LncRNAs are involved in a wide range of biological processes, such as chromatin remodeling, mRNA splicing, mRNA editing and translation. They can either upregulate or downregulate gene expression, and play key roles in the progression of various human cancers. However, the functional mechanisms of most lncRNAs still remain unknown at present. This paper aims to provide space for the understanding of lncRNAs by proposing a new method to obtain protein-coding genes (PCGs) regulated by lncRNAs, thus identifying candidate cancer-related lncRNAs using bioinformatics approaches. This study presents a method based on sample correlation, which is applied to the expression profiles of lncRNAs and PCGs in prostate cancer in combination with protein interaction data to build a lncRNA-PCG bipartite network. Candidate cancer-related lncRNAs were extracted from the bipartite network by using a random walk. 14 prostate cancer-related lncRNAs were acquired from the LncRNADisease database and MNDR, of which 6 lncRNAs were present in our network. As one of the seed nodes, ENSG00000234741 achieved the highest score among them. The other two cancer-related lncRNAs (ENSG00000225937 and ENSG00000236830) were ranked within the top 30. In addition, the top candidate lncRNA ENSG00000261777 shares an intron with DDX19, and interacts with IGF2 P1, indicating its involvement in prostate cancer. In this paper, we described a new method for predicting candidate lncRNA targets, and obtained candidate therapeutic targets using this method. We hope that this study will bring a new perspective in future lncRNA studies.