Background: Clear cell renal cell carcinoma (ccRCC) is the most predominant histology of renal cancer, representing 75% of all cases and accounting for the majority of associated deaths. To gain insight into the impact of genomic alterations on the functional modules that drive ccRCC tumorigenesis, we leveraged comprehensive proteogenomic characterization of 110 treatment-naïve renal cell carcinoma (RCC) and 84 paired-matched normal adjacent tissue (NAT) samples.
Methods: We utilized an integrated proteogenomic approach, performing whole genome sequencing (WGS), whole exome sequencing (WES), DNA methylation profiling for all tumors; RNA-seq, proteomic, and phosphoproteomic characterization was performed for all samples.
Results: WGS analysis revealed arm-level loss of chromosome 3p as a frequent event in ccRCC, with 61% of tumors showing evidence of 3p chromosomal translocation events. Comparative profiling of ccRCC and NATs samples identified pathways associated with immune response, EMT, and glycolysis to be up-regulated in ccRCC, while TCA cycle, fatty acid metabolism, and oxidative phosphorylation were down-regulated. Examination of mRNA-protein correlation revealed a non-linear relationship in cellular processes including Warburg Effect-related metabolism, as well as the tumor-specific trend of higher sample-wise correlation associating with prognostically-defined aggressive features of ccRCC. Analysis of differential phosphosite occupancy between tumors and NAT showed MAPK/ERK signalling and G2/M stalling to be up-regulated across the majority of ccRCC cases. We deconvoluted immune and stromal cell gene signatures in the tumor microenvironment (TME), with consensus clustering of the TME compositions identifying four immune-based subtypes: Inflamed CD8+, Inflamed CD8-, VEGF Immune Desert, and Metabolic Immune Desert. Integrated transcriptomic and proteomic profiling of the ccRCC subtypes revealed unique, discriminatory signalling pathways associated with immune exhaustion, cancer-associated fibroblast-related signalling, angiogenesis, and metabolic activity.
Conclusions: Our results link the functional impact of genomic alterations at the protein level, and provides evidence for rational treatment selection stemming from proteomic, phosphoproteomic, and tumor microenvironment signatures.