Method for predicting value of neoadjuvant chemotherapy of breast cancer

Abstract

The invention provides a method for predicting value of breast cancer neoadjuvant chemotherapy. The method comprises the following steps: S1, collecting information of a patient who is diagnosed as breast cancer for the first time and is subjected to neoadjuvant chemotherapy; s2, performing MAT condition analysis on different clinical pathological characteristic groups; s3, checking the curative effect of neoadjuvant chemotherapy of the patient; s4, comparing the prediction effects of the MAT, the TG and the TC by using an ROC curve; on the basis of 221 patients diagnosed with breast cancer parallel neoadjuvant chemotherapy from January 2019 to December 2020, MAT is constructed through statistical analysis to predict the relation between breast cancer neoadjuvant chemotherapy effects, evaluation of the breast cancer neoadjuvant chemotherapy effects is facilitated, accurate evaluation of neoadjuvant chemotherapy reaction is achieved, and the clinical application prospect is broad. Important information is further provided for the influence of systemic treatment on breast cancer biology and prognosis.

Description

Technical Field

[0001] The present invention relates to the technical field of medical analysis, and particularly relates to a method for predicting the value of neoadjuvant chemotherapy for breast cancer. Background Art

[0002] The latest global statistical data shows that breast cancer has become the most common cancer among women, accounting for 32%, and ranks second in terms of mortality, accounting for 15%. Adipocytes are the largest component of breast tissue. Adipose tissue plays a crucial role as an energy storage depot and can also act as endocrine cells to produce various bioactive substances. Excessive adipose tissue may lead to the occurrence and development of malignant tumors in organs such as the endometrium, breast, esophagus, liver, colon, and ovary.

[0003] An adverse crosstalk is established during the tumor progression between breast cancer and adipose tissue, especially the MAT closest to breast cancer cells. Compared with the indicators reflecting systemic obesity such as BMI and local obesity indicators, MAT can better reflect the fat content in the female breast.

[0004] Preoperative neoadjuvant chemotherapy (NACT) is an important part of the comprehensive treatment of breast cancer and has been proven to have great clinical value for locally advanced and inoperable breast cancer. Its role in the treatment process of breast cancer has received more and more extensive attention and research. However, there is currently a lack of effective means to evaluate the treatment effect of preoperative neoadjuvant chemotherapy (NACT). Summary of the Invention

[0005] The purpose of the present invention is to solve the deficiencies existing in the prior art and to propose a method for predicting the value of neoadjuvant chemotherapy for breast cancer by analyzing the breast adipose index.

[0006] To achieve the above object, the present invention adopts the following technical solution: A method for predicting the value of neoadjuvant chemotherapy for breast cancer, comprising the following steps: S1: Collect information of patients initially diagnosed with breast cancer and undergoing neoadjuvant chemotherapy; The inclusion criteria for the patients: ① For all patients, breast core needle biopsy was performed under B-ultrasound guidance before neoadjuvant chemotherapy, and the specimens were sent for pathological examination and diagnosed as breast invasive ductal carcinoma by the pathology department; ② All the collected patients received a complete neoadjuvant chemotherapy regimen and subsequent surgery according to the guidelines; ③ Breast MRI evaluation was performed before the start of neoadjuvant chemotherapy. The exclusion criteria for the patients: ① Patients with serious heart, liver, kidney diseases, diabetes, etc.; ② Patients with distant metastases or combined with other malignant tumors; ③ Special types of breast cancer such as inflammatory breast cancer, mucinous adenocarcinoma, medullary carcinoma, sarcoma, etc.; ④ Patients with missing items in the research data.

[0007] The information includes MAI, ER, PR, lymph node status, age, tumor size, Ki-67, Her-2; S2: Analysis of MAT in different groups with different clinicopathological characteristics; The experimenter divided the patients into two groups according to each different pathological characteristic of the patients: the patients were divided into two groups based on the median age of the patients, divided into two groups according to the tumor size, divided into two groups according to the lymph node status, divided into two groups according to the ER status, divided into two groups according to the PR status, divided into two groups according to the Her-2 status, divided into two groups according to the Ki-67 status, and divided into two groups according to the MP grading; The experimenter observed and compared the MAT means in each group with different clinicopathological characteristics to compare whether there were statistical differences. It was found that the MAT values of patients with age ≥ 49, ER positive, PR positive, and high MP grading were higher, and the P value was statistically significant.

[0008] S3: Examine the efficacy of neoadjuvant chemotherapy for patients; It includes the following steps: S31: Use X 2 to examine and analyze the relationship between each clinicopathological factor and the efficacy of neoadjuvant chemotherapy for breast cancer; The experimenter used X 2 to examine and analyze the relationship between each clinicopathological factor and the efficacy of neoadjuvant chemotherapy for breast cancer, drew a univariate analysis chart of the response of neoadjuvant chemotherapy for breast cancer, and analyzed the relationship between each factor and the efficacy of neoadjuvant chemotherapy for breast cancer through the chart.

[0009] S32: Multivariate analysis of the efficacy of neoadjuvant chemotherapy for patients based on the nomogram of the Logistic regression model; It includes the following steps: S321: Classify patient information; Among them, the continuous variable is MAI, and the categorical variables are ER, PR, lymph node status, age, tumor size, Ki-67, Her-2; S322: Construct a Logistic regression model according to patient information; Using the logistic regression method in the generalized linear model, age, tumor size, ER, PR, MAI, Ki-67, Her-2, and lymph node status are all incorporated into the model formula; The model formula = age + tumor size + ER + PR + MAI + Ki-67 + lymph node status + Her-2; When the ER is positive, ER = 1; when the ER is negative, ER = 0; when the PR is positive, PR = 1; when the PR is negative, PR = 0; when the age is greater than the median age of the patient, the age is marked as 1; when the age is less than the median age of the patient, the age is marked as 0; when the lymph node status is continuous, the lymph node status is marked as 0; when the lymph node status is discrete, the lymph node status is marked as 1; when the tumor size is greater than 3 cm, the tumor size is marked as 1; when the tumor size is less than 3 cm, the tumor size is marked as 0; when Ki-67 is less than 15, Ki-67 is marked as 0; when Ki-67 is greater than 15, Ki-67 is marked as 1; when Her-2 is positive, Her-2 = 1; when Her-2 is negative, Her-2 = 0.

[0010] S323: Verify the Logistic regression model; Calculate the C-index using the Cstat() function in the DescTools package and compare the magnitudes of the C-indices; Draw a calibration curve and compare the positional relationship between the calibration curve and the diagonal line.

[0011] S324: Draw a nomogram based on the Logistic model; S4: Use the ROC curve to compare the prediction effects of MAT, TG, and TC; The experimenter imports the patient information in R language, then calculates the ROC curves of MAT, TG, and TC respectively, further draws the ROC curves of MAT, TG, and TC in the same chart, and then compares the AUC values between MAT and TG, MAT and TC, and TG and TC respectively. Through the AUC values among the three, compare the efficacy of MAT, TG, and TC in predicting the NACT response of breast cancer.

[0012] Compared with the prior art, the beneficial effects of the present invention are as follows: Neoadjuvant chemotherapy is increasingly applied to breast cancer, especially for downstaging the primary breast tumor and metastatic axillary lymph nodes. The accurate assessment of the neoadjuvant chemotherapy response by MAT provides important information on the impact of systemic treatment on the biology and prognosis of breast cancer and provides guidance for further treatment. Brief Description of the Drawings

[0013] Figure 1 It is a flowchart of the steps of a method for predicting the value of neoadjuvant chemotherapy for breast cancer according to the present invention; Figure 2 It is a histogram of the MAT numerical distribution; Figure 3 It is a normal P-P plot (A) and a normal Q-Q plot (B) of the MAT numerical distribution; Figure 4 Table of the relationship between MAT and clinicopathological features; Figure 5 Table of univariate analysis of neoadjuvant chemotherapy for breast cancer patients; Figure 6 Logistic regression model for multivariate analysis of the efficacy of neoadjuvant chemotherapy in R software; Figure 7 Predictive nomogram of NACT efficacy based on Logistic regression; Figure 8 Calibration curve; Figure 9 ROC curves of MAT, TG, and TC for predicting the efficacy of neoadjuvant chemotherapy. Detailed implementation manner

[0014] To further understand the purpose, structure, features, and functions of the present invention, the following is a detailed description in conjunction with embodiments.

[0015] As Figure 1 shown, a method for predicting the value of neoadjuvant chemotherapy for breast cancer includes the following steps: S1: Collect information of patients initially diagnosed with breast cancer and receiving neoadjuvant chemotherapy; The inclusion criteria for the patients: ① All patients underwent breast core needle biopsy under B-ultrasound guidance before neoadjuvant chemotherapy, and the specimens were sent for pathological examination. They were diagnosed as breast invasive ductal carcinoma by the pathology department; ② All the collected patients received a complete neoadjuvant chemotherapy regimen and subsequent surgery according to the guidelines; ③ Breast MRI was evaluated before the start of neoadjuvant chemotherapy. The exclusion criteria for the patients: ① Patients with severe heart, liver, kidney diseases, diabetes, etc.; ② Patients with distant metastases or combined with other malignant tumors; ③ Special types of breast cancer such as inflammatory breast cancer, mucinous adenocarcinoma, medullary carcinoma, sarcoma; ④ Patients with missing items in the research data.

[0016] The information includes MAI, ER, PR, lymph node status, age, tumor size, Ki-67, Her-2; S2: Analyze the MAT situation for different groups with different clinicopathological characteristics; The experimenter divided the patients into two groups according to each different pathological characteristic of the patients: The patients were divided into two groups based on the median age of the patients, divided into two groups according to the tumor size, divided into two groups according to the lymph node status, divided into two groups according to the ER situation, divided into two groups according to the PR situation, divided into two groups according to the Her-2 situation, divided into two groups according to the Ki-67 situation, and divided into two groups according to the MP grading. The experimenter further analyzed the characteristics of the MAT values of these patients in depth and found that the average value of MAT was 0.6916, the median was 0.6996, the standard deviation was 0.1146, the 25th percentile was 0.6159, the 50th percentile was 0.6996, and the 75th percentile was 0.7748.

[0017] As Figure 2 , 3 shown, the experimenter observed the MAT values of each group of patients and plotted a histogram of MAT distribution. It was found that the distribution was similar to the normal distribution, with a skewness of -0.406, a standard deviation of skewness of 0.164, a kurtosis of -0.345, and a standard deviation of kurtosis of 0.326. Further, the P-P plot and Q-Q plot of the MAT values were drawn, and it was obtained that the MAT values approximately followed a normal distribution. These data indicate that in breast cancer patients, the MAT values are approximately normally distributed. Verifying the normal distribution of MAT values from multiple angles shows that the MAT values conform to the central limit theorem, which can simplify the analysis and improve the reliability of the conclusions.

[0018] As Figure 4 shown, the experimenter observed and compared the MAT means in each group grouped by different clinicopathological characteristics to determine whether there were statistically significant differences. It was found that patients with age ≥ 49, ER positive, PR positive, and high MP grade had higher MAT values, and the P values were statistically significant.

[0019] S3: Examine the efficacy of neoadjuvant chemotherapy for patients; It includes the following steps: S31: Use X 2 to test and analyze the relationship between various clinicopathological factors and the efficacy of neoadjuvant chemotherapy for breast cancer; As Figure 5 shown, the experimenter used X 2 to test and analyze the relationship between various clinicopathological factors and the efficacy of neoadjuvant chemotherapy for breast cancer, and found that patients with high MAT values (P < 0.001), large tumors (P < 0.0001), many lymph nodes (P < 0.0001), ER positive (P < 0.001), PR positive (P < 0.001), Her-2 negative (P < 0.001), and low expression of Ki-67 (P = 0.028) had poor tumor regression effects in neoadjuvant chemotherapy. A univariate analysis chart of the response to neoadjuvant chemotherapy for breast cancer was drawn to analyze the relationship between each factor and the efficacy of neoadjuvant chemotherapy for breast cancer through the chart.

[0020] S32: Perform a multivariate analysis of the efficacy of neoadjuvant chemotherapy for patients based on the nomogram of the Logistic regression model; It includes the following steps: S321: Classify patient information; Among them, the continuous variable is MAI, and the categorical variables are ER, PR, lymph node status, age, tumor size, Ki-67, and Her-2; S322: Construct a Logistic regression model based on patient information; As Figure 6 shown, using the logistic regression method in the generalized linear model, age, tumor size, ER, PR, MAI, Ki-67, Her-2, and lymph node status are all incorporated into the model formula; The model formula = age + tumor size + ER + PR + MAI + Ki-67 + lymph node status + Her-2; The value obtained according to the model formula represents the predicted efficacy probability.

[0021] When the ER is positive, ER = 1; when the ER is negative, ER = 0; when the PR is positive, PR = 1; when the PR is negative, PR = 0; when the age is greater than the median age of the patients, age is marked as 1; when the age is less than the median age of the patients, age is marked as 0; when the lymph node status is continuous, the lymph node status is marked as 0; when the lymph node status is discrete, the lymph node status is marked as 1; when the tumor size is greater than 3 cm, the tumor size is marked as 1; when the tumor size is less than 3 cm, the tumor size is marked as 0; when Ki-67 is less than 15, Ki-67 is marked as 0; when Ki-67 is greater than 15, Ki-67 is marked as 1; when Her-2 is positive, Her-2 = 1; when Her-2 is negative, Her-2 = 0.

[0022] The results suggest that patients with high MAT values (P < 0.0001), larger tumors (P = 0.0258), and more lymph nodes (P = 0.0004) have poor tumor regression effects in neoadjuvant chemotherapy.

[0023] S323: Validate the Logistic regression model; The C-index is calculated through the Cstat() function of the DescTools package. The C-index is 0.886, slightly smaller than 1, indicating strong model discrimination ability and showing that this model is reliable; As Figure 8 shown, draw a calibration curve and compare the positional relationship between the calibration curve and the diagonal line.

[0024] The calibration curve fits well with the diagonal line, indicating that the model we established is relatively accurate and reliable.

[0025] As Figure 7As shown, S324: Draw a nomogram based on the Logistic model; For the said nomogram, according to the patient's age, tumor size, etc., scores can be found on the corresponding axes, and the scores of all variables are added together. The value corresponding to the total score on the bottom probability axis is the predicted efficacy probability.

[0026] S4: Use the ROC curve to compare the prediction effects of MAT, TG, and TC; As Figure 9 shown, the experimenter imports the patient information in R language, then calculates the ROC curves of MAT, TG, and TC respectively, further draws the ROC curves of MAT, TG, and TC in the same chart, and then compares the AUC values between MAT and TG, MAT and TC, and TG and TC respectively. Through the AUC values among the three, the efficacy of MAT, TG, and TC in predicting the NACT response of breast cancer is compared. In the ROC curve graph, the curve with a position closer to the upper left corner has a higher prediction efficacy. The AUC value ranges from 0.5 (no discrimination) to 1 (perfect discrimination). It is found that the prediction value of MAT for the NACT efficacy of breast cancer patients is higher than that of TG and TC, and the AUC values are 0.778, 0.584, and 0.550 respectively.

[0027] The present invention has been described by the above related embodiments. However, the above embodiments are only examples for implementing the present invention. It must be pointed out that the disclosed embodiments do not limit the scope of the present invention. On the contrary, modifications and refinements made without departing from the spirit and scope of the present invention fall within the patent protection scope of the present invention.

Claims

1. A method for predicting the value of neoadjuvant chemotherapy for breast cancer, characterized in that: It includes the following steps: S1: Collect the information of patients with breast cancer as the first diagnosis and who have received neoadjuvant chemotherapy; S2: Analyze the MAT situation of different groups classified by clinicopathological characteristics; S3: Examine the efficacy of neoadjuvant chemotherapy for patients; It includes the following steps: S31: Use X 2 The relationships between various clinicopathological factors and the efficacy of neoadjuvant chemotherapy for breast cancer were examined and analyzed; S32: Perform multivariate analysis of the efficacy of neoadjuvant chemotherapy for patients based on the nomogram of the Logistic regression model; Specifically, it includes the following steps: S321: Classify the patient information; Among them, the continuous variable is MAI, and the categorical variables are ER, PR, lymph node status, age, tumor size, Ki-67, Her-2; S322: Construct a Logistic regression model according to the patient information; Using the logistic regression method in the generalized linear model, age, tumor size, ER, PR, MAI, Ki-67, Her-2, and lymph node status are all incorporated into the model formula; The model formula = age + tumor size + ER + PR + MAI + Ki-67 + lymph node status + Her-2; S323: Verify the Logistic regression model; Calculate the C-index through the Cstat() function of the DescTools package and compare the magnitudes of the C-indices; Draw a calibration curve and compare the positional relationship between the calibration curve and the diagonal line; S324: Draw a nomogram based on the Logistic model; S4: Use the ROC curve to compare the prediction effects of MAT, TG, and TC; For a method for predicting the value of neoadjuvant chemotherapy for breast cancer according to claim 1, wherein: The criteria for including the patients: ① For all patients, under the guidance of B-ultrasound, a breast core needle biopsy was performed before neoadjuvant chemotherapy, and the specimens were sent for pathological examination, and the patients were diagnosed as invasive ductal carcinoma of the breast by the pathology department; ② All the patients collected received a complete neoadjuvant chemotherapy regimen and subsequent surgery according to the guidelines; ③ Breast MRI was evaluated before the start of neoadjuvant chemotherapy; The exclusion criteria for the patients: ① Patients with serious heart, liver, kidney diseases, diabetes, etc.; ② Patients with distant metastases or combined with other malignant tumors; ③ Special types of breast cancer such as inflammatory breast cancer, mucinous adenocarcinoma, medullary carcinoma, sarcoma, etc.; ④ Patients with missing research data items.

2. The method for predicting the value of neoadjuvant chemotherapy for breast cancer according to claim 1, characterized in that: The specific content of step S2: The experimenter divides the patients into two groups according to each different pathological characteristic of the patients: divide the patients into two groups based on the median age of the patients, divide the patients into two groups according to the tumor size, divide the patients into two groups according to the lymph node status, divide the patients into two groups according to the ER situation, divide the patients into two groups according to the PR situation, divide the patients into two groups according to the Her-2 situation, divide the patients into two groups according to the Ki-67 situation, and divide the patients into two groups according to the MP grading; The experimenter observes the MAT values of each group of patients, draws a MAT histogram, and further draws a P-P plot and a Q-Q plot of the MAT values, and obtains that the MAT values are approximately normally distributed; The experimenter observes and compares the MAT means in each group classified by different clinicopathological characteristics to compare whether there are statistical differences.

3. The method for predicting the value of neoadjuvant chemotherapy for breast cancer according to claim 1, wherein: The specific content of step S31 is: The experimenter uses X 2 Examined and analyzed the relationship between various clinicopathological factors and the efficacy of neoadjuvant chemotherapy for breast cancer, and drew a univariate analysis chart of the response to neoadjuvant chemotherapy for breast cancer. Analyzed the relationship between each factor and the efficacy of neoadjuvant chemotherapy for breast cancer through the chart analysis.

4. The method for predicting the value of neoadjuvant chemotherapy for breast cancer according to claim 1, characterized in that: The specific content of step S4 is as follows: The experimenter imports the patient information in R language, then calculates the ROC curves of MAT, TG, and TC respectively, further plots the ROC curves of MAT, TG, and TC in the same chart, and then compares the AUC values between MAT and TG, MAT and TC, and TG and TC respectively. Through the AUC values among the three, the efficacy of MAT, TG, and TC in predicting the NACT response of breast cancer is compared.

Images