Fluorescence in situ hybridization (FISH) and array-comparative genomic hybridization (a-CGH) from percutaneous needle biopsy compared to renal mass histology.

471 Background: Image-guided percutaneous needle biopsies are increasingly utilized for the diagnosis of renal tumors. Histologic diagnosis of renal mass subtypes, including malignant clear cell (ccRCC), papillary (pRCC), chromophobe (chrRCC) renal cell carcinoma, and benign oncocytomas (OC) can be challenging due to the low cellularity and damaged architecture of needle biopsy specimens. However, each subtype exhibits unique genetic aberrations that can assist in histologic classification and potentially assist in guiding management decisions. We report our initial experience correlating renal mass histology to subtype-associated genomic alterations detected by FISH and a-CGH of percutaneous needle biopsy specimens. METHODS In an ongoing IRB-approved blinded study, 17 samples obtained from 15 patients with known renal masses were submitted to FISH (FReCaD) and targeted a-CGH (UroGenRA-Kidney). Specimens were classified using a subtype classification decision tree algorithm based on the presence/absence of genomic abnormalities. The results were compared to biopsy histology or surgical specimen when available. RESULTS Histology revealed ccRCC in 6 patients, pRCC in 4, OC in 2, Angiomyolipoma in 1, and unclassified type RCC in 2. In 6 of 9 cases FISH achieved a diagnosis, which correlated with histology in 4. FISH incorrectly classified as ccRCC two cases with pRCC on histology. A-CGH was diagnostic in 14 of 15 cases and correlated with histology in 13. In one case, a-CGH showed a genomic profile not consistent with ccRCC, pRCC, chrRCC, or OC according to the algorithm. CONCLUSIONS The addition of genetic tumor tissue studies to complement histology from biopsy tissues may supplement or improve the accuracy of classification and biological characterization of renal tumor biopsies and influence treatment planning. In our initial experience, a-CGH showed better correlation with histology in subtype classification of malignant and benign renal masses than FISH. Prospective testing will be required to validate these results.