TMPRSS2-ERG fusion transcripts in matched urine and needle rinse material after biopsy for the detection of prostate cancer: really a step forward?

Prostate cancer (PCa) is the most common malignancy in America, Western Europe, and Australia, and there are large discrepancies in incidence rates. In a few countries in South America and Africa and in Sweden, PCa is the leading cause of cancer deaths (1 ). Prostatespecific antigen (PSA) is the best available tumor marker. It has an irreplaceable role for PCa follow-up and prognosis but has severe limitations for diagnosing PCa (2 ). The low diagnostic specificity and positive predictive value of only approximately 25% for screening (3 ) encourages opponents of PSA to be overly critical and to neglect some of its contributions (e.g., dramatic reduction of metastatic disease and a mortality rate reduced by approximately 20% (4 ). Other problems, such as a reduced diagnostic sensitivity and the overdetection with a 4g/L PSA cutoff and subsequent overtreatment (5 ), support the need for new biomarkers that overcome these obstacles. Although the vast majority of patients with PCa will not die from the disease, the goal is to detect aggressive and lifethreatening PCa. Therefore, scientists are searching for more PCa-specific markers that can preferentially detect aggressive cancer. As recent examples, a subform of free PSA, [ 2]proPSA (6 ), and the urinary PCa antigen 3 (PCA3) have already shown some promising results (7 ). The discovery of fusion transcripts for the TMPRSS2 (transmembrane protease, serine 2) and ERG [v-ets erythroblastosis virus E26 oncogene homolog (avian)] genes (i.e., TMPRSS2-ERG) in PCa tissue was one of the most important developments in the field of PCa research in the last decade (8). Meanwhile, other gene fusions, such as SLC45A3-ERG (solute carrier family 45, member 3–ERG), TMPRSS2-ETV1 (TMPRSS2–ets variant 1), SLC45A3-ETV1, and NDRG1-ERG (N-myc downstream regulated 1–ERG) have been described (9 ). If there is a detectable TMPRSS2-ERG gene fusion, it is almost certainly cancer associated (10 ); however, the prevalence of the TMPRSS2-ERG gene fusion is only about 50% in PCa patients (9 ), and its occurrence in some premalignant prostate tissue lesions (10 ) suggests that such a test will hardly function as a screening tool. In contrast, the high diagnostic specificity indicates that the TMPRSS2-ERG gene fusion could be used as a noninvasive adjunct marker in urine, in combination with other markers such as PSA and PCA3, to improve overall diagnostic accuracy (11 ). Therefore, a urinary assay for the TMPRSS2-ERG fusion gene that requires stabilization of the urine sample after collection is currently in development (12 ). Thus, this approach prevents possible errors due to inappropriate sample collection or analyte preparation. In a study reported in this issue of Clinical Chemistry, Bories et al. combined the data obtained from a biopsy rinse material and TMPRSS2-ERG fusion data obtained from urine sediments after digital rectal examination to evaluate 57 patients undergoing prostate biopsy (13 ). After the sampling of 10 –12 cores, the biopsy needle (without any tissue) was taken at the end of a biopsy procedure and rinsed repeatedly into a tube containing a reagent that prevents RNA degradation. Total RNA was then isolated, cDNA produced, and quantitative real-time PCR performed. The data showed a 69% diagnostic sensitivity for the urinary TMPRSS2-ERG results, a 62% diagnostic sensitivity for the biopsy rinse material, and an increased diagnostic sensitivity (89%) when these data were combined. A comparison with tissue TMPRSS2-ERG data was not performed. It would have been desirable to compare data from urine, biopsy rinse material, and tissue collected from the biopsy cores, or, better yet, from the whole prostate after radical prostatectomy in PCa patients. The authors showed the feasibility of detecting TMPRSS2-ERG in biopsy rinse material (13 ); however, 1 Department of Urology, Charité-Universitätsmedizin Berlin, Berlin, Germany; 2 Berlin Institute for Urologic Research, Berlin, Germany. * Address correspondence to this author at: University Hospital Charité, Charitéplatz 1, 10117 Berlin, Germany. Fax 49-30-450-5159504; e-mail cstephan6@ yahoo.com. Received October 18, 2012; accepted October 22, 2012. Previously published online at DOI: 10.1373/clinchem.2012.196360 3 Nonstandard abbreviations: PCa, prostate cancer; PSA, prostate-specific antigen; PCA3, PCa antigen 3. 4 Human genes: TMPRSS2, transmembrane protease, serine 2; ERG, v-ets erythroblastosis virus E26 oncogene homolog (avian); SLC45A3, solute carrier family 45, member 3; ETV1, ets variant 1; NDRG1, N-myc downstream regulated 1. Clinical Chemistry 59:1 9–10 (2013) Editorials