Identification of small molecule drugs and development of a novel autophagy-related prognostic signature for kidney renal clear cell carcinoma

Abnormal autophagic levels have been implicated in the pathogenesis of multiple cancers, however, its role in tumors is complex and has not yet been explored clearly. Hence, we aimed to explore the prognostic values of autophagy-related genes (ARGs) for kidney renal clear cell carcinoma (KIRC). Differentially expressed ARGs and transcription factors (TFs) were identified in KIRC patients obtaining from the The Cancer Genome Atlas (TCGA) database. Then, networks between TFs and ARGs, gene ontology functional annotations and Kyoto Encyclopedia of Genes and Genomes pathway enrichment analysis were conducted. Next, we performed consensus clustering, COX regression analysis and Lasso regression analysis to identify the prognostic ARGs. Finally, an individual prognostic index (PI,riskScore) was established. Based on TCGA cohort and ArrayExpress cohort, Survival analysis, ROC curve, independent prognostic analysis, and clinical correlation analysis were also performed to evaluate this PI. Based on differentially expressed ARGs, KIRC patients were successfully divided into two clusters (P = 5.916e-04). AS for PI, it was constructed based on 11 ARGs and significantly classified KIRC patients into high-risk group and low-risk group in terms of OS (P = 4.885e-15 for TCGA cohort,P = 6.366e-03 for ArrayExpress cohort). AUC of its ROC curve reached 0.747 for TCGA cohort and 0.779 for ArrayExpress cohort. What's more, this PI was proven to be a valuable independent prognostic factor in both univariate and multivariate COX regression analysis (P < .001). Prognostic nomograms were also performed to visualize the relationship between individual predictors and survival rates in patients with KIRC. By means of connectivity map database, emetine, cephaeline and co-dergocrine mesilate related to ARGs were found to be negatively correlated with KIRC. This study provided an effective PI for KIRC and also displayed networks between TFs and ARGs. KIRC patients were successfully divided into two clusters based on differentially expressed ARGs. Besides, small molecule drugs related to ARGs were also identified for KIRC.