Metabolic Classifications of Renal Cell Carcinoma Reveal Intrinsic Connections with Clinical and Immune Characteristics.
Li Le,Chao Zheng,Waikeong Un,Xiao Jun,Ge Yue,Wang Yanan,Xiong Zezhong,Ma Sheng,Wang Zhihua,Hu Zhiquan,Zeng Xing
DOI: https://doi.org/10.1186/s12967-023-03978-y
IF: 8.44
2023-01-01
Journal of Translational Medicine
Abstract:BACKGROUND:Kidney cancer undergoes a dramatic metabolic shift and has demonstrated responsiveness to immunotherapeutic intervention. However, metabolic classification and the associations between metabolic alterations and immune infiltration in Renal cell carcinoma still remain elucidative.METHODS:Unsupervised consensus clustering was conducted on the TCGA cohorts for metabolic classification. GESA, mRNAsi, prognosis, clinical features, mutation load, immune infiltration and differentially expressed gene differences among different clusters were compared. The prognosis model and nomograms were constructed based on metabolic gene signatures and verified using external ICGC datasets. Immunohistochemical results from Human Protein Atlas database and Tongji hospital were used to validate gene expression levels in normal tissues and tumor samples. CCK8, apoptosis analysis, qPCR, subcutaneously implanted murine models and flowcytometry analysis were applied to investigate the roles of ACAA2 in tumor progression and anti-tumor immunity.RESULTS:Renal cell carcinoma was classified into 3 metabolic subclusters and the subcluster with low metabolic profiles displayed the poorest prognosis, highest invasiveness and AJCC grade, enhanced immune infiltration but suppressive immunophenotypes. ACAA2, ACAT1, ASRGL1, AKR1B10, ABCC2, ANGPTL4 were identified to construct the 6 gene-signature prognosis model and verified both internally and externally with ICGC cohorts. ACAA2 was demonstrated as a tumor suppressor and was associated with higher immune infiltration and elevated PD-1 expression of CD8+ T cells.CONCLUSIONS:Our research proposed a new metabolic classification method for RCC and revealed intrinsic associations between metabolic phenotypes and immune profiles. The identified gene signatures might serve as key factors bridging tumor metabolism and tumor immunity and warrant further in-depth investigations.