve in this application Figure 1 .

[0087] Figure 3 This is the random forest analysis curve in this application Figure 2 .

[0088] Figure 4 N in this application (PCA) ROC curve of the N value obtained by the value calculation formula and the N value formula.

[0089] Figure 5 This is a correlation analysis chart between the quantitative stage values ​​in this application and the N value calculated using the N value formula (N1).

[0090] Figure 6 This is a correlation analysis chart of the quantitative stage values ​​in this application and the N value calculated using the N value calculation formula (N2).

[0091] Figure 7 This is a correlation analysis chart of the quantitative stage values ​​in this application and the N value calculated using the N value calculation formula (N3).

[0092] Figure 8 This is a correlation analysis chart of the quantitative stage values ​​in this application and the N value calculated using the N value calculation formula (N4).

[0093] Figure 9 The quantitative stage value and N in this application (PCA) Correlation analysis diagram of N values ​​obtained by calculating the value calculation formula. DETAILED DESCRIPTION

[0094] The present application is further described in detail below with reference to the embodiments.

[0095] A method for constructing a prediction model for distant metastasis of nasopharyngeal carcinoma, referring to Figure 1 ,include:

[0096] (1) Selecting the sample population

[0097] The sample population was pathologically diagnosed with nasopharyngeal carcinoma, which was undifferentiated or poorly differentiated squamous cell carcinoma, with a single primary tumor type. Based on the updated version of the cancer staging system published by the American Joint Committee on Cancer in 2017, the TNM staging system was standardized to classify the sample population as stage II or above, and the patients were prepared for radical intensity-modulated radiation therapy at least one week later.

[0098] Patients with pregnancy, autoimmune diseases, infections, allergies and other immune-related diseases, and other cancers were excluded from the sample population;

[0099] In the sample population, patients with TNM stage II and III disease had no distant metastases, while some patients with TNM stage IV disease had distant metastases. Therefore, patients with TNM stage IV disease were divided into the IVA group and the IVB group. The IVA group consisted of patients with TNM stage IV disease but no distant metastases, while the IVB group consisted of patients with distant metastases. Patients with TNM stage II and III disease and those in the IVA group were combined into the II-IVA group (control group), while patients with distant metastases were assigned to the IVB group (metastasis group). There were 140 patients in the control group and 48 patients in the metastasis group.

[0100] (2) Collecting sample data

[0101] The age and body mass index (BMI) of the IVB group and the II-IVA group were statistically analyzed. The age of the patients in the IVB group was 50.46±10.96 years, and the BMI was 23.14±2.94; the age of the patients in the II-IVA group was 50.42±10.48 years, and the BMI was 23.39±3.41.

[0102] Peripheral blood was collected from the IVB group and the II-IVA group. The collected blood samples were subjected to biochemical tests and flow cytometry tests. Data of 119 test indicators were obtained for the IVB group and the II-IVA group, respectively. The specific test indicators and test data are shown in Table 1.

[0103] Table 1. Test indicators and test results of the quartile range

[0104]

[0105] Note: Abs.count indicates absolute count index; the index in brackets is the full English name of the index and the index unit; MM is the mitochondrial mass index, which is a relative index; MMP is the percentage of mitochondrial low membrane potential, which is MMP low Abbreviation for %.

[0106] (3) Screening and testing indicators

[0107] The 119 detection indicators were analyzed for differences using the t-test method in SPSS software, and 18 difference indicators with significant differences between the two groups were screened out. The screening results are shown in Table 2.

[0108] Table 2 Screening results of differential analysis of detection indicators

[0109]

[0110] Referring to Table 2, the 18 difference indicators are:

[0111] Resting Treg: resting regulatory T cells, the percentage of Treg cells in an unactivated state;

[0112] T4Tef.MM: effector helper T cell mitochondrial mass value;

[0113] CK: creatine kinase;

[0114] LDH: lactate dehydrogenase;

[0115] FIB: fibrinogen concentration;

[0116] T8Tn.MMP: Initial CD8 ＋ T cell mitochondrial low membrane potential percentage;

[0117] T8Tcm.MMP: percentage of mitochondrial low membrane potential in central memory killer T cells;

[0118] Na + : sodium ion concentration value;

[0119] MONO: percentage of monocytes;

[0120] PDW: platelet distribution width;

[0121] CD8 + .MMP: percentage of low mitochondrial membrane potential in killer T cells;

[0122] RBC: red blood cells;

[0123] HGB: hemoglobin;

[0124] CD4 + : percentage of helper T cells;

[0125] MCHC: mean RBC hemoglobin concentration;

[0126] CD4 + .MMP: percentage of low mitochondrial membrane potential in helper T cells;

[0127] RDW-CV: red blood cell distribution density CV value;

[0128] RDW-SD: red blood cell distribution density SD value.

[0129] The above 18 difference indicators were subjected to random forest analysis, with the number of decision trees set to 6000, the cross-validation fold to 10, and the variable reduction amplitude to 1.1. The analysis results are shown in Table 3.

[0130] Table 3 Random forest analysis results of difference index

[0131]

[0132] According to the above random forest analysis results, the random forest curve is obtained (refer to Figure 2-Figure 3 ), the five main indicators with the highest contribution rates were screened out, which are:

[0133] T4Tef.MM: effector helper T cell mitochondrial mass value;

[0134] LDH: lactate dehydrogenase activity;

[0135] Resting Treg: The percentage of resting regulatory T cells, i.e., Treg cells in an unactivated state;

[0136] CK: creatine kinase activity;

[0137] FIB: fibrinogen concentration.

[0138] (4) Analyze indicator data:

[0139] The data of the five main indicators were processed by min-max standardization using SPSS software to obtain the standardized data of the main indicators and eliminate the influence of unit differences, including the following steps:

[0140] S1. Import the data of the five main indicators into SPSS software and statistically output the maximum and minimum values ​​of each main indicator;

[0141] S2. The calculation formula for the standardized data of the main indicator is:

[0142] (Current main indicator data - minimum value) / (maximum value - minimum value).

[0143] For example, the maximum value of Resting Treg in the sample population is 6.94% and the minimum value is 0.19%. When the Resting Treg value of a random sample is 2.35%, the standardized data of Resting Treg in the sample is (2.35-0.19) / (6.94-0.19)=0.3200.

[0144] According to the sample population, the maximum value of Resting Treg was 6.94% and the minimum was 0.19%; the maximum value of T4Tef.MM was 14.77 and the minimum was 0.43; the maximum value of CK was 409.5 U / L and the minimum was 24.3 U / L; the maximum value of LDH was 822.40 U / L and the minimum was 104.3 U / L; the maximum value of FIB was 8.04 g / L and the minimum was 0.94 g / L. The standardized data of the main indicators of the sample population were calculated as follows:

[0145] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S001 (Group II-IVA) were 0.3200, 0.2801, 0.1698, 0.0978, and 0.3225, respectively;

[0146] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S002 (Group II-IVA) were 0.0696, 0.3988, 0.0852, 0.0253, and 0.3366, respectively;

[0147] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S003 (Group II-IVA) were 0.2800, 0.7220, 0.0511, 0.0454, and 0.2127, respectively;

[0148] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S004 (Group II-IVA) were 0.2430, 0.1946, 0.0701, 0.0453, and 0.2662, respectively;

[0149] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S005 (Group II-IVA) were 0.0133, 0.1230, 0.1620, 0.0802, and 0.3704, respectively;

[0150] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S006 (Group II-IVA) were 0.3585, 0.5270, 0.0727, 0.0387, and 0.2113, respectively;

[0151] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S007 (Group II-IVA) were 0.1881, 0.7116, 0.0896, 0.0557, and 0.2042, respectively;

[0152] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S008 (Group II-IVA) were 0.2178, 0.4846, 0.0389, 0.0295, and 0.4282, respectively;

[0153] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S009 (Group II-IVA) were 0.1259, 0.5331, 0.0997, 0.0503, and 0.2113, respectively;

[0154] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S010 (Group II-IVA) were 0.1259, 0.5331, 0.0997, 0.0503, and 0.2113, respectively;

[0155] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S011 (Group II-IVA) were 0.2978, 0.2395, 0.1038, 0.1037, and 0.2859, respectively;

[0156] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S012 (Group II-IVA) were 0.3852, 0.5065, 0.1646, 0.0492, and 0.1352, respectively;

[0157] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S013 (Group II-IVA) were 0.2815, 0.2273, 0.3712, 0.0734, and 0.1901, respectively;

[0158] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S014 (Group II-IVA) were 0.0681, 0.3514, 0.1809, 0.0856, and 0.2225, respectively;

[0159] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S015 (Group II-IVA) were 0.0904, 0.2709, 0.2420, 0.1218, and 0.3394, respectively;

[0160] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S016 (Group II-IVA) were 0.2756, 0.8326, 0.1046, 0.0709, and 0.2169, respectively;

[0161] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S017 (Group II-IVA) were 0.2830, 0.3027, 0.2897, 0.0446, and 0.2493, respectively;

[0162] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S018 (Group II-IVA) were 0.2474, 0.3241, 0.2139, 0.1347, and 0.4183, respectively;

[0163] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S019 (Group II-IVA) were 0.8148, 0.4622, 0.1976, 0.1408, and 0.2352, respectively;

[0164] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S020 (Group II-IVA) were 0.1733, 0.2895, 0.0999, 0.0539, and 0.3859, respectively;

[0165] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S021 (Group II-IVA) were 0.1111, 0.1791, 0.2139, 0.0639, and 0.1634, respectively;

[0166] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S022 (Group II-IVA) were 0.0919, 0.1625, 0.1882, 0.0465, and 0.2282, respectively;

[0167] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S023 (Group II-IVA) were 0.0859, 0.3019, 0.6103, 0.0994, and 0.3606, respectively;

[0168] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S024 (Group II-IVA) were 0.0370, 0.1015, 0.1576, 0.0361, and 0.2423, respectively;

[0169] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S025 (Group II-IVA) were 0.3319, 0.0919, 0.0480, 0.0099, and 0.2197, respectively;

[0170] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S026 (Group II-IVA) were 0.2178, 0.4727, 0.0844, 0.0597, and 0.5254, respectively;

[0171] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S027 (Group II-IVA) were 0.2741, 0.1863, 0.1319, 0.0320, and 0.3592, respectively;

[0172] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S028 (Group II-IVA) were 0.4474, 0.0230, 0.2342, 0.1912, and 0.7986, respectively;

[0173] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S029 (Group II-IVA) were 0.0000, 0.1103, 0.1703, 0.0769, and 0.5718, respectively;

[0174] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S030 (Group II-IVA) were 0.1704, 0.1739, 0.2303, 0.0574, and 0.3070, respectively;

[0175] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S031 (Group II-IVA) were 0.0800, 0.2327, 0.1386, 0.1173, and 0.2521, respectively;

[0176] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S032 (Group II-IVA) were 0.3007, 1.0000, 0.1636, 0.1043, and 0.4465, respectively;

[0177] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S033 (Group II-IVA) were 0.4370, 0.2921, 0.0498, 0.0024, and 0.3746, respectively;

[0178] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S034 (Group II-IVA) were 0.6637, 0.3751, 0.0584, 0.0316, and 0.4887, respectively;

[0179] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S035 (Group II-IVA) were 0.5896, 0.9400, 0.0550, 0.0182, and 0.3451, respectively;

[0180] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S036 (Group II-IVA) were 0.1052, 0.4808, 0.1911, 0.1224, and 0.3972, respectively;

[0181] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S037 (Group II-IVA) were 0.0430, 0.0049, 0.1542, 0.2128, and 0.3268, respectively;

[0182] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S038 (Group II-IVA) were 0.0889, 0.1262, 0.1931, 0.0756, and 0.1803, respectively;

[0183] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S039 (Group II-IVA) were 0.2696, 0.3491, 0.0914, 0.0649, and 0.1958, respectively;

[0184] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S040 (Group II-IVA) were 0.2326, 0.2648, 0.1695, 0.1127, and 0.5493, respectively;

[0185] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S041 (Group II-IVA) were 0.1985, 0.2173, 0.1038, 0.0274, and 0.3465, respectively;

[0186] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S042 (Group II-IVA) were 0.1511, 0.0351, 0.2298, 0.1115, and 0.5789, respectively;

[0187] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S043 (Group II-IVA) were 0.6815, 0.4788, 0.1591, 0.0361, and 0.1620, respectively;

[0188] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S044 (Group II-IVA) were 0.1022, 0.0000, 0.8009, 0.1224, and 0.3211, respectively;

[0189] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S045 (Group II-IVA) were 0.0800, 0.5186, 0.0506, 0.1050, and 0.2704, respectively;

[0190] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S046 (Group II-IVA) were 0.2341, 0.1466, 0.1950, 0.0914, and 0.3282, respectively;

[0191] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S047 (Group II-IVA) were 0.2385, 0.5441, 0.1355, 0.0655, and 0.2225, respectively;

[0192] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S048 (Group II-IVA) were 0.1304, 0.2688, 0.1602, 0.0000, and 0.2282, respectively;

[0193] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S049 (Group II-IVA) were 0.0770, 0.0962, 0.2368, 0.1218, and 0.3465, respectively;

[0194] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S050 (Group II-IVA) were 0.3585, 0.3193, 0.1472, 0.0689, and 0.3070, respectively;

[0195] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S051 (Group II-IVA) were 0.4104, 0.5108, 0.0314, 0.0597, and 0.2662, respectively;

[0196] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S052 (Group II-IVA) were 0.2356, 0.2860, 0.5330, 0.1390, and 0.4183, respectively;

[0197] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S053 (Group II-IVA) were 0.2044, 0.2242, 0.2534, 0.0632, and 0.2704, respectively;

[0198] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S054 (Group II-IVA) were 0.1911, 0.2919, 0.2461, 0.1125, and 0.2056, respectively;

[0199] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S055 (Group II-IVA) were 0.1407, 0.6425, 0.2009, 0.0703, and 0.2352, respectively;

[0200] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S056 (Group II-IVA) were 0.0622, 0.2831, 0.2404, 0.0588, and 0.5366, respectively;

[0201] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S057 (Group II-IVA) were 0.3630, 0.1637, 0.1316, 0.1152, and 0.2282, respectively;

[0202] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S058 (Group II-IVA) were 0.2207, 0.2478, 0.1976, 0.0911, and 0.2620, respectively;

[0203] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S059 (Group II-IVA) were 0.3837, 0.1572, 1.0000, 0.1749, and 0.5042, respectively;

[0204] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S060 (Group II-IVA) were 0.2267, 0.5722, 0.1280, 0.1014, and 0.3732, respectively;

[0205] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S061 (Group II-IVA) were 0.6578, 0.3546, 0.2124, 0.0951, and 0.1366, respectively;

[0206] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S062 (Group II-IVA) were 0.2504, 0.1221, 0.1944, 0.1327, and 0.3380, respectively;

[0207] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S063 (Group II-IVA) were 0.2000, 0.1454, 0.2721, 0.1383, and 0.3310, respectively;

[0208] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S064 (Group II-IVA) were 0.1274, 0.2891, 0.0636, 0.0734, and 0.4563, respectively;

[0209] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S065 (Group II-IVA) were 0.5985, 0.2384, 0.2165, 0.1053, and 0.3563, respectively;

[0210] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S066 (Group II-IVA) were 0.6681, 0.3177, 0.1007, 0.1337, and 0.1859, respectively;

[0211] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S067 (Group II-IVA) were 0.2711, 0.1490, 0.0698, 0.0312, and 0.4831, respectively;

[0212] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S068 (Group II-IVA) were 0.2370, 0.2858, 0.1233, 0.1330, and 0.4690, respectively;

[0213] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S069 (Group II-IVA) were 0.0948, 0.3033, 0.2648, 0.1074, and 0.2986, respectively;

[0214] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S070 (Group II-IVA) were 0.1570, 0.2490, 0.0607, 0.1078, and 0.3070, respectively;

[0215] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S071 (Group II-IVA) were 0.2341, 0.6809, 0.1685, 0.1432, and 0.5507, respectively;

[0216] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S072 (Group II-IVA) were 0.0622, 0.4256, 0.1472, 0.1707, and 0.3028, respectively;

[0217] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S073 (Group II-IVA) were 0.0652, 0.2229, 0.2142, 0.0617, and 1.0000, respectively;

[0218] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S074 (Group II-IVA) were 0.1970, 0.2509, 0.0958, 0.0823, and 0.5282, respectively;

[0219] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S075 (Group II-IVA) were 0.2667, 0.3893, 0.1342, 0.0664, and 0.4028, respectively;

[0220] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S076 (Group II-IVA) were 0.0074, 0.0517, 0.3518, 0.1334, and 0.3493, respectively;

[0221] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S077 (Group II-IVA) were 0.0415, 0.1705, 0.1205, 0.0813, and 0.4901, respectively;

[0222] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S078 (Group II-IVA) were 0.6385, 0.2503, 0.1687, 0.0678, and 0.2211, respectively;

[0223] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S079 (Group II-IVA) were 0.0415, 0.3580, 0.0161, 0.0944, and 0.4465, respectively;

[0224] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S080 (Group II-IVA) were 0.0563, 0.2406, 0.1771, 0.0691, and 0.4324, respectively;

[0225] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S081 (Group II-IVA) were 0.4459, 0.3697, 0.2095, 0.0790, and 0.1606, respectively;

[0226] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S082 (Group II-IVA) were 0.0741, 0.0016, 0.0151, 0.0230, and 0.6507, respectively;

[0227] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S083 (Group II-IVA) were 0.0459, 0.0237, 0.3030, 0.1419, and 0.2099, respectively;

[0228] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S084 (Group II-IVA) were 0.0933, 0.0217, 0.0013, 0.0414, and 0.7620, respectively;

[0229] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S085 (Group II-IVA) were 0.0904, 0.0138, 0.2056, 0.1107, and 0.5000, respectively;

[0230] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S086 (Group II-IVA) were 0.0904, 0.0985, 0.0989, 0.0416, and 0.2155, respectively;

[0231] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S087 (Group II-IVA) were 0.2489, 0.1187, 0.1885, 0.0720, and 0.1493, respectively;

[0232] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S088 (Group II-IVA) were 0.1867, 0.6767, 0.1462, 0.1441, and 0.3901, respectively;

[0233] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S089 (Group II-IVA) were 0.3926, 0.1462, 0.2181, 0.1022, and 0.2155, respectively;

[0234] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S090 (Group II-IVA) were 0.6044, 0.4799, 0.0937, 0.1341, and 0.2930, respectively;

[0235] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S091 (Group II-IVA) were 0.1511, 0.3920, 0.1109, 0.1047, and 0.4535, respectively;

[0236] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S092 (Group II-IVA) were 0.4963, 0.2890, 0.0304, 0.0092, and 0.4972, respectively;

[0237] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S093 (Group II-IVA) were 0.3526, 0.6781, 0.1194, 0.0571, and 0.3592, respectively;

[0238] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S094 (Group II-IVA) were 0.3156, 0.7598, 0.1134, 0.0515, and 0.2099, respectively;

[0239] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S095 (Group II-IVA) were 0.1970, 0.5034, 0.1558, 0.0936, and 0.4690, respectively;

[0240] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S096 (Group II-IVA) were 0.1007, 0.3274, 0.2090, 0.0476, and 0.3465, respectively;

[0241] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S097 (Group II-IVA) were 0.2963, 0.1756, 0.0449, 0.0423, and 0.4620, respectively;

[0242] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S098 (Group II-IVA) were 0.3348, 0.3233, 0.3603, 0.1370, and 0.2620, respectively;

[0243] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S099 (Group II-IVA) were 0.0430, 0.0014, 0.2461, 0.0930, and 0.4718, respectively;

[0244] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample number S100 (group II-IVA) were 0.1793, 0.4968, 0.0867, 0.0524, and 0.3831, respectively;

[0245] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample number S101 (group II-IVA) were 0.1541, 0.0134, 0.2728, 0.0408, and 0.3366, respectively;

[0246] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S102 (Group II-IVA) were 0.4059, 0.2107, 0.0592, 0.0096, and 0.4972, respectively;

[0247] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S103 (Group II-IVA) were 0.1748, 0.2431, 0.1415, 0.0712, and 0.5775, respectively;

[0248] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S104 (Group II-IVA) were 0.3096, 0.2070, 0.8583, 0.1730, and 0.1577, respectively;

[0249] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample number S105 (group II-IVA) were 0.2400, 0.2334, 0.0781, 0.0829, and 0.5042, respectively;

[0250] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S106 (Group II-IVA) were 0.4281, 0.4009, 0.0981, 0.1149, and 0.5127, respectively;

[0251] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S107 (Group II-IVA) were 0.0089, 0.1905, 0.5148, 0.1503, and 0.4887, respectively;

[0252] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S108 (Group II-IVA) were 0.1911, 0.6595, 0.0854, 0.0787, and 0.4901, respectively;

[0253] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample number S109 (group II-IVA) were 0.3363, 0.3584, 0.0408, 0.0870, and 0.4062, respectively;

[0254] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample number S110 (group II-IVA) were 0.2741, 0.2586, 0.1488, 0.1270, and 0.7113, respectively;

[0255] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample number S111 (group II-IVA) were 0.1630, 0.1960, 0.1877, 0.1103, and 0.3085, respectively;

[0256] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S112 (Group II-IVA) were 0.0415, 0.5215, 0.0802, 0.1319, and 0.3197, respectively;

[0257] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample number S113 (group II-IVA) were 0.4163, 0.2844, 0.2420, 0.0503, and 0.7606, respectively;

[0258] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample number S114 (group II-IVA) were 0.2267, 0.4877, 0.0898, 0.0813, and 0.3183, respectively;

[0259] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample number S115 (group II-IVA) were 0.1467, 0.0359, 0.4561, 0.0806, and 0.3563, respectively;

[0260] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample number S116 (group II-IVA) were 0.3437, 0.2578, 0.3118, 0.1292, and 0.2775, respectively;

[0261] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S117 (Group II-IVA) were 0.3541, 0.4947, 0.0979, 0.0894, and 0.4254, respectively;

[0262] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample number S118 (group II-IVA) were 0.4044, 0.5402, 0.2832, 0.1366, and 0.3268, respectively;

[0263] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample number S119 (group II-IVA) were 0.4089, 0.6773, 0.0000, 0.0766, and 0.4944, respectively;

[0264] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample number S120 (group II-IVA) were 0.4711, 0.1901, 0.1018, 0.0703, and 0.4620, respectively;

[0265] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample number S121 (group II-IVA) were 0.1156, 0.2254, 0.2038, 0.0886, and 0.2099, respectively;

[0266] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S122 (Group II-IVA) were 0.7274, 0.2102, 0.1607, 0.1633, and 0.7014, respectively;

[0267] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample number S123 (group II-IVA) were 0.1600, 0.3851, 0.1363, 0.1156, and 0.4183, respectively;

[0268] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample number S124 (group II-IVA) were 0.2519, 0.6910, 0.0805, 0.1217, and 0.3817, respectively;

[0269] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample number S125 (group II-IVA) were 1.0000, 0.1720, 0.0527, 0.0929, and 0.2169, respectively;

[0270] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample number S126 (group II-IVA) were 0.6119, 0.0614, 0.0532, 0.0522, and 0.5761, respectively;

[0271] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample number S127 (group II-IVA) were 0.1956, 0.3837, 0.0781, 0.0791, and 0.4352, respectively;

[0272] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample number S128 (group II-IVA) were 0.2726, 0.3397, 0.2290, 0.1991, and 0.3479, respectively;

[0273] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample number S129 (group II-IVA) were 0.0504, 0.4191, 0.1423, 0.0549, and 0.9690, respectively;

[0274] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample number S130 (group II-IVA) were 0.1674, 0.3551, 0.1664, 0.1799, and 0.7028, respectively;

[0275] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S131 (Group II-IVA) were 0.3822, 0.1671, 0.0768, 0.0309, and 0.4408, respectively;

[0276] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S132 (Group II-IVA) were 0.4119, 0.2686, 0.2051, 0.1185, and 0.3732, respectively;

[0277] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S133 (Group II-IVA) were 0.1496, 0.0864, ​​0.2448, 0.1067, and 0.3028, respectively;

[0278] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample number S134 (group II-IVA) were 0.2430, 0.4145, 0.1976, 0.0911, and 0.2620, respectively;

[0279] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample number S135 (group II-IVA) were 0.5615, 0.1937, 0.0810, 0.0783, and 0.2423, respectively;

[0280] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample number S136 (group II-IVA) were 0.2430, 0.4669, 0.1134, 0.0313, and 0.2352, respectively;

[0281] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S137 (Group II-IVA) were 0.3185, 0.1804, 0.1249, 0.1239, and 0.2803, respectively;

[0282] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample number S138 (group II-IVA) were 0.5644, 0.2432, 0.1293, 0.0934, and 0.5155, respectively;

[0283] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample number S139 (group II-IVA) were 0.4163, 0.8106, 0.1436, 0.0649, and 0.3859, respectively;

[0284] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample number S140 (group II-IVA) were 0.2993, 0.4072, 0.1547, 0.0698, and 0.3028, respectively;

[0285] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S141 (IVB group) were 0.0148, 0.0964, 0.2534, 0.3225, and 0.6507, respectively;

[0286] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S142 (IVB group) were 0.0904, 0.0589, 0.1386, 0.0961, and 0.2620, respectively;

[0287] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S143 (IVB group) were 0.1496, 0.0145, 0.0750, 0.0705, and 0.2662, respectively;

[0288] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S144 (IVB group) were 0.0474, 0.0905, 0.1711, 0.0816, and 0.4620, respectively;

[0289] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S145 (IVB group) were 0.1526, 0.1317, 0.0974, 0.0656, and 0.5493, respectively;

[0290] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S146 (IVB group) were 0.0148, 0.3037, 0.2889, 0.1301, and 0.3366, respectively;

[0291] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S147 (IVB group) were 0.3126, 0.4201, 0.0195, 0.0248, and 0.5690, respectively;

[0292] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S148 (IVB group) were 0.2919, 0.4661, 0.0496, 0.1486, and 0.9676, respectively;

[0293] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S149 (IVB group) were 0.2933, 0.2039, 0.1786, 0.1143, and 0.5845, respectively;

[0294] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample number S150 (IVB group) were 0.4133, 0.2090, 0.1565, 0.3891, and 0.9183, respectively;

[0295] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S151 (IVB group) were 0.2593, 0.1169, 0.0890, 0.8681, and 0.3944, respectively;

[0296] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S152 (IVB group) were 0.1585, 0.0679, 0.1269, 0.2253, and 0.3901, respectively;

[0297] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S153 (IVB group) were 0.0978, 0.2346, 0.0589, 0.0585, and 0.5507, respectively;

[0298] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S154 (IVB group) were 0.0400, 0.1418, 0.1044, 0.3616, and 0.5817, respectively;

[0299] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S155 (IVB group) were 0.1393, 0.2618, 0.1952, 0.1147, and 0.6085, respectively;

[0300] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S156 (IVB group) were 0.0681, 0.1042, 0.0890, 0.8681, and 0.3944, respectively;

[0301] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S157 (IVB group) were 0.0489, 0.1209, 0.2056, 0.1021, and 0.7310, respectively;

[0302] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S158 (IVB group) were 0.0430, 0.1370, 0.0465, 0.0737, and 0.3930, respectively;

[0303] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S159 (IVB group) were 0.0993, 0.0519, 0.0623, 0.1518, and 0.2704, respectively;

[0304] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample number S160 (IVB group) were 0.1096, 0.0749, 0.1238, 0.1692, and 0.2972, respectively;

[0305] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S161 (IVB group) were 0.1630, 0.1628, 0.0569, 0.0958, and 0.5775, respectively;

[0306] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S162 (IVB group) were 0.0044, 0.0128, 0.1687, 0.0815, and 0.0000, respectively;

[0307] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S163 (IVB group) were 0.1244, 0.0665, 0.1166, 0.1337, and 0.3225, respectively;

[0308] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S164 (IVB group) were 0.1837, 0.0742, 0.0940, 0.1306, and 0.2930, respectively;

[0309] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S165 (IVB group) were 0.0681, 0.0902, 0.0818, 0.1416, and 0.5577, respectively;

[0310] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S166 (IVB group) were 0.0963, 0.1872, 0.0447, 0.0806, and 0.6099, respectively;

[0311] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S167 (IVB group) were 0.1022, 0.0847, 0.0935, 0.6106, and 0.3380, respectively;

[0312] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S168 (IVB group) were 0.0800, 0.0791, 0.0706, 0.0558, and 0.4099, respectively;

[0313] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S169 (IVB group) were 0.0933, 0.0533, 0.1472, 0.2865, and 0.2099, respectively;

[0314] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S170 (IVB group) were 0.0785, 0.1035, 0.0561, 0.8897, and 0.6803, respectively;

[0315] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S171 (IVB group) were 0.1200, 0.0861, 0.1776, 1.0000, and 0.6239, respectively;

[0316] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S172 (IVB group) were 0.1156, 0.1272, 0.1656, 0.1246, and 0.3324, respectively;

[0317] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S173 (IVB group) were 0.0548, 0.0819, 0.1433, 0.2285, and 0.3197, respectively;

[0318] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S174 (IVB group) were 0.0193, 0.0045, 0.1166, 0.0954, and 0.2930, respectively;

[0319] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S175 (IVB group) were 0.0800, 0.0686, 0.0654, 0.0664, and 0.3408, respectively;

[0320] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S176 (IVB group) were 0.0830, 0.1070, 0.0761, 0.0284, and 0.6648, respectively;

[0321] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S177 (IVB group) were 0.1037, 0.0637, 0.0280, 0.2700, and 0.4042, respectively;

[0322] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S178 (IVB group) were 0.0904, 0.1126, 0.1231, 0.1406, and 0.4535, respectively;

[0323] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S179 (IVB group) were 0.1274, 0.1098, 0.1251, 0.1242, and 0.4211, respectively;

[0324] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample number S180 (IVB group) were 0.1659, 0.2172, 0.1732, 0.1604, and 0.7000, respectively;

[0325] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S181 (IVB group) were 0.1674, 0.1649, 0.1280, 0.2174, and 0.4634, respectively;

[0326] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S182 (IVB group) were 0.1778, 0.1174, 0.1833, 0.2262, and 0.4648, respectively;

[0327] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S183 (IVB group) were 0.0415, 0.0582, 0.1267, 0.0996, and 0.4028, respectively;

[0328] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S184 (IVB group) were 0.2133, 0.1258, 0.1005, 0.0277, and 0.4014, respectively;

[0329] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S185 (IVB group) were 0.1289, 0.1119, 0.1199, 0.3498, and 0.4662, respectively;

[0330] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S186 (IVB group) were 0.1852, 0.0526, 0.0415, 0.0738, and 0.4479, respectively;

[0331] The normalized data of Resting Treg, T4Tef.MM, CK, LDH, and FIB of sample No. S187 (IVB group) were 0.1037, 0.0958, 0.1659, 0.0402, and 0.3408, respectively;

[0332] The normalized data of Resting Treg, T4Tef.MM, CK, LDH and FIB of sample No. S188 (IVB group) were 0.0563, 0.3039, 0.0465, 0.0737 and 0.3930, respectively.

[0333] The standardized data of the main indicators were subjected to PCA dimensionality reduction analysis using SPSS, and five principal components (PCs) were obtained, as shown in Table 4.

[0334] Table 4 Total variance explained

[0335]

[0336] Table 5 Composition matrix

[0337]

[0338] Referring to Table 5, the load values ​​of each main indicator variable on different principal components can be seen from the component matrix table. According to the relevant knowledge of mathematical statistics, the main component load matrix can be calculated based on the mathematical relationship between the main component load matrix, the factor load matrix and the eigenvalue, and the coefficient of each main indicator in the five principal components can be obtained. The calculation formula is: .

[0339] Among them, U is the coefficient, A is the load value of the main index on the main component, and λ is the main component eigenvalue. For example, the coefficient of RestingTreg in the expression of PC1 (take two significant figures after the decimal point) is .

[0340] The calculation results of the coefficients of the main indicators in each principal component expression are shown in Table 6.

[0341] Table 6 Composition coefficient table

[0342]

[0343] Referring to Table 6, according to the component coefficient table of the main components, the five main components are obtained y 1-y The expression for 5 is:

[0344]

[0345] in, x 1 - x 5 These are the normalized values ​​of Resting Treg, T4Tef.MM, CK, LDH, and FIB.

[0346] The test data of patients in group IVB and group II-IVA are brought into the principal component y 1 -y In the expression of 5, the principal component is calculated y 1 -y 5, and some calculation results are shown in Table 7.

[0347] Refer to Table 4-Table 6, take the variance percentage corresponding to the five principal components as their respective coefficients, and add them up to get N （PCA） Value calculation formula:

[0348] ;

[0349] The principal components of patients in group IVB and group II-IVA were y 1 -y Substitute the value of 5 into N （PCA） In the value calculation formula, the patient's N （PCA） Value, N of IVB group （PCA） The interquartile range of the values ​​was 0.13 (0.10, 0.17), and the N （PCA） The interquartile range of the values ​​is 0.22 (0.16, 0.29). Some calculation results are shown in Table 7.

[0350] Table 7 Principal components y 1 -y Partial calculation values ​​of 5 and N （PCA） The value of the formula obtained by calculating the part N （PCA） value

[0351]

[0352] (5) Establish calculation formula

[0353] Referring to Table 3, based on the average reduced Gini coefficient used in random forest analysis, the impurity of nodes in decision trees is measured. The smaller the average reduced Gini coefficient, the higher the node purity (the more homogeneous the sample category), indicating that the feature contributes less to reducing model impurity and has a lower importance for model evaluation. To avoid overfitting, indicators with an average reduced Gini coefficient greater than 3 are selected and included in the formula calculation, including:

[0354] T4Tef.MM: effector helper T cell mitochondrial mass value;

[0355] LDH: lactate dehydrogenase activity;

[0356] Resting Treg: The percentage of resting regulatory T cells, i.e., Treg cells in an unactivated state;

[0357] CK: creatine kinase activity;

[0358] FIB: fibrinogen concentration;

[0359] RBC: red blood cell concentration;

[0360] T8Tn.MMP: Initial CD8 ＋ Percentage of T cell mitochondria with low membrane potential;

[0361] T8Tcm.MMP: Percentage of mitochondrial low membrane potential in central memory killer T cells.

[0362] Table 8 Median values ​​of the eight indicators included in the calculation formula in the IVB group and the II-IVA group

[0363]

[0364] Referring to Table 8, the median values ​​of Resting Treg, T4Tef.MM, CK and RBC in the IVB group were lower than those in the II-IVA group; the median values ​​of LDH, FIB, T8Tn.MMP and T8Tcm.MMP in the IVB group were higher than those in the II-IVA group; （PCA） The median value of the IVB group (0.13) was lower than that of the II-IVA group (0.22). It can be considered that the data of RestingTreg, T4Tef.MM, CK and RBC were significant for N (PCA) The calculated results of the values ​​provide positive correlation contributions. The data of LDH, FIB, T8Tn.MMP and T8Tcm.MMP have a positive impact on N (PCA)The calculation results of the values ​​provide negative correlation contributions, so Resting Treg, T4Tef.MM, CK and RBC are classified as the numerator part of the calculation formula, and LDH, FIB, T8Tn.MMP, T8Tcm.MMP and BMI are classified as the denominator part of the calculation formula.

[0365] According to the article "Obesity-dependent selection of driver mutations in cancer" (published by Tang C et al. in the journal Nature Genetics in 2024, Vol. 56, No. 11, pp. 2318-2321), BMI is significantly correlated with tumor metastasis, and BMI is negatively correlated with tumor metastasis. Therefore, BMI is included in the formula calculation together with the above 8 indicators, and BMI is classified as the denominator of the calculation formula.

[0366] Refer to Table 1,

[0367] The interquartile range of resting Treg was 1.45 (0.81, 2.22);

[0368] The interquartile range of T4Tef.MM was 3.77 (2.06, 5.78);

[0369] The interquartile range of CK was 77.70 (57.13, 101.38);

[0370] The interquartile range of RBC values ​​was 4.66 (4.33, 5.02);

[0371] The interquartile range of LDH was 169.10 (147.20, 198.00);

[0372] The interquartile range of FIB was 3.54 (2.84, 4.35);

[0373] The interquartile range of T8Tn.MMP was 80.33 (61.65, 90.09);

[0374] The interquartile range of T8Tcm.MMP was 33.56 (24.51, 44.87).

[0375] The BMI of patients in group IVB was 23.14±2.94; that of patients in group II-IVA was 23.39±3.41.

[0376] Since there are large differences between the values ​​of Resting Treg, T4Tef.MM, CK, RBC, LDH, FIB, T8Tn.MMP, T8Tcm.MMP and BMI, it is necessary to perform calculations such as lg (indicator), indicator / 100 and indicator / 1000 on these nine indicators to unify the magnitude of the values ​​of the nine indicators; in order to avoid the situation where lg (0) cannot be calculated and lg (0~1) leads to a negative result, lg (indicator) is changed to lg (indicator + 1).

[0377] According to the establishment principle of the above formula, the following four N value calculation formulas are established:

[0378] Formula 1: ;

[0379] Formula 2: ;

[0380] Formula 3: ;

[0381] Formula 4: ;

[0382] Among them, Resting Treg, T4Tef.MM, CK, RBC, LDH, FIB, T8Tn.MMP, T8Tcm.MMP and BMI are all used in the above formulas 1-4 in the form of numerical values ​​without units.

[0383] The test data of the patients in the IVB group and the II-IVA group were substituted into formulas 1-4, respectively, and the N values ​​of the IVB group and the II-IVA group under each formula were calculated. The calculation results are shown in Table 9.

[0384] Table 9 Interquartile ranges of N values ​​for Group II-IVA and Group IVB calculated using different formulas

[0385]

[0386] Reference Figure 4 , draw N （PCA） The ROC curve of the N value calculation formula and the four N value calculation formulas is used to calculate the AUC value corresponding to the ROC curve. （PCA） The AUC value of the value calculation formula is 78.11, the AUC value of N1 is 91.53, the AUC value of N2 is 86.68, the AUC value of N3 is 91.38, and the AUC value of N4 is 92.02. In comparison, the AUC value of N4 is the largest, indicating that the prediction accuracy of N4 is the highest, so N4 is the preferred choice.

[0387] (6) Verify the calculation formula:

[0388] The TNM staging results of the sample population were quantified in the form of integral data to obtain quantitative staging values. The scoring basis is shown in Table 10.

[0389] Table 10 Quantitative numerical scoring table for patients in different TNM groups

[0390]

[0391] Among them, T stage is distinguished according to the size and invasion range of the primary tumor:

[0392] T0: no evidence of primary tumor;

[0393] T1: The tumor is confined to the nasopharynx or invades the oropharynx and / or nasal cavity but does not invade the parapharyngeal space;

[0394] T2: The tumor invades the parapharyngeal space. Invasion of the parapharyngeal space indicates that the tumor has broken through the pharyngeal wall and spread to the surrounding soft tissues.

[0395] T3: The tumor invades the skull base and / or paranasal sinuses;

[0396] T4: The tumor invades intracranial structures and / or cranial nerves, hypopharynx, orbit, masticatory space, pterygoid process, or cervical spine.

[0397] N staging is based on the location, size, and number of cervical lymph node metastases. In this application, N1-N4 are used to refer to different N value calculation formulas. To avoid confusion between different N value calculation formulas and different N staging, C0-C3 are used to distinguish different N staging:

[0398] C0: No regional lymph node metastasis.

[0399] C1: Unilateral lymph node metastasis with the shortest diameter ≤ 3 cm and located above the supraclavicular fossa.

[0400] C2: Unilateral or bilateral lymph node metastasis with the shortest diameter greater than 3 cm but ≤ 6 cm, or the metastatic lymph nodes are located above the supraclavicular fossa and are accompanied by central necrosis / extracapsular invasion.

[0401] C3: The shortest diameter of the metastatic lymph node is greater than 6 cm, or the metastatic lymph node is located in the supraclavicular fossa.

[0402] M stage is divided according to whether the tumor has metastasized to distant organs (such as bones, lungs, liver, etc.):

[0403] M0: no distant metastasis;

[0404] M1: Distant metastasis is present.

[0405] The TNM staging classification of NPC patients is shown in Table 11.

[0406] Table 11 TNM classification of nasopharyngeal carcinoma

[0407]

[0408] Among them, Tis is carcinoma in situ, which belongs to the pre-invasive cancer stage. The tumor cells are only confined to the mucosal epithelial layer and have not yet broken through the basement membrane to infiltrate and grow into deep tissues.

[0409] Substitute the scores of each scoring item in Table 10 into Table 11 to obtain the scores of different TNM stages. The specific scores are shown in Table 12.

[0410] Table 12 TNM classification score table for nasopharyngeal carcinoma

[0411]

[0412] Referring to Table 12, the quantitative TNM staging values ​​of 188 patients were statistically analyzed, and the quantitative TNM staging values ​​of patients in the IVB group were 150.42±14.21, and the quantitative TNM staging values ​​of patients in the II-IVA group were 45.54±18.65.

[0413] The TNM stage quantitative values ​​of patients were compared with N by Spearman nonparametric analysis. （PCA） The correlation between the value calculation formula and the N value calculated by formula 1-4 is shown in the analysis results. Figure 5-Figure 9 .

[0414] Record N separately （PCA） The value calculation formula and formula 1-4 S value, p Value and ρ value:

[0415] Formula 1 S The value is 1.73e+6, p The value is 6.49e-17, ρ The value is -0.56;

[0416] Formula 2 S The value is 1.61e+6, p The value is 5.61e-11, ρ The value is -0.45;

[0417] Formula 3 S The value is 1.74e+6, p The value is 1.32e-17, ρ The value is -0.57;

[0418] Formula 4 SThe value is 1.75e+6, p The value is 3.67e-18, ρ The value is -0.58;

[0419] N （PCA） Calculation formula S The value is 1.41e+6, p The value is 1.42e-4, ρ The value is -0.27;

[0420] in, S The value is the test statistic of the Spearman rank correlation coefficient, which is used to measure the monotonic relationship between two variables. S The larger the value, the greater the difference in order between the two variables; p The value is the significance judgment, p <0.05 was considered a significant difference; ρ The coefficient with a value between -1 and 1 represents the slope of the phase line and can be used to judge negative correlation or positive correlation. The closer the absolute value is to 1, the stronger the correlation is, and the closer it is to 0, the less correlation there is.

[0421] N （PCA） The value calculation formula and formula 1-4 ρ The values ​​are all less than 0, indicating that N （PCA） The N values ​​calculated by the formula and the four N value calculation formulas are negatively correlated with the patient's TNM stage score. ρ The value is closest to -1, and p The minimum value, S The value is the largest, indicating that the negative correlation between the N value of N4 and the patient's TNM staging score is the strongest, and the difference between the N value of N4 and the patient's TNM staging score is the most significant. Therefore, N4 is the preferred choice in the prediction formula for distant metastasis of nasopharyngeal carcinoma.

[0422] The N value of the metastasis group calculated by N4 was 0.11-0.16, and the N value of the control group ranged from 0.19 to 0.29. It can be considered that nasopharyngeal carcinoma patients with N values ​​between 0.19-0.29 are likely to have no distant metastasis of tumor cells, and nasopharyngeal carcinoma patients with N values ​​between 0.11-0.16 are likely to have distant metastasis of tumor cells.

[0423] This specific embodiment is merely an explanation of the present application and is not a limitation of the present application. After reading this specification, those skilled in the art may make non-creative modifications to the present embodiment as needed, but as long as they are within the scope of protection of the present application, they are protected by the patent law.

Claims

1. A method for constructing a prediction model for distant metastasis of nasopharyngeal carcinoma, characterized in that: include: (1) Patients with nasopharyngeal carcinoma (TNM stage II or above) were selected as the sample population, and the sample population was divided into a metastasis group and a control group based on the presence of distant metastasis of nasopharyngeal carcinoma; (2) Collect the test indicators of the sample population, conduct difference analysis and random forest analysis on the test indicators, and screen out the top five main indicators with the highest contribution rate; Then, the main index is processed by min-max standardization to obtain the standardized data of the main index, and the standardized data of the main index is subjected to PCA dimensionality reduction analysis to obtain the expression of the main component and N （PCA） Value calculation formula; 5 principal components y 1 -y The expressions for 5 are: in, x 1 - x 5 These are the normalized values ​​of Resting Treg, T4Tef.MM, CK, LDH, and FIB; N （PCA） The value calculation formula is: N (PCA) =0.31× y 1+0.23× y 2+0.19× y 3+0.15× y 4+0.13× y 5； Among them, N （PCA） The value is the distant metastasis index of nasopharyngeal carcinoma calculated by PCA dimensionality reduction analysis; (3) Establish the calculation formula: The characteristic index of reducing the average Gini coefficient by more than 3 was obtained by random forest analysis. （PCA） The value of N is calculated by the formula （PCA） The N value calculation formula composed of characteristic indicators and BMI is determined by the value result: in, Resting Treg is the percentage of resting regulatory T cells, i.e., Treg cells in an unactivated state; T4Tef.MM is the mitochondrial mass value of effector helper T cells; CK is creatine kinase activity; RBC is the red blood cell concentration; LDH is lactate dehydrogenase activity; FIB is fibrinogen concentration; T8Tn.MMP is also T8Tn.MMP low % abbreviation, representing the initial CD8 ＋ Percentage of T cell mitochondria with low membrane potential; T8Tcm.MMP is also T8Tcm.MMP low %, which is the percentage of low mitochondrial membrane potential in central memory killer T cells; BMI is body mass index; Draw N separately （PCA） The ROC curve of the N value calculation formula and the N value calculation formula, when the AUC value of the ROC curve of the N value calculation formula is greater than N （PCA） When the N value calculation formula is used, the N value calculation formula is the nasopharyngeal carcinoma distant metastasis prediction formula, and is verified; (4) Output the N value calculated by the nasopharyngeal carcinoma distant metastasis prediction formula to evaluate the risk of distant metastasis of nasopharyngeal carcinoma.

2. The method for constructing a prediction model for distant metastasis of nasopharyngeal carcinoma according to claim 1, characterized in that: The prediction formula for distant metastasis of nasopharyngeal carcinoma is: 。

Images