Identification and prognostic modeling of clinical factors associated with Gastric Cancer based on the TCGA database

In order to study the relationship between various factors in clinical data and survival time of gastric cancer patients, a clinically relevant prognostic model was established. First, the clinical data of gastric cancer were downloaded from the TCGA database, and all the data were randomly divided into a training set and a test set, and the statistically significant (p<0.05) factors were screened using COX regression analysis, and the C index was calculated to evaluate the predictive ability of the model and to calculate the probability of survival after 3 years and 5 years based on the risk scores, and finally the survival curves and the relationship between each factor and the prognosis of survival of gastric cancer patients were plotted. Finally, the survival curves and the relationship between each factor and the survival prognosis of gastric cancer patients were drawn. The results showed that two factors, radiation and pathological_stage, were statistically significant. To assess the predictive ability of the model, its C-index was calculated as 0.704, which means that the predictive ability of the model is moderate. The risk score was calculated, and based on the risk score, the survival probability after 3 years was 0.848, and the survival probability after 5 years was 0.808. The results of the survival analyses showed that the survival time of the low-risk group was significantly longer than that of the high-risk group, and the life expectancy of males was shorter than that of females, and the higher the cancer's stage, the the shorter the survival time, and the lower the stage the higher the probability of survival for the same survival time. In conclusion, constructing a clinically relevant prognostic model for gastric cancer is a better guide to study the factors affecting the length of survival time of gastric cancer patients.