Platinum Priority – Prostate Cancer
Editorial by Ian G. Mills on pp. 568–569 of this issue

Characterization of 1577 Primary Prostate Cancers Reveals Novel Biological and Clinicopathologic Insights into Molecular Subtypes

By: Scott A. Tomlins a b c d , Mohammed Alshalalfa e , Elai Davicioni e , Nicholas Erho e , Kasra Yousefi e , Shuang Zhao f , Zaid Haddad e , Robert B. Den g , Adam P. Dicker g , Bruce J. Trock h , Angelo M. DeMarzo h , Ashley E. Ross h , Edward M. Schaeffer h , Eric A. Klein i , Cristina Magi-Galluzzi i , R. Jeffrey Karnes j , Robert B. Jenkins k and Felix Y. Feng a d f

European Urology, Volume 68 Issue 4, October 2015, Pages 555-567

Published online: 27 October 2015

Keywords: Prostate cancer, , Microarray, Prognosis

Abstract Full Text Full Text PDF (3,7 MB) Patient Summary



Prostate cancer (PCa) molecular subtypes have been defined by essentially mutually exclusive events, including ETS gene fusions (most commonly involving ERG) and SPINK1 overexpression. Clinical assessment may aid in disease stratification, complementing available prognostic tests.


To determine the analytical validity and clinicopatholgic associations of microarray-based molecular subtyping.

Design, setting, and participants

We analyzed Affymetrix GeneChip expression profiles for 1577 patients from eight radical prostatectomy cohorts, including 1351 cases assessed using the Decipher prognostic assay (GenomeDx Biosciences, San Diego, CA, USA) performed in a laboratory with Clinical Laboratory Improvements Amendment certification. A microarray-based (m-) random forest ERG classification model was trained and validated. Outlier expression analysis was used to predict other mutually exclusive non-ERG ETS gene rearrangements (ETS+) or SPINK1 overexpression (SPINK1+).

Outcome measurements

Associations with clinical features and outcomes by multivariate logistic regression analysis and receiver operating curves.

Results and limitations

The m-ERG classifier showed 95% accuracy in an independent validation subset (155 samples). Across cohorts, 45% of PCas were classified as m-ERG+, 9% as m-ETS+, 8% as m-SPINK1+, and 38% as triple negative (m-ERG/m-ETS/m-SPINK1). Gene expression profiling supports three underlying molecularly defined groups: m-ERG+, m-ETS+, and m-SPINK1+/triple negative. On multivariate analysis, m-ERG+ tumors were associated with lower preoperative serum prostate-specific antigen and Gleason scores, but greater extraprostatic extension (p< 0.001). m-ETS+ tumors were associated with seminal vesicle invasion (p= 0.01), while m-SPINK1+/triple negative tumors had higher Gleason scores and were more frequent in Black/African American patients (p< 0.001). Clinical outcomes were not significantly different among subtypes.


A clinically available prognostic test (Decipher) can also assess PCa molecular subtypes, obviating the need for additional testing. Clinicopathologic differences were found among subtypes based on global expression patterns.

Patient summary

Molecular subtyping of prostate cancer can be achieved using extra data generated from a clinical-grade, genome-wide expression-profiling prognostic assay (Decipher). Transcriptomic and clinical analysis support three distinct molecular subtypes: (1) m-ERG+, (2) m-ETS+, and (3) m-SPINK1+/triple negative (m-ERG/m-ETS/m-SPINK1). Incorporation of subtyping into a clinically available assay may facilitate additional applications beyond routine prognosis.

Take Home Message

Extra gene expression profiling data from a clinically available prognostic assay support three molecularly and clinically distinct molecular subtypes: (1) m-ERG+, (2) m-ETS+, and (3) m-SPINK1+/triple negative (m-ERG/m-ETS/m-SPINK1). Incorporation of molecular subtyping may complement purely prognostic assays.

Keywords: Prostate cancer, ERG, ETS, SPINK1, Microarray, Prognosis.

1. Introduction

Prostate cancer (PCa) is clinically and molecularly heterogeneous. PCa genome and transcriptome characterization has identified molecular subtypes defined by essentially mutually exclusive genetic/transcriptomic events [1]. For example, approximately 50% of PCa foci from prostate-specific antigen (PSA)-screened Caucasian cohorts harbor rearrangements between the 5′ untranslated region of androgen-responsive genes (most commonly TMPRSS2) and members of the ETS transcription factor family [2] and [3]. Fusions involving the ETS gene ERG are the most common (referred to as ERG+; comprising ∼90% of all ETS fusions), while mutually exclusive gene fusions involving non-ERG ETS genes, including ETV1, ETV4, ETV5, and FLI1, are infrequent (referred to as ERG/ETS+ or ETS+; collectively comprising ∼10% of all ETS fusions) [4]. Given the rarity of ETS+ PCa, it is unclear whether these tumors are molecularly and clinicopathologically similar to ERG+ tumors. Approximately 10% of PCas, which are nearly exclusively negative for ERG or other ETS gene fusions (ETS), harbor marked SPINK1 overexpression, consistent with a unique molecular subtype (SPINK1+) [3] and [5]. Although we and others have validated antibodies against ERG and SPINK1, and fluorescence in situ hybridization (FISH) or RNA in situ hybridization (RISH) assays against ETV1, ETV4, and ETV5[4], [6], and [7], routine comprehensive subtyping remains challenging and is cost-prohibitive given the lack of current clinical indications.

High-throughput PCa transcriptome characterization has identified prognostic biomarkers that have been translated to clinically available multigene prognostic tests compatible with routine formalin-fixed, paraffin-embedded (FFPE) clinical biopsy or radical prostatectomy (RP) specimens [8], [9], [10], and [11]. Such tests must account for disease multifocality, as most men with PCa actually harbor multiple genetically independent tumor clones that may have variable morphology (including Gleason score) and molecular alterations [12]. For example, 40–70% of RP samples harbor PCa foci with divergent TMPRSS2:ERG gene fusion status in distinct tumor foci, consistent with multiclonality [13], [14], and [15]. Conflicting reports on associations between PCa molecular subtype–defining lesions—such as TMPRSS2:ERG fusions and SPINK1 overexpression—and prognosis have been reported. Prognostic associations are confounded by cohort differences (PSA screened vs unscreened; biopsy- vs transurethral resection (TURP)-detected; treatment modality; and definition of “poor” outcome) and detection methodologies [16] and [17]. Nevertheless, inclusion of molecular subtyping in clinically available prognostic tests may provide additional information beyond routine prognosis in the post-RP setting, including assessment of multiclonality/multifocality [18], [19], and [20] and predictive applications given clinical trials incorporating ETS status (NCT01576172). In addition, it is unclear if prognostic signatures perform equally for different molecular PCa subtypes.

The goal of this study was to determine if PCa molecular subtyping could be performed from the extra data generated by a clinically available prognostic assay (Decipher; GenomeDx, San Diego, CA, USA) that utilizes genome-wide microarray expression profiles from FFPE tissues to determine a prognostic score using the expression of 22 genes [9]. To this end, we developed and validated computational tools for molecular subtyping using Decipher-generated microarray expression data. We then determined clinicopathologic and prognostic associations from these microarray-derived subtypes using 1577 RP samples.

2. Materials and methods

2.1. PCa samples

A total of 1577 patient PCa expression profiles (1351 from FFPE tissue) were analyzed from eight RP cohorts: Mayo Clinic (MCI and II) [9] and [21], Thomas Jefferson University (TJU) [22], Cleveland Clinic (CCF) [23], Johns Hopkins (JHMI), Memorial Sloan Kettering (MSKCC) [24], Erasmus MC (EMC) [25], and the German National Cancer Registry (DKFZ) [26] (Supplementary Table 1). In each of the eight RP cohorts, a single tumor focus per patient was profiled; Supplementary Table 1 includes the selection criteria. The 1351 FFPE samples were processed, assessed, and analyzed using the Decipher clinical assay in the GenomeDx Biosciences Laboratory (San Diego, CA, USA), which has Clinical Laboratory Improvements Amendment (CLIA) certification. The remaining 226 samples from three cohorts used RNA extracted in research laboratories from fresh-frozen or unfixed tissue preserved in RNAlater. These samples were profiled in microarray core facilities of major teaching hospitals and universities, although not to clinical-grade standards. Data analysis was performed as for the Decipher clinical assay.

For the development of microarray-based classifiers, MCI was used as a discovery cohort. As previously reported [27], 407/580 MCI patients had ERG status determined by FISH; this cohort was split into a training set of 252 patients and a validation set of 155 patients for training and validation of the microarray-based ERG classifier (m-ERG). The other cohorts without FISH or immunohistochemistry (IHC) ERG status (or assessment of non-ERG ETS genes or SPINK1 expression) were used for classifier evaluation. The Supplementary Material provides additional details.

2.2. Microarray data processing

RNA extraction and microarray expression data generation using the Affymetrix Human Exon 1.0 ST arrays as part of the Decipher assay, including generation of the 22-gene prognostic score, were described previously [9], [21], [24], [25], and [26]. The Supplementary Material provides additional details.

2.3. Development of m-ERG classification models

We developed a random forest (RF) supervised model (m-ERG) to predict FISH-assessed ERG rearrangement status using the MCI cohort. The RF model was developed for a training subset of tumor patient profiles (n = 252) combined with 29 benign prostate-tissue profiles (from the MSKCC cohort) before assessment using the validation subset of MCI patient profiles (n = 155) with known FISH-ERG status. The m-ERG model generated scores ranging from 0 to 1, with higher scores indicating increased likelihood of ERG rearrangement presence. On the basis of cutoff optimization methods [28], a m-ERG score above 0.6 was used to define m-ERG+ profiles from the training subset before application in the validation subset.

2.4. Development of ETV1, ETV4, ETV5, FLI1, and SPINK1 microarray-based classification models

FISH and/or IHC data were not available for the other non-ERG ETS family members or SPINK1 in the cohorts assessed, precluding gold-standard validation of classifiers for these alterations. Hence, to develop classifiers we performed unsupervised outlier analysis using the extremevalues R package (R Project for Statistical Computing, Vienna, Austria) on expression of core probe sets (those in canonical exons) for each gene using the entire MCI cohort (discovery) to define an expression threshold to classify each sample as an outlier (or not) for each gene. This outlier detection method estimates a model distribution for the discovery population (MCI) and uses regression analysis to identify outliers as observations that are unlikely to be drawn from the same distribution. The minimum value of outliers in MCI was then set as the cutpoint to classify samples from the evaluation cohorts as outliers. Patients with outlier profiles defined as just described were annotated as m-ETV1+, m-ETV4+, m-ETV5+, m-FLI1+, or m-SPINK1+.

2.5. PCa molecular subtyping

In this study, we initially classified patient profiles into four previously reported subtypes according to results for the m-ERG, m-ETS, and m-SPINK1 models. Tumor profiles with a high m-ERG score (m-ERG+) and m-ETV1, m-ETV4, m-ETV5, m-FLI1, and m-SPINK1 status were classified as the m-ERG+ subtype. Profiles with m-ETV1+, m-ETV4+, m-ETV5+ or m-FLI1+, and m-ERG status were classified as the m-ETS+ subtype, and those with m-SPINK1+ and m-ERG status as the m-SPINK1+ subtype. Finally, patient profiles with m-ERG, m-ETV1, m-ETV4, m-ETV5, m-FLI1, and m-SPINK1 status were classified as triple negative. The four subtypes from this step were used to characterize the clinical and molecular characteristics of each subtype. m-ERG+/m-ETS+ or m-ERG+/m-SPINK1+ profiles were considered as conflict cases and were assessed separately.

2.6. Statistical analysis

Statistical analyses were performed in R v3.0. All statistical tests were two-sided using a significance level of p< 0.05. Univariate and multivariate logistic regression analyses were performed to evaluate the statistical associations between microarray-defined molecular subtypes and clinical variables including age, race/ethnicity, preoperative PSA, surgical margin status (SMS), extraprostatic extension (EPE), seminal vesicle invasion (SVI), lymph node involvement, and Gleason score. The multiple cohorts were considered as a random effect in the multivariate regression model to remove individual cohort bias.

3. Results

3.1. Clinical characteristics of the study cohorts

To develop and validate computational tools for basic PCa molecular subtyping by gene expression generated as part of the Decipher prognostic assay, we pooled RP samples from 1577 patients from eight cohorts profiled using Affymetrix Human Exon 1.0 ST arrays (Supplementary Table 1). These cohorts represent the spectrum of RP-treated PCa from low- to high-risk localized disease. Overall, 61% of patients in the pooled cohort had one or more adverse pathology finding (APF; RP Gleason score ≥8, >pT2 or pN1); however, APF incidence ranged from 5% to 89% among individual cohorts (Supplementary Table 1). The majority of patients were Caucasian (89%) and had aggressive PCa. Most of the patients received postoperative hormonal and/or radiation therapy. Patients with metastasis had a median follow-up of 62 mo (range 3–213 mo), and patients with no metastasis had a median follow-up of 129 mo (range 1–280 mo). Detailed clinicopathologic information for the pooled cohort is provided in Supplementary Table 2.

3.2. m-ERG model development and validation

The most informative microarray probe sets for the m-ERG model were identified via a multistep procedure. First, expression clustering of the 132 ERG locus probe sets in the training set (n = 252) revealed that most are highly correlated and informative of FISH ERG status (Fig. 1 A and Supplementary Fig. 1A). Filtering of redundant and noninformative features (eg, not expressed above background) was performed before training an RF classifier for predicting FISH ERG status. The final model used expression values for three ERG locus probe sets and two probe sets associated with FISH-ERG+ but low ERG expression; the area under the receiver operating curve (AUC) for predicting FISH ERG status in the training set using the final m-ERG model was 0.98. In the validation subset (n = 155 profiles, not used for training), the m-ERG model had an AUC of 0.94 and overall accuracy of 95% (Fig. 1B). Most misclassified patients had low ERG expression, consistent with previous reports that very small subsets of ERG fusion-positive tumors (as detected by FISH) do not overexpress ERG protein [29]. Validation data for benign tissue, cell line controls, and technical replicates are reported in the Supplementary Material and Supplementary Figure 1. Application of the model to 1170 patient profiles from the seven cohorts not used for training or testing revealed that 550 (47%) were classified as m-ERG+ (38–64% across cohorts), as shown in Figure 2.


Fig. 1

Development and validation of microarray-based prostate cancer (PCa) molecular subtyping using genome-wide expression profiling data from the Decipher assay. (A) Development of a microarray-based ERG rearrangement classifier (m-ERG) from genome-wide expression profiling data from the Decipher assay. Unsupervised clustering of the training subset (n = 252 samples) from the discovery cohort (Mayo Clinic I) was performed using gene expression from ERG exon (orange) and intron (green) probe sets (5′ end at the bottom). Expression of five summarized features was used to train a random forest (RF) classifier based on known fluorescence in situ hybridization (FISH) assessment of ERG rearrangement status (F-ERG). m-ERG and F-ERG status for each profiled sample are indicated in the header according to the legend. (B) m-ERG scores in the validation subset (n = 155) of the discovery cohort are plotted with stratification by F-ERG status. The predefined m-ERG+/ERG score cutoff is indicated by the blue dashed line. Classification results are shown in the contingency table. (C) Development of microarray-based classifiers for other ETS gene rearrangements and SPINK1 overexpression using outlier analysis. Beeswarm plots show core-level expression of ETV1, ETV4, ETV5, FLI1, and SPINK1 in the discovery (Disc; n = 580 samples) and evaluation cohorts (Eval; n = 997 samples). Outlier analysis was used to define indicated cutoff scores for m-ETV1, m-ETV4, m-ETV5, m-FLI1, and m-SPINK1 classifiers in the discovery cohort, and these were then applied to the evaluation cohort.


Fig. 2

Distribution of molecular prostate cancer subtypes across assessed cohorts. In each cohort, the percentage of samples in each microarray-defined subtype (m-ERG+, m-ETS1+, m-SPINK1+, and m-ERG/m-ETS/m-SPINK1 [triple negative]) is shown. Cohorts are ordered according to the frequency of adverse pathology findings (APF). Mayo Clinic I was used as the discovery cohort, and the remaining cohorts were used as the evaluation cohort.

3.3. Development of ETS and SPINK1 microarray classifiers

We next sought to classify patients based on SPINK1 overexpression, or outlier expression of non-ERG ETS genes (ETV1, ETV4, ETV5, and FLI1) using overexpression as a surrogate for rearrangement of the respective ETS gene. Heat maps of all probe sets for the ETV1, ETV4, ETV5, and SPINK1 loci showed that a subset of patients overexpressed each gene, as expected (Supplementary Fig. 2). Outlier analysis was first performed for the MCI cohort to define outlier threshold cutpoints for each gene (Section 2.4) that were then applied to the remaining evaluation cohorts as no gold-standard data were available for the profiled data sets (Fig. 1C). In the MCI cohort, microarray outlier analysis classified 5% of samples as m-ETV1+, 1.7% as m-ETV4+, 0.5% as m-ETV5+, 1% as m-FLI+, and 7.7% as m-SPINK1+ (Supplementary Table 3). Performance data for these classifiers for benign tissue, cell line controls, and technical replicates are reported in the Supplementary Material.

3.4. Molecular subtyping of 1577 RP patient specimens using the microarray-based classifiers

Across the 1577 profiles from eight cohorts, microarray analysis classified 46% as m-ERG+, 8% as m-ETV1+, 1.1% as m-ETV4+, 1.6% as m-ETV5+, 0.6% as m-FLI+, and 8% as m-SPINK1+; 36% (n = 575) lacked any outlier expression and were considered triple negative (Supplementary Table 3). In addition, 3% of patient profiles had outlier expression for two or more markers (m-ERG+/m-ETS+ or m-ERG+/m-SPINK1+ profiles; Supplementary Table 3), which we consider as conflict cases (Section 4). To focus on cases with clearly defined subtypes, we considered these conflict cases separately and collapsed the four ETS family members into one group (given the low numbers of individual m-ETV1+, m-ETV4+, m-ETV5+, and m-FLI1+ profiles even in this large cohort). This analysis resulted in four molecular subtypes, m-ERG+ (45%), m-ETS+ (9%), m-SPINK1+ (8%), and triple negative (38%) (Fig. 2) at frequencies consistent with distributions in other predominantly Caucasian cohorts assessed by for these individual subtypes by gene expression, FISH, and/or IHC [2] and [30].

3.5. Gene expression clustering of PCa molecular subtypes

The low frequency of ETS+ and SPINK1+ subtypes has precluded comprehensive molecular and clinicopathologic evaluation in large cohorts to determine whether these minor PCa subtypes represent distinct molecular subtypes, or are best classified as ERG+ and triple negative, respectively. Hence, we first assessed whether m-ETS+ tumors (or m-SPINK1+ tumors) show global transcriptional profiles more similar to m-ERG+ or triple negative tumors. We first defined expression centroids for m-ERG+ and triple negative tumors using transcriptome-wide differential expression analysis. To define m-ERG+ and triple negative expression centroids, we selected all probe sets with AUC >0.75 for discrimination of these two subtypes (n = 360 probe sets). Calculation of the distance between each m-ETS+ or m-SPINK1+ sample and the m-ERG+ and triple negative centroids demonstrated that 98% (117/119) of m-SPINK1+ tumors had cluster distances closer to the triple negative centroid. By contrast, 35% of m-ETS+ tumors (48/139) had cluster distances closer to the m-ERG+ centroid, while 65% of m-ETS+ tumors were closer to the triple negative centroid (Fig. 3A). Defining m-ERG+ and triple negative expression centroids using other gene sets and fuzzy c-means clustering supported these results (Supplementary Table 4 and Supplementary Figure 3). Together these analyses demonstrate that m-SPINK1+ tumors are highly similar to triple negative tumors, while m-ETS+ tumors are distinct from m-ERG+ tumors.


Fig. 3

Gene expression profiling supports high similarity of m-SPINK1+ and triple negative subtypes, unlike m-ETS+ and m-ERG+ prostate cancer. (A) m-ERG+ and triple negative expression centroids were generated by identifying all probe sets (n = 360) with an area under the curve (AUC) of >0.7 for discriminating m-ERG+ and triple negative samples across all cohorts (n = 1531 samples). To assess the relationship of m-SPINK1+ and m-ETS+ prostate cancer to m-ERG+ and triple negative, the relative closeness of each m-SPINK1+ or m-ETS+ sample to the m-ERG+/ triple negative centroids is plotted (a larger value indicates greater similarity). (B) Clustering of subtype-defining gene expression demonstrates overlap of m-SPINK1+ and triple negative prostate cancer and unique profiles of m-ETS+ prostate cancer. Expression of the most predictive genes for each subtype (AUC >0.70 for discrimination from all other subtypes, n = 360) were used for clustering all profiled samples (n = 1531). Benign specimens (yellow) from the German National Cancer Registry cohort clustered separately from all prostate cancer samples.

To gain additional insight into subtype relationships, the most predictive genes for each subtype were defined according to the AUC for discrimination of each subtype from the others. Seventy six, 15, 14, and three genes had AUC >0.7 for m-ERG+, m-ETS+, m-SPINK1+, and triple negative subtypes, respectively (Supplementary Table 5). Clustering expression of these discriminatory genes across all samples demonstrated two main dendrogram branches corresponding to m-ERG+ and triple negative predictive genes. While m-ETS+ tumors shared expression of m-ERG+-predictive genes and expressed a unique subset of genes, the expression pattern of m-SPINK1+ tumors was highly similar to that of triple negative PCa (Fig. 3B). As expected, benign samples clustered separately from all tumor samples. Subtype-specific genes are described in the Supplementary Material.

3.6. Clinical associations of PCa molecular subtypes

On univariate analysis, race, preoperative PSA, Gleason score, EPE, and SVI status were non-uniformly distributed across microarray-defined subtypes (Supplementary Table 6). We used multinomial multivariate analysis to compare clinical and pathologic characteristics among subtypes (Table 1). Compared to the triple negative subtype, m-ERG+ PCa was associated with lower preoperative PSA (odds ratio [OR] 0.47, p< 0.001) and lower Gleason score (OR 0.43, p< 0.001) but was nearly twice as likely to have EPE (OR 1.80, p< 0.001) and nearly five times more likely to occur in Caucasian men (p< 0.001; Table 1). The m-ETS+ subtype was more likely to have SVI compared to both triple negative (OR 2.27, p = 0.004) and m-ERG+ PCa (OR 1.96, p = 0.01; Table 1). Both triple negative and m-SPINK1+ tumors had significantly higher preoperative PSA (OR 2.12, p< 0.001, and OR 1.73, p= 0.05) and Gleason scores (OR 2.3, p< 0.001, and OR 3.0, p< 0.001), and were more common in Black/African American patients (OR 5.44, p= 0.002, and OR 16.87, p< 0.001, respectively) compared to m-ERG+ tumors. Interestingly, m-SPINK1+ was significantly associated with lack of SMS compared to m-ERG+ (OR 0.58, p= 0.006). Together, these clinicopathologic associations are consistent with our transcriptome analysis demonstrating that m-SPINK1+ and triple negative subtypes are highly similar, while m-ERG+ and m-ETS+ have distinct features.

Table 1

Multinomial multivariate logistic regression analysis between clinicopathologic variables and molecular subtypes across 1531 profiled samples with triple negative and m-ERG+ subtypes as references

Variablem-ERG+m-ETS+m-SPINK+ANOVA p value
OR (95% CI)MVA p valueOR (95% CI)MVA p valueOR (95% CI)MVA p value
Reference: triple negative
Preoperative PSA0.47 (0.33–0.68)<0.0010.48 (0.26–0.88)0.0210.81 (0.44–1.51)0.42<0.001
Race (Black/African American)0.18 (0.07–0.52)0.0020.21 (0.03–1.6)0.12 3.10(1.23–7.82)0.02<0.001
Extraprostatic extension1.80 (1.34–2.41)<0.0011.23 (0.75–2.01)0.340.76 (0.46–1.26)0.37<0.001
Seminal vesicle invasion1.16 (0.83–1.62)0.242.27 (1.35–3.82)0.0040.84 (0.47–1.53)0.510.01
Pathologic Gleason score <70.96 (0.61–1.51)0.930.75 (0.31–1.81)0.580.89 (0.39–2.04)0.79<0.001
Pathologic Gleason score >70.43 (0.32–0.6)<0.0010.75 (0.45–1.26)0.461.31 (0.78–2.21)0.39<0.001
Surgical margin status1.18 (0.89–1.56)0.291.27 (0.79–2.04)0.530.69 (0.42–1.12)0.130.13
Age1.00 (0.98–1.02)0.710.97 (0.94–1.00)0.0451.01 (0.97–1.05)0.620.24
Lymph node involvement1.27 (0.77–2.11)0.421.6 (0.78–3.27)0.221.16 (0.49–2.76)0.730.60
m-ETS+m-SPINK+Triple negativeANOVA p value
OR (95% CI)MVA p valueOR (95% CI)MVA p valueOR (95% CI)MVA p value
Reference: m-ERG +
Preoperative PSA1.01 (0.54–1.88)0.921.73 (0.91–3.27)0.052.12 (1.47–3.06)<0.001<0.001
Race (Black/African American)1.12 (0.13–9.88)0.6116.87 (5.13–55.48)<0.0015.44 (1.94–15.29)0.002<0.001
Extraprostatic extension0.68 (0.42–1.11)0.130.42 (0.25–0.7)0.0010.56 (0.41–0.75)<0.001<0.001
Seminal vesicle invasion1.96 (1.18–3.24)0.010.73 (0.4–1.32)0.370.86 (0.62–1.2)0.240.014
Pathologic Gleason score <70.78 (0.33–1.85)0.50.93 (0.41–2.13)0.961.04 (0.66–1.64)0.93<0.001
Pathologic Gleason score >71.74 (1.05–2.88)0.053.01 (1.77–5.13)<0.0012.30 (1.68–3.15)<0.001<0.001
Surgical margin status1.08 (0.68–1.72)0.910.58 (0.36–0.95)0.0060.85 (0.64–1.12)0.290.12
Age0.97 (0.94–1.00)0.081.01 (0.98–1.05)0.711.00 (0.98–1.02)0.710.23
Lymph node involvement1.25 (0.62–2.53)0.590.91 (0.38–2.17)0.570.78 (0.47–1.3)0.420.60

ANOVA = analysis of variance; OR = odds ratio; CI = confidence interval; MVA = multivariate analysis; PSA = prostate-specific antigen (reference <20 ng/ml).

3.7. Impact of molecular subtypes on prognosis

To evaluate the impact of molecular subtyping on prognosis, we assessed the ability of the subtypes to predict patient outcomes such as biochemical recurrence (BCR), metastasis, and prostate cancer–specific mortality (PCSM) after RP (Supplementary Table 7). Receiver operating characteristic analysis showed that the subtypes do not discriminate well for these endpoints (AUC ∼0.5). Likewise, across all cohorts (excluding the MCI cohort used for development), Decipher [9] showed similar discrimination (as measured by AUC) for metastasis in all four subtypes (Fig. 4). Other prognostic signatures such as the cell cycle progression score [8], the genomic prostate score (GPS) [10], and the Penney et al signature [31], which can be derived from our global gene expression data, also showed similar discrimination for metastasis in all subtypes except the microarray-derived GPS signature, which was not discriminative in the m-SPINK+ subtype (Supplementary Fig. 4). Lastly, Kaplan-Meier analyses in the MCII cohort failed to show significant differences in time to events for BCR (Fig. 5A) and metastasis (Fig. 5B) endpoints among the subtypes. However, a trend towards significance was observed for association of the triple negative subtype with worse PCSM compared to the other subtypes (Fig. 5C). Together, these results support limited prognostic utility for molecular subtypes in the post-RP setting, in which most patients receive postoperative hormonal and/or radiation therapy.


Fig. 4

Performance of a multigene prostate cancer prognostic predictor (Decipher) is similar across molecular subtypes. The Decipher score (greater score predicts greater aggressiveness) for each profiled sample in the pooled cohorts (n = 997, excluding Mayo Clinic I as it was used for Decipher discovery) is plotted, stratified by assigned molecular subtypes. Patients who developed metastasis or not are indicated by different colored points, and median scores per subtype are indicated by bars. The area under the curve (AUC) for Decipher score prediction of metastasis development in each subtype (along with the 95% confidence interval) is given. AUCs for each subtype were significantly greater than expected by chance (p< 0.0001 for all subtypes).


Fig. 5

Kaplan-Meier analysis demonstrates similar PCa outcome measures across molecular subtypes. Kaplan-Meier analysis was performed for all Mayo Clinic II cohort samples (case cohort, n = 235 samples) stratified by assigned molecular subtype for (A) biochemical recurrence (BCR), (B) metastasis (MET), and (C) prostate cancer–specific mortality (PCSM)-free survival. Log rank p values are given, along with the percentage of each subtype experiencing each outcome. RP = radical prostatectomy.

4. Discussion

High-throughput technologies such as DNA microarrays and next-generation sequencing have greatly increased our understanding of PCa molecular alterations, including the definition of molecular subtypes and the identification of prognostic gene-expression signatures. Although well-validated antibodies and FISH/RISH assays have been developed for research and clinical applications, comprehensive subtyping using these assays remains challenging and has not been applied to large translational research cohorts. Likewise, the lack of current clinical indications in PCa makes comprehensive molecular subtyping cost-prohibitive in routine clinical practice. This lack of large, comprehensively subtyped clinical cohorts has hindered thorough evaluation of subtype-specific clinicopathologic associations and molecular features. However, the development and uptake of clinically available, FFPE-compatible gene expression prognostic assays [8], [9], [10], and [11] suggest that expression profiling data will be available for tens of thousands of patients in the immediate future.

In this study, we sought to determine whether the extra gene expression data generated as part of the clinically available Decipher assay, which derives a 22-gene prognostic score from genome-wide microarray expression profiling, can be used to determine molecular subtypes. To this end, we built computational models to predict the most common PCa molecular subtypes defined by alterations resulting in marked transcript overexpression: ERG, ETV1, ETV4, ETV5, and FLI1 (due to rearrangement) and SPINK1 (unknown mechanism). Our m-ERG classifier showed 95% accuracy in predicting FISH-ERG status in an independent validation set, similar to the accuracy reported for ERG IHC [29], [32], and [33] as used diagnostically in challenging cases [34]. In the pooled cohort, 45% of patients were predicted as m-ERG+, similar to 47% ERG rearrangement–positive frequency reported from a meta-analysis of more than 10 000 PCa samples [35]. We also demonstrated the robustness of this m-ERG classifier using PCa/normal prostate tissue pairs and technical replicates. Hence, using this validated classifier, the Decipher prognostic assay can also assess ERG status without the cost or delay of separate IHC- or FISH-based evaluation.

We also developed microarray-based classifiers for gene fusions involving other non-ERG ETS genes (m-ETS), as well as SPINK1 overexpression (m-SPINK1). As gold-standard FISH/IHC data for assessing the performance of these classifiers were unavailable, findings from these analyses should be considered exploratory, representing a limitation of our study. We also identified a total of 3% of cases with m-ERG+/m-ETS+ or m-ERG+/m-SPINK1+ profiles. In our experience, ERG, non-ERG ETS, and SPINK1 subtype-defining alterations are nearly always mutually exclusive, and the co-occurrence observed is most likely due to either misclassification (given the lack of gold-standard training data for non-ERG classifiers) or profiling of collisions between genetically distinct tumor clones (which may appear morphologically indistinguishable), although exceptionally rare examples of focal SPINK1 expression in otherwise ERG+ tumors have been reported [7], [36], and [37]. As shown by multivariate analysis (Supplementary Table 8), conflict cases identified here show similar clinicopathologic associations as m-ERG+ PCa, consistent with enrichment of m-ERG+ tumors in these conflict cases. Thus, studies are ongoing to generate gold-standard data for these non–ERG-based classifiers. Importantly, however, for clinicopathologic assessment of our microarray-defined subtypes, only cases with clearly defined single subtypes were included, limiting the impact of these conflict cases on our findings.

Our combined cohort comprised more than 1500 PCa patient profiles and allowed us to explore clinicopathologic and molecular correlates from these microarray-defined subtypes that have not been addressed in smaller or less comprehensive studies. Multivariate analysis revealed that m-ERG+ status was significantly associated with lower Gleason score, lower preoperative serum PSA, and European American race. These findings were in keeping with other large RP cohorts [35] and further support the validity of our approach. Interestingly, although m-ETS+ PCa was associated with lower PSA when compared to triple negative PCa, this subtype was specifically associated with increased SVI when compared to both m-ERG+ and triple negative. m-SPINK1+ PCa was specifically associated with Black/African American race, in line with recent findings from an IHC-based RP study assessing molecular subtypes and race [38].

At the molecular level, we also attempted to address whether m-ETS+ and m-SPINK1+ PCas are similar to m-ERG+ or triple negative tumors, respectively, or represent distinct subtypes. Results revealed that most m-SPINK1+ PCas cluster with triple negative according to global and supervised gene expression, unlike m-ETS+ PCa, which shares molecular overlap with both triple negative and m-ERG+ subtypes. Clinicopathologic associations similarly demonstrate the differences between m-ERG+ and other m-ETS+ PCas, as well as the similarity between m-SPINK1+ and triple negative PCa. Thus, we believe that PCa can be grouped into three clinically and molecularly distinct groups (m-ERG+, m-ETS+ and m-SPINK1+/triple negative), although detection of SPINK1 may still be useful as a single gene marker, particularly given the frequency of its overexpression in Black/African American patients (even compared to triple negative).

A limitation of our study is the lack of assessment of other relevant genomic lesions that occur across (eg, PTEN deletion) or within specific molecular subtypes (eg, SPOP mutations or CHD1 deletions/mutations in ERG PCa). Efforts are ongoing to develop classifiers for these events; however, they are more challenging to detect in gene expression data than outlier over-expression based events. Likewise, although our study suggests that ERG+ and non-ERG ETS+ PCa subtypes should not be combined in clinicopathologic analyses, several thousand more samples will need to be profiled to provide information on whether the non-ERG ETS+ subtype should be stratified by individual alterations (eg, ETV1+ vs ETV4+). Additional limitations include the lack of central histology review and the inclusion of a small subset of fresh-frozen samples (n = 226) assessed outside the CLIA laboratory. All cohorts were from centers with expert genitourinary pathologists and we have shown high concordance of Decipher profiles from matched fresh-frozen and FFPE samples assessed here ([39] and data not shown). Hence, although these factors may limit our ability to observe associations, it is likely that our results are generalizable, and several individual clinicopathologic associations we observed (eg, less frequent ERG+ in Black/African American men) are consistent with prior studies.

The recognition of overtreatment has led to an enormous interest in the development of prognostic biomarkers, including several commercially available gene expression–based prognostic assays applicable to routine biopsy or RP specimens [9], [10], and [8]. Whether such assays are similarly prognostic across previously defined molecular subtypes had not been assessed. Our results, which show a limited effect of subtyping on prognosis and prognostic assay performance, are consistent with large FISH/IHC-based studies of ERG and SPINK1 status in RP cohorts that identified a lack of prognostic ability to predict postsurgical outcomes. Why subtypes lack prognostic ability despite strong associations with known prognostic pathologic parameters (m-ETS+ associated with SVI) and the reasons for potential conflicting prognostic associations (m-ERG+ associated with Gleason <7 and EPE) will require additional research. We and others have hypothesized that subtype-defining lesions may play a more important role in tumor initiation and local growth characteristics, rather than in the factors that drive post-resection recurrence [35] and [40], suggesting prognostic or predictive applications in non-RP cohorts (as discussed below).

Despite the lack of impact on post-RP prognosis, we anticipate that incorporation of molecular subtyping into a clinically available prognostic assay has several areas of potential near-term clinical utility. For example, ERG status has been reported as prognostic in several non-RP cohorts. Most notably, in assessing a cohort of 217 active surveillance (AS) patients, Berg et al [41] reported that patients with any ERG+ cores at diagnosis (by IHC) were more than twice as likely to progress compared to ERG patients; ERG+ was the most significant predictor of AS progression in multivariate Cox regression analysis. These findings are in keeping with the hypothesis that ERG rearrangements may drive local growth as described above, and hence molecular subtyping may be particularly relevant in the AS setting. Although Decipher has received Medicare coverage in the USA for post-RP prognosis, efforts to apply this assay to diagnostic biopsy specimen are ongoing. Given the use of other prognostic assays that assess biopsy specimens (such as Oncotype DX and Prolaris), largely in patients considering AS, application of Decipher in this setting will allow assessment of the impact of m-ERG, m-ETS, and mSPINK1/triple negative subtypes on prognosis in this setting without the need for additional IHC/FISH/RISH-based subtyping.

Likewise, incorporation of ERG status into prognostic tests may have utility in evaluating multifocality or clonality. PCa is commonly multifocal, where a single prostate may harbor multiple genetically distinct tumor foci (as has been demonstrated through ERG status) that may be indistinguishable by routine histology. For example, consecutive biopsies with discordant microarray subtypes in patients on AS would indicate that different tumor clones were sampled, as has been shown by IHC-based ERG assessment in a patient who developed an aggressive interval cancer while on AS [20]. Of critical importance, our approach can now be used to profile multiple foci at RP (or multiple involved prostate biopsy cores) and directly assess the impact of true multifocality (as indicated by discordant subtypes) on the Decipher prognostic score and other derived prognostic signatures. Although other prognostic assays have reported robustness to multifocality [10], molecular subtyping was not incorporated and hence it is unclear if separate areas of the same tumor focus or truly genetically independent tumors (as would be indicated by discordant subtypes) were profiled. Given the need for prognostic assays to reflect the most aggressive tumor focus, even if not sampled in the assessed biopsy specimen, our approach can be used to directly assess the robustness of prognostic assays to true multifocality.

Lastly, the ability to robustly detect molecular subtypes facilitates prespecified molecular subgroup analyses or enrichment of patient populations for clinical trials [20], which although common in precision medicine approaches in other cancers, are not routinely used in PCa. Importantly, ETS rearrangements are nearly always early clonal alterations, suggesting that subtypes identified in diagnostic biopsy or RP samples will be maintained through advanced disease [1], [2], and [3]. As an example of the potential predictive utility of PCa molecular subtypes, Galletti et al [42] demonstrated that in a pilot cohort of 34 men with metastatic castration-resistant PCa treated with docetaxel chemotherapy, men with ERG+ primary tumors (by IHC) were nearly twice as likely to show resistance (by lack of PSA response) than men with ERG tumors. Likewise, preclinical data support the targeting of specific subtype-defining alterations (eg, targeting PARP in ERG+ or ETS+ and targeting EGFR in SPINK1+ PCa) [43] and [44], culminating in ongoing clinical trials that require ERG and ETV1 status evaluation (NCT01576172). Importantly, several molecularly defined subtypes in other cancers (KRAS/ALK/EGFR mutant lung cancers) show little to no association with prognosis, but have distinct clincopathologic features and became predictive and clinically useful after the development of targeted therapies [45] and [46].

5. Conclusions

Several lines of evidence support PCa molecular subtypes defined largely by mutually exclusive genomic/transcriptomic events. The most common subtype-defining lesion—ERG rearrangement—has been evaluated clinically by FISH and IHC. We validated the use of extra data from a clinically available prognostic assay based on gene expression (Decipher) for assessing ERG rearrangement status. In addition, although clinicopathologic and molecular associations for m-ERG+ versus m-ERG PCa are well described, little is known about less frequent subtypes. We developed classifiers for subtypes defined by rearrangements of other ETS genes or SPINK1 overexpression, and explored associations in over 1500 PCa samples. Importantly, gene expression profiles and clinicopathologic associations support three general molecular subtypes (m-ERG+, m-ETS+, and m-SPINK1+/triple negative), providing comprehensive support for distinguishing ERG+ and ETS+ PCa. Of note, although IHC/FISH/RISH assays have been developed to assess these molecular subtypes, our microarray-based classifiers are derived from extra data generated as part of the Decipher assay, a prognostic assay performed in a CLIA-certified laboratory. Hence, a potential advantage of this assay compared to other assays is the inclusion of molecular subtyping information without the need for additional testing, delay, or cost. Taken together, our findings demonstrate the validity of PCa molecular subtyping using extra data from a gene expression–based prognostic assay and identify novel clinicopathologic and molecular correlates to these subtypes. Although we found that molecular subtypes are not prognostic in the post-prostatectomy setting (and do not impact the performance of currently available prognostic signatures), we anticipate that molecular subtyping will complement purely prognostic tests in several areas of PCa management including non-RP cohorts.

Author contributions: Scott A. Tomlins had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Tomlins, Alshalalfa, Davicioni, Erho, Feng.

Acquisition of data:.

Analysis and interpretation of data: Tomlins, Alshalalfa, Davicioni, Erho, Haddad, Zhao, Feng.

Drafting of the manuscript: Tomlins, Alshalalfa, Davicioni, Feng.

Critical revision of the manuscript for important intellectual content: Tomlins, Alshalalfa, Davicioni, Erho, Yousefi, Zhao, Haddad, Den, Dicker, Trock, DeMarzo, Ross, Schaeffer, Klein, Magi-Galluzzi, Karnes, Jenkins, Feng.

Statistical analysis: Tomlins, Alshalalfa, Yousefi.

Obtaining funding: None.

Administrative, technical, or material support: None.

Supervision: None.

Other: None.

Financial disclosures: Scott A. Tomlins certifies that all conflicts of interest, including specific financial interests and relationships and affiliations relevant to the subject matter or materials discussed in the manuscript (eg, employment/affiliation, grants or funding, consultancies, honoraria, stock ownership or options, expert testimony, royalties, or patents filed, received, or pending), are the following: Scott A. Tomlins is listed as a co-inventor of a patent issued to the University of Michigan on the detection of ETS gene fusions in prostate cancer and a patent filed by the University of Michigan on SPINK1 in prostate cancer. In both cases the University of Michigan has licensed the diagnostic field of use to Hologic/Gen-Probe, which has sublicensed some rights to Ventana Medical Systems. Scott A. Tomlins has received honoraria from and served as a consultant to Ventana Medical Systems. Mohammed Alshalalfa, Elai Davicioni, Nicholas Erho, Kasra Yousefi, and Zaid Haddad are employees of GenomeDx. Felix Y. Feng has served on an advisory board for GenomeDx. The remaining authors have nothing to disclose.

Funding/Support and role of the sponsor: Scott A. Tomlins and Felix Y. Feng are supported by the Prostate Cancer Foundation and the A. Alfred Taubman Medical Research Institute. GenomeDX funded specimen collection and performed the Decipher assay. The sponsor participated in the design and conduct of the study; data collection, management, analysis, and interpretation; and manuscript preparation, review, and approval.

Appendix A. Supplementary data


  • [1] M.A. Svensson, C.J. LaFargue, T.Y. MacDonald, et al. Testing mutual exclusivity of ETS rearranged prostate cancer. Lab Invest. 2011;91:404-412 Crossref
  • [2] M.A. Rubin, C.A. Maher, A.M. Chinnaiyan. Common gene rearrangements in prostate cancer. J Clin Oncol. 2011;29:3659-3668 Crossref
  • [3] M.A. Rubin. ETS rearrangements in prostate cancer. Asian J Androl. 2012;14:393-399 Crossref
  • [4] C.E. Barbieri, S.A. Tomlins. The prostate cancer genome: perspectives and potential. Urol Oncol Semin Orig Investig. 2014;32 53.e15–22
  • [5] S.A. Tomlins, D.R. Rhodes, J. Yu, et al. The role of SPINK1 in ETS rearrangement-negative prostate cancers. Cancer Cell. 2008;13:519-528 Crossref
  • [6] B. Furusato, S.-H. Tan, D. Young, 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 Crossref
  • [7] L.P. Kunju, S. Carskadon, J. Siddiqui, S.A. Tomlins, A.M. Chinnaiyan, N. Palanisamy. Novel RNA hybridization method for the in situ detection of ETV1, ETV4, and ETV5 gene fusions in prostate cancer. Appl Immunohistochem Mol Morphol. 2014;22:e32-e40 Crossref
  • [8] J. Cuzick, G.P. Swanson, G. Fisher, et al. Prognostic value of an RNA expression signature derived from cell cycle proliferation genes in patients with prostate cancer: a retrospective study. Lancet Oncol. 2011;12:245-255 Crossref
  • [9] N. Erho, A. Crisan, I.A. Vergara, et al. Discovery and validation of a prostate cancer genomic classifier that predicts early metastasis following radical prostatectomy. PLoS One. 2013;8:e66855 Crossref
  • [10] E.A. Klein, M.R. Cooperberg, C. Magi-Galluzzi, et al. A 17-gene assay to predict prostate cancer aggressiveness in the context of Gleason grade heterogeneity, tumor multifocality, and biopsy undersampling. Eur Urol. 2014;66:550-560 Crossref
  • [11] J. Davis. Novel commercially available genomic tests for prostate cancer: a roadmap to understanding their clinical impact. BJU Int. 2014;114:320-322
  • [12] R. Arora, M.O. Koch, J.N. Eble, T.M. Ulbright, L. Li, L. Cheng. Heterogeneity of Gleason grade in multifocal adenocarcinoma of the prostate. Cancer. 2004;100:2362-2366 Crossref
  • [13] R. Mehra, B. Han, S.A. Tomlins, 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 Crossref
  • [14] M. Barry, S. Perner, F. Demichelis, M.A. Rubin. TMPRSS2-ERG fusion heterogeneity in multifocal prostate cancer: clinical and biologic implications. Urology. 2007;70:630-633 Crossref
  • [15] S. Perner, F. Demichelis, R. Beroukhim, et al. TMPRSS2:ERG fusion-associated deletions provide insight into the heterogeneity of prostate cancer. Cancer Res. 2006;66:8337-8341 Crossref
  • [16] F. Demichelis, M.A. Rubin. TMPRSS2-ETS fusion prostate cancer: biological and clinical implications. J Clin Pathol. 2007;60:1185-1186 Crossref
  • [17] S.A. Tomlins, A. Bjartell, A.M. Chinnaiyan, et al. ETS gene fusions in prostate cancer: from discovery to daily clinical practice. Eur Urol. 2009;56:275-286 Crossref
  • [18] S. Minner, M. Gärtner, F. Freudenthaler, et al. Marked heterogeneity of ERG expression in large primary prostate cancers. Mod Pathol. 2013;26:106-116 Crossref
  • [19] M.C. Haffner, T. Mosbruger, D.M. Esopi, et al. Tracking the clonal origin of lethal prostate cancer. J Clin Invest. 2013;123:4918-4922 Crossref
  • [20] M.C. Haffner, A.M. De Marzo, S. Yegnasubramanian, J.I. Epstein, H.B. Carter. Diagnostic Challenges of Clonal Heterogeneity in Prostate Cancer. J Clin Oncol. 2014;32:2013-2015
  • [21] R.J. Karnes, E.J. Bergstralh, E. Davicioni, et al. Validation of a genomic classifier that predicts metastasis following radical prostatectomy in an at risk patient population. J Urol. 2013;190:2047-2053 Crossref
  • [22] R.B. Den, F.Y. Feng, T.N. Showalter, et al. Genomic prostate cancer classifier predicts biochemical failure and metastases in patients after postoperative radiation therapy. Int J Radiat Oncol. 2014;89:1038-1046 Crossref
  • [23] E.A. Klein, K. Yousefi, Z. Haddad, et al. A genomic classifier improves prediction of metastatic disease within 5 years after surgery in node-negative high-risk prostate cancer patients managed by radical prostatectomy without adjuvant therapy. Eur Urol. 2015;67:778-786 Crossref
  • [24] B.S. Taylor, N. Schultz, H. Hieronymus, et al. Integrative genomic profiling of human prostate cancer. Cancer Cell. 2010;18:11-22 Crossref
  • [25] J.L. Boormans, H. Korsten, A.J.C. Ziel-Van Der Made, et al. Identification of TDRD1 as a direct target gene of ERG in primary prostate cancer. Int J Cancer. 2013;133:335-345 Crossref
  • [26] J.C. Brase, M. Johannes, H. Mannsperger, et al. TMPRSS2-ERG-specific transcriptional modulation is associated with prostate cancer biomarkers and TGF-beta signaling. BMC Cancer. 2011;11:507 Crossref
  • [27] T. Nakagawa, T.M. Kollmeyer, B.W. Morlan, et al. A tissue biomarker panel predicting systemic progression after PSA recurrence post-definitive prostate cancer therapy. PLoS One. 2008;3:e2318 Crossref
  • [28] M.D. Ruopp, N.J. Perkins, B.W. Whitcomb, E.F. Schisterman. Youden index and optimal cut-point estimated from observations affected by a lower limit of detection. Biometrical J. 2008;50:419-430 Crossref
  • [29] S. Minner, M. Enodien, H. Sirma, 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 Crossref
  • [30] C. Kumar-Sinha, S.A. Tomlins, A.M. Chinnaiyan. Recurrent gene fusions in prostate cancer. Nat Rev Cancer. 2008;8:497-511 Crossref
  • [31] K.L. Penney, J.A. Sinnott, K. Fall, et al. mRNA expression signature of Gleason grade predicts lethal prostate cancer. J Clin Oncol. 2011;29:2391-2396 Crossref
  • [32] M. Braun, D. Goltz, Z. Shaikhibrahim, 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 Crossref
  • [33] K. Park, S.A. Tomlins, K.M. Mudaliar, et al. Antibody-based detection of ERG rearrangement-positive prostate cancer. Neoplasia. 2010;12:590-598
  • [34] S.A. Tomlins, N. Palanisamy, J. Siddiqui, A.M. Chinnaiyan, L.P. Kunju. Antibody-based detection of erg rearrangements in prostate core biopsies, including diagnostically challenging cases: ERG staining in prostate core biopsies. Arch Pathol Lab Med. 2013;136:935-946
  • [35] A. Pettersson, R.E. Graff, S.R. Bauer, 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 Crossref
  • [36] S. Smith, S.A. Tomlins. Prostate cancer subtyping biomarkers and outcome: is clarity emerging?. Clin Cancer Res. 2014;20:4733 Crossref
  • [37] R.J. Flavin, A. Pettersson, W.K. Hendrickson, et al. SPINK1 protein expression and prostate cancer progression. Clin Cancer Res. 2014;20:4904 Crossref
  • [38] F. Khani, J.M. Mosquera, K. Park, et al. Evidence for molecular differences in prostate cancer between African American and Caucasian Men. Clin Cancer Res. 2014;20:4925 Crossref
  • [39] D. Abdueva, M. Wing, B. Schaub, T. Triche, E. Davicioni. Quantitative expression profiling in formalin-fixed paraffin-embedded samples by Affymetrix microarrays. J Mol Diagn. 2010;12:409-417 Crossref
  • [40] G. Schaefer, J.-M. Mosquera, R. Ramoner, et al. Distinct ERG rearrangement prevalence in prostate cancer: higher frequency in young age and in low PSA prostate cancer. Prostate Cancer Prostatic Dis. 2013;16:132-138
  • [41] K.D. Berg, B. Vainer, F.B. Thomsen, et al. ERG protein expression in diagnostic specimens is associated with increased risk of progression during active surveillance for prostate cancer. Eur Urol. 2014;66:861-862
  • [42] G. Galletti, A. Matov, H. Beltran, et al. ERG induces taxane resistance in castration-resistant prostate cancer. Nat Commun. 2014;5:5548 Crossref
  • [43] B. Ateeq, S.A. Tomlins, B. Laxman, et al. Therapeutic targeting of SPINK1-positive prostate cancer. Sci Transl Med. 2011;3:72ra17 Crossref
  • [44] J.C. Brenner, B. Ateeq, Y. Li, et al. Mechanistic rationale for inhibition of poly(ADP-ribose) polymerase in ETS gene fusion-positive prostate cancer. Cancer Cell. 2011;19:664-678 Crossref
  • [45] F.A. Shepherd, C. Domerg, P. Hainaut, et al. Pooled analysis of the prognostic and predictive effects of KRAS mutation status and KRAS mutation subtype in early-stage resected non-small-cell lung cancer in four trials of adjuvant chemotherapy. J Clin Oncol. 2013;31:2173-2181 Crossref
  • [46] A.T. Shaw, B.Y. Yeap, M. Mino-Kenudson, et al. Clinical features and outcome of patients with non-small-cell lung cancer who harbor EML4-ALK. J Clin Oncol. 2009;27:4247-4253 Crossref


a Michigan Center for Translational Pathology, University of Michigan Medical School, Ann Arbor, MI, USA

b Department of Pathology, University of Michigan Medical School, Ann Arbor, MI, USA

c Department of Urology, University of Michigan Medical School, Ann Arbor, MI, USA

d Comprehensive Cancer Center, University of Michigan Medical School, Ann Arbor, MI, USA

e GenomeDx Bioscience Inc., Vancouver, British Columbia, Canada

f Department of Radiation Oncology, University of Michigan Medical School, Ann Arbor, MI, USA

g Kimmel Cancer Center, Jefferson Medical College of Thomas Jefferson University, Philadelphia, PA, USA

h James Buchanan Brady Urological Institute, Johns Hopkins School of Medicine, Baltimore, MD, USA

i Glickman Urological & Kidney Institute, Cleveland Clinic, Cleveland, OH, USA

j Department of Urology, Mayo Clinic, Rochester, MN, USA

k Department of Pathology and Laboratory Medicine, Mayo Clinic, Rochester, MN, USA

Corresponding authors. Departments of Pathology and Urology, Michigan Center for Translational Pathology, University of Michigan Medical School, 1524 BSRB, 109 Zina Pitcher Place, Ann Arbor, MI 48109-2200, USA. Tel. +1 734 7641549; Fax: +1 734 6477950.Department of Radiation Oncology, Michigan Center for Translational Pathology, University of Michigan Medical School, 1500 East Medical Center Drive, UHB2C490-SPC5010, Ann Arbor, MI 48109-5010, USA. Tel. +1 734 9364302; Fax: +1 734 9367859.

Place a comment

Your comment *

max length: 5000