Prognosis model of endometrial cancer and construction method

Abstract

The invention relates to the technical field of biomedicine, in particular to an endometrial cancer prognosis model and a construction method thereof, and the technical key points are as follows: the model is a gene-gene interaction model, gene-gene interaction is taken as a variable, a COX-ph model is used for screening prognosis independent correlation variables, and weights are given; the subject working curve is used for model prediction of the probability of the lifetime of the patient of 1-5 years; the method comprises the following steps: carrying out RNA-seq sequencing on an endometrial cancer tumor tissue, and constructing a stable prediction model (PMID: 35436725) by taking a product of expression quantities of two standardized genes as measurement of gene-gene interaction. According to the invention, RNA-seq sequencing is carried out on tumor tissues of endometrial cancer, so that the gene expression quantity and gene-gene interaction are effectively evaluated, and the detection cost can be remarkably reduced while a robust prediction result is provided for a patient; moreover, the data scale in the scheme of the invention is relatively large, so that a guarantee can be provided for the robustness of the model.

Description

Technical Field

[0001] This invention patent relates to the field of biomedical technology. Specifically, it relates to a prognostic model for endometrial cancer and a construction method thereof. Background Art

[0002] Endometrial cancer is a common gynecological tumor. Accurately predicting the prognosis of patients is helpful for formulating individualized treatment strategies. At present, there are few prediction models that can reliably predict the overall survival of endometrial cancer patients, and there is no rigorous internal and external cross-validation.

[0003] At present, the following technical disadvantages exist in the published overall survival prognosis prediction models for endometrial cancer:

[0004] 1. The lack of external validation of the model leads to overfitting of the model, and the prediction efficiency for the random population may be overestimated;

[0005] 2. The prediction efficiency of the model is not high, and there is only a single evaluation method (such as calculating the area under the curve AUC), which affects the robustness of the model;

[0006] 3. Only focus on the main effects of genes.

[0007] Combined with the clinical situation, constructing a prediction model for overall survival has important clinical significance. At present, data such as transcriptome, genome, serology, imaging, basic clinical information, and multi-omics data can all be used for the construction of prediction models. Moreover, next-generation sequencing technology is more favored by clinicians because of its direct detection of gene expression levels in tumor tissues and easy operation.

[0008] Therefore, the present invention provides a prognostic model for endometrial cancer and a construction method thereof to solve the above problems. Summary of the Invention

[0009] The purpose of the present invention is to solve the problems in the prior art proposed in the above background art, and provides a prognostic model for endometrial cancer and a construction method thereof.

[0010] The above object of the present invention is achieved as follows:

[0011] The solution of the present invention provides a construction method for a prognostic model of endometrial cancer. The model is a gene-gene interaction model. Gene-gene interactions are used as variables, and the COX-ph model is used to screen prognostic independent related variables and assign weights; the receiver operating characteristic curve is used to predict the probability of the patient's survival period from 1 to 5 years for the model; the method is as follows:

[0012] By performing RNA-seq sequencing on endometrial cancer tumor tissues, the product of the expression levels of two genes after standardization was used as a measure of gene-gene interaction to construct a robust prediction model (PMID: 35436725).

[0013] Further, the genes include B4GALNT3, PALM3, and SIX1;

[0014] The method also includes calculating a patient score, specifically: for the samples of RNA-seq sequencing, the obtained Count value of gene expression was normalized to TPM value, and then logarithmic transformation was performed, i.e., log(TPM + 1); then the final score was calculated according to the following calculation formula:

[0015] Transcriptome score = -0.535×B4GALNT3 + 0.289×PALM3 + 0.374×SIX1 + 0.118×PALM3×SIX1;

[0016] Final score = 0.35×Transcriptome score + 0.522×tumor stage + 0.031×Age.

[0017] Further, the method is used to predict the overall survival of endometrial cancer patients.

[0018] The difficulty and significance of the present invention in solving technical problems are as follows:

[0019] Constructing a prediction model for overall survival has important clinical significance. Currently, data such as transcriptome, genome, serology, imaging, clinical basic information, and multi-omics data can all be used for the construction of prediction models. Second-generation sequencing technology is more favored by clinicians due to its direct detection of gene expression levels in tumor tissues and ease of operation. Due to the establishment and improvement of the previous database, the inventors of the present application used the transcriptome sequencing and clinical follow-up data of 546 cases of endometrial cancer to construct a prediction model for overall survival. We used gene-gene interaction as a variable and used the COX-ph model to screen for prognostic independent related variables and assign weights. The receiver operating characteristic curve was used to evaluate the probability of the model predicting the survival of patients for 1 - 5 years, and the clinical decision curve was used to evaluate the robustness of the prediction model.

[0020] Compared with the prior art, the present invention has the following beneficial effects:

[0021] 1. By performing RNA-seq sequencing on endometrial cancer tumor tissues, the present invention effectively evaluates gene expression levels and gene-gene interactions, significantly reducing the detection cost while providing robust prediction results for patients;

[0022] 2. The data scale in the solution of the present invention is large, which can guarantee the robustness of the model;

[0023] 3. The solution of the present invention has been tested by a variety of existing statistical parameters and is more comprehensive than the existing prognostic models;

[0024] 4. In the solution of the present invention, through statistical analysis, it is found that the prediction accuracy of the model of the present invention is high and the model performance is robust;

[0025] 5. In the solution of the present invention, the prognosis of patients is accurately judged by detecting the expression of a few genes, which is more economical than the conventional whole (20,000) gene sequencing. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 shows the relationship between gene effect and patient survival period in the embodiment of the present invention;

[0027] Figure 2 shows that in the embodiment of the present invention, genes with different protein levels in normal tissues and tumors are determined, the main functions of B4GALNT3 and the interaction between SIX1 and PALM3 are incorporated into the Cox-ph model, and the regression coefficients of age and tumor stage are combined to construct the final prognostic model;

[0028] Figure 3 shows the ability of the model in the embodiment of the present invention to predict prognosis;

[0029] Figure 4 shows the clinical benefit of determining the final model in the embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0030] In order to make the objectives, technical solutions and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0031] The implementation of the present invention will be described in detail below with reference to specific embodiments.

[0032] Refer to Figures 1-4 as shown below, the following is a preferred embodiment provided by the present invention.

[0033] The embodiment of the present invention provides a method for constructing a prognostic model for endometrial cancer. This model is a gene-gene interaction model. By taking gene-gene interactions as variables and using the COX-ph model to screen for independent prognostic variables and assign weights; the receiver operating characteristic curve is used to predict the probability of patient survival at 1-5 years. The specific construction method is as follows: By performing RNA-seq sequencing on endometrial cancer tumor tissues, the product of the expression levels of two genes after standardization is used as a measure of gene-gene interaction to construct a robust prediction model (PMID: 35436725).

[0034] Among them, the genes detected by this model include B4GALNT3, PALM3, and SIX1;

[0035] The calculation method of the patient score for this model: For the samples of RNA-seq sequencing, the obtained gene expression Count values are normalized to TPM values, and then logarithmically transformed, that is, log(TPM + 1); then calculate the final score according to the following calculation formula:

[0036] Transcriptome score = -0.535×B4GALNT3 + 0.289×PALM3 + 0.374×SIX1 + 0.118×PALM3×SIX1;

[0037] Final score = 0.35×Transcriptome score + 0.522×tumor stage + 0.031×Age.

[0038] The model method of the present invention can be used to predict the overall survival of endometrial cancer patients.

[0039] The core point of the above embodiment of the present invention is: the calculation method of the above Final score, including the specific genes (B4GALNT3, PALM3, and SIX1) and weights screened by the inventors of the present application for modeling. Compared with the prior art, the solution of the present invention significantly increases the number of candidate genes and greatly increases the candidate genes applicable to each data set. Through RNA-seq sequencing of endometrial cancer tumor tissues, the gene expression levels and gene-gene interactions are effectively evaluated, while providing a robust prediction result for patients, significantly reducing the detection cost (that is: using the expression of a small number of genes to accurately judge the prognosis of patients, which is more economical than the conventional sequencing of all (20,000) genes).

[0040] During the research process, we calculated the Transcriptome score and Final score for millions of gene pairs formed by pairwise pairing of more than 8,000 genes according to the above formula. The prognostic value of the Final score formed by each gene pair was examined by Cox regression, and the ROC curve was used to calculate the area under the curve corresponding to each Final score. Finally, the three genes B4GALNT3, PALM3, and SIX1 were identified to construct a prognostic model for endometrial cancer.

[0041] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, and improvements made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A method for constructing a prognostic model for endometrial cancer, characterized in that: The model is a gene-gene interaction model, which takes gene-gene interaction as a variable, uses the COX-ph model to screen prognostic independent related variables and assigns weights; the receiver operating curve is used to predict the probability of 1-5 years patient survival by the model; the method is: By performing RNA-seq sequencing on endometrial cancer tumor tissues, the product of the standardized expression levels of two genes was used as a measure of gene-gene interaction to construct a robust prediction model (PMID: 35436725).

2. The method for constructing a prognostic model for endometrial cancer according to claim 1, characterized in that: The genes include B4GALNT3, PALM3, and SIX1; The method further includes calculating a patient score, specifically: for samples sequenced by RNA-seq, the Count value of gene expression is standardized to a TPM value, and then logarithmically transformed, i.e., log(TPM+1); and then the final score is calculated according to the following calculation formula: Transcriptome score=-0.535×B4GALNT3+0.289×PALM3+0.374×SIX1+0.118×PALM3×SIX1; Final score=0.35×Transcriptome score+0.522×tumor stage+0.031×Age.

3. A method for constructing a prognostic model for endometrial cancer according to any one of claims 1 to 2, characterized in that: The method is used to predict the overall survival of patients with endometrial cancer.

Images