miRNAs as diagnostic biomarkers for renal cell carcinoma subtypes by chromogenic in situ hybridization.

Objective: To determine whether a limited number of miRNAs could accurately classify the most common RCC subtypes, oncocytoma and unclassified RCC. In addition, we aimed to determine the diagnostic utility of our miRNA classifier by in situ hybridization (ISH). Methods: We extracted RNA from 93 formalin-fixed paraffin-embedded (FFPE) tissues including 30 ccRCC, 28 pRCC, 30 chRCC, 4 unclassified RCC tumours and 11 oncocytomas. We measured absolute expression of 6 miRNAs by qRT-PCR. Receiver operator characteristic curves were constructed and the areas under the curve were calculated to assess performance. We also tested miRNA expression ISH in an independent set of 98 FFPE renal tumours. Data and results:We developed a two-step miRNA classifier. In the first step, expressions of selected miRNAs were found to discriminate clear cell and papillary RCC from chromophobe RCC and oncocytoma. Two miRNAs were able to discriminate clear cell and papillary RCC from chromophobe RCC and oncocytoma (AUC: 0.91, p<0.0001). In the second step, the absolute expression of miR-126 was used to distinguish clear cell from papillary RCC (AUC: 0.98, p <0.0001) and two miRNAs were used to differentiate chromophobe from oncocytoma (AUC: 0.83, p =0.003). Select miRNAs were also useful in pointing to distinct histological types in unclassified RCC. ISH revealed that miRNAs display a nuclear staining pattern capable of distinguishing ccRCC, pRCC, chRCC and oncocytoma. Conclusions: miRNA expression can distinguish between RCC subtypes and oncocytoma. miRNA assessment by ISH is a clinically useful diagnostic tool that can complement current methods for renal tumour classification. [ABSTRACT FROM AUTHOR]