Virtual karyotyping with SNP microarrays reduces uncertainty in the diagnosis of renal epithelial tumors

Jill M Hagenkord,Anil V Parwani,Leif Peterson,Jose M Gonzalez-Berjon,Federico A Monzon,Rajiv Dhir,Robert Amato,Karla Alvarez,Zoran Gatalica,Maureen A Lyons-Weiler

doi:10.1186/1746-1596-3-44

Abstract

BackgroundRenal epithelial tumors are morphologically, biologically, and clinically heterogeneous. Different morphologic subtypes require specific management due to markedly different prognosis and response to therapy. Each common subtype has characteristic chromosomal gains and losses, including some with prognostic value. However, copy number information has not been readily accessible for clinical purposes and thus has not been routinely used in the diagnostic evaluation of these tumors. This information can be useful for classification of tumors with complex or challenging morphology. 'Virtual karyotypes' generated using SNP arrays can readily detect characteristic chromosomal lesions in paraffin embedded renal tumors and can be used to correctly categorize the common subtypes with performance characteristics that are amenable for routine clinical use.MethodsTo investigate the use of virtual karyotypes for diagnostically challenging renal epithelial tumors, we evaluated 25 archived renal neoplasms where sub-classification could not be definitively rendered based on morphology and other ancillary studies. We generated virtual karyotypes with the Affymetrix 10 K 2.0 mapping array platform and identified the presence of genomic lesions across all 22 autosomes.ResultsIn 91% of challenging cases the virtual karyotype unambiguously detected the presence or absence of chromosomal aberrations characteristic of one of the common subtypes of renal epithelial tumors, while immunohistochemistry and fluorescent in situ hybridization had no or limited utility in the diagnosis of these tumors.ConclusionThese results show that virtual karyotypes generated by SNP arrays can be used as a practical ancillary study for the classification of renal epithelial tumors with complex or ambiguous morphology.

Highlights

Renal epithelial tumors are morphologically, biologically, and clinically heterogeneous
Classic morphology cohort The chromosomal copy number aberrations and loci of loss of heterozygosity (LOH) derived using the SNP arrays were in agreement with those reported in the literature for each subtype
*Array Results vs. Final Diagnosis field summarizes the potential impact on diagnosis had the SNP array karyotype been available

Summary

Introduction

Renal epithelial tumors are morphologically, biologically, and clinically heterogeneous. Copy number information has not been readily accessible for clinical purposes and has not been routinely used in the diagnostic evaluation of these tumors. This information can be useful for classification of tumors with complex or challenging morphology. 'Virtual karyotypes' generated using SNP arrays can readily detect characteristic chromosomal lesions in paraffin embedded renal tumors and can be used to correctly categorize the common subtypes with performance characteristics that are amenable for routine clinical use. Morphological distinction among the renal cell tumor subtypes is generally straightforward after routine pathologic examination of the tissue. A subset of cases are diagnosed as renal cell carcinoma, unclassified, due to the presence of non-specific features that can be seen in all of the subtypes (such as granular, oncocytic or sarcomatoid morphology) [3] or the presence of more than one subtype in the same tumor [2]

Objectives

Methods

Results

Conclusion