Chapter 22 - Downregulation of the Raf kinase inhibitory protein (RKIP) in clear cell renal cell carcinoma associates with poor prognosis

Abstract

Clear cell renal cell carcinoma (ccRCC) is the major cause of kidney cancer death. Downregulation of the Raf kinase inhibitory protein (RKIP) tumor suppressor likely contributes to ccRCC progression and fatality; however, this possibility has not been thoroughly examined. By analyzing publically available ccRCC datasets (TCGA and others), we report here a robust association of RKIP downregulation with ccRCC progression. Reductions in RKIP mRNA expression occur following the oncogenic processes from normal kidney tissues (n=23) to ccRCC (n=23) and to distant metastases (n=9). Using the ccRCC dataset (n=530) organized by KM Plotter and the TCGA cohort (n=533), we detected associations of RKIP downregulation with fatality (P=1.3e−07 and 2e−06 respectively) and recurrence (TCGA, P=3e−05). High levels of RKIP expression are an independent predictor of long-term survival (HR 0.541, 95% CI 0.384–0.761, P=.000427) and delays in recurrence (HR 0.56, 95% CI 0.374–0.84, P=.005052) after adjusting for age at diagnosis, tumor grade, gender, tumor size (T), and metastasis. CcRCCs with high RKIP expression are reversely associated with adverse tumor features: grade (OR 0.626, 95% CI 0.397–0.974, P=.0398), tumor size (T) (OR 0.47, 95% CI 0.303–0.729, P=.000751), and distant metastasis (OR 0.535, 95% CI 0.313–0.935, P=.0245). Additionally, RKIP downregulations display an overlap with typical ccRCC mutations in VHL, PBRM1, SETD2, and BAP1 particularly in ccRCCs at risk of fatality. Furthermore, 10 co-expressed genes relative to RKIP reduction have been identified (Spearman's correlation ≥0.62); alterations in all 10 genes (KM Plotter) and 7 of the 10 genes (TCGA) are significantly associated with reductions in overall survival. These genes together with RKIP are clustered in three groups and play roles in fatty acid degradation, pyruvate, and other metabolism; these observations are well in line with ccRCC being associated with metabolic changes. Of note, RKIP and its co-expressed genes form a multigene panel (RKIPGNpanel), which robustly stratifies the risk of ccRCC fatality (P=3e−11) and is an independent risk factor of ccRCC death (HR 2.78, 95% CI 1.939–3.999, P=9.63e−06) after adjusting for age at diagnosis, tumor grade, gender, tumor size, and metastasis. Taken together, we provide a comprehensive set of evidence supporting RKIP as a valuable target in ccRCC diagnosis and therapy.