Prognostic marker gene and random survival forest model for predicting recurrence of II-stage colorectal cancer

Abstract

The invention discloses a prognostic marker gene for predicting recurrence of II-stage colorectal cancer and application. The invention provides a model for predicting the recurrence risk of a patientaccording to gene expression information of tumors of II-stage colorectal cancer patients. The AUC value of the model for predicting the five-year recurrence risk of the II-stage colorectal cancer patients is 0.993, and patients with high recurrence risk and patients with low recurrence risk can be remarkably separated in a test set. For the establishment and selection of the model, a random survival forest model is used, variable screening is carried out according to the minimum depth value of the maximum sub-tree where variables are located, and an important variables are selected to reestablish the model, so that the variable dimension of the model is greatly reduced. After the patients are divided into a high recurrence risk group and a low recurrence risk group by using the random survival forest model in a test set, no-recurrence lifetimes of the patients of the two groups have significant difference.

Description

[0001] (1) Technical field [0002] The invention relates to a method for constructing a recurrence prediction model of stage II colorectal cancer and a prognostic marker gene. [0003] (2) Background technology [0004] The treatment of colorectal cancer is mainly surgical treatment, supplemented by comprehensive treatment methods such as chemotherapy, radiotherapy, targeted therapy, and immunotherapy. For patients with early colorectal cancer, high-quality radical surgery for colorectal cancer can bring significant benefits to patients. However, some patients with early colorectal cancer have local recurrence and metastatic recurrence after radical surgery. The prognosis of these recurrence patients is often poor, and some studies have found that the RFS (recurrence-free survival) of colorectal cancer patients after radical surgery is shorter, Its overall survival is also shorter. Therefore, the prediction of recurrence risk after radical surgery for early colorectal cancer...

AI Technical Summary

Problems solved by technology

However, many current clinical studies have found that the definition of high-risk stage II is not accurate. Many high-risk stage II patients have no recurrence, while some general-risk stage II patients have recurrence and metastasis. This may be related to the traditional high-risk factors. The clinicopathological characteristics of the patient are not considered, and the biological characteristics of the tumor itself are not considered. However, the current gene chip technology and high-throughput sequencing technology can allow researchers to better mine the gene expression information of the tumor, and then reflect the biology of the tumor. feature

Images