Construction And Verification Of Recurrence Risk Predicting Model For Prostate Cancer Patients After Radical Surgery Based On Expression Profile Data

Objective: If the recurrence-free survival(RFS)situation can be monitored in real time for prostate cancer patients after radical surgery,it will have a great significance of the patient’s postoperative treatment strategy,the development of clinical research and the adjustment of the follow-up plan.Our research aims to establish and validate a RFS prognostic prediction model based on gene expression profile data in public databases Methods: A total of 5 independent cohorts were included in this study,including 1,046 expression profile data containing RFS prognostic information.We used univariate analysis to calculate candidate genes related to RFS in five cohort studies and took intersections.Multivariate Cox regression analysis was used to calculate the corresponding coefficient(co-efficient,co-ef)of each candidate gene,and combined the candidate gene expression to construct a RFS prognosis prediction model.Receiver Operating Characteristic(ROC)was used to evaluate the predictive value of the model.In addition,Pearson correlation and KEGG information database(Kyoto Encyclopedia of Genes and Genomes)pathway enrichment analyses were used to explore the underlying mechanism of these key genes affecting PCa progression.Finally,we compared the expression differences of these key candidate genes between different clinical subgroups,such as relapse,age,and tumor grade,etc.Results: According to the results of the univariate Cox regression analysis of the five independent cohort data to obtain the intersection,a total of 7 consensus genessignificantly correlated with the prognosis of RFS were identified(P < 0.01).Co-ef and candidate gene expression data provided by multivariate Cox regression analysis were used to construct a RFS prognosis prediction model for PCa patients.Kaplan-Meier(K-M)analysis results suggest that in all cohorts,patients identified as low-risk by the RFS prognostic prediction model have better RFS prognosis than high-risk patients.AUC enclosed with the coordinate axis proves that the model has excellent prediction performance.In addition,the research also analyzed the correlation between the expression of these 7 candidate genes and clinicopathological characteristics in the public database and clinical samples of this center.Using co-expression analysis and KEGG pathway enrichment analysis,we speculated the potential mechanisms of how these critical genes affected PCa progress or prognosis.Conclusions: The current study uses public database released data and sample resources collected from our center to construct and successfully verify a RFS predicting model,and reveals the potential mechanisms of how the critical genes affect the progression or prognosis of PCa,providing a theoretical basis for promoting individualized medicine for PCa patients.