Collaborative Review – Prostate Cancer

Genomic Predictors of Outcome in Prostate Cancer

By: Peter J. Boström a , Anders S. Bjartell b , James W.F. Catto c , Scott E. Eggener d , Hans Lilja e f g , Stacy Loeb h , Jack Schalken i , Thorsten Schlomm j and Matthew R. Cooperberg k

European Urology, Volume 68 Issue 6, December 2015, Pages 1033-1044

Published online: 01 December 2015

Keywords: Prostate cancer, Prognostic, Predictive, Outcome, Gene panel, Expression panel, Genetic test, RNA, Cell cycle, DNA

Abstract Full Text Full Text PDF (544 KB) Patient Summary



Given the highly variable behavior and clinical course of prostate cancer (PCa) and the multiple available treatment options, a personalized approach to oncologic risk stratification is important. Novel genetic approaches offer additional information to improve clinical decision making.


To review the use of genomic biomarkers in the prognostication of PCa outcome and prediction of therapeutic response.

Evidence acquisition

Systematic literature review focused on human clinical studies reporting outcome measures with external validation. The literature search included all Medline, Embase, and Scopus articles from inception through July 2014.

Evidence synthesis

An improved understanding of the genetic basis of prostate carcinogenesis has produced an increasing number of potential prognostic and predictive tools, such as transmembrane protease, serine2:v-ets avian erythroblastosis virus E26 oncogene homolog (TMPRSS2:ERG) gene fusion status, loss of the phosphatase and tensin homolog (PTEN) gene, and gene expression signatures utilizing messenger RNA from tumor tissue. Several commercially available gene panels with external validation are now available, although most have yet to be widely used. The most studied commercially available gene panels, Prolaris, Oncotype DX Genomic Prostate Score, and Decipher, may be used to estimate disease outcome in addition to clinical parameters or clinical nomograms. ConfirmMDx is an epigenetic test used to predict the results of repeat prostate biopsy after an initial negative biopsy. Additional future strategies include using genetic information from circulating tumor cells in the peripheral blood to guide treatment decisions at the initial diagnosis and at subsequent decision points.


Major advances have been made in our understanding of PCa biology in recent years. Our field is currently exploring the early stages of a personalized approach to augment traditional clinical decision making using commercially available genomic tools. A more comprehensive appreciation of value, limitations, and cost is important.

Patient summary

We summarized current advances in genomic testing in prostate cancer with a special focus on the estimation of disease outcome. Several commercial tests are currently available, but further understanding is needed to appreciate the potential benefits and limitations of these novel tests.

Take Home Message

Personalized genetic profiling of primary and metastatic tumor cells is available for routine clinical decision making. Greater understanding of the potential long-term benefits and limitations of these tests and their optimal use in clinical decision making is important.

Keywords: Prostate cancer, Prognostic, Predictive, Outcome, Gene panel, Expression panel, Genetic test, RNA, Cell cycle, DNA.

1. Introduction

Prostate cancer (PCa) is the most common urologic malignancy and the second leading cause of male cancer-related deaths in many developed countries [1] . A personalized approach, including the prediction of individual patient outcomes and therapeutic responses, is important in all cancers but especially for PCa, given the variability in disease behavior, the diversity of treatment options, and the risk of treatment-related impairment of quality of life [2] . Novel genomic technologies, such as microarray analyses and next-generation sequencing, have improved our understanding of the biology of PCa. Consequently, the scientific community is faced with an explosion of data, new challenges, and opportunities in biomarker discovery and validation [3] . With improved approaches to biomarker research, combined with lower cost and more efficient techniques, the potential of a personalized genomic approach for clinical decision making has recently been made possible.

Among the most prominent topic in PCa genetics is the characterization of somatic genomic alterations in tumor tissue for the prognosis and prediction of treatment response. Novel approaches include genetic analyses from peripheral blood, either germline analyses or characterization of DNA/RNA from circulating tumor cells (CTCs), or free circulating nucleic acids. The genetic landscape, key genetic alterations, epigenetic events, and microRNAs (miRNAs) in PCa have been reviewed [4], [5], [6], and [7].

In this paper, we focus on the value of genomic markers in the personalized prediction of PCa outcome and response to various therapeutic interventions. Due to the breadth of the topic and recent high-quality reviews, we have specifically focused on genomic tests that are already available or approaching the point of clinical use [4], [5], [6], [7], [8], [9], and [10].

2. Evidence acquisition

A literature review was performed according to the Preferred Reporting Items for Systematic Reviews and Meta-analysis criteria. Figure 1 presents the process of identifying references [11] . The first author performed a Medline, Embase, and Scopus search of all articles from inception through July 2014 using the keywords prostate cancer and genetics and prognostic. Genetic PCa outcome studies with the following criteria were prioritized: human clinical studies, clinical outcome end points (biochemical progression, clinical progression, disease-specific survival [DSS], and overall survival), and external validation cohorts. Articles of interest and review articles were surveyed and verified for any missed reports. All authors oversaw and approved the final literature review and selection.


Fig. 1 Preferred Reporting Items for Systematic Reviews and Meta-analysis flow diagram presenting the steps of the literature search and the selection process of the articles.

3. Evidence synthesis

3.1. Clinically relevant genes and genetic pathways in prostate cancer

3.1.1. TMPRSS2:ERG fusion

In 2005, Tomlins and coworkers reported a novel frequent chromosomal rearrangement in PCa, a fusion between transmembrane protease, serine 2 (TMPRSS2) gene and v-ets avian erythroblastosis virus E26 oncogene homolog (ERG) gene or other ETS (E26 transformation specific) transcription factors, until now recognized as the most frequent gene-specific alterations in PCa [12] and [13]. ETS fusion–type cancers are believed to represent a genetically distinct subset of PCa characterized by deletions of the phosphatase and tensin homolog (PTEN) gene and of chromosome 3p, whereas deletions of 5q and 6q prevail in fusion-negative cancers [14], [15], [16], and [17]. Although gene fusions in general, and specifically ETS fusions, have been associated with the early onset of PCa [18] and [19], the clinical utility of the gene fusion as a prognostic or predictive tool is still unclear.

Many studies have investigated the association of TMPRSS2:ERG fusion status and outcome in PCa ( Table 1 ). Ten studies reported the prognostic value of the gene fusion in radical prostatectomy (RP) cohorts [19], [20], [21], [22], [23], [24], [25], [26], [27], and [28]. In 6 of the 10 studies, TMPRSS2:ERG fusion status was not associated with outcome after surgery [19], [21], [22], [24], [27], and [29]. In one study, patients with rearrangement had an 8.6-fold increased risk for biochemical recurrence (BCR), and in another study, fusion status was predictive of BCR risk in a small selected cohort of Gleason 7 cases [23] and [28]. In contrast, one study demonstrated lower BCR risk after RP among patients with the TMPRSS2:ERG fusion [25] . Overall, a meta-analysis including 5074 men following RP found no significant association with BCR or lethal disease [24] . One study investigated the outcome after intensity-modulated radiation therapy but found no association between fusion status and BCR.

Table 1 Studies reporting TMPRSS2:ERG fusion status and outcome after various treatment modalities

Study No. of cases Tissue type Detection method Intervention ERG rearrangement rate, % ERG rearrangement association with clinical parameters Main results
Radical prostatectomy and external-beam radiation therapy cohorts
 Steurer et al [19] 9567 RP FISH/IHC RP 54 NR Rearrangement associated with low-grade tumors in younger patients. ERG not associated with outcome.
 FitzGerald et al [20] 214 RP FISH, SNP genotyping RP 36 No Rearrangement not predictive of DSS, but cases with multiple fusion copies had trend toward poorer survival
 Gopalan et al [21] 521 RP FISH RP 46 Lower Gleason score Rearrangement not associated with outcome
 Hoogland et al [22] 509 RP IHC RP 55 Lower PSA ERG staining not associated with BCR or local recurrence risk
 Nam et al [23] 165 RP RT-PCR RP 49 No Rearrangement independently predictive of BCR (HR: 8.6)
 Pettersson et al [24] 1292 RP IHC RP 49 Higher stage, lower PSA Rearrangement not associated with outcome
 Saramäki et al [25] 150 RP FISH RP 33 No Rearrangement independently associated with lower BCR risk
 Boormans et al [26] 112 RP RT-PCR RP 42 No TMPRSS2:ERG (Exon0)–ERG fusion associated with lower risk of BCR compared with Exon1 fusion
 Minner et al [27] 2891 RP FISH/IHC RP 52 No ERG IHC positivity not predictive of BCR risk
 Nam et al [28] 26 RP RT-PCR RP 42 NA Rearrangement independently associated with recurrence risk
 Dal Pra et al [29] 118 (IHC)

126 (aCGH)
Biopsy IHC aCGH IMRT 21 (aCGH), 50 (IHC) Higher T stage Rearrangement not associated with BCR risk after IMRT
Watchful waiting, active surveillance, and ADT cohorts
 Attard et al [30] 445 TURP FISH WW 30 Higher Gleason score, higher stage, higher PSA Rearrangement independent predictor of poor DSS and OS
 Demichelis et al [31] 111 TURP FISH WW 15 Higher Gleason score Rearrangement associated with higher risk of metastatic progression and PCa death in univariate analysis
 Hägglöf et al [32] 350 TURP IHC WW 40 Higher Gleason score and higher PSA ERG IHC positivity independently predictive of poor DSS
 Qi et al [33] 224 TURP FISH/IHC WW 23 Higher PSA Rearrangement/ERG IHC positivity independently associated with PCa death risk (HR: 2.1)
 Bismar et al [34] 152 (no. 1); 160 (no. 2) TURP IHC AS/RP/EBRT (no. 1), ADT (no. 2) 26 Higher Gleason score and higher tumor volume ERG IHC positivity associated with longer time to CRPC among androgen-deprived patients
 Boormans et al [35] 85 Node metastasis

RT-PCR ADT 59 No Rearrangement was not associated with duration of ADT response or outcome
 Leinonen et al [36] 178 Biopsy FISH ADT 34 Ki-67 proliferation index, age, and tumor volume Rearrangement not associated with disease progression
 Berg et al [37] 265 Biopsy IHC AS 38 Higher tumor volume in biopsies and higher clinical stage ERG positivity independently associated with progression risk (HR: 2.45)
 Lin et al [38] 387 Urine RT-PCR AS NA Higher Gleason score and higher tumor volume Urine-detected rearrangement associated with positive repeat biopsy

aCGH = array comparative genomic hybridization; ADT = androgen-deprivation therapy; AS = active surveillance; BCR = biochemical recurrence; CRPC = castration-resistant prostate cancer; DSS = disease-specific survival; FISH = fluorescence in situ hybridization; HR = hazard ratio; IHC = immunohistochemistry; IMRT = intensity-modulated radiation therapy; NA = not applicable; NR = not reported; OS = overall survival; PCa = prostate cancer; PSA = prostate-specific antigen; RT-PCR = reverse transcriptase polymerase chain reaction; RP = radical prostatectomy; SNP = single nucleotide polymorphism; TURP = transurethral resection of prostate; WW = watchful waiting.

Nevertheless, when investigated beyond the gene fusion status, some additional prognostic information has been reported. FitzGerald and coworkers did not observe a significant association between TMPRSS2:ERG fusion status and outcome, but patients with increased copy numbers of the fusion gene showed poorer survival [20] . Furthermore, Boormans and coworkers reported fusion gene transcript-specific data; that is, TMPRSS2:ERG (Exon0)–ERG fusion was associated with a lower risk of BCR compared with Exon1 fusion [26] .

In contrast to studies in cohorts treated with curative intent, the presence of TMPRSS2:ERG fusion had an independent negative impact on outcome in four watchful waiting (WW) cohorts and on a cohort of patients with castration-resistant PCa (CRPC) undergoing palliative transurethral resection of the prostate (TURP) [30], [31], [32], [33], and [34]. Therefore, one could speculate that TMPRSS2:ERG fusion status is a predictor of response to androgen-deprivation therapy (ADT). However, this hypothesis was not supported by Boormans and coworkers. They found no association between TMPRSS2:ERG fusion status with ADT response or outcome in PCa patients with lymph node metastases (N1) treated with ADT [35] . Similarly, Leinonen and coworkers found no association between TMPRSS2:ERG fusion status and outcome among ADT-treated patients [36] . A recent study investigated TMPRSS2:ERG fusion status from biopsies of 265 active surveillance (AS) patients and found that TMPRSS2:ERG fusion-positive patients had a significantly higher risk of disease progression (hazard ratio: 2.45) compared with fusion-negative patients [37] . However, another study of PCa patients on AS showed that urinary TMPRSS2:ERG and the prostate cancer antigen 3 (PCA3) gene were not significant independent predictors of biopsy reclassification on multivariable analysis [38] .

In addition to its own potential prognostic value, TMPRSS2:ERG fusion status may modify the interpretation of other PCa biomarkers in outcome prediction. Barwick and coworkers noted that the expression of several genes was affected by TMPRSS2:ERG fusion status [39] . In fusion-positive cases, upregulated genes were related to mismatch base repair and histone deacetylation, whereas genes involved in insulinlike growth factor (IGF) and Janus kinase/signal transducers and activators of transcription (JAK-STAT) signaling were downregulated [39] . In addition Brase et al showed the TMPRSS2-ERG gene fusion results in the modulation of certain transcriptional patterns and well-known PCa biomarkers like CRISP3 and TDRD1 that were found to be associated with the gene fusions [40] . Karnes and coworkers did not detect a direct association between TMPRSS2:ERG fusion status and outcome, but classifying the cohort according to TMPRSS2:ERG fusion status had a significant impact on the predictive value of other investigated markers [41] . Similarly, TMPRSS2:ERG fusion status was noted to significantly affect the prognostic value of a 36-gene expression panel [42] . Taken together, although the true prognostic value of TMPRSS2:ERG fusion status itself has not been proven, fusion status is a key genomic event and should be taken into consideration when the prognostic value of other genomic events is investigated.

3.1.2. PTEN

PTEN deleted on chromosome 10 is one of the most frequently mutated genes in human cancer. It dephosphorylates lipid-signaling intermediates, resulting in deactivation of PI3K signaling, and thus controls proliferation and growth [43] . In a landmark study by Saal and coworkers in 2007, PTEN loss was associated with poor outcome in a variety of cancers including PCa and cancer of the urinary bladder [44] . The prognostic value of PTEN in PCa was investigated in a few studies ( Table 2 ). In 649 PCa patients, Leinonen and coworkers demonstrated a higher frequency of PTEN loss in more advanced cases (CRPC compared with RP cases) and that PTEN loss was associated with shorter progression-free survival time but notably only in ERG-positive cases [45] . Similarly, in another study the prognostic value of PTEN was clearly associated with TMPRSS2:ERG fusion status [46] . In a large cohort including 4699 RP specimens and 57 CRPC cases, Krohn and coworkers also demonstrated that PTEN loss was associated with adverse clinicopathologic factors and a higher risk of BCR [47] . Contrary to the findings of Leinonen et al, PTEN had similar prognostic utility in ERG-positive and -negative cases. In a study among conservatively managed PCa patients by Reid et al, PTEN loss without TMPRSS2:ERG fusion was associated with poor cancer-specific survival, which is in contrast to other studies where PTEN loss and TMPRSS2:ERG fusion defined the patients with the worst survival [48] .

Table 2 Studies investigating association of PTEN mutations and outcome of prostate cancer

Study No. of cases Tissue type Detection method Mutation/Negative IHC staining rate, % PTEN mutation and association with clinical parameters Main results
Leinonen et al [45] 326 (RP), 166 (ADT), 177 (CRPC), 32 (CRPC mets) RP, biopsy, TURP, autopsy IHC 15 (RP)

45 (CRPC)

67 (CRPC mets)
NR PTEN loss more frequent in CRPC than RP specimens. PTEN loss associated with shorter PFS in ERG-positive cases
Yoshimoto et al [46] 125 RP FISH 45 NR Homozygous PTEN deletion independently associated with BCR risk. Significant prognostic association between ERG and PTEN
Krohn et al [47] 4699 (RP), 57 (CRPC) RP, CRPC (TURP) FISH/IHC Deletion (20)

Negative/Weak IHC (30)
Advanced stage, high Gleason score, lymph node metastasis, and positive surgical margin PTEN loss independent predictor of poorer PFS. ERG status did not affect predictive value of PTEN
Reid et al [48] 308 (conservative management) TURP FISH 17 Advanced stage, high Gleason score, lymph node metastasis, and positive surgical margin PTEN loss alone not predictive, but patients with PTEN loss and normal ERG status had significantly poorer PCa survival
McCall et al [49] 68 matched castration sensitive and resistant TURP FISH/IHC 23 (castration sensitive)

52 (CRPC)
No Low PTEN staining in IHC associated with poor PCa-specific survival among castration-sensitive cases

ADT = androgen deprivation therapy; BCR = biochemical recurrence; CRPC = castration-resistant prostate cancer; FISH = fluorescence in situ hybridization; IHC = immunohistochemistry; mets = metastases; NR = not reported; PCa = prostate cancer; PFS = progression-free survival; RP = radical prostatectomy; TURP = transurethral resection of prostate.

As yet, the predictive role of PTEN status in castration-sensitive and resistant cancers has only been evaluated in one study [49] . McCall et al investigated PTEN status by fluorescent in situ hybridization and immunohistochemistry in matched tumor pairs (one before and one after ADT relapse). They noted that loss of PTEN expression in the nucleus was independently associated with poor DSS but only in the castration-sensitive tumor specimens. PTEN-negative tumors were recently shown to have shorter survival in the post-docetaxel abiraterone treatment setting compared with cases with preserved PTEN expression [50] .

3.2. Gene/expression panels

Cancer is a complex disease, and it is unlikely a single genetic abnormality will sufficiently reflect events in a tumor to give enough prognostic information for clinical decisions. Most authors suggest that a combination of multiple genetic markers will be necessary. Panels evaluate differential expression of multiple genes between patient groups of interest (eg, biochemical relapse vs no relapse after RP). These panels may be selected using prior knowledge by including key carcinogenic pathways in PCa (eg, cell cycle regulation, apoptosis) [51] or filtered from thousands of unselected genes to distinguish gene-phenotype correlates [52], [53], and [54].

These studies face many challenges including the risk of chance associations given the quantity of data. Therefore experienced biostatistical support and appropriate external validations are essential before widespread clinical applications can be considered. Approved principles of study design include blinded marker analyses and randomly selected cases (in retrospective studies) [55] . Study reporting may be negatively affected by several potential biases, and therefore adherence to standard criteria, such as Reporting Recommendations for Tumor Marker Prognostic Studies (REMARK), is essential for providing evidence on the clinical utility of biomarkers in oncology [56] . The biomarkers to be included in clinical decision making have to provide additional independent prognostic information or additive value together with established clinical and pathologic variables in a multivariate setting like the Memorial Sloan Kettering Cancer Center or Cleveland Clinic nomograms or Cancer of the Prostate Risk Assessment Postsurgical (CAPRA-S) risk stratification for PCa.

3.2.1. Discovery studies

Table 3 lists studies reporting the prognostic value of gene/expression panels for clinically significant end points. Most have investigated prediction of outcome after RP using different end points, such as risk for biochemical failure [42], [51], [52], [57], and [58], metastatic progression [52], [53], [59], and [60], and DSS [54], [60], and [61]. A few studies investigated TURP tissue to predict the outcome of men undergoing conservative treatment [51], [62], and [63].

Table 3 Studies investigating predictive value of gene/expression panels in prostate cancer

Study No. of cases Tissue type No. of genes analyzed No. of genes in final set End point Mean follow-up, yr Main results Commercial application
Discovery studies
 Cuzick et al [51] 366 (RP)

337 (TURP)
RP TURP 126 31 BCR (RP) DSS (TURP) 9.4/9.8 Expression panel independently predictive for BCR (RP) or PCa mortality (TURP, conservative management) Prolaris
 Erho et al [53] Discovery (n = 359)

Validation (n = 186)
RP NR 22 Metastatic PFS 16.9 Nested case-control study (cases with metastasis, controls with or without PSA relapse after RP). Expression panel had AUC of 0.75 (validation) for prediction of metastasis Decipher
 Talantov et al [57] Discovery (n = 138)

Validation (n = 158)
RP 1200 3 BCR 6.0 BCR risk after RP. Predictive value of combined expression and Kattan postoperative nomogram better than clinical nomogram alone (AUC 0.77 vs 0.67) No
 Sboner et al [63] Discovery (n = 186)

Validation (n = 90)
TURP 6100 18 DSS ≥10 Case-control study (indolent vs lethal PCa in WW cohort). Expression panel not better than clinical model predicting outcome No
 Irshad et al [62] Discovery, 2 sets (n = 29/25)

Validation, 2 sets (n = 131/28)
Various 377 3 BCR NA Several discovery cohorts validated in TURP WW cohorts. Three-gene model had better prediction (AUC: 0.86) than Gleason (0.82) or D’Amico classification (AUC: 0.72) No
 Gasi Tandefelt et al [42] Discovery (n = 48)

Validation (n = 127)
RP NR 36 BCR 10 BCR risk analysis after RP. Expression panel predictive for BCR risk, but only in subgroup of ERG fusion-positive cases No
 Penney et al [64] 358 (TURP)

109 (RP/TURP)
TURP RP 6100 157 DSS ≥10 Expression panel improved prediction of PCa mortality among Gleason 7 cases after conservative management or RP No
 Nakagawa et al [59] Three sets; n = 213 in each RP 1021 17 MFS, DSS NR Case-control study (systemic progression vs PSA relapse only vs no evidence of disease after RP). Expression panel predictive of systemic progression and DSS No
 Wu et al [52] Discovery (n = 209)

Validation (n = 306)
RP 1536 32 BCR, MFS 12.7 Expression panel offered independent predictive value and improved postoperative nomograms in prediction of BCR and freedom from metastasis after RP No
 Chen et al [58] Discovery (n = 78)

Validation (n = 79)
RP 22 283 7 BCR 4.3 Seven-gene panel predictive of BCR in univariate analysis No
 Cheville et al [60] n = 157 RP 38 2 (with ERG and aneuploidy) MFS and DSS NR Case-control study (metastasis/PCa death within 5 yr after RP vs no events, matched for Gleason/TNM/PSA/SM status). Expression panel had AUC of 0.81 (validation: 0.79) for prediction of metastasis or PCa death No
External validation studies
 Cooperberg et al [65] 413

353 (second validation)
RP NA 31 BCR 7.1 CCP score independent predictor of BCR after RP. Combined genetic and clinical CAPRA score outperformed individual scores Prolaris
 Bishoff et al [67] Set 1 (283)

Set 2 (176)

Set 3 (123)
Bx NA 31 BCR MFS 5.1 (1)

7.3 (2)

11.0 (3)
Validation of expression panel from biopsies among patients undergoing RP. Panel independent predictor of BCR and strongest predictor of metastatic progression in univariate analyses Prolaris
 Cuzick et al [66] 349 Bx NA 31 DSS 11.8 Conservatively managed cohort. Expression panel strongest predictor of DSS when compared with clinical parameters Prolaris
 Freedland et al [69] 141 Bx NA 31 BCR DSS 4.8 From biopsies to predict failure after EBRT. Gene panel improved predictive value when added to clinical parameters Prolaris
 Cooperberg et al [61] 185 RP NA 22 DSS 6.4 Case-control study (high risk PCa, PCa death vs no PCa death). Combined high CAPRA and CCP scores predict high risk for PCa death Decipher
 Ross et al [71] 85 RP NA 22 MFS NR Case-control study (BCR after RP, followed by metastasis vs no metastasis). Expression panel more predictive than clinical nomograms both in ROC analysis (AUC: 0.82) and decision-curve analysis Decipher
 Karnes et al [72] 219 RP NA 22 MFS 6.7 Case-cohort study to validate 22-gene expression panel for high-risk RP patients. AUC: 0.79 for 5-yr metastasis risk Decipher
 Den et al [73] 139 RP NA 22 BCR, MFS 7.4 Radiation patients after RP (pT3 or positive SMs). Expression panel independent predictor of BCR and metastasis risk. Additive predictive value when genetic and clinical models combined Decipher
 Klein et al [54] Set 1(n = 441)

Set 2 (n = 167)

Set 3 (n = 395)
RP (1 and 3)

Bx (2)
727 17 MFS (1), adverse RP pathology (2 and 3) NR Expression panel independently predicted adverse RP pathology (high grade/high stage) from biopsies. Inclusion of expression panel improved net benefit when combined with CAPRA in decision-curve analysis Oncotype DX
 Cullen et al [76] 431 Bx NA 17 Adverse RP pathology

5.2 Test associated with BCR risk in univariate analysis and after adjusting for NCCN risk groups Oncotype DX

AUC = area under the curve; BCR = biochemical recurrence; Bx = biopsy; CAPRA = Cancer of the Prostate Risk Assessment; CCP = cell cycle progression; DSS = disease-specific survival; EBRT = external beam radiation; MFS = metastasis-free survival; NA = not applicable; NCCN = National Comprehensive Cancer Network; NR = not reported; OS = overall survival; PCa = prostate cancer; PSA = prostate-specific antigen; RP = radical prostatectomy; SM = surgical margin; TURP = transurethral resection of prostate; WW = watchful waiting.

The design of discovery studies included several approaches: single- and multicenter studies and correlation of gene panel data to outcomes of the full cohort or selected subgroups [42], [51], [52], [57], [58], and [62] or a case-control population selected on a particular outcome [53], [59], [60], and [63]. All except one study reported that the applied expression panel offered significant prognostic information in the particular study cohort. Sboner and coworkers studied TURP tissue from WW patients, but the gene signature failed to improve the prognostic value of a model including clinicopathologic parameters [63] . Studies have utilized different methodological approaches to assess the value of genomic tests. These approaches included traditional statistical methods (survival analyses, multivariable models with other clinicopathologic variables, and receiver operating characteristic analysis) [42], [51], [52], [53], [59], [60], [62], [63], and [64]. In some studies, results from expression panels were combined with other variables or a nomogram to determine if genomic data added prognostic information above the baseline models [54], [57], [58], [61], and [65].

3.2.2. External validation studies

A 46-gene expression panel (31 cell-cycle progression genes and 15 housekeeping genes) initially reported by Cuzick and coworkers in 2011 was validated in four studies and is commercially available as the Prolaris test (Myriad Genetics, Salt Lake City, UT, USA). The test was first validated from biopsy and TURP specimens in a conservatively managed cohort, and the gene panel significantly predicted PCa death in a multivariate model [66] . The panel was externally validated in two RP studies (one analyzing pre-RP biopsy tissue and one using RP tissue) including a total of >1300 patients and was noted to be an independent prognostic factor for BCR and metastatic progression [65] and [67]. When added to a multivariable score reflecting post-RP clinical and pathologic risk (CAPRA-S score) [68] , the gene classifier provided incremental prognostic value beyond standard clinical models (concordance index for combined genetic/clinical model was 0.77 versus 0.73 for the clinical model alone) [65] .

A combined model incorporating CAPRA-S and a cell cycle progression score also performed better than either alone on decision-curve analysis. Similarly, in an external-beam radiation therapy (EBRT) cohort, the panel was an independent prognostic factor after adjusting for clinical variables [69] . The potential impact of Prolaris was investigated in one study where physicians were surveyed about treatment recommendations in 305 men with newly diagnosed PCa [70] . In 65% of the cases, the treatment recommendation changed after the genetic test, and in 40% there was reduction in treatment burden (interventional treatment changed to noninterventional). Although this study shows genomic tests can have a significant impact on treatment decisions, follow-up data were not reported to determine the long-term impact of these changes in management. Furthermore, the test remains very expensive (approximately $3400), and available data on cost effectiveness are limited.

In 2013 Erho et al reported in a case-control study that a 22-gene panel predicted survival after RP [53] . This panel has also been externally validated in multiple cohorts and is commercially available as the Decipher genetic test (GenomeDX Biosciences, Vancouver, BC, Canada). Four studies reported the utilization of this gene panel to predict BCR, metastatic progression, or DSS after RP plus or minus EBRT [61], [71], [72], and [73]. The prognostic accuracy was highest when the genomic classifier and clinical models (CAPRA-S) were combined [61] . In another study, including 85 high-risk RP patients, the 22-gene panel was the only variable associated with metastatic progression in a multivariable model and had a favorable net benefit compared with clinical models (CAPRA-S and Stephenson postoperative nomogram) [68] and [74] in decision-curve analysis [71] and [72]. The test also improved prediction of BCR and metastatic progression risk in a cohort of 139 men undergoing EBRT after RP [73] . The impact of Decipher was evaluated in a clinical utility study where 21 uro-oncologists were presented 24 patient cases (12 potential candidates for adjuvant and 12 for salvage EBRT) and were asked for treatment recommendations with and without information from the genetic test [75] . The recommendation changed in 43% of the adjuvant cases and 53% in the salvage setting, suggesting a potentially significant impact on treatment decisions after RP. However, the long-term impact of these changes in management is unknown.

Another commercially available test, Oncotype DX Genomic Prostate Score (GPS; Genomic Health Inc, Redwood City, CA, USA), is a 17-gene expression panel that has been investigated as a predictor for the risk of recurrence, PCa death, and especially adverse pathology at RP [54] . For the latter, biopsy tissue was used to derive a gene panel and estimate the risk of high-grade (Gleason ≥4 + 3) and/or high-stage disease (pT3 or higher). The panel was validated in a cohort of 395 RP patients, and the Genomic Prostate Score was an independent predictor of unfavorable pathology in models including individual clinical parameters (age, prostate-specific antigen [PSA], clinical stage, and biopsy Gleason score) or a multivariable pretreatment clinical risk model (CAPRA score) [54] . The test was further recently validated on biopsies from 431 patients with very low-, low-, or intermediate-risk PCa. The test was significantly associated with adverse pathologic features and also independently predicted time to BCR after adjusting for risk as well as time to metastases [76] .

It should be noted that although these three PCa expression panels include a total of 85 genes, there is virtually no overlap between the tests. The panels in Prolaris and Decipher have only one gene in common. Importantly, as yet there are no comparative data testing these panels in the same patient cohort.

3.3. Epigenetic signature

A comprehensive next-generation sequencing study of Gu and coworkers recently underscored the prognostic value of global- and gene-specific epigenetic alterations in PCa [77] . A methylation marker genetic test, ConfirmMDx (MDxHealth), utilizes methylation analysis of glutathione S-transferase pi 1 (GSTP1), adenomatous polyposis coli (APC), and Ras association (RalGDS/AF-6) domain family member 1 (RASSF1) genes from negative biopsies to estimate the likelihood of a repeat biopsy also being negative [78] . The test achieved a 90% negative predictive value (NPV) within 30 mo of the initial biopsy. In a recent validation trial, 88% NPV was reported, and the test was the most significant predictor of biopsy results [79] . The impact of the epigenetic test on rebiopsy rates was recently surveyed in five centers, and among 138 patients with a negative ConfirmMDx assay, only six patients (4%) underwent repeat biopsies [80] .

3.4. Copy number variation

Copy number variation (CNV) refers to gains or losses of certain areas of somatic DNA that potentially have carcinogenic consequences (eg, activation of oncogenes or inactivation of tumor suppressor genes) [4] . Overall, PCa is characterized by loss of genomic material [81] . The prognostic role of CNV may be analyzed with different approaches, by either investigating specific genetic gains or deletions, or by analyzing the overall burden of CNV. For example, Tsuchiya et al investigated specific chromosome 8 abnormalities, and loss of 8p22 was associated with an increased risk of BCR and metastatic progression [82] . Liu et al studied the 20 most significant CNVs (15 deletions, 5 amplifications) in two RP cohorts and noted two CNVs (gain of area of v-myc avian myelocytomatosis viral oncogene homolog [MYC], deletion of PTEN) were significantly associated with PCa death [83] . Similar findings were reported in patients undergoing radiation therapy [84] .

Recent advances in high-throughput methodology have allowed investigations of the overall CNV burden and outcome. Taylor and coworkers analyzed RP cohorts for CNV utilizing unsupervised clustering and identified six patient clusters according to the degree of CNV. When analyzed for risk of BCR, CNV clusters had a significant association with outcome in univariate analysis [81] . The “simplest” approach was reported by Hieronymus and coworkers, who studied the association between percentage of CNV from intact somatic DNA and outcome after RP [85] . A significant difference was noted for BCR and metastatic progression risks in patients with ≥5.4% altered tumor DNA. The degree of altered DNA was also an independent predictor of BCR on multivariable analysis of the whole cohort and a subcohort of Gleason 7 tumors.

Paris and coworkers utilized array comparative genomic hybridization to identify specific DNA-based biomarkers (eg, loss at 8p23.2 and gain at 11q13.1). They suggested a combined set of 39 loci termed Genomic Evaluators of Metastatic Prostate Cancer (GEMCaP). In the discovery study, the GEMCaP set of markers was associated with disease recurrence and metastasis [86] . Later the GEMCaP was demonstrated to offer additional prognostic information above the Kattan nomogram for disease recurrence in high-risk node-negative PCa cases after RP (nomogram accuracy 65% vs accuracy of nomogram and GEMCaP 78%) [87] .

According to these studies, CNV analyses may have a prognostic role in PCa patients, but standardization of methods and additional validation studies are required before clinical applications may be planned.

3.5. Genetic information from nucleic acids in peripheral blood and circulating tumor cells

In addition to genetic information available from germline DNA and tumor tissue–derived DNA and RNA, peripheral blood is a potential source for genomic tumor characterization using free circulating nucleic acids, whole blood transcripts, or CTCs.

In 2007 Bastian and coworkers reported an increasing quantity of circulating cell-free DNA was independently associated with the risk of BCR after RP [88] . In November 2012, two separate studies reported on gene expression profiling from blood RNA in patients with CRPC. Ross and coworkers examined a six-gene panel in CRPC patients with significantly improved prognostic value compared with a clinical model alone [89] . Olmos et al used a similar approach but divided the CRPC cohort into four groups according to microarray data analyzed from blood messenger RNA (mRNA) [90] . One patient group had a significantly poorer survival, identified by a nine-gene panel. Specific miRNAs are found, not only in tumor tissue, but also in the plasma of PCa patients; miRNA-375 and miRNA-141 are reported to be associated with advanced disease [91] .

Recently Danila et al investigated the detection of CTCs and the expression of five genes frequently detected in PCa cells (but not in peripheral mononuclear cells) utilizing reverse transcriptase-polymerase chain reaction (RT-PCR) to detect transcripts from peripheral blood [92] . Both unfavorable CTC count (five or more cells) and detection of two or more gene transcripts had similar significant prognostic value for risk of PCa death, and when combined, additional prognostic value was demonstrated. With a similar approach, kallikrein-related peptidase 3 (KLK3), PCA3, and TMPRSS2:ERG mRNA could be detected in the peripheral blood of CRPC patients but not in healthy controls [93] . Also, decreased expression levels of these genes were noted after docetaxel treatment, suggesting a potential role for treatment monitoring.

Peripheral blood genetic information may also be useful to predict therapeutic response in CRPC. Recently, Antonarakis et al reported that a splice variant of the androgen receptor (AR-V7) could be detected in CTCs, and AR-V7–positive patients were less likely to respond to abiraterone or enzalutamide and had a poorer survival [94] . Confirmatory studies are awaited. In addition to specific genetic changes found in CTCs, the pretherapy CTC count has been demonstrated to predict response, and a decrease in the number of CTCs after therapy has greater predictive value than the classic 50% PSA decrease. This was observed after treatment with both docetaxel and abiraterone [95] .

3.6. Discussion

After years of intense research, we are finally witnessing progress in the field of PCa genomics and the emergence of commercially available genetic tests to assist clinical decision making. Because information on these tests is available not only to PCa specialists but to all physicians and patients, it is important to understand their potential implications, optimal use, and limitations. Genetic prediction tools may also add significant costs to the PCa diagnostic and therapeutic algorithms, but these costs might be justified if indeed they lead to a reduction in unnecessary treatments for localized disease or a more appropriate selection of therapy for advanced disease.

An important aspect of biomarker and genetics research is the heterogeneity of PCa both within a single tumor locus (intrafocal heterogeneity) and between different tumor deposits (interfocal heterogeneity) [96], [97], and [98]. In addition to intra/interfocal heterogeneity, a field effect of genetic changes should also be considered because cancer-related genetic changes are also detected in benign areas of the same prostate [99] . This is the underlying premise of new tests designed to predict the risk of finding cancer on repeat biopsy for men with a negative biopsy [79] , as well as biopsy-based tissue tests designed to predict whole-gland pathologic features. Genomic analysis of tumor tissue may aid in overcoming the challenges of sampling error and the variability of traditional pathologic grading. Standard pathologic evaluation, such as Gleason grading, is subjective and associated with significant inter- (and also intra-) observer variability that may have a significant impact on an individual patient's treatment recommendations [100] .

Genetic prognostication has potential applications in every step of PCa care. Commercially available epigenetic ConfirmMDx may be of value when repeat biopsies are considered after negative initial prostate biopsies. One of the most important is the appropriate selection of men to AS versus treatments with curative intent. To offer AS safely, the risk of underestimating the metastatic and local invasive potential of the individual tumor has to be minimized. In addition to improved biopsy techniques and imaging, genomic tests may be used to estimate the potential of tumor progression. The Oncotype DX Genomic Prostate Score was investigated in this setting and found to provide additional information to clinical parameters and nomograms. Even after RP, the risk of recurrence and metastatic progression is highly variable, and the addition of genomic information to traditional variables appears to improve prognostic accuracy modestly.

All three commercially available gene panels described in detail in this review (Prolaris, Decipher, and Oncotype DX Genomic Prostate Score) have been evaluated in terms of potential prognostic value after RP. The future will tell if this additional information is considered sufficient by the urologic community and PCa patients to change practice. Although clinical studies have suggested potential benefits with these tests, real clinical use and long-term data are needed to judge the added value.

In addition to general prognostic information, prediction of response to specific treatment modalities (eg, adjuvant/salvage radiation, ADT, novel systemic agents) is of great importance. Due to an ever-expanding number of treatment options in CRPC, involving very different mechanisms and significant costs, there is a great need for markers to predict therapeutic response, typically seen in a minority of patients. With multiple sequentially delivered treatments, longitudinal monitoring of disease status is needed. In this setting, promise exists for sampling free circulating DNA and RNA or CTCs in peripheral blood, but further work is necessary to validate the findings before widespread clinical use. The issue of tumor cell heterogeneity in CTCs has yet to be explored.

4. Conclusions

Major advances in PCa genetics have occurred in recent years, and in the near future personalized genetic profiling of primary and metastatic tumor cells may become readily available for routine clinical decision making. Many new genetic-based tests are newly available or in late stages of clinical development, with potential applications in PCa decisions ranging from the need for repeat biopsy to initial treatment selection, decisions about secondary therapy, and selection of treatment for advanced disease. Greater understanding of the potential long-term benefits and limitations of these tests is important, and how exactly they should be used in clinical practice to optimize decision making must be the subject of future prospective studies.

Author contributions: Peter J. Boström 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: Boström, Cooperberg.

Acquisition of data: Boström, Bjartell, Catto, Eggener, Lilja, Loeb, Schalken, Schlomm, Cooperberg.

Analysis and interpretation of data: Boström, Bjartell, Catto, Eggener, Lilja, Loeb, Schalken, Schlomm, Cooperberg.

Drafting of the manuscript: Boström.

Critical revision of the manuscript for important intellectual content: Boström, Bjartell, Catto, Eggener, Lilja, Loeb, Schalken, Schlomm, Cooperberg.

Statistical analysis: None.

Obtaining funding: None.

Administrative, technical, or material support: None

Supervision: None.

Other (specify): None

Financial disclosures: Peter J. Boström 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: Peter J. Boström advises and consults for Astellas, Amgen, and Janssen. Scott E. Eggener advises and consults for Genomic Health and MDx Health; he advises, consults, and provides research support for Myriad. Hans Lilja holds patents for free PSA, hK2, and intact PSA assays and is named on a patent application for a statistical method to detect PCa. The method was commercialized by OPKO. Lilja will receive royalties from any sales of the test. Stacy Loeb lectures and advises Bayer. Jack Schalken receives speaker honoraria from Astellas and consultancy payments from Hologic Genprobe. Thorsten Schlomm advises and consults for Genomic Health. Matthew R. Cooperberg conducts ongoing research for Genomic Health, GenomeDx, and Myriad Genetics (no personal financial income), and he advises and consults for Myriad Genetics. Anders S. Bjartell and James W.F. Catto have nothing to disclose.

Funding/Support and role of the sponsor: None.


  • [1] L.A. Torre, F. Bray, R.L. Siegel, J. Ferlay, J. Lortet-Tieulent, A. Jemal. Global cancer statistics, 2012. CA Cancer J Clin. 2015;65:87-108 Crossref
  • [2] M.G. Sanda, R.L. Dunn, J. Michalski, et al. Quality of life and satisfaction with outcome among prostate-cancer survivors. N Engl J Med. 2008;358:1250-1261 Crossref
  • [3] T. Schlomm, A. Erbersdobler, M. Mirlacher, G. Sauter. Molecular staging of prostate cancer in the year 2007. World J Urol. 2007;25:19-30 Crossref
  • [4] C.E. Barbieri, C.H. Bangma, A. Bjartell, et al. The mutational landscape of prostate cancer. Eur Urol. 2013;64:567-576 Crossref
  • [5] A.D. Choudhury, R. Eeles, S.J. Freedland, et al. The role of genetic markers in the management of prostate cancer. Eur Urol. 2012;62:577-587 Crossref
  • [6] C. Jerónimo, P.J. Bastian, A. Bjartell, et al. Epigenetics in prostate cancer: biologic and clinical relevance. Eur Urol. 2011;60:753-766
  • [7] J.W. Catto, A. Alcaraz, A.S. Bjartell, et al. MicroRNA in prostate, bladder, and kidney cancer: a systematic review. Eur Urol. 2011;59:671-681 Crossref
  • [8] J.R. Schoenborn, P. Nelson, M. Fang. Genomic profiling defines subtypes of prostate cancer with the potential for therapeutic stratification. Clin Cancer Res. 2013;19:4058-4066 Crossref
  • [9] H. Easwaran, H.C. Tsai, S.B. Baylin. Cancer epigenetics: tumor heterogeneity, plasticity of stem-like states, and drug resistance. Mol Cell. 2014;54:716-727 Crossref
  • [10] C. Chao, M. Chi, M. Preciado, M.H. Black. Methylation markers for prostate cancer prognosis: a systematic review. Cancer Causes Control. 2013;24:1615-1641 Crossref
  • [11] D. Moher, A. Liberati, J. Tetzlaff, D.G. Altman. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. J Clin Epidemiol. 2009;62:1006-1012 Crossref
  • [12] S.A. Tomlins, D.R. Rhodes, S. Perner, et al. Recurrent fusion of TMPRSS2 and ETS transcription factor genes in prostate cancer. Science. 2005;310:644-648 Crossref
  • [13] 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
  • [14] J.C. King, J. Xu, J. Wongvipat, et al. Cooperativity of TMPRSS2-ERG with PI3-kinase pathway activation in prostate oncogenesis. Nat Genet. 2009;41:524-526 Crossref
  • [15] A. Krohn, A. Seidel, L. Burkhardt, et al. Recurrent deletion of 3p13 targets multiple tumour suppressor genes and defines a distinct subgroup of aggressive ERG fusion-positive prostate cancers. J Pathol. 2013;231:130-141 Crossref
  • [16] F. Demichelis, S.R. Setlur, R. Beroukhim, et al. Distinct genomic aberrations associated with ERG rearranged prostate cancer. Genes Chromosomes Cancer. 2009;48:366-380 Crossref
  • [17] L. Burkhardt, S. Fuchs, A. Krohn, et al. CHD1 is a 5q21 tumor suppressor required for ERG rearrangement in prostate cancer. Cancer Res. 2013;73:2795-2805 Crossref
  • [18] J. Weischenfeldt, R. Simon, L. Feuerbach, et al. Integrative genomic analyses reveal an androgen-driven somatic alteration landscape in early-onset prostate cancer. Cancer Cell. 2013;23:159-170 Crossref
  • [19] S. Steurer, P.S. Mayer, M. Adam, et al. TMPRSS2-ERG fusions are strongly linked to young patient age in low-grade prostate cancer. Eur Urol. 2014;66:978-981 Crossref
  • [20] L.M. FitzGerald, I. Agalliu, K. Johnson, et al. Association of TMPRSS2-ERG gene fusion with clinical characteristics and outcomes: results from a population-based study of prostate cancer. BMC Cancer. 2008;8:230 Crossref
  • [21] A. Gopalan, M.A. Leversha, J.M. Satagopan, et al. TMPRSS2-ERG gene fusion is not associated with outcome in patients treated by prostatectomy. Cancer Res. 2009;69:1400-1406 Crossref
  • [22] A.M. Hoogland, G. Jenster, W.M. van Weerden, 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 Crossref
  • [23] R.K. Nam, L. Sugar, W. Yang, et al. Expression of the TMPRSS2:ERG fusion gene predicts cancer recurrence after surgery for localised prostate cancer. Br J Cancer. 2007;97:1690-1695 Crossref
  • [24] 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
  • [25] O.R. Saramäki, A.E. Harjula, P.M. Martikainen, R.L. Vessella, T.L. Tammela, T. Visakorpi. TMPRSS2:ERG fusion identifies a subgroup of prostate cancers with a favorable prognosis. Clin Cancer Res. 2008;14:3395-3400
  • [26] J.L. Boormans, K. Porkka, T. Visakorpi, J. Trapman. Confirmation of the association of TMPRSS2(exon 0):ERG expression and a favorable prognosis of primary prostate cancer. Eur Urol. 2011;60:183-184 Crossref
  • [27] 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
  • [28] R.K. Nam, L. Sugar, Z. Wang, 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 Crossref
  • [29] A. Dal Pra, E. Lalonde, J. Sykes, et al. TMPRSS2-ERG status is not prognostic following prostate cancer radiotherapy: implications for fusion status and DSB repair. Clin Cancer Res. 2013;19:5202-5209 Crossref
  • [30] G. Attard, J. Clark, L. Ambroisine, et al. Duplication of the fusion of TMPRSS2 to ERG sequences identifies fatal human prostate cancer. Oncogene. 2008;27:253-263 Crossref
  • [31] F. Demichelis, K. Fall, S. Perner, et al. TMPRSS2:ERG gene fusion associated with lethal prostate cancer in a watchful waiting cohort. Oncogene. 2007;26:4596-4599 Crossref
  • [32] C. Hägglöf, P. Hammarsten, K. Strömvall, et al. TMPRSS2-ERG expression predicts prostate cancer survival and associates with stromal biomarkers. PLoS One. 2014;9:e86824
  • [33] M. Qi, X. Yang, F. Zhang, et al. ERG rearrangement is associated with prostate cancer-related death in Chinese prostate cancer patients. PLoS One. 2014;9:e84959 Crossref
  • [34] T.A. Bismar, M. Dolph, L.H. Teng, S. Liu, B. Donnelly. ERG protein expression reflects hormonal treatment response and is associated with Gleason score and prostate cancer specific mortality. Eur J Cancer. 2012;48:538-546 Crossref
  • [35] J.L. Boormans, K.G. Hermans, A.C. Made, et al. Expression of the androgen-regulated fusion gene TMPRSS2-ERG does not predict response to endocrine treatment in hormone-naïve, node-positive prostate cancer. Eur Urol. 2010;57:830-835 Crossref
  • [36] K.A. Leinonen, T.T. Tolonen, H. Bracken, et al. Association of SPINK1 expression and TMPRSS2:ERG fusion with prognosis in endocrine-treated prostate cancer. Clin Cancer Res. 2010;16:2845-2851 Crossref
  • [37] 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;66:851-860 Crossref
  • [38] D.W. Lin, L.F. Newcomb, E.C. Brown, et al. Urinary TMPRSS2:ERG and PCA3 in an active surveillance cohort: results from a baseline analysis in the Canary Prostate Active Surveillance Study. Clin Cancer Res. 2013;19:2442-2450 Crossref
  • [39] B.G. Barwick, M. Abramovitz, M. Kodani, et al. Prostate cancer genes associated with TMPRSS2-ERG gene fusion and prognostic of biochemical recurrence in multiple cohorts. Br J Cancer. 2010;102:570-576 Crossref
  • [40] J.C. Brase, M. Johannes, H. Mannsperger, et al. TMPRSS2-ERG-specific transcriptional modulation is associated with prostate cancer biomarkers and TGF-β signaling. BMC Cancer. 2011;11:507 Crossref
  • [41] R.J. Karnes, J.C. Cheville, C.M. Ida, et al. The ability of biomarkers to predict systemic progression in men with high-risk prostate cancer treated surgically is dependent on ERG status. Cancer Res. 2010;70:8994-9002 Crossref
  • [42] D. Gasi Tandefelt, J.L. Boormans, H.A. van der Korput, G.W. Jenster, J. Trapman. A 36-gene signature predicts clinical progression in a subgroup of ERG-positive prostate cancers. Eur Urol. 2013;64:941-950 Crossref
  • [43] P. Cairns, K. Okami, S. Halachmi, et al. Frequent inactivation of PTEN/MMAC1 in primary prostate cancer. Cancer Res. 1997;57:4997-5000
  • [44] L.H. Saal, P. Johansson, K. Holm, et al. Poor prognosis in carcinoma is associated with a gene expression signature of aberrant PTEN tumor suppressor pathway activity. Proc Natl Acad Sci U S A. 2007;104:7564-7569 Crossref
  • [45] K.A. Leinonen, O.R. Saramäki, B. Furusato, et al. Loss of PTEN is associated with aggressive behavior in ERG-positive prostate cancer. Cancer Epidemiol Biomarkers Prev. 2013;22:2333-2344 Crossref
  • [46] M. Yoshimoto, A.M. Joshua, I.W. Cunha, 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 Crossref
  • [47] A. Krohn, T. Diedler, L. Burkhardt, et al. Genomic deletion of PTEN is associated with tumor progression and early PSA recurrence in ERG fusion-positive and fusion-negative prostate cancer. Am J Pathol. 2012;181:401-412 Crossref
  • [48] A.H. Reid, G. Attard, L. Ambroisine, 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 Crossref
  • [49] P. McCall, C.J. Witton, S. Grimsley, K.V. Nielsen, J. Edwards. Is PTEN loss associated with clinical outcome measures in human prostate cancer?. Br J Cancer. 2008;99:1296-1301 Crossref
  • [50] R. Ferraldeschi, D. Nava Rodrigues, R. Riisnaes, S. Miranda, I. Figueiredo, P. Rescigno. PTEN protein loss and clinical outcome from castration-resistant prostate cancer treated with abiraterone acetate. Eur Urol. 2015;67:795-802 Crossref
  • [51] 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
  • [52] C.L. Wu, B.E. Schroeder, X.J. Ma, et al. Development and validation of a 32-gene prognostic index for prostate cancer progression. Proc Natl Acad Sci U S A. 2013;110:6121-6126 Crossref
  • [53] 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
  • [54] 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
  • [55] M.S. Pepe, Z. Feng, H. Janes, P.M. Bossuyt, J.D. Potter. Pivotal evaluation of the accuracy of a biomarker used for classification or prediction: standards for study design. J Natl Cancer Inst. 2008;100:1432-1438 Crossref
  • [56] L.M. McShane, D.F. Hayes. Publication of tumor marker research results: the necessity for complete and transparent reporting. J Clin Oncol. 2012;30:4223-4232 Crossref
  • [57] D. Talantov, T.A. Jatkoe, M. Böhm, et al. Gene based prediction of clinically localized prostate cancer progression after radical prostatectomy. J Urol. 2010;184:1521-1528 Crossref
  • [58] X. Chen, S. Xu, M. McClelland, et al. An accurate prostate cancer prognosticator using a seven-gene signature plus Gleason score and taking cell type heterogeneity into account. PLoS One. 2012;7:e45178 Crossref
  • [59] 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
  • [60] J.C. Cheville, R.J. Karnes, T.M. Therneau, et al. Gene panel model predictive of outcome in men at high-risk of systemic progression and death from prostate cancer after radical retropubic prostatectomy. J Clin Oncol. 2008;26:3930-3936 Crossref
  • [61] M.R. Cooperberg, E. Davicioni, A. Crisan, R.B. Jenkins, M. Ghadessi, R.J. Karnes. Combined value of validated clinical and genomic risk stratification tools for predicting prostate cancer mortality in a high-risk prostatectomy cohort. Eur Urol. 2015;67:326-333 Crossref
  • [62] S. Irshad, M. Bansal, M. Castillo-Martin, et al. A molecular signature predictive of indolent prostate cancer. Sci Transl Med. 2013;5:202ra122 Crossref
  • [63] A. Sboner, F. Demichelis, S. Calza, et al. Molecular sampling of prostate cancer: a dilemma for predicting disease progression. BMC Med Genomics. 2010;3:8 Crossref
  • [64] 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
  • [65] M.R. Cooperberg, J.P. Simko, J.E. Cowan, et al. Validation of a cell-cycle progression gene panel to improve risk stratification in a contemporary prostatectomy cohort. J Clin Oncol. 2013;31:1428-1434 Crossref
  • [66] J. Cuzick, D.M. Berney, G. Fisher, et al. Prognostic value of a cell cycle progression signature for prostate cancer death in a conservatively managed needle biopsy cohort. Br J Cancer. 2012;106:1095-1099 Crossref
  • [67] J.T. Bishoff, S.J. Freedland, L. Gerber, et al. Prognostic utility of the cell cycle progression score generated from biopsy in men treated with prostatectomy. J Urol. 2014;192:409-414 Crossref
  • [68] M.R. Cooperberg, J.F. Hilton, P.R. Carroll. The CAPRA-S score: a straightforward tool for improved prediction of outcomes after radical prostatectomy. Cancer. 2011;117:5039-5046 Crossref
  • [69] S.J. Freedland, L. Gerber, J. Reid, et al. Prognostic utility of cell cycle progression score in men with prostate cancer after primary external beam radiation therapy. Int J Radiat Oncol Biol Phys. 2013;86:848-853 Crossref
  • [70] E.D. Crawford, M.C. Scholz, A.J. Kar, et al. Cell cycle progression score and treatment decisions in prostate cancer: results from an ongoing registry. Curr Med Res Opin. 2014;30:1025-1031 Crossref
  • [71] A.E. Ross, F.Y. Feng, M. Ghadessi, et al. A genomic classifier predicting metastatic disease progression in men with biochemical recurrence after prostatectomy. Prostate Cancer Prostatic Dis. 2014;17:64-69
  • [72] 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
  • [73] 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 Biol Phys. 2014;89:1038-1046 Crossref
  • [74] A.J. Stephenson, P.T. Scardino, J.A. Eastham, et al. Postoperative nomogram predicting the 10-year probability of prostate cancer recurrence after radical prostatectomy. J Clin Oncol. 2005;23:7005-7012 Crossref
  • [75] K. Badani, D.J. Thompson, C. Buerki, et al. Impact of a genomic classifier of metastatic risk on postoperative treatment recommendations for prostate cancer patients: a report from the DECIDE study group. Oncotarget. 2013;4:600-609
  • [76] J. Cullen, I.L. Rosner, T.C. Brand, et al. A biopsy-based 17-gene Genomic Prostate Score predicts recurrence after radical prostatectomy and adverse surgical pathology in a racially diverse population of men with clinically low- and intermediate-risk prostate cancer. Eur Urol. 2015;68:123-131
  • [77] L. Gu, S.C. Frommel, C.C. Oakes, et al. BAZ2A (TIP5) is involved in epigenetic alterations in prostate cancer and its overexpression predicts disease recurrence. Nat Genet. 2015;47:22-30
  • [78] G.D. Stewart, L. Van Neste, P. Delvenne, et al. Clinical utility of an epigenetic assay to detect occult prostate cancer in histopathologically negative biopsies: results of the MATLOC study. J Urol. 2013;189:1110-1116 Crossref
  • [79] A.W. Partin, L. Van Neste, E.A. Klein, et al. Clinical validation of an epigenetic assay to predict negative histopathological results in repeat prostate biopsies. J Urol. 2014;192:1081-1087 Crossref
  • [80] K.J. Wojno, F.J. Costa, R.J. Cornell, et al. Reduced rate of repeated prostate biopsies observed in ConfirmMDx clinical utility field study. Am Health Drug Benefits. 2014;7:129-134
  • [81] B.S. Taylor, N. Schultz, H. Hieronymus, et al. Integrative genomic profiling of human prostate cancer. Cancer Cell. 2010;18:11-22 Crossref
  • [82] N. Tsuchiya, J.M. Slezak, M.M. Lieber, E.J. Bergstralh, R.B. Jenkins. Clinical significance of alterations of chromosome 8 detected by fluorescence in situ hybridization analysis in pathologic organ-confined prostate cancer. Genes Chromosomes Cancer. 2002;34:363-371 Crossref
  • [83] W. Liu, C.C. Xie, C.Y. Thomas, et al. Genetic markers associated with early cancer-specific mortality following prostatectomy. Cancer. 2013;119:2405-2412 Crossref
  • [84] G. Zafarana, A.S. Ishkanian, C.A. Malloff, et al. Copy number alterations of c-MYC and PTEN are prognostic factors for relapse after prostate cancer radiotherapy. Cancer. 2012;118:4053-4062 Crossref
  • [85] H. Hieronymus, N. Schultz, A. Gopalan, et al. Copy number alteration burden predicts prostate cancer relapse. Proc Natl Acad Sci U S A. 2014;111:11139-11144 Crossref
  • [86] P.L. Paris, A. Andaya, J. Fridlyand, et al. Whole genome scanning identifies genotypes associated with recurrence and metastasis in prostate tumors. Hum Mol Genet. 2004;13:1303-1313 Crossref
  • [87] P.L. Paris, V. Weinberg, G. Albo, et al. A group of genome-based biomarkers that add to a Kattan nomogram for predicting progression in men with high-risk prostate cancer. Clin Cancer Res. 2010;16:195-202 Crossref
  • [88] P.J. Bastian, G.S. Palapattu, S. Yegnasubramanian, et al. Prognostic value of preoperative serum cell-free circulating DNA in men with prostate cancer undergoing radical prostatectomy. Clin Cancer Res. 2007;13:5361-5367 Crossref
  • [89] R.W. Ross, M.D. Galsky, H.I. Scher, et al. A whole-blood RNA transcript-based prognostic model in men with castration-resistant prostate cancer: a prospective study. Lancet Oncol. 2012;13:1105-1113 Crossref
  • [90] D. Olmos, D. Brewer, J. Clark, et al. Prognostic value of blood mRNA expression signatures in castration-resistant prostate cancer: a prospective, two-stage study. Lancet Oncol. 2012;13:1114-1124 Crossref
  • [91] J.C. Brase, M. Johannes, T. Schlomm, et al. Circulating miRNAs are correlated with tumor progression in prostate cancer. Int J Cancer. 2011;128:608-616 Crossref
  • [92] D.C. Danila, A. Anand, N. Schultz, et al. Analytic and clinical validation of a prostate cancer-enhanced messenger RNA detection assay in whole blood as a prognostic biomarker for survival. Eur Urol. 2014;65:1191-1197 Crossref
  • [93] S. Dijkstra, G.H. Leyten, S.A. Jannink, et al. KLK3, PCA3, and TMPRSS2-ERG expression in the peripheral blood mononuclear cell fraction from castration-resistant prostate cancer patients and response to docetaxel treatment. Prostate. 2014;74:1222-1230 Crossref
  • [94] E.S. Antonarakis, C. Lu, H. Wang, et al. AR-V7 and resistance to enzalutamide and abiraterone in prostate cancer. N Engl J Med. 2014;371:1028-1038 Crossref
  • [95] D.C. Danila, M. Fleisher, H.I. Scher. Circulating tumor cells as biomarkers in prostate cancer. Clin Cancer Res. 2011;17:3903-3912 Crossref
  • [96] D. Brocks, Y. Assenov, S. Minner, et al. Intratumor DNA methylation heterogeneity reflects clonal evolution in aggressive prostate cancer. Cell Rep. 2014;8:798-806 Crossref
  • [97] 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
  • [98] M. Gerlinger, J.W. Catto, T.F. Orntoft, F.X. Real, E.C. Zwarthoff, C. Swanton. Intratumour heterogeneity in urologic cancers: from molecular evidence to clinical implications. Eur Urol. 2015;67:729-737 Crossref
  • [99] R.M. Väänänen, H. Lilja, L. Kauko, et al. Cancer-associated changes in the expression of TMPRSS2-ERG, PCA3, and SPINK1 in histologically benign tissue from cancerous vs noncancerous prostatectomy specimens. Urology. 2014;83 511.e1–7
  • [100] J.K. McKenney, J. Simko, M. Bonham, et al. The potential impact of reproducibility of Gleason grading in men with early stage prostate cancer managed by active surveillance: a multi-institutional study. J Urol. 2011;186:465-469 Crossref


a Department of Urology, Turku University Hospital, Turku, Finland

b Department of Urology, Skåne University Hospital Malmö, Lund University, Lund Sweden

c Academic Urology Unit, University of Sheffield, Sheffield, UK

d Section of Urology, University of Chicago, Chicago, IL, USA

e Departments of Laboratory Medicine, Surgery (Urology), and Medicine, Memorial Sloan Kettering Cancer Center, New York, NY, USA

f Nuffield Department of Surgical Sciences, University of Oxford, Oxford, UK

g Institute of Biomedical Technology, University of Tampere, Tampere, Finland

h Department of Urology and Population Health, New York University and Manhattan Veterans Affairs Medical Center, New York, NY, USA

i Department of Urology, Radboud University Medical Center, Nijmegen, the Netherlands

j Martini-Clinic, Prostate Cancer Center, University Medical Center Hamburg-Eppendorf, Hamburg, Germany

k Departments of Urology and Epidemiology and Biostatistics, UCSF Helen Diller Family Comprehensive Cancer Center, San Francisco, CA, USA

Corresponding author. Department of Urology, Turku University Hospital, Kiinamyllynkatu 4–8, 20520, Turku, Finland. Tel. +358 44 2059004.

Place a comment

Your comment *

max length: 5000