Premium
Expression profiling of microdissected matched prostate cancer samples reveals CD166/MEMD and CD24 as new prognostic markers for patient survival
Author(s) -
Kristiansen Glen,
Pilarsky Christian,
Wissmann Christoph,
Kaiser Simone,
Bruemmendorf Thomas,
Roepcke Stefan,
Dahl Edgar,
Hinzmann Bernd,
Specht Thomas,
Pervan Janja,
Stephan Carsten,
Loening Stefan,
Dietel Manfred,
Rosenthal André
Publication year - 2005
Publication title -
the journal of pathology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 2.964
H-Index - 184
eISSN - 1096-9896
pISSN - 0022-3417
DOI - 10.1002/path.1676
Subject(s) - prostate cancer , immunohistochemistry , prostate , biology , cd24 , gene , gene expression profiling , gene expression , proportional hazards model , pathological , pathology , cancer research , pca3 , cancer , oncology , medicine , phenotype , genetics
Abstract In order to screen for differentially expressed genes that might be useful in diagnosis or therapy of prostate cancer we have used a custom made Affymetrix GeneChip containing 3950 cDNA fragments. Expression profiles were obtained from 42 matched pairs of mRNAs isolated from microdissected malignant and benign prostate tissues. Applying three different bioinformatic approaches to define differential gene expression, we found 277 differentially expressed genes, of which 98 were identified by all three methods. Fourteen per cent of these genes were not found in other expression studies, which were based on bulk tissue. Resultant candidate genes were further validated by quantitative RT‐PCR, mRNA in situ hybridization and immunohistochemistry. AGR2 was over‐expressed in 89% of prostate carcinomas, but did not have prognostic significance. Immunohistologically detected over‐expression of MEMD and CD24 was identified in 86% and 38.5% of prostate carcinomas respectively, and both were predictive of PSA relapse. Combined marker analysis using MEMD and CD24 expression proved to be an independent prognostic factor (RR = 4.7, p = 0.006) in a Cox regression model, and was also superior to conventional markers. This combination of molecular markers thus appears to allow improved prediction of patient prognosis, but should be validated in larger studies. Copyright © 2004 Pathological Society of Great Britain and Ireland. Published by John Wiley & Sons, Ltd.