Comparison of Immunohistochemistry with FISH for Detection of TMPRSS2-ERG Rearrangement in Prostate Cancers in African American and Caucasian Populations.
Robin M Elliott, Margaret A Leversha, Jaya M Satagopan, Qin C Zhou, Maria Dudas, Ying-bei Chen, Hikmat Al-Ahmadie, Samson Fine, Satish Tickoo, James A Eastham, Peter T Scardino, Michael Koscuiszka, Peng Lee, Iman Osman, William L Gerald, Victor E Reuter, Anuradha Gopalan. Memorial Sloan-Kettering Cancer Center, New York; New York University, NY
Background: Gene rearrangements of TMPRSS2 and ERG occur in approximately 13% of prostate cancers in African-Americans (AA) and 30-50% of those in Caucasians (CA). Screening for the rearrangement may provide prognostic information and may be used for patient stratification to guide treatment. Because screening by FISH may be expensive and time consuming, an immunohistochemical method for screening would be advantageous, especially in the AA population, which has a significantly lower prevalence of TMPRSS2-ERG rearrangement.
Design: The study cohort included 87 AA and 71 CA men. Gene rearrangement status was determined by interphase FISH using a 3 color break-apart probe containing BAC clones against 3'ERG, 3' and 5' TMPRSS2. FISH data were available for 78 AA and 67 CA men. Overexpression of the truncated ERG protein product was assessed using a primary monoclonal antibody (dilution 1:250) obtained from Epitomics (San Diego, CA). ERG antibody (ERG AB) data were available for 87 AA and 71 CA men. FISH was not evaluable in 9 AA and 4 CA cases due to technical difficulties. A pathologist blinded to the results of the FISH analysis assessed nuclear reactivity in tumor cells and scored reactivity as positive or negative.
Results: ERG expression by IHC was shown in 22% (19 of 87) of AA men and 42% (30 of 71) of CA men. Gene rearrangement was shown by FISH in 13% (10 of 78) of AA men and 37% (25 of 67) CA men. All AA cases were rearranged through deletion. IHC data were available for all 78 AA men who had data by FISH, and in these, 10 cases were positive by both ERG AB and FISH, while 7 cases were positive by ERG AB and negative by FISH. Both IHC and FISH data were available for 67 CA men, and in these, 25 cases were positive by both ERG AB and FISH, while 5 were positive by ERG AB and negative by FISH. None of the ERG AB negative cases were positive by FISH in either group. There was no difference in intensity of staining between tumors with translocation and those with deletion. FISH and ERG AB were discordant in 41% of AA cases.
Conclusions: 1. In patients with positive FISH, 100% showed ERG AB expression.
2. Fewer cases were 'not evaluable' by the ERG AB assay.
3. The value of FISH is in the distinction between translocation and deletion; however in the AA population this is probably unnecessary because all cases are rearranged through deletion.
4. Our study validates the difference in ERG expression between AA and CA patients.
Category: Genitourinary (including renal tumors)
Monday, February 28, 2011 1:00 PM
Poster Session II # 102, Monday Afternoon