Elsevier

European Urology

Volume 65, Issue 3, March 2014, Pages 534-542
European Urology

Platinum Priority – Prostate Cancer
Editorial by Scott A. Tomlins on pp. 543–545 of this issue
Prospective Multicentre Evaluation of PCA3 and TMPRSS2-ERG Gene Fusions as Diagnostic and Prognostic Urinary Biomarkers for Prostate Cancer

https://doi.org/10.1016/j.eururo.2012.11.014Get rights and content

Abstract

Background

Prostate cancer antigen 3 (PCA3) and v-ets erythroblastosis virus E26 oncogene homolog (TMPRSS2-ERG) gene fusions are promising prostate cancer (PCa) specific biomarkers that can be measured in urine.

Objective

To evaluate the diagnostic and prognostic value of Progensa PCA3 and TMPRSS2-ERG gene fusions (as individual biomarkers and as a panel) for PCa in a prospective multicentre setting.

Design, setting, and participants

At six centres, post–digital rectal examination first-catch urine specimens prior to prostate biopsies were prospectively collected from 497 men. We assessed the predictive value of Progensa PCA3 and TMPRSS2-ERG (quantitative nucleic acid amplification assay to detect TMPRSS2-ERG messenger RNA [mRNA]) for PCa, Gleason score, clinical tumour stage, and PCa significance (individually and as a marker panel). This was compared with serum prostate-specific antigen and the European Randomised Study of Screening for Prostate Cancer (ERSPC) risk calculator. In a subgroup (n = 61) we evaluated biomarker association with prostatectomy outcome.

Outcome measurements and statistical analysis

Univariate and multivariate logistic regression analysis and receiver operating curves were used.

Results and limitations

Urine samples of 443 men contained sufficient mRNA for marker analysis. PCa was diagnosed in 196 of 443 men. Both PCA3 and TMPRSS2-ERG had significant additional predictive value to the ERSPC risk calculator parameters in multivariate analysis (p < 0.001 and resp. p = 0.002). The area under the curve (AUC) increased from 0.799 (ERSPC risk calculator), to 0.833 (ERSPC risk calculator plus PCA3), to 0.842 (ERSPC risk calculator plus PCA3 plus TMPRSS2-ERG) to predict PCa. Sensitivity of PCA3 increased from 68% to 76% when combined with TMPRSS2-ERG. TMPRSS2-ERG added significant predictive value to the ERSPC risk calculator to predict biopsy Gleason score (p < 0.001) and clinical tumour stage (p = 0.023), whereas PCA3 did not.

Conclusions

TMPRSS2-ERG had independent additional predictive value to PCA3 and the ERSPC risk calculator parameters for predicting PCa. TMPRSS2-ERG had prognostic value, whereas PCA3 did not. Implementing the novel urinary biomarker panel PCA3 and TMPRSS2-ERG into clinical practice would lead to a considerable reduction of the number of prostate biopsies.

Introduction

The gold standard for the diagnosis of prostate cancer (PCa) is based on the histopathologic evaluation of prostate biopsies, an invasive procedure with significant morbidity. Because localised PCa often does not present with symptoms, the selection of men qualifying for prostate biopsies relies on serum prostate-specific antigen (PSA) testing and digital rectal examination (DRE). PSA is currently the only widely used serum biomarker for PCa. However, PSA has a low specificity of 25–40% in the so-called grey area of PSA levels 4.0–10.0 ng/ml, resulting in a high negative biopsy rate [1], [2]. Widespread PSA testing also leads to the diagnosis of clinically insignificant tumours, resulting in potential overtreatment, causing morbidity and leading to unnecessary increased health care costs. In the ongoing search for more specific biomarkers for PCa, prostate cancer antigen 3 (PCA3) and v-ets erythroblastosis virus E26 oncogene homolog (ERG, also known as TMPRSS2-ERG) gene fusion transcripts have been identified as promising urinary novel biomarkers [3], [4].

The Progensa PCA3 test is approved by the US Food and Drug Administration and commercially available to aid in the decision of taking repeat prostate biopsies. A PCA3 cut-off score of 35 is generally used. However, most recent studies show that a lower cut-off score of 25 might be preferable [5], [6], [7], [8]. The results of a possible correlation with established prognostic factors (histologic Gleason grade and tumour stage) are conflicting [5], [8], [9], [10], [11]. The consensus in most papers is that PCA3 is often negative in patients with indolent cancer, yet in the clinically significant cancers there is no evidence for an association with histopathologic prognostic factors (stage, grade).

In PCa, transmembrane protease, serine 2 (TMPRSS2) can be fused to ERG (a member of the ETS family of oncogenes) [4]. These TMPRSS2-ERG gene fusions are highly specific for PCa and are present in approximately half of white PCa patients [12]. PCA3 and TMPRSS2-ERG are also found occasionally in high-grade prostatic intraepithelial neoplasia, in prostate glands in which PCa is also mostly found [4]. In 2006, the TMPRSS2-ERG gene fusion transcripts were successfully detected in urine samples [13]. This urine test had a sensitivity of 37% and a specificity of 93% for the prediction of PCa on prostate biopsy [14]. The prognostic value of this urine test has not yet been assessed.

Considering the heterogeneous character of the disease, the preferred approach in the diagnostic process of PCa will likely be the use of a panel of biomarkers. In 2007, it was shown that the combined use of PCA3 and TMPRSS2-ERG gene fusion transcripts improved sensitivity significantly [14]. This was recently confirmed by Tomlins et al.; however, that study was not conducted prospectively [15]. The aim of our study was to evaluate the diagnostic and prognostic predictive value of Progensa PCA3 plus TMPRSS2-ERG gene fusions (as individual biomarkers and as a panel) in a prospective multicentre setting, in accordance with the Standards for Reporting of Diagnostic Accuracy criteria [16].

Section snippets

Study design: clinical study

Inclusion criteria were men scheduled for prostate biopsy based on elevated serum PSA levels (≥3 ng/ml), a family history of PCa, or an abnormal DRE. Exclusion criteria were a history of PCa, medical therapy known to affect serum PSA levels, symptoms of urinary tract infection, prostate biopsy within 3 mo prior to enrolment, or invasive treatment for benign prostatic hyperplasia within 6 mo prior to enrolment. Subjects were recruited at six urology centres in the Netherlands (Radboud University

Results

Urine samples of 497 men were collected between September 2009 and July 2011, of which 54 samples were excluded for containing too much (in)organic precipitate or inadequate amounts of prostate cells (<1000 copies PSA mRNA). Thus, in total, the urine samples of 443 men were analysed successfully (89%). Table 1 shows the patient characteristics. Median serum PSA was 7.4 ng/ml, of which 274 men (62%) had a serum PSA level 4.0–10.0 ng/ml and 140 men (32%) had a serum PSA level ≥10.0 ng/ml. Of the

Discussion

In our study, PCA3 was a highly accurate biomarker for predicting PCa, exceeding the performance of the widely used serum PSA. The optimal PCA3 cut-off score is still subject to debate. Several studies suggested lowering the cut-off score from 35 to 25 [5], [6], [7], [8]. In this study, the sensitivity for diagnostically clinically significant PCa increased remarkably when lowering the cut-off score to 25 (68% vs 83%). Concurrently, the specificity decreased from 58% to 51%. Defining the

Conclusions

In this prospective multicentre study we evaluated two novel urinary biomarkers for PCa; Progensa PCA3 and TMPRSS2-ERG had independent additional predictive value for PCA3 and ERSPC risk calculator parameters for predicting PCa. In addition, TMPRSS2-ERG had prognostic value. Implementing the novel biomarker panel PCA3 and TMPRSS2-ERG into clinical practice would lead to a considerable reduction of prostate biopsies.

References (26)

  • H. Nakanishi et al.

    PCA3 molecular urine assay correlates with prostate cancer tumor volume: implication in selecting candidates for active surveillance

    J Urol

    (2008)
  • D. Hessels et al.

    Predictive value of PCA3 in urinary sediments in determining clinico-pathological characteristics of prostate cancer

    Prostate

    (2010)
  • M.P. van Gils et al.

    Detailed analysis of histopathological parameters in radical prostatectomy specimens and PCA3 urine test results

    Prostate

    (2008)
  • Cited by (295)

    View all citing articles on Scopus

    Please visit www.eu-acme.org/europeanurology to read and answer questions on-line. The EU-ACME credits will then be attributed automatically.

    View full text