The Diagnostic Utility of Novel Immunohistochemical Marker ERG in the Workup of Prostate Biopsies With “Atypical Glands Suspicious for Cancer”

A diagnosis of “atypical glands suspicious for cancer” (ATYP) in prostate needle biopsy is associated with a 40% to 50% risk of finding prostate carcinoma (PCa) in subsequent biopsies. Many studies have attempted to identify clinical, histologic, or molecular characteristics of ATYP that correlated with the risk of PCa in follow-up biopsies. TMPRSS2:ERG gene rearrangement is the most common chromosomal alteration and is highly specific for PCa. Recently, 2 studies reported that positive immunohistochemical (IHC) stains with an ERG antibody highly correlated with the TMPRSS2:ERG gene rearrangement status. We evaluated the use of this antibody as an IHC marker on prostate biopsies with an initial ATYP diagnosis to determine whether positive ERG IHC was associated with increased PCa detection in subsequent biopsies, which therefore might be useful for stratifying ATYP prostate biopsies. ERG IHC was performed on 103 biopsies with initial ATYP diagnosis. Positive ERG IHC staining was detected in 16 of the 103 cases (15.5%) of the ATYP prostate biopsies. Of these 16 ERG-positive cases, the atypical glands were positive for ERG in 9 cases. In the remaining 7 cases, positive ERG staining was found in glands other than ATYP glands, including high-grade prostatic intraepithelial neoplasia and morphologically benign glands. ERG IHC was negative in other benign prostate lesions, including simple atrophy, partial atrophy, proliferative inflammatory atrophy, basal cell hyperplasia, postatrophic hyperplasia, and squamous metaplasia. In subsequent follow-up biopsies, PCa was detected in 7 of the 16 (43.8%) ERG-positive cases and in 42 of the 87 (48.3%) ERG-negative cases (P=0.952 by &khgr;2 test). In biopsies with ERG-positive ATYP glands, cancer was found in 5 of 9 (55.6%) cases in subsequent biopsies. This is the first study to investigate the use of ERG IHC in difficult prostate biopsies. ERG IHC was positive in a small percentage (15.5%) of the ATYP prostate biopsies, and positive ERG staining did not correlate with the increased cancer detection in subsequent prostate biopsies. Therefore, ERG IHC is not useful for stratifying ATYP prostate biopsies to identify patients who have increased risk for PCa in repeat biopsies. Furthermore, positive ERG staining is not entirely specific for PCa and can occasionally be found in high-grade prostatic intraepithelial neoplasia and benign glands that are not associated with PCa in prostate biopsies.