[1690] Virtual-Karyotyping with SNP Arrays Is a Useful Approach for the Diagnosis of Morphologically Difficult Renal Cell Neoplasms

HJ Kim, SS Shen, LD Truong, JY Ro, AG Ayala, K Alvarez, Z Gatalica, J Bridge, FA Monzon. The Methodist Hospital Research Institute, Houston, TX; Inje University College of Medicine, Sanggye Paik Hospital, Seoul, Korea; Creighton University, Omaha, NE; University of Nebraska, Omaha, NE

Background: Renal epithelial neoplasms are morphologically and clinically heterogeneous and have characteristic chromosomal imbalances that can be used for classification. Even after ancillary testing, approximately 7% of tumors remain uncategorized [renal cell carcinoma (RCC) unclassified], limiting optimal patient management. In this study, we applied virtual-karyotyping derived from SNP microarrays (v-karyotype) to a cohort of morphologically difficult renal epithelial neoplasms to determine if this approach can assist in achieving a specific histopathologic classification.
Design: Twenty-six morphologically difficult cases defined as cases in which seven genitourinary surgical pathologists could not achieve a consensus diagnosis were subjected to v-karyotyping using the Affymetrix GeneChip 10K 2.0 or 250K Nsp mapping arrays. Subsequently, the v-karyotype data obtained was reviewed to determine if it could be diagnostically informative.
Results: Five cases were removed due to 100% concordance among pathologists and 3 cases failed to yield adequate v-karyotypes. Eighteen cases with no consensus diagnosis (13) or a consensus diagnosis of RCC unclassified (5) were selected for final analysis. From the no consensus cases, 8 cases reached a majority-vote diagnosis and 5 cases did not. All cases without a majority-vote diagnosis or with a diagnosis of RCC unclassified were able to be classified based on the presence of characteristic chromosomal abnormalities (e.g. loss of 3p for clear cell RCC). Virtual karyotypes confirmed the majority-vote diagnosis in 6 of 8 cases (3 clear cell with -3p, 2 papillary with +7/+17 and 1 chromophobe with loss of multiple chromosomes); two cases showed disagreement with the majority-vote: 1 case with papillary morphology and loss of 3p on the v-karyotype and one oncocytoma in which the v-karyotype was not characteristic of any reported subtype.
Conclusions: Our results show that morphologically difficult renal neoplasms can be classified based on the chromosomal imbalances detected by virtual-karyotyping with SNP arrays. Virtual karyotyping is a useful approach for the diagnosis of renal tumors, including cases diagnosed as RCC unclassified and yields classification information that can be used to determine optimal patient management.
Category: Special Category for 2009 - Pan-genomic/Pan-proteomic approaches to Cancer

Monday, March 9, 2009 2:00 PM

Platform Session: Section G, Monday Afternoon