TMPRSS2-ERG, the most common gene fusion in prostate cancer, is associated with expression of a truncated protein product of the oncogene ERG. A novel anti-ERG monoclonal antibody has been recently characterized. We investigated the correlation between ERG rearrangement assessed by fluorescence in situ hybridization (FISH) and ERG expression detected by immunohistochemistry in a large cohort of patients treated with radical prostatectomy for clinically localized prostate cancer. Thirteen tissue microarrays comprising 305 tumors and a subset of 112 samples of nonneoplastic prostatic tissue were assessed for ERG rearrangement status by FISH and for ERG expression by immunohistochemistry. Accuracy of ERG detection by immunohistochemistry in predicting ERG status as assessed by FISH (criterion standard) was calculated in terms of sensitivity, specificity, positive and negative predictive values. Of 305 tumor foci, 103 (34%) showed ERG rearrangement by FISH. ERG was detected by immunohistochemistry in 100 (33%) cases, 99 of which were FISH positive. ERG detection by immunohistochemistry demonstrated a sensitivity and specificity of 96% and 99%, respectively, with positive and negative predictive values of 99% and 98%, respectively. None of the 112 samples of nonneoplastic prostatic tissue was rearranged by FISH or showed any ERG expression. In conclusion, ERG detection by immunohistochemistry in prostate cancer was highly predictive of ERG rearrangement as assessed by FISH in a large cohort of prostatectomy patients. Given the high yield and the easier task of performing immunohistochemistry vs. FISH, ERG assessment by immunohistochemistry may be useful for characterizing ERG status in prostate cancer.