Gastric cancer prognosis method based on apoptosis index

Abstract

The invention relates to a cancer prognosis data processing technology, and aims to provide a gastric cancer prognosis method based on apoptosis index, which comprises the following steps: A, acquiring pathological detection data of a gastric cancer pathological tissue sample, and constructing a source database; respectively establishing a regression model Y1 and a prediction model A1 by taking the survival time as a dependent variable; c, based on a nonlinear regression method, taking the whole regression model Y1 as an independent variable of nonlinear regression, establishing a regression model Y2, and calculating a regression coefficient and a random error of the regression model Y2; correcting the prediction model A1 by using a regression coefficient and a random error to obtain a prediction model A2; and D, predicting the survival time by using the prediction model A2 to obtain a prognosis conclusion. The method is simple and easy to implement, has good prognosis accuracy and can be effectively popularized and popularized.

Description

Technical Field

[0001] The present invention relates to the technical field of cancer prognosis data processing, and in particular to a gastric cancer prognosis method based on apoptosis index. Background Art

[0002] Gastric cancer is one of the malignant tumors that endanger life and health worldwide. In recent years, with the increase of social pressure, the spread of bad living habits, and the malignant changes in some dietary structures, the number of new gastric cancer patients and deaths has increased sharply worldwide. The incidence and mortality of gastric cancer are high, and most patients have no obvious specific symptoms in the early and middle stages, resulting in most patients being in the middle and late stages of poor differentiation or undifferentiation, accompanied by lymph node metastasis, with a 5-year survival rate of only 20-30%. The relevant prognosis of gastric cancer can be discovered in time and assist in adjusting the treatment plan, which is very important for the treatment and cure of gastric cancer. At present, most gastric cancer prognosis is still carried out by the traditional TNM staging method. At the same time, studies have found that the dysregulation of apoptosis and uncontrolled cell proliferation in almost all cancer cells play an important role in the process of tumorigenesis. Moreover, the apoptosis index and tumor size of gastric cancer tumor cells can not only be used as independent prognostic indicators alone, but also highlight specific advantages when used as a combined indicator, but there are few related reports and studies. In addition, in recent years, with the further popularization of big data, the rise of various data storage systems and data analysis systems has made it possible to realize gastric cancer prognosis methods based on various algorithms and data analysis. By introducing relevant algorithms into AI data systems, rapid and convenient prognosis of gastric cancer can be achieved, providing a strong guarantee for reducing the economic burden on patients and improving the accuracy of prognosis. Summary of the invention

[0003] The purpose of the present invention is to provide a gastric cancer prognosis method based on apoptosis index, which provides a new and popularizable idea and direction for gastric cancer prognosis.

[0004] To achieve the above-mentioned purpose, the technical solution adopted by the present invention is: a gastric cancer prognosis method based on apoptosis index, comprising the following steps:

[0005] A. Obtain pathological test data of gastric cancer pathological tissue samples and construct a source database including tumor size and tumor cell apoptosis index:

[0006] B. Process the data of the source database, use tumor size and tumor cell apoptosis index as independent variables, and survival time as dependent variable, and establish regression model Y1 and prediction model A1 respectively;

[0007] C. Based on the nonlinear regression method, the regression model Y1 is used as the independent variable of nonlinear regression to establish the regression model Y2, and the regression coefficient and random error of the regression model Y2 are calculated; and the prediction model A1 is corrected by using the regression coefficient and random error to obtain the prediction model A2;

[0008] D. Use prediction model A2 to predict survival time and draw prognostic conclusions.

[0009] The beneficial effects of the present invention are mainly reflected in the formation of a prognostic model based on the tumor cell apoptosis index and tumor size. On the basis of tumor size, the coordinated prognosis of the tumor cell apoptosis index is added to achieve data fitting of the special correlation between the two. Not only is the method relatively simple and easy, but also it has good prognostic accuracy and can be effectively promoted and popularized. DETAILED DESCRIPTION

[0010] The present invention will be described in detail below with reference to embodiments.

[0011] Example 1

[0012] Methods for obtaining tumor cell apoptosis index, detection methods of cell apoptosis include morphological detection, transmission electron microscopy, TUNEL, DNA ladder determination and flow cytometry, etc. Among them, TUNEL is based on the characteristic that endogenous nuclease is activated in apoptotic cells and cuts its own chromosomal DNA into gaps or 3'-OH terminal fragments containing 180-200bp. It uses TdT to transfer deoxynucleotides (such as dUTP) labeled with markers (such as digoxin, biotin, fluorescein, etc.) to the 3'-OH terminal and then detects apoptotic cells by immunohistochemistry.

[0013] Here we take the TUNEL method as an example, and the specific steps include:

[0014] S1. Fix the excised pathological tissue with 10% neutral formalin for 36-48 hours to obtain paraffin-embedded tissue specimens;

[0015] S2. Paraffin-embedded tissue specimens were sliced. After dewaxing, the specimens were immersed in a methanol solution containing 0.3% H2O2 for 30 minutes and washed with PBS buffer. The specimens were digested with proteinase K (10ug / ml) for 50 minutes and washed with PBS buffer. The tissue specimens were covered with 1×TdT buffer and incubated for 30 minutes. TdT reaction mixture (1×TdT Buffer, 10mmol / L DTT, 100mmol / L dATP,

[0016] 1mmol / L Biotin-16-dUTP, 25mmol / L CoCL2, 400U / ul TdT) were covered on the tissue sections and reacted at 37°C for 60 minutes, and then washed with PBS buffer; horseradish peroxidase-labeled streptavidin was covered on the tissue sections and reacted for 15 minutes, and then washed with PBS buffer; DAB was used for color development, hematoxylin was used for light counterstaining, xylene was used for transparency, and the sections were sealed;

[0017] S3. Observe the sections under a microscope to determine the field of view containing at least 1000 tumor cells, count the number of apoptotic tumor cells, and calculate the tumor cell apoptosis index. The tumor cell apoptosis index is determined as a percentage, which is P;

[0018] Example 2

[0019] A gastric cancer prognosis method based on apoptosis index comprises the following steps:

[0020] A. Obtain pathological test data of at least 1,000 gastric cancer pathological tissue samples from different patients, and construct a source database including tumor size and tumor cell apoptosis index. The tumor size is based on the longest direction and the actual measured value. The tumor cell apoptosis index is obtained according to the method described in Example 1: During the data acquisition period, the patient's survival status is continuously tracked for 5 years;

[0021] Cancer is a multi-step disease that develops through different stages of abnormal gene accumulation caused by continuous cell division and incorrect DNA replication. Therefore, the development of the tumor is directly related to the number of proliferation cycles after the tumor occurs. The higher cell loss rate caused by tumors with a higher tumor cell apoptosis index compared to tumors with a lower tumor cell apoptosis index requires the tumor to proliferate faster to obtain its unique size or volume. In addition, clinical testing also requires that the tumor must reach a "critical volume" and tumors with a higher tumor cell apoptosis index must undergo more tumor cell replications to reach this "critical volume" compared to tumors with a lower tumor cell apoptosis index. Therefore, tumors with a higher tumor cell apoptosis index have more proliferation cycles and thus obtain more abnormal genes, which ultimately makes tumor metastasis more likely to occur. Therefore, both the tumor cell apoptosis index and tumor size have a more significant impact on patient survival.

[0022] During the construction of the sample source database, the inventors obtained a total of 1,200 gastric cancer pathological tissue samples from patients. Due to the large amount of sample data, individual sample data are no longer listed one by one in this article. The overall distribution is shown in Table 1 below:

[0023] Table 1 Distribution of tumor size, tumor cell apoptosis index P and survival time

[0024]

[0025] B. The data of the source database were processed, and the correlation between tumor size and tumor cell apoptosis index and survival prognosis was analyzed respectively, and the analysis method was carried out by combining range analysis and variance analysis. The correlation between the two and survival time was obtained under binary conditions, and divided into thousandths. The correlation between tumor size was 0.756, and the correlation between tumor cell apoptosis index was 0.244.

[0026] Let the dependent variable be Y, the independent variable be xi, i is 1 to n, n is the number of influencing factors (tumor size and tumor cell apoptosis index) (the extreme value is 2, and adaptive adjustment can also be made by introducing more independent variables), that is, xi represents the i-th influencing factor. Based on the multivariate nonlinear regression theory, firstly, according to the relationship between the dependent variable Y and the independent variable xi, the regression model is established respectively, that is, Y = f i (x i ) Formula (1)

[0027] The relationship between tumor size and survival time is shown in formula (2):

[0028]

[0029] The relationship between tumor cell apoptosis index and survival time is shown in formula (3):

[0030]

[0031] The prediction model is shown in formula (4), denoted as model A1;

[0032]

[0033] Select 2 case data from the source database:

[0034] A: tumor size 40 mm, tumor cell apoptosis index 1.375, actual survival time 255 days;

[0035] B: tumor size 13.6 mm, tumor cell apoptosis index 2.485, actual survival time 1020 days;

[0036] The above model A1 was used to estimate the survival of cases A and B: A≈324 days, B≈1171 days. The excess error of A was about 27%, and the excess error of B was about 15%.

[0037] The overall error rate is more than 20% and less than 30%, and the average error is 21%. A1 can be used directly as a prediction model, but the error is still too large, and model A1 needs to be corrected.

[0038] C. Based on the pathological data of the above source database, the regression model Y1 is used as an independent variable of nonlinear regression to establish a regression model Y2, and the regression coefficient and random error of the regression model Y2 are calculated; and the prediction model A1 is corrected using the regression coefficient and random error to obtain the prediction model A2;

[0039] The model constructed in formula (2) is used as a variable for multivariate nonlinear regression. Let X1 = f1(x1), X2 = f2(x2), ..., Xn = fn(xn), and the random error is ξ. Then the nonlinear regression model of the dependent variable Y and the respective variables (X1, X2, ..., Xn) can be expressed as formula (5):

[0040] Y=K1X1+K2X2+…+K n X n +ξ(5)

[0041] In formula (5), K1, K2, ..., Kn are regression coefficients, which are based on at least five sets of pathological data and then analyzed by regression analysis using SPSS software. The error coefficient ξ is obtained by regression analysis of a set of pathological data, and its value is 0.146.

[0042] Based on the PCPM software simulating the changes in the patient cycle, the prediction model A1 was modified to obtain a more accurate prediction model; the modified model is shown below and is recorded as model A2:

[0043]

[0044] D. Use prediction model A2 to re-prognose the survival time of case A and B and the newly added case C:

[0045] A: tumor size 40 mm, tumor cell apoptosis index 1.375, actual survival time 255 days;

[0046] B: tumor size 13.6 mm, tumor cell apoptosis index 2.485, actual survival time 1020 days;

[0047] C: tumor size 25 mm, tumor cell apoptosis index 1.7, actual survival time 1203 days;

[0048] The above model A2 was used to make the survival prognosis for cases A, B and C: A≈283 days, B≈1050 days, and C≈1166 days. The excess error of A was about 11%, the excess error of B was about 3%, and the under-error of C was about 5%.

[0049] The overall error rate is less than 15%, and the average error is about 6%, which can draw better conclusions on survival prognosis.

[0050] The above-mentioned embodiments only express the specific implementation of the present invention, and the description thereof is relatively specific and detailed, but it cannot be understood as limiting the scope of the present invention. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, which all belong to the protection scope of the present invention.

Claims

1. A gastric cancer prognosis method based on apoptosis index, characterized in that: The following steps are involved: A. Obtain pathological test data of gastric cancer pathological tissue samples and construct a source database including tumor size and tumor cell apoptosis index: B. Process the data of the source database, use tumor size and tumor cell apoptosis index as independent variables, and survival time as dependent variable, and establish regression model Y1 and prediction model A1 respectively; C. Based on the nonlinear regression method, the regression model Y1 is used as the independent variable of nonlinear regression to establish the regression model Y2, and the regression coefficient and random error of the regression model Y2 are calculated; and the prediction model A1 is corrected by using the regression coefficient and random error to obtain the prediction model A2; D. Use prediction model A2 to predict survival time and draw prognostic conclusions.

2. The gastric cancer prognosis method based on apoptosis index according to claim 1, characterized in that: In step A and step C, obtaining the tumor cell apoptosis index includes the following steps: S1. Fix the pathological tissue with 10% neutral formalin for 36 hours to obtain paraffin-embedded tissue specimens; S2. Paraffin-embedded tissue specimens were sliced. After dewaxing, the specimens were immersed in a methanol solution containing 0.3% H2O2 for 30 minutes and washed with PBS buffer. The specimens were digested with proteinase K (10ug / ml) for 50 minutes and washed with PBS buffer. The tissue specimens were covered with 1×TdT buffer and incubated for 30 minutes. TdT reaction mixture (1×TdT Buffer, 10mmol / L DTT, 100mmol / L dATP, 1mmol / L Biotin-16-dUTP, 25mmol / L CoCL2, 400U / ul TdT) were covered on the tissue sections and reacted at 37°C for 60 minutes, and then washed with PBS buffer; horseradish peroxidase-labeled streptavidin was covered on the tissue sections and reacted for 15 minutes, and then washed with PBS buffer; DAB was used for color development, hematoxylin was used for light counterstaining, xylene was used for transparency, and the sections were sealed; S3. Observe the sections under a microscope to determine the field of view containing at least 1000 tumor cells, count the number of apoptotic tumor cells, and derive the tumor cell apoptosis index.

Images