Development and validation of lncRNAs-based nomogram for prediction of biochemical recurrence in prostate cancer by bioinformatics analysis

Background: Early biochemical recurrence (BCR) was considered as a sign for clinical recurrence and metastasis of prostate cancer (PCa). The purpose of the present study was to identify a lncRNA-based nomogram that can predict BCR of PCa accurately. Materials and methods: Bioinformatics analysis, such as propensity score matching (PSM) and differentially expressed genes (DEGs) analyses were used to identify candidate lncRNAs for further bioinformatics analysis. LASSO Cox regression model was used to select the most significant prognostic lncRNAs and construct the lncRNAs signature for predicting BCR in discovery set. Additionally, a nomogram based on our lncRNAs signature was also formulated. Both lncRNAs signature and nomogram were validated in test set. GSEA was carried out to identify various gene sets which share a common biological function, chromosomal location, or regulation. Results: A total of 457 patients with sufficient BCR information were included in our analysis. Finally, a five lncRNAs signature significantly associated with BCR was identified in discovery set (HR=0.44, 95%CI: 0.27-0.72, C-index = 0.63) and validated in test set (HR=0.22, 95%CI: 0.09-0.56, C-index = 0.65). Additionally, the lncRNAs-based nomogram showed significant performance for predicting BCR in both discovery set (C-index = 0.74) and test set (C-index = 0.78). Conclusion: In conclusion, our lncRNAs-based nomogram is a reliable prognostic tool for BCR in PCa patients. In addition, the present study put forward the direction for the further investigation on the mechanism of PCa progression.