Diagnostic ability of four cell-free miRNA signatures pre- and post-nephrectomy concordantly found in the tumor and blood from patients with renal cell carcinoma.

Abstract

e16577 Background: Renal cell carcinoma (RCC) has shown an increase in incidence based on continued incidental finding of these tumors by imaging. There is a need for reliable biomarkers like MicroRNAs (miRNA) that are released by the tumor cells and can be detected in assays using blood or urine samples. The first aim of the present pilot study is to determine the diagnostic ability of cell-free miRNA (cfmiR) biomarkers released by RCC tumor cells in urine and plasma samples. The secondary aim was to determine cfmiRs utility in monitoring RCC before and after radical or partial nephrectomy. Methods: We profiled tumor tissues (n = 11), pre-operative (pre-P n = 18; pre-U n = 17) and post-operative (post-P n = 18; post-U n = 17) plasma and urine paired samples from 18 RCC patients with a median follow-up of 18.4 months. As a control, we utilized plasma (n = 73) and urine (n = 16) samples taken from normal healthy donors (NHD). All specimens (n = 170) were processed and analyzed using HTG EdgeSeq miRNA whole transcriptome assay. All of the samples were normalized and DESeq2. Only miRNAs with a FC < -1.5 or > 1.5, FDR < 0.05, normalized counts > 30 were considered Results: We assessed urine, plasma, and tissue for 2083 miRNAs. The pre-U profiles from patients with RCC and NHD were compared to find differentially expressed (DE) cfmiRs. We found 182 cfmiRs DE in pre-U RCC, of which 106 were upregulated and 76 were downregulated. Similarly, we found 830 cfmiRs DE in the pre-P from RCC compared to NHD, of which 192 were upregulated and 638 were downregulated. We then searched for the top 100 miRNAs most frequently detected and identified in the tumor and in pre-P and pre-U samples. Forty miRNAs were consistently found and highly detected in all of the specimens. Of those 40 miRNAs, 33 cfmiRs were found DE in pre-P and 9 cfmiRs significantly decreased in post-P samples after surgery to the level values observed in the plasma from NHD. In the pre-P and pre-U samples from RCC patients, let-7a-5p, let-7b-5p, miR-23b-3p, and miR-30d-5p were found to be consistently upregulated compared to their respective controls. By using receiving operating characteristic (ROC) curves we assessed the area under the curve (AUC) of all the four cfmiRs in detecting RCC patients. The values of AUC for the four cfmiRs detected in pre-P ranged from 76.2-81% [sensitivity, 61.1-83.3%; specificity, 74-86.3%] and in pre-U samples ranged from 76.1-82.4% [sensitivity, 64.7-70.6%; specificity, 100%]. We observed that the four cfmiRs significantly decreased in the post-U samples from RCC patients after surgery to the level values observed in urine from NHD. Conclusions: Our results propose a four cfmiR signature as a potential diagnostic/monitoring urine biomarker that is also detectable in the plasma and tumor tissues from RRC. Further studies to validate these cfmiRNAs as biomarkers for RCC in blood and urine are ongoing.