Please note that we have stopped the regular imports of Gene Expression Omnibus (GEO) data into ArrayExpress. This may not be the latest version of this experiment.
E-GEOD-2748 - Transcription profiling of human papillary renal cell carcinoma to identify molecular PRCC subclasses with morphologic correlation
Released on 20 October 2007, last updated on 3 May 2014
BACKGROUND: Despite the moderate incidence of papillary renal cell carcinoma; (PRCC), there is a disproportionately limited understanding of its underlying genetic programs. There is no effective therapy for metastatic PRCC, and patients are often excluded from kidney cancer trials. A morphological classification of PRCC into Type 1 and Type 2 tumors has been recently proposed, but its biological relevance remains uncertain. PATIENTS AND METHODS. We studied the gene expression profiles of 34 cases of PRCC using Affymetrix HGU133 Plus 2.0 arrays (54,675 probe sets) using both unsupervised and supervised analysis. Comparative genomic microarray analysis (CGMA) was used to infer cytogenetic aberrations, and pathways were ranked with a curated database. Expression of selected genes was validated by immunohistochemistry in 34 samples, with 15 independent tumors. RESULTS. We identified 2 highly distinct molecular PRCC subclasses with morphologic correlation. The first class, with excellent survival, corresponded to 3 histological subtypes: Type 1, low-grade Type 2 and mixed Type 1/low-grade Type 2 tumors. The second class, with poor survival, corresponded to high-grade Type 2 tumors (n=11). Dysregulation of G1/S and G2/M checkpoint genes were found in Class 1 and Class 2 tumors respectively, alongside characteristic chromosomal aberrations. We identified a 7-transcript predictor that classified samples on cross-validation with 97% accuracy. Immunohistochemistry confirmed high expression of cytokeratin 7 in Class 1 tumors, and of topoisomerase IIa in Class 2 tumors. CONCLUSIONS. We report 2 molecular subclasses of PRCC, which are biologically and clinically distinct, which may be readily distinguished in a clinical setting. Experiment Overall Design: 34 individual samples analyzed, no technical replicates. Based on histologic features and Fuhrman grading system, we designated four categories of tumors: type 1 (n = 14) characterized by small tumor cells of low Fuhrman grade (2); type 2A (n = 4) characterized by large eosinophilic tumor cells of low Fuhrman grade (2); combined type 1 and 2A (n = 5); and type 2B (n = 11) characterized by large eosinophilic tumor cells with Fuhrman grade.
transcription profiling by array, co-expression, disease state
A molecular classification of papillary renal cell carcinoma. Ximing J Yang, Min-Han Tan, Hyung L Kim, Jonathon A Ditlev, Mark W Betten, Carolina E Png, Eric J Kort, Kunihiko Futami, Kyle A Furge, Masayuki Takahashi, Hiro-Omi Kanayama, Puay Hoon Tan, Bin Sing Teh, Chunyan Luan, Kim Wang, Michael Pins, Maria Tretiakova, John Anema, Richard Kahnoski, Theresa Nicol, Walter Stadler, Nicholas G Vogelzang, Robert Amato, David Seligson, Robert Figlin, Arie Belldegrun, Craig G Rogers, Bin Tean Teh. Cancer Res 65(13):5628-37 (2005)