Abstract
A precise stratification of our patients is essential and can support clinicians to determine the right therapy. The aim of this study was to identify clinically relevant genes using The Cancer Genome Atlas (TCGA) datasets. A comprehensive pan-cancer analysis of 30 distinct tumor entities (N = 9022) identified the largely unknown gene Downstream neighbor of SON (DONSON) to be particularly associated with unfavorable overall survival in clear cell renal cell carcinoma (KIRC). This prognostic potential of DONSON was validated in an independent KIRC cohort via quantitative real-time PCR (n = 152). Further, DONSON protein expression was evaluated via immunohistochemical staining followed by quantitative image analysis using the image analysis software QuPath on a renal cancer tissue microarray (n = 270). Interestingly, DONSON overexpression was preferentially associated with poor survival in 9 of the 30 entities, suggesting tumor-independent oncogenic properties of this largely unknown gene. A particularly strong association of DONSON to an aggressive phenotype was evident in KIRC and proved to be a strong independent predictor of unfavorable overall survival in two additional cohorts on the mRNA and protein level. In our KIRC cell culture model, we observed a substantial attenuation of proliferative activity and migration capacity of the KIRC cells Caki1 and 769p. In conclusion, we identified DONSON as a robust biomarker for risk stratification in KIRC in three independent cohorts and provide evidence that DONSON is linked to a malignant phenotype in the KIRC cell culture model.