Background: Clear cell renal cell carcinoma (ccRCC) is an aggressive urological cancer that originates from the proximal tubular epithelium. As one of the most common post-translational modification, protein arginine methylation plays a pivotal role in various cancer-associated biological functions, especially in cancer immunity. Therefore, constructing a protein arginine methylation-related prognostic signature would be beneficial in guiding better personalized clinical management for patients with ccRCC. Methods: Based on the multi-omics profiling of the expression levels of eight protein arginine methyltransferases (PRMTs) in 763 ccRCC samples (from TCGA, CPTAC, EMBL, and ICGC databases), we established a scoring system with machine-learning algorithms to quantify the modification patterns on clinical and immunological characterizations of individual ccRCC patient, which was termed as PRMTScore. Moreover, we utilized two external clinical cohorts receiving immunotherapy (n=302) to validate the reliability of the PRMTScore system. Multiplex immunohistochemistry (mIHC) was performed to characterize the cellular composition of 30 paired ccRCC samples. The proteomic profiling of 232 ccRCC samples obtained from Fudan University Shanghai Cancer Center (FUSCC) was analyzed to validate the protein expression of PRMT5 in ccRCC. Finally, CCK-8, transwell, and wound healing assays were conducted to elucidate the role of PRMT5 in ccRCC in vitro. Results: A total of 763 ccRCC patients with available multi-omics profiling were stratified into two clusters (PRMTCluster A and B) with distinctive prognosis, genomic alterations, tumor microenvironment (TME) characteristics, and fundamental biological mechanisms. Subsequently, protein arginine methylation-related prognostic signature (PRMTScore) was constructed and consisted of SLC16A12, HRH2, F2RL3, and SAA1. The PRMTScore showed remarkable differences in outcomes, immune and stromal fractions, expressions of immune checkpoints, the abundance of immune cells, and immunotherapy response in ccRCC patients. Additionally, preliminary insights unveiled the tumor-suppressive role of PRMT5 in ccRCC, and the signal of PRMT5 low significantly predicted aggressive prognosis and the high abundance of PD1 + CD8 + cells in ccRCC.
Conclusion:We constructed a PRMTScore system, which showed the potent ability to assess the prognosis, TME characteristics, and immunotherapy response for patients with ccRCC. Moreover, this is the first study to propose that PRMT5 acts as a cancer suppressor in ccRCC.