Advances in Anatomic Pathology:
Clinical Applications of Novel ERG Immunohistochemistry in Prostate Cancer Diagnosis and Management
Shah, Rajal B. MD
Division of Urologic Pathology, Miraca Life Sciences Research Institute, Irving, TX
The author has no funding or conflicts of interest to disclose.
All figures can be viewed online in color at http://www.anatomicpathology.com.
Reprints: Rajal B. Shah, MD, Division of Urologic Pathology, Miraca Life Sciences Research Institute, 6655 North MacArthur Blvd, Irving, TX 75039 (e-mail: rshah@MiracaLS.com).
TMPRSS2:ERG gene fusions, the most common molecular subtype of ETS family gene fusions occur in ∼50% of prostate carcinomas (PCas) and ∼20% of high-grade prostatic intraepithelial neoplasia (HGPIN) intermingled with adjacent PCa demonstrating identical gene fusions. ERG gene fusions have not yet been demonstrated in isolated benign prostate tissue, isolated high-grade prostatic intraepithelial neoplasia, or benign cancer mimics. Taken together, ERG gene fusions are the most prostate cancer-specific biomarker yet identified and define a specific molecular subtype of PCa with important clinical and biological implications. ERG gene fusions result in the overexpression of a chimeric fusion transcript that encodes a truncated ERG protein product. Recently, N-terminal epitope-targeted mouse (9FY) and C-terminal-targeted rabbit monoclonal (EPR 3864) ERG antibodies are commercially available and are increasingly utilized as a surrogate for TMPRSS2:ERG gene fusions. Until recently, because of lack of availability of reliable ERG antibody, the most commonly utilized methods for studying ERG aberrations in PCa specimens included fluorescence in situ hybridization or reverse transcriptase polymerase chain reaction. The knowledge gleaned from these studies has significantly improved our understanding of molecular biology of ERG gene fusions. With availability of highly specific anti-ERG monoclonal antibodies, there are now unprecedented opportunities to explore and validate clinical applications of ERG antibody in routine pathology practice, which has just started. This review provides a brief background of molecular biology of ERG gene fusions in PCa and focuses on characterizing the current state of ERG oncoprotein and determining the role of ERG immunohistochemistry in the diagnosis and biological stratification of prostate cancer.
BIOLOGY OF ETS GENE FUSIONS IN PROSTATE CANCER
Recurrent chromosomal rearrangements in prostate carcinoma (PCa) were discovered through the power of an unconventional bioinformatics approach termed as the “Cancer Outlier Profile Analysis” algorithm to analyze DNA microarray studies.1,2 Using the results of Cancer Outlier Profile Analysis of many prostate cancer profiling studies, in 2005, Tomlins et al2 for the first time discovered recurrent chromosomal rearrangements in PCa that demonstrated fusion of the 5′ untranslated region of the androgen-regulated gene TMPRSS2 with ERG or ETV1, 2 members of ETS transcription factor family genes. Subsequent studies confirmed ETS gene fusions in the majority (∼50%) of prostate-specific antigen-screened PCa surgical cohorts.1,3–5 Fusions between TMPRSS2 and ERG represent ∼90% of all ETS gene fusions.1–3,5 In addition to the most common TMPRSS2:ERG rearrangements, several other novel 5′ promoter or other upstream sequences of androgen-inducible genes (HERV_K22q11.23, SLC45A3, C15orf21, HNRPA2B1, KLK2, CANT1) and 3′ ETS transcription factors genes (ETV4, ETV5, and ELK4) have also been identified, which comprise about 5% to 10% of all gene fusions in PCas.1,6,7 Previous studies have also demonstrated that rearrangements of ERG at the chromosomal level are highly specific to PCa and it is an early molecular event seen in ∼18% of high-grade prostatic intraepithelial neoplasia (HGPIN) immediately adjacent to cancer demonstrating identical gene fusions.8–11 HGPIN lesions expressing TMPRSS2:ETS rearrangements are invariably associated with invasive cancer, suggesting that they are a subset of true neoplastic precursors for TMPRSS2:ETS-positive PCa. Clinically localized PCa is typically a multifocal disease, with heterogeneous rearrangement for TMPRSS2:ETS fusions between different tumor foci.12 In this schema of multifocal disease, a primary focus rearranged for TMPRSS2:ETS may progress and become capable of dissemination and give rise to metastatic disease. All metastatic disease foci retain similar TMPRSS2:ETS rearrangement like the primary focus, indicating that ETS rearrangement occurs before progression to metastatic disease and metastatic disease arise through the clonal expansion of a single focus of primary PCa capable of dissemination.13 In summary, ETS gene fusions in combination with PTEN loss or AKT overexpression has been implicated to play a critical role in prostate carcinogenesis.1,14 ERG gene fusions have yet not been demonstrated in benign prostate tissue, isolated HGPIN, or benign cancer mimics.8–11,15 Taken together, ERG gene fusions are the most prostate cancer-specific biomarker yet identified and define a specific molecular subtype of PCa with important implications in the diagnosis and management of PCa.
Until recently, because of lack of availability of reliable ERG antibody, the most commonly utilized methods for studying ERG aberrations in PCa specimens included fluorescence in situ hybridization (FISH) or reverse transcriptase polymerase chain reaction. These approaches have yielded significant body of knowledge in understanding the molecular biology of ERG gene fusions. Availability of highly specific anti-ERG monoclonal antibodies offers now unprecedented opportunities to explore and validate clinical applications of ERG in routine pathology practice from the knowledge learned through the previous studies.
ANTI-ERG ANTIBODY AS A SURROGATE FOR ERG GENE FUSIONS IN PROSTATE CANCER
TMPRSS2:ERG gene fusions result in the overexpression of chimeric fusion transcripts that encode a truncated ERG protein product. Recently, 2 studies reported monoclonal anti-ERG antibodies that could be utilized in immunohistochemistry (IHC) to detect the ERG protein that was present at high level in fusion-positive cancer cells.16,17 Park et al16 characterized a rabbit anti-ERG monoclonal antibody. A positive IHC with this antibody highly correlated with the ERG gene rearrangement status determined by FISH, with 96% sensitivity and specificity for determining ERG rearrangement in PCa.16 Chaux et al18 and Falzarano et al19 recently validated this observation and demonstrated that ERG immunohistochemical expression has a high accuracy for defining the TMPRSS2:ERG fusion status in PCa. The sensitivity and specificity of ERG antibody for determining ERG rearrangement in these studies were 86% and 89% and 96% and 99%, respectively. Van Leenders et al20 found ERG IHC was highly concordant with ERG mRNA overexpression with sensitivity of 100% and specificity of 85%. Table 1 summarizes the published studies showing type of ERG antibody utilized and its sensitivity and specificity for the detection of ERG gene fusions in PCa. Overall, emerging data overwhelmingly suggest that ERG oncoprotein detection in PCa is highly concordant with ERG gene fusion status and can be reliably utilized as a surrogate of ERG gene fusions in prostate cancer diagnosis and management. As these gene fusions are highly specific to prostate cancer, the ERG IHC promises to be a prostate cancer-specific biomarker that can be utilized in routine clinical practice to augment prostate cancer workup.
CLINICAL APPLICATIONS OF ERG IMMUNOHISTOCHEMISTRY
ERG in the Diagnosis of Limited Prostate Carcinoma in Prostate Needle Biopsy
Diagnosis of limited PCa can be challenging in prostate biopsies. Basal cell markers including high–molecular-weight cytokeratin and p63, and prostate cancer marker α-methylacyl-CoA-racemase (AMACR) (P504S), individually or as a part of PIN-4 cocktail, are currently the most commonly utilized IHC markers in clinical practice.22,23 However, these markers are neither 100% sensitive nor specific for the diagnosis of PCa. The diagnosis of PCa has typically relied upon lack of expression of basal cell markers but several noncancerous lesions including partial atrophy, atypical adenomatous hyperplasia (adenosis), and HGPIN, may have an incomplete, or absent basal cell layer.24–26 Similarly, a small proportion of histologically unequivocally low Gleason grade PCa may demonstrate basal cell expression secondary to the retention of basal cells by early invasive carcinoma, because of intermingling of or nearby HGPIN glands or rarely because of expression of p63-positive basal cell phenotype.27,28 AMACR is preferentially overexpressed in ∼80% of PCa detected in prostate biopsies.29,30 However, its expression is also found in the majority of HGPINs, in significant proportion of adenosis (atypical adenomatous hyperplasia), nephrogenic adenoma and partial atrophy, and occasionally even in morphologically benign glands.23,24,29,30 Therefore, a tumor marker that demonstrates better specificity for PCa and is not expressed in noncancerous lesions may complement basal cell markers and AMACR and greatly facilitate the identification of limited PCa in prostate biopsies.
Van Leenders et al20 analyzed clinical utility of ERG IHC in 83 consecutive prostate cancer needle biopsies. ERG protein expression was identified in 51/83 PCa (61%) on needle biopsies. ERG expression was more frequent in tumors infiltrating ≥2 needle biopsies or occupying ≥50% of a single biopsy. Expression of ERG also occurred in 11/21 (52%) high-grade prostate intraepithelial neoplasia lesions. In 5/87 (6%) needle biopsies containing benign secretory glands, weak ERG staining was focally observed. In all of these cases, respective glands were adjacent to PCas. The authors concluded that ERG IHC strongly correlated with ERG mRNA overexpression and was specific for prostate cancer on needle biopsies.20 Yaskiv et al31 studied a total of 77 prostate needle biopsies containing cancer occupying <1 mm of the length of only 1 core of the entire biopsy set and stained with the double stain containing ERG and P63 antibodies. ERG positivity and its staining intensity in cancerous and other noncancerous lesions were evaluated. ERG expression was detected in 42% (32 of 77) of cases, with strong, moderate, and weak staining intensity in 72%, 16%, and 12% of cases. The staining was uniform in 84% of cases and heterogeneous in 16% of cases with different staining intensities in ≥10% of cancerous cells. HGPIN was present in 17 cases, and in 5 (29%) cases ERG was positive in HGPIN glands, which were all immediately adjacent to or intermingled with ERG-positive cancerous glands. In 4 additional cases, positive ERG staining was found in morphologically benign glands, which were also immediately adjacent to or intermingled with ERG-positive cancerous glands. All other benign lesions distant from cancerous glands, including simple and partial atrophy, were negative for ERG. P63 was negative in all cancerous glands and positive in noncancerous lesions. The authors concluded that the P63/ERG double immunostain combines the high sensitivity of P63 and the high specificity of ERG and may be potentially useful in the workup of difficult prostate biopsies.31 Tomlins et al32 evaluated ERG staining by IHC in a large cohort of both retrospective and prospectively collected prostate biopsies and demonstrated positivity in 44% of PCa, 18% of HGPIN, and 11% of atypical foci. In positive cancer foci, ERG was expressed uniformly in almost all cases, and ERG staining was exceedingly rare in benign glands. The authors concluded that overall, ERG seems to be more specific than AMACR for PCa; hence, ERG staining in atypical focus (where HGPIN or PINATYP can be excluded) supports a diagnosis of PCa, irrespective of AMACR staining.32 In summary, these studies utilizing ERG IHC in diagnostic settings, validate previously published genomic findings and support that ERG has high specificity for the prostate cancer detection with important implications for prostate biopsy interpretation. A representative example of limited PCa stained with ERG antibody is represented in Figure 1.
Utility of ERG in Resolving an Atypical Glands Suspicious for Prostate Cancer (ATYP) Diagnosis
One of the challenges encountered during biopsy evaluation is the diagnosis of atypical glands suspicious for cancer (ATYP), which typically require IHC for further diagnostic workup. Only a few studies have examined the significance of ERG in this setting.32,33 We studied 84 ATYP cases using multiplex ERG/AMACR/high–molecular-weight cytokeratin/p63 IHC to determine clinical utility of ERG in resolving an ATYP diagnosis.33 A final diagnosis of benign, ATYP and cancer was rendered after the review of morphology and all markers in 3, 30, and 51 cases, respectively. Of 51 cancer diagnoses, 45% and 94% were positive for ERG and AMACR, respectively. Of 30 atypical diagnoses, 10% and 67% were positive for ERG and AMACR, respectively. Of 3 benign diagnoses, none and 83% were positive for ERG and AMACR, respectively. All ERG-positive atypical cases were classified as “HGPIN with adjacent ATYP.” ERG was expressed in adjacent noncancer glands of 20% of PCas, whereas AMACR was expressed in noncancer glands in all diagnostic categories in 40% cases. In ERG-positive ATYP focus, the expression was predominantly uniform within the focus with minimal staining heterogeneity. Overall, our study further suggests that ERG has a low sensitivity but high specificity for prostate cancer detection. Therefore, ERG positivity in small atypical glands where the diagnosis of HGPIN is excluded, can be utilized to establish a definitive cancer diagnosis in the majority of ATYP cases.
Utility of ERG in Resolving an ATYP Diagnosis Beyond That Provided by Traditional α-Methylacyl-CoA-Racemase and Basal Cell Markers
An important clinical question remains: Is a positive ERG staining used merely to confirm a malignant diagnosis that could otherwise be established based on routine hematoxylin and eosin histology and traditionally utilized PIN-4 cocktail antibodies composed of α-AMACR and basal cell markers? Alternatively, could a positive ERG staining be used to convert an atypical diagnosis to cancer in cases that otherwise would not be diagnostic of PCa based on histology and traditionally utilized AMACR and basal cell markers? In our experience addressed in the earlier section of utility of ERG in resolving an “ATYP” diagnosis, traditionally utilized AMACR and basal markers were adequate to resolve ATYP diagnosis in the vast majority of cases.33 However, because of high specificity of ERG for prostate cancer, ERG positivity in small atypical glands where the diagnosis of HGPIN was excluded helped establish a definitive cancer diagnosis in small proportion of additional ATYP cases where either morphology and/or traditional markers were not deemed adequate enough to offer a definitive cancer diagnosis. In this series, 12/48 (28%) of atypical diagnoses based on morphology, AMACR, and basal cell markers were changed to cancer after incorporating a positive ERG staining. These cases were morphologically suspicious for cancer and in all cases AMACR was expressed but a definitive diagnosis of cancer could not be rendered because of either quantitatively or qualitatively less than optimal morphology or inconclusive basal cell IHC. In 8 of these 12 cases, the number of atypical glands was 3 or 4. In 1 case composed of a relatively large number of infiltrative appearing atypical glands with exclusively small acinar architecture, it demonstrated rare basal cells. In 2 cases, the atypical glands had a partially atrophic appearance and did not have prominent nuclear atypia (representative example shown in Fig. 2). In 1 case the atypical glands were at the edge of a biopsy with staining and biopsy edge artifacts. A positive AMACR staining was not sufficient in these cases to make a definitive cancer diagnosis as AMACR is also known to be expressed in a significant proportion of benign prostate cancer mimics. In this study 67% of lesions which were classified as atypical and noncancer glands in 40% of all diagnostic categories demonstrated AMACR expression, indicating that AMACR expression is not cancer specific and by itself is not sufficient to convert an atypical diagnosis to cancer. On the contrary, ERG expression in small proportion of benign or PIN glands has been predominantly reported in the setting of adjacent cancer; therefore, ERG positivity in small atypical glands where the diagnosis of PINATYP or HGPIN is excluded is virtually diagnostic of PCa.33
ERG-expressing High-grade Prostatic Intraepithelial Neoplasia and ATYP Lesions as a Predictive Biomarker of Prostate Carcinoma Risk Stratification in Subsequent Prostate Biopsy
Previous FISH-based evaluations of the genomic rearrangements associated with PCa development have consistently revealed that about 20% of HGPIN lesions in proximity to PCa with identical ERG gene fusions are also positive for ERG rearrangement.6,10,11 Evaluation of ERG oncoprotein expression in whole mount sections using clone 9FY has revealed a strong concordance between focally ERG-positive PIN and homogenously ERG-positive PCa in 96.5% of cases.8 These findings indicate a clonal relationship between gene fusion-positive HGPIN and PCa and potential implications for utilization of ERG as a marker for PCa risk stratification in patients with HGPIN diagnosis. Two studies have examined the significance of ERG-positive HGPIN for future cancer risk stratification and came to contrasting conclusions. Gao et al34 studied a total of 162 patients with HGPIN with follow-up repeat biopsies and stratified them into 2 groups: one with ERG rearrangements rate ≥1.6% (n=59) and the other with an ERG rearrangement rate <1.6% (n=1=103). A total of 56 (of 59, 94.9%) HGPIN cases with an ERG rearrangement rate ≥1.6% and 5 (of 103, 4.9%) HGPIN cases with an ERG rearrangement rate <1.6% were diagnosed with PCa during repeat biopsy follow-up (P<0.001). Authors concluded that the presence of ERG rearrangement in HGPIN lesions detected on initial biopsy warrants repeat biopsy.34 He et al,35 however, did not find the utility of ERG to stratify cancer risks associated with HGPIN. Patients with initial HGPIN in biopsies and at least 1 follow-up prostate biopsy were included and were immunostained for ERG. The ERG staining results were then correlated with the PCa risk in subsequent biopsies. The repeat biopsies showed PCa and noncancer lesions (benign, HGPIN, atypical glands suspicious for cancer) in 36 patients (38%) and 58 patients (62%), respectively. ERG immunostain was positive in 5 (5.3%) biopsies with HGPIN, in which PCa was found in 2 (40%) subsequent biopsies. Of 89 biopsies with negative ERG staining, PCa was found in 34 (38%) repeat biopsies. The cancer detection rate was not different between ERG-positive and ERG-negative cases (P=0.299). The authors concluded that ERG expression is distinctly uncommon in isolated HGPIN (5.3%) and positive ERG expression is not associated with increased cancer detection in subsequent repeat biopsies.35
A repeat prostate biopsy after 3 to 6 months is recommended for patients with ATYP diagnosis because of its high predictive value for cancer in repeat biopsy. In a study of follow-up biopsies from 103 patients with a preliminary diagnosis of ATYP, ERG expression was detected in 9 patients. Five of these (55.6%) had cancer on repeat biopsies, compared with 48.3% of patients with ERG-negative preliminary biopsies. The authors concluded that ERG expression is unlikely to help identify patients suitable for subsequent biopsies. Of note is that in the study the repeat biopsies were not directed to the ERG-positive ATYP sites.36 Overall, additional biopsy studies assessing ERG-positive HGPIN and ATYP sites are thus needed to evaluate the utility of measuring ERG expression as a PCa risk stratification in subsequent biopsies in patients with HGPIN or ATYP more thoroughly.
Utility of ERG in the Work of Atypical Cribriform Lesion of the Prostate in Needle Biopsy
Atypical cribriform lesion (ACL) of the prostate gland is defined as cribriform or rarely solid proliferation of prostate glands populated with cytologically malignant cells with preservation of basal cells.37,38 It may represent cribriform HGPIN or intraductal carcinoma of the prostate (IDC-P). IDC-P is almost always associated with high-grade and high-volume invasive carcinoma, whereas cribriform HGPIN is a putative neoplastic precursor lesion and recent studies have shown the significance of HGPIN as a predictive marker of cancer has reduced significantly in the range of 25%.38 The diagnosis of HGPIN now does not mandate a repeat biopsy within the first year of diagnosis whether it is focal in nature, defined by <2 cores involvement.39 In addition, PCa with IDC-P component has a significantly worse prognosis than PCa without a component of IDC-P.38 Therefore, the distinction of cribriform PIN from IDC-P is of paramount importance because of its widely differing clinical significance. We evaluated ERG gene fusions in 16 cribriform HGPIN (non–cancer-associated ACLs) and 48 IDC-P lesions (cancer-associated ACLs) in totally embedded radical prostatectomy specimens.15 IDC-P lesions were further subdivided into 2 groups: Group A showing classic criteria of IDC-P, including pleomorphic nuclei and/or intraluminal necrosis and group B which had low-grade features difficult to separate from HGPIN. All isolated cribriform HGPIN lesions lacked ERG gene rearrangements whereas IDC-P lesions regardless of their morphologic spectrum (type A or B) were highly enriched in these gene fusions. ERG gene rearrangement was observed in 75% (36/48) of IDC-P, of which 64% (23/36) were through deletion and 36% (13/36) through insertion. Notably, EDel 2+, a subtype of gene fusions that has been shown to have an adverse outcome,15,40 was identified in 17% (6/36) of the IDC-P in this cohort. Therefore, these data demonstrate that all cancer-associated ACLs, essentially represent an intraductal spread of prostate cancer and ERG IHC has potential utility in stratification of an ACL lesion encountered in prostate biopsy. Essentially, all ERG-expressing ACLs, especially when they are associated with adjacent PCa represent examples of IDC-P.38 A representative example demonstrating utility of ERG IHC in the workup of an ACL in prostate biopsy is represented in Figure 3.
Utility of ERG Immunohistochemistry in the Evaluation of Metastatic Tumor of Unknown Origin
ERG is known to be expressed in endothelial cells, and oncogenic ERG gene fusions occur in subsets of prostatic carcinoma, acute myeloid leukemia, and Ewing sarcoma.41,42 Miettinen et al42 utilized monoclonal ERG antibody, CPDR ERG-MAb in a broad range of vascular endothelial (n=250), other mesenchymal (n=973), and epithelial tumors (n=657) to determine the use of ERG IHC in surgical pathology. Only immunostains with ERG-positive normal endothelia (internal control) were considered valid, and only nuclear staining was considered to be positive. In adult tissues, ERG was restricted to endothelial cells and to a subset of bone marrow precursors, but early fetal mesenchyme and subpopulations of fetal cartilage were also positive. In vascular tumors, ERG was expressed in endothelia of all hemangiomas and lymphangiomas, and typically extensively expressed in 96 of 100 angiosarcomas, 42 of 43 epithelioid hemangioendotheliomas, and all 26 Kaposi sarcomas. Among nonvascular mesenchymal tumors, only blastic extramedullary myeloid tumors (7 of 10) and rare Ewing sarcomas (2 of 29) were positive. Among epithelial tumors, 30 of 66 prostatic adenocarcinomas showed focal-to-extensive ERG positivity, with no immunoreactivity in the normal prostate. Other carcinomas and epithelial tumors (n=643) were ERG negative, with the exception of 1 of 42 large cell undifferentiated pulmonary carcinomas and 1 of 27 mesotheliomas, each of which showed focal nuclear ERG positivity. On the basis of the above observations, the authors concluded that ERG is a highly specific new marker for benign and malignant vascular tumors. Among epithelial tumors, ERG shows a great promise as a marker to identify prostatic carcinoma in both primary and metastatic settings.42 Minner et al43 analyzed a total of 11,483 tumors and 72 different normal tissue types using a tissue microarray format. Strong nuclear ERG overexpression was found in 36.7% of PCa samples as well as in various vascular tumors, including Kaposi sarcomas (91.7%), angiosarcomas (100%), and hemangiomas (90.9%). Moderate to strong nuclear ERG expression was also observed in thymoma (6.1%). Weak to moderate ERG staining was found in a small number of squamous cell carcinomas of the skin, squamous carcinomas of the lung, malignant mesotheliomas, carcinosarcomas of the uterus, gastrointestinal stromal tumors, hepatocellular carcinomas, teratomas of the testis, anaplastic carcinomas of the thyroid, giant cell tumors of the tendon sheath, and benign fibrous histiocytomas of the skin. The authors concluded the high specificity of ERG expression in both normal and neoplastic tissue suggests a very narrow biological role for ERG in highly selected tissues.43 In summary, among epithelial tumors in appropriate clinical setting, strong and diffuse ERG expression would essentially support the diagnosis of PCa. Similarly, in the setting of small cell carcinoma of unknown origin, ERG positivity would support the prostatic origin of small cell carcinoma.44,45
Utility of ERG Antibody as a Prognostic Prostate Carcinoma Biomarker
The prognostic association of ERG alterations remains uncertain in prostate cancer. Numerous genomic studies have examined the association of ERG and prostate cancer and have found variable results. Several studies have found independent association with poor outcomes,5,40,46,47 whereas some studies have found no association48 and paradoxically have shown association of ERG gene fusions and favorable outcomes.17,49 Pettersson et al50 utilized random-effects models and performed a meta-analysis of 47 previously published studies of prostatectomy series comprised of 5074 men followed for biochemical recurrence (1623 events) and 2049 men followed for lethal disease (131 events) to estimate associations between rearrangement and outcomes. TMPRSS2:ERG was associated with stage at diagnosis [risk ratio (RR) T3≥T2, 1.23; 95% confidence interval (CI), 1.16-1.30] but not with biochemical recurrence (RR, 1.00; 95% CI, 0.86-1.17) or lethal disease (RR, 0.99; 95% CI, 0.47-2.09). In addition, authors examined associations of ERG protein overexpression with biochemical recurrence and lethal disease defined by distant metastasis or cancer-specific mortality in a separate cohort of 1180 men treated with radical prostatectomy. During a median follow-up of 12.6 years, 266 of 1180 men treated with radical prostatectomy experienced recurrence and 85 men developed lethal disease. There was no significant association between ERG overexpression and biochemical recurrence (hazard ratio, 0.99; 95% CI, 0.78-1.26) or lethal disease (hazard ratio, 0.93; 95% CI, 0.61-1.43). These results suggest that TMPRSS2:ERG, or ERG overexpression, is associated with tumor stage but does not strongly predict recurrence or mortality among men treated with radical prostatectomy.50 Minner et al51 examined both ERG oncoprotein expression and genomic rearrangements from 2805 patients with prostate cancer over a mean follow-up period of 72 months. None of the patients received neoadjuvant therapy. In this study, ERG oncoprotein status also did not correlate with biochemical recurrence. Similarly, Hoogland et al52 did not find prognostic value of ERG for any of the clinical end points in univariate and multivariate analyses.
It is important to note, however, that to study prognostic associations in prostate cancer, use of ERG antibody may not be an optimal approach, as ERG expression does not stratify underlying mechanisms of ERG rearrangements in PCa. Notably, several studies have demonstrated poor outcomes when ERG rearrangements occur through deletion or deleted and amplified (EDel2+) mechanisms.13,40 Several studies have also shown that prognostic association becomes stronger when ERG is multiplexed with PTEN, a critical tumor suppressor gene involved in prostate carcinogenesis.53,54
Utility of ERG as a Predictive Prostate Carcinoma Biomarker of Therapeutic Outcome
The role of ERG as a predictive biomarker of therapeutic outcome is being explored and may represent a new paradigm in prostate cancer stratification and treatment. The ERG is part of a network, which includes proteins and protein complexes known to have critical functions in PCa, such as the androgen receptor, C-MYC, NKX3.1, and P13K-PTEN axis.55,56 Emerging studies in preclinical models suggest a possible role of therapeutic targeting of the ERG network through the inhibition of poly (ADP-ribose) polymerase (PARP).55 Preliminary data demonstrate that overexpression of ERG, the predominant ETS gene fusion product, results in resistance to radiotherapy in prostate cancer cell models. In addition, ERG interacts with PARP1, and that inhibition of PARP1 potentiates DNA damage preferentially in ETS-positive cells, abrogates ETS-mediated transcription, and reverses ETS-associated radioresistance.55 Indeed, future clinical trials aiming to study the role of ERG as a predictive biomarker of therapeutic outcome are needed to verify utility of ERG as a predictive biomarker of therapeutic outcome.
A flow chart summarizing potential applications of ERG antibody in clinical practice is represented in Image 1.
COMPARISONS OF 2 MONOCLONAL ERG (C-TERMINUS AND N-TERMINUS) ANTIBODIES
A recent study comparing EPR 3864 and 9FY demonstrated similar levels of sensitivity and specificity for detecting the ERG gene fusions. Both primary and secondary metastasis sites from 278 specimens (including 29 lymph nodes, 30 sites of metastasis, and 13 benign prostate tissues) were evaluated for ERG rearrangement status by FISH and ERG protein expression by both antibodies. ERG 9FY detected 272 of 278 (97.8%) cases, with 98.1% specificity and 97.8% sensitivity. ERG 3864 detected 274 of 278 (98.6%) cases, with 99.4% specificity and 97.5% sensitivity.21 However, it should be noted that C-terminus antibody clone EPR 3864 is the most widely utilized ERG antibody in published studies (Table 1).
LIMITATIONS OF ERG IMMUNOHISTOCHEMISTRY AS DIAGNOSTIC BIOMARKER
Despite several promising clinical applications, ERG has several important limitations as a diagnostic PCa biomarker that needs to be addressed. Although positive ERG establishes the PCa diagnosis in the majority of cases, negative ERG expression offers no value in the workup of atypical cases as ERG overall has a low sensitivity of PCa detection. In addition, ERG expression should be interpreted with caution when the small atypical glands are either intermingled or closely associated with HGPIN glands. As ERG is expressed in small proportion of HGPIN glands, a diagnosis of HGPIN or PINATYP cannot be ruled out with certainty in such cases. Whether ERG protein expression in such situations is a marker of unsampled adjacent PCa, still remains to be addressed. Similarly, interfocal tumor heterogeneity for ERG expression observed within multifocal PCas may also potentially affect the utilization of ERG as a diagnostic, prognostic, or predictive PCa biomarker.
Overall, emerging data suggest that ERG oncoprotein detection in PCa is highly concordant with ERG gene fusion status and can be reliably utilized as a surrogate of ERG gene fusions in prostate cancer diagnosis and management. Some observed differences between ERG protein and FISH-based analysis could be explained potentially by technical or biological issues. Overall, ERG is expressed in ∼45% of limited cancer cases in prostate biopsies. It is also expressed in a subset of high-grade PIN and benign glands, majority of which are intermingled or immediately adjacent to cancer. Because of high specificity of ERG for PCa, ERG positivity in small atypical glands where the diagnosis of HGPIN is excluded helps establish a definitive cancer diagnosis in a small proportion of additional ATYP cases where either morphology and/or traditional markers may not be deemed adequate enough to offer a definitive cancer diagnosis. In conclusion, emerging studies suggest that ERG IHC has significant applications in both PCa diagnosis and management.
The author thanks Kristin Bell for assisting with the artwork and Suzanne Ridner for editorial assistance.
1. Shah RB, Chinnaiyan AM. The discovery of common recurrent transmembrane protease serine 2 (TMPRSS2)-erythroblastosis virus E26 transforming sequence (ETS) gene fusions in prostate cancer: significance and clinical implications. Adv Anat Pathol. 2009;16:145–153
2. Tomlins SA, Rhodes DR, Perner S, et al. Recurrent fusion of TMPRSS2 and ETS transcription factor genes in prostate cancer. Science. 2005;310:644–648
3. Mehra R, Tomlins SA, Shen R, et al. Comprehensive assessment of TMPRSS2 and ETS family gene aberrations in clinically localized prostate cancer. Mod Pathol. 2007;20:538–544
4. Mosquera JM, Mehra R, Regan MM, et al. Prevalence of TMPRSS2-ERG fusion prostate cancer among men undergoing prostate biopsy in the United States. Clin Cancer Res. 2009;15:4706–4711
5. Perner S, Demichelis F, Beroukhim R, et al. TMPRSS2:ERG fusion-associated deletions provide insight into the heterogeneity of prostate cancer. Cancer Res. 2006;66:8337–8341
6. Han B, Mehra R, Dhanasekaran SM, et al. A fluorescence in situ
hybridization screen for E26 transformation-specific aberrations: identification of DDX5-ETV4 fusion protein in prostate cancer. Cancer Res. 2008;68:7629–7637
7. Helgeson BE, Tomlins SA, Shah N, et al. Characterization of TMPRSS2:ETV5 and SLC45A3:ETV5 gene fusions in prostate cancer. Cancer Res. 2008;68:73–80
8. Furusato B, Tan SH, Young D, et al. ERG oncoprotein expression in prostate cancer: clonal progression of ERG-positive tumor cells and potential for ERG-based stratification. Prostate Cancer Prostatic Dis. 2010;13:228–237
9. Han B, Mehra R, Lonigro RJ, et al. Fluorescence in situ
hybridization study shows association of PTEN deletion with ERG rearrangement during prostate cancer progression. Mod Pathol. 2009;22:1083–1093
10. Mosquera JM, Perner S, Genega EM, et al. Characterization of TMPRSS2-ERG fusion high-grade prostatic intraepithelial neoplasia and potential clinical implications. Clin Cancer Res. 2008;14:3380–3385
11. Perner S, Mosquera JM, Demichelis F, et al. TMPRSS2-ERG fusion prostate cancer: an early molecular event associated with invasion. Am J Surg Pathol. 2007;31:882–888
12. Mehra R, Han B, Tomlins SA, et al. Heterogeneity of TMPRSS2 gene rearrangements in multifocal prostate adenocarcinoma: molecular evidence for an independent group of diseases. Cancer Res. 2007;67:7991–7995
13. Mehra R, Tomlins SA, Yu J, et al. Characterization of TMPRSS2-ETS gene aberrations in androgen-independent metastatic prostate cancer. Cancer Res. 2008;68:3584–3590
14. Rosen P, Sesterhenn IA, Brassell SA, et al. Clinical potential of the ERG oncoprotein in prostate cancer. Nat Rev Urol. 2012;9:131–137
15. Han B, Suleman K, Wang L, et al. ETS gene aberrations in atypical cribriform lesions of the prostate: implications for the distinction between intraductal carcinoma of the prostate and cribriform high-grade prostatic intraepithelial neoplasia. Am J Surg Pathol. 2010;34:478–485
16. Park K, Tomlins SA, Mudaliar KM, et al. Antibody-based detection of ERG rearrangement-positive prostate cancer. Neoplasia. 2010;12:590–598
17. Petrovics G, Liu A, Shaheduzzaman S, et al. Frequent overexpression of ETS-related gene-1 (ERG1) in prostate cancer transcriptome. Oncogene. 2005;24:3847–3852
18. Chaux A, Albadine R, Toubaji A, et al. Immunohistochemistry for ERG expression as a surrogate for TMPRSS2-ERG fusion detection in prostatic adenocarcinomas. Am J Surg Pathol. 2011;35:1014–1020
19. Falzarano SM, Zhou M, Carver P, et al. ERG gene rearrangement status in prostate cancer detected by immunohistochemistry. Virchows Arch. 2011;459:441–447
20. van Leenders GJ, Boormans JL, Vissers CJ, et al. Antibody EPR3864 is specific for ERG genomic fusions in prostate cancer: implications for pathological practice. Mod Pathol. 2011;24:1128–1138
21. Braun M, Goltz D, Shaikhibrahim Z, et al. ERG protein expression and genomic rearrangement status in primary and metastatic prostate cancer—a comparative study of two monoclonal antibodies. Prostate Cancer Prostatic Dis. 2012;15:165–169
22. Hameed O, Humphrey PA. Immunohistochemistry in diagnostic surgical pathology of the prostate. Semin Diagn Pathol. 2005;22:88–104
23. Varma M, Jasani B. Diagnostic utility of immunohistochemistry in morphologically difficult prostate cancer: review of current literature. Histopathology. 2005;47:1–16
24. Przybycin CG, Kunju LP, Wu AJ, et al. Partial atrophy in prostate needle biopsies: a detailed analysis of its morphology, immunophenotype, and cellular kinetics. Am J Surg Pathol. 2008;32:58–64
25. Shah RB, Kunju LP, Shen R, et al. Usefulness of basal cell cocktail (34betaE12+p63) in the diagnosis of atypical prostate glandular proliferations. Am J Clin Pathol. 2004;122:517–523
26. Shah RB, Zhou M, LeBlanc M, et al. Comparison of the basal cell-specific markers, 34betaE12 and p63, in the diagnosis of prostate cancer. Am J Surg Pathol. 2002;26:1161–1168
27. Oliai BR, Kahane H, Epstein JI. Can basal cells be seen in adenocarcinoma of the prostate?: An immunohistochemical study using high molecular weight cytokeratin (clone 34betaE12) antibody. Am J Surg Pathol. 2002;26:1151–1160
28. Osunkoya AO, Hansel DE, Sun X, et al. Aberrant diffuse expression of p63 in adenocarcinoma of the prostate on needle biopsy and radical prostatectomy: report of 21 cases. Am J Surg Pathol. 2008;32:461–467
29. Kunju LP, Chinnaiyan AM, Shah RB. Comparison of monoclonal antibody (P504S) and polyclonal antibody to alpha methylacyl-CoA racemase (AMACR) in the work-up of prostate cancer. Histopathology. 2005;47:587–596
30. Kunju LP, Rubin MA, Chinnaiyan AM, et al. Diagnostic usefulness of monoclonal antibody P504S in the workup of atypical prostatic glandular proliferations. Am J Clin Pathol. 2003;120:737–745
31. Yaskiv O, Zhang X, Simmerman K, et al. The utility of ERG/P63 double immunohistochemical staining in the diagnosis of limited cancer in prostate needle biopsies. Am J Surg Pathol. 2011;35:1062–1068
32. Tomlins SA, Palanisamy N, Siddiqui J, et al. Antibody-based detection of ERG rearrangements in prostate core biopsies, including diagnostically challenging cases: ERG staining in prostate core biopsies. Arch Pathol Lab Med. 2012;136:935–946
33. Shah RB, Tadros Y, Brummell B, et al. The diagnostic use of ERG in resolving an “atypical glands suspicious for cancer” diagnosis in prostate biopsies beyond that provided by basal cell and alpha-methylacyl-CoA-racemase markers. Hum Pathol. 2012 Nov 14. [Epub ahead of print]
34. Gao X, Li LY, Zhou FJ, et al. ERG rearrangement for predicting subsequent cancer diagnosis in high-grade prostatic intraepithelial neoplasia and lymph node metastasis. Clin Cancer Res. 2012;18:4163–4172
35. He H, Osunkoya AO, Carver P, et al. Expression of ERG protein, a prostate cancer specific marker, in high grade prostatic intraepithelial neoplasia (HGPIN): lack of utility to stratify cancer risks associated with HGPIN. BJU Int. 2012 Oct 9. [Epub ahead of print]
36. He H, Magi-Galluzzi C, Li J, et al. The diagnostic utility of novel immunohistochemical marker ERG in the workup of prostate biopsies with “atypical glands suspicious for cancer”. Am J Surg Pathol. 2011;35:608–614
37. Shah RB, Magi-Galluzzi C, Han B, et al. Atypical cribriform lesions of the prostate: relationship to prostatic carcinoma and implication for diagnosis in prostate biopsies. Am J Surg Pathol. 2010;34:470–477
38. Shah RB, Zhou M. Atypical cribriform lesions of the prostate: clinical significance, differential diagnosis and current concept of intraductal carcinoma of the prostate. Adv Anat Pathol. 2012;19:270–278
39. Epstein JI, Herawi M. Prostate needle biopsies containing prostatic intraepithelial neoplasia or atypical foci suspicious for carcinoma: implications for patient care. J Urol. 2006;175:820–834
40. Attard G, Clark J, Ambroisine L, et al. Duplication of the fusion of TMPRSS2 to ERG sequences identifies fatal human prostate cancer. Oncogene. 2008;27:253–263
41. Wang WL, Patel NR, Caragea M, et al. Expression of ERG, an Ets family transcription factor, identifies ERG-rearranged Ewing sarcoma. Mod Pathol. 2012;25:1378–1383
42. Miettinen M, Wang ZF, Paetau A, et al. ERG transcription factor as an immunohistochemical marker for vascular endothelial tumors and prostatic carcinoma. Am J Surg Pathol. 2011;35:432–441
43. Minner S, Luebke AM, Kluth M, et al. High level of Ets-related gene expression has high specificity for prostate cancer: a tissue microarray study of 11 483 cancers. Histopathology. 2012;61:445–453
44. Guo CC, Dancer JY, Wang Y, et al. TMPRSS2-ERG gene fusion in small cell carcinoma of the prostate. Hum Pathol. 2011;42:11–17
45. Han B, Mehra R, Suleman K, et al. Characterization of ETS gene aberrations in select histologic variants of prostate carcinoma. Mod Pathol. 2009;22:1176–1185
46. Nam RK, Sugar L, Wang Z, et al. Expression of TMPRSS2:ERG gene fusion in prostate cancer cells is an important prognostic factor for cancer progression. Cancer Biol Ther. 2007;6:40–45
47. Demichelis F, Fall K, Perner S, et al. TMPRSS2:ERG gene fusion associated with lethal prostate cancer in a watchful waiting cohort. Oncogene. 2007;26:4596–4599
48. Gopalan A, Leversha MA, Satagopan JM, et al. TMPRSS2-ERG gene fusion is not associated with outcome in patients treated by prostatectomy. Cancer Res. 2009;69:1400–1406
49. Hermans KG, Boormans JL, Gasi D, et al. Overexpression of prostate-specific TMPRSS2(exon 0)-ERG fusion transcripts corresponds with favorable prognosis of prostate cancer. Clin Cancer Res. 2009;15:6398–6403
50. Pettersson A, Graff RE, Bauer SR, et al. The TMPRSS2:ERG rearrangement, ERG expression, and prostate cancer outcomes: a cohort study and meta-analysis. Cancer Epidemiol Biomarkers Prev. 2012;21:1497–1509
51. Minner S, Enodien M, Sirma H, et al. ERG status is unrelated to PSA recurrence in radically operated prostate cancer in the absence of antihormonal therapy. Clin Cancer Res. 2011;17:5878–5888
52. Hoogland AM, Jenster G, van Weerden WM, et al. ERG immunohistochemistry is not predictive for PSA recurrence, local recurrence or overall survival after radical prostatectomy for prostate cancer. Mod Pathol. 2012;25:471–479
53. Reid AH, Attard G, Ambroisine L, et al. Molecular characterisation of ERG, ETV1 and PTEN gene loci identifies patients at low and high risk of death from prostate cancer. Br J Cancer. 2010;102:678–684
54. Yoshimoto M, Joshua AM, Cunha IW, et al. Absence of TMPRSS2:ERG fusions and PTEN losses in prostate cancer is associated with a favorable outcome. Mod Pathol. 2008;21:1451–1460
55. Brenner JC, Ateeq B, Li Y, et al. Mechanistic rationale for inhibition of poly(ADP-ribose) polymerase in ETS gene fusion-positive prostate cancer. Cancer Cell. 2011;19:664–678
56. Roach M III, DeSilvio M, Valicenti R, et al. Whole-pelvis, “mini-pelvis,” or prostate-only external beam radiotherapy after neoadjuvant and concurrent hormonal therapy in patients treated in the Radiation Therapy Oncology Group 9413 trial. Int J Radiat Oncol Biol Phys. 2006;66:647–653
This article has been cited 2 time(s).
Human PathologyFrequent TMPRSS2-ERG rearrangement in prostatic small cell carcinoma detected by fluorescence in situ hybridization: the superiority of fluorescence in situ hybridization over ERG immunohistochemistryHuman Pathology
Science Translational MedicineMolecular Clues Assist in the Cancer ClinicScience Translational Medicine
ERG antibody; prostate cancer; TMPRSS2:ERG gene fusions
© 2013 Lippincott Williams & Wilkins, Inc.
Highlight selected keywords in the article text.