Evaluation of a Novel ERG Antibody in Prostate Cancer and Correlation with TMPRSS2-ERG Gene Rearrangement Status by FISH, ACGH and ERG mRNA Expression.
Anuradha Gopalan, Maria Dudas, Margaret A Leversha, Barry S Taylor, Nikolaus Schultz, Haley Hieronymous, Yingbei Chen, Hikmat Al-Ahmadie, Samson W Fine, Satish K Tickoo, Charles L Sawyers, William L Gerald, Victor E Reuter. MSKCC, New York
Background: TMPRSS2-ERG gene rearrangements are present in 30-50% of prostate cancer (PC) and lead to overexpression of a truncated ERG protein. The presence of the rearrangement may have prognostic significance and assist in patient stratification to guide therapy. An antibody based test for detection of rearrangement would therefore have significant clinical utility. We studied a recently characterized ERG antibody on a well annotated set of prostate cancer samples in which FISH, copy number and ERG mRNA expression data were available.
Design: Immunohistochemistry for ERG protein using a monoclonal antibody (clone EPR3864; Epitomics, Burlingame, CA) was performed on tissue microarrays containing 145 cases of PC. Positive or negative nuclear expression and intensity of staining (1-3) was recorded. TMPRSS2-ERG gene rearrangement status was determined by interphase FISH using a 3 color breakapart probe containing BAC clones against 3'ERG, 3' and 5' TMPRSS2. FISH data were available in 95 cases. In 50 cases with both ERG antibody (ERG AB) and FISH data, TMPRSS2-ERG rearrangement status derived using combined aCGH and whole-transcript outlier expression inferred from exon expression arrays was also available for correlation.
Results: Of 94 cases where FISH and ERG AB data were available, 3 (3%) were ERG (+) and FISH (-). One of these had expression and copy data available, both of which supported the ERG AB result. None of the ERG AB (-) cases were FISH (+). No difference in staining intensity was observed between cases rearranged by translocation or deletion. ERG mRNA overexpression was present in 48 cases in the dataset. 45 were evaluable, of which 37 were ERG AB (+) and 8 ERG AB (-). Four of the 8 ERG AB (-) cases were evaluable by FISH and all were negative for the rearrangement while all 8 were fusion positive by aCGH. The discrepancy between ERG AB/ FISH and aCGH/ expression is likely due to tumor heterogeneity and sampling.
Conclusions: Antibody based detection of TMPRSS2-ERG rearrangement shows a high concordance with FISH
The greater discordance between ERG AB and mRNA/ aCGH likely reflects tumor heterogeneity and different areas sampled for genomic studies and TMA based assays
Category: Genitourinary (including renal tumors)
Monday, February 28, 2011 8:00 AM
Platform Session: Section A, Monday Morning