Key points are not available for this paper at this time.
Abstract Background Renal cell carcinoma (RCC), the most common malignancy of the renal epithelium, accounts for over 90% of kidney cancer cases, with clear cell RCC (ccRCC) representing about 75% of these diagnoses. MicroRNAs (miRNAs), short non-coding RNAs of 21 to 23 nucleotides, play significant roles in RNA silencing and post-transcriptional regulation of gene expression. They are also crucial in tumorigenesis. We aimed to identify differentially expressed miRNA between ccRCC tumor and normal-adjacent samples, aiming to find miRNA signatures for ccRCC diagnostics and prognosis. Methods We collected tumor and normal-adjacent kidney samples from the Dartmouth Renal Tumor Biobank spanning 1994 to 2009. Samples were mechanically and enzymatically disassociated and preserved at -80 Celsius until processing. The single-stranded sequencing data was processed using the smrnaseq (v2.2.4) Nextflow analysis pipeline to map miRNAs to the reference database. The preprocessed counts data was imported into R (v4.3.3). We employed strict quality control measures excluding samples with low counts with a mean log-transcripts per million of less than -2, a read length below 25% within the range of 20-25 nucleotides, a microRNA fraction less than 20%, non-primate reads exceeding 70%, or a mapping rate greater than 1.5 times the interquartile range. Differential expression analysis was conducted using the Bioconductor DESeq2 package (v1.42.1), comparing tumor and normal samples as well as across different survival time categories. miRNAs with a false discovery rate (FDR) 0.05 and log2 fold change 2 were considered differentially expressed (DE) after adjustment for batch effects, tumor grade, sex, and tumor stage. An overrepresentation enrichment analysis was performed for the DE microRNAs using the miRNA Enrichment Analysis and AnnotationTool (miEAA). Survival analysis, utilizing stepwise selection in both directions, was conducted using identified DE miRNAs while adjusting for the covariates above plus age at diagnosis. Mature and hairpin miRNAs were analyzed separately for DE and survival. Results Among 213 samples with mature microRNA data, 198 samples (excluding one control) passed the quality control. These comprised 153 tumor samples and 45 normal samples, originating from a cohort of 159 patients. Among these patients, 13.8% have survived longer than 5 years, while 51.6% survived between 1 to 5 years, and 34.6% less than 1 year. 35 mature microRNAs were found to be differentially expressed between tumor and normal samples. Three mature miRNAs were under-expressed among samples with longer survival compared to those with short to medium survival. The DE mature miRNAs were overrepresented in vascular diseases (FDR=4.44 E-4), carcinoma (FDR=1.87 E-3) and clear cell renal cell carcinoma (FDR=5.65 E-3). miR-187-3p, miR-224-5p, miR-155-5p, miR-514a-3p, miR-490-5p were significantly associated with survival after adjustment (p 0.05). Out of the 219 samples analyzed for hairpin microRNA, 200 samples (excluding one control) passed quality control (155 tumor samples and 45 normal samples from 160 patients). Among these patients, 13.7% survived longer than 5 years, 51.9% survived between 1 to 5 years, and 34.4% survived less than 1 year. Differential expression analysis identified 28 hairpin microRNAs showing significant differences between tumor and normal samples. Notably, the differentially expressed hairpin microRNAs were overrepresented in renal cell carcinoma (FDR=5.38e-4) and renal clear cell carcinoma (FDR=0.01), as well as pathways associated with transcription factor TGFB1 (FDR=7.8×e−3), while being underrepresented in pathways associated with transcription factor KDM5B (FDR=9.5×e−3). Furthermore, hsa-mir-21, hsa-mir-155, hsa-mir-1291, and hsa-mir-506 were found to be significantly associated with survival after adjustment (p 0.05). Conclusions This exploratory analysis identified distinct miRNA differential expression between ccRCC tumors and normal tissue, suggesting a potential miRNA signature for ccRCC. Our findings will be further explored using miRNA clustering analysis, functional analysis, network analysis, and differential transcript usage analysis. DOD CDMRP Funding: yes
Zhang et al. (Mon,) studied this question.