Clear cell renal carcinoma is the most frequent type of kidney cancer, with an increasing incidence rate worldwide. In this research, we used a proteotranscriptomic approach to differentiate normal and tumor tissues in clear cell renal cell carcinoma (ccRCC). Using transcriptomic data of patients with malignant and paired normal tissue samples from gene array cohorts, we identified the top genes over-expressed in ccRCC. We collected surgically resected ccRCC specimens to further investigate the transcriptomic results on the proteome level. The differential protein abundance was evaluated using targeted mass spectrometry (MS). We assembled a database of 558 renal tissue samples from NCBI GEO and used these to uncover the top genes with higher expression in ccRCC. For protein level analysis 162 malignant and normal kidney tissue samples were acquired. The most consistently upregulated genes were IGFBP3, PLIN2, PLOD2, PFKP, VEGFA, and CCND1 (p < 10-5 for each gene). Mass spectrometry further validated the differential protein abundance of these genes (IGFBP3, p = 7.53 × 10-18; PLIN2, p = 3.9 × 10-39; PLOD2, p = 6.51 × 10-36; PFKP, p = 1.01 × 10-47; VEGFA, p = 1.40 × 10-22; CCND1, p = 1.04 × 10-24). We also identified those proteins which correlate with overall survival. Finally, a support vector machine-based classification algorithm using the protein-level data was set up. We used transcriptomic and proteomic data to identify a minimal panel of proteins highly specific for clear cell renal carcinoma tissues. The introduced gene panel could be used as a promising tool in the clinical setting.
Keywords: biomarker; diagnostics; kidney cancer; mass spectrometry; proteomics.