Many recent studies have described miRNAs expression profile in RCC and adjacent non-tumoral tissue confirming the different pattern. Although significant overlap between miRNAs identified by independent studies exist, regarding the number and type of up-/down-regulated miRNAs, data are in part conflicting [1, 2, 24, 25]. Circulating miRNAs may possibly serve as a new class of powerful and non-invasive biomarkers for RCC patients. Regarding this hypothesis, we compared the miRNAs expression profiles of RCC patients' and healthy controls' serum samples. Based on the qRT-PCR arrays approach we were able to describe 30 differentially expressed miRNAs, 19 of these miRNAs were up-regulated and 11 were down-regulated. Such expression profile is a bit controversial, as the global miRNA down-regulation in RCC tissue samples has been repeatedly described [1–4]. The possible relation of miRNA levels between tissue and corresponding serum is still not clearly understood - some researchers postulate that circulating miRNAs may not always be directly associated with the changes occuring in tumor tissues but may also reflect indirect effects and could be secreted by non-tumoral cells. Also mechanisms were described how extracellular miRNAs can potentially interact with recipient cells via a number of different processes, including: direct fusion, internalization or receptor-mediated interactions. There are likely to be other mechanisms, especially for vesicle-free miRNAs, but all await further investigation to provide convincing evidence of their involvement in inter-cellular miRNA exchange [26, 27]. However, this proposed potential of tumor cells to actively uptake miRNAs from circulation can partly explain the opposite trends of miRNA expression levels in tissue compared to blood serum. Several dysregulated miRNAs identified in our study have been described to have altered expression levels in plasma or serum of various cancers recently: up-regulated mir-425* [28, 29], let-7a  or let-7f .
In the validation phase of this study, we tested 3 candidate miRNAs (miR-378, miR-451, miR-150) in the independent cohort of RCC patients. The up-regulation of miR-378 and down-regulation of miR-451 expression levels between serum of RCC patients and healthy controls reached statistical significance in validation study. Analytical characteristics of miR-378 (sensitivity of 70%, specificity of 60%) and miR-451 (sensitivity of 81%, specificity of 77%) proved that both miR-378 and miR-451 are potent in discriminating RCC from healthy control serum. Furthermore, combination of serum miR-378 and miR-451 levels, yielded sensitivity of 81% and specificity 83%, proved to be even more powerful discrimination tool. To our knowledge, the only study concerning circulating miRNAs in renal cell carcinoma identified circulating miR-1233 as a potential biomarker for RCC patients, but although they performed the validation phase on a large cohort of RCC patients from three different study centers, the diagnostic information was below their expectations (AUC of 0.588, sensitivity of 77%, specificity of 37.6%) .
Although our observations are promising, and miR-378/miR-451 analytical characteristics reached values for clinical utility, large-scale prospective studies aiming their evaluation in the renal benign neoplasms and early stages of the RCC are necessary to validate them as biomarkers for early diagnosis, and the analysis of blood serum collections from each patient to evaluate miRNA biomarker dynamics are needed to prove their potential for early detection of relapse in RCC patients.