This web only provides the extract of this article. If you want to read the figures and tables, please reference the PDF full text on Blackwell Synergy. Thank you.

- Original Article -

Detection of TMPRSS2:ERG fusion gene in circulating prostate cancer cells

Xueying Mao¹, Greg Shaw^1,2, Sharon Y. James¹, Patricia Purkis¹, Sakunthala C. Kudahetti¹, Theodora Tsigani¹, Saname Kia¹, Bryan D. Young¹, R. Tim D. Oliver¹, Dan Berney³, David M. Prowse², Yong-Jie Lu¹

¹Medical Oncology and ²Molecular Oncology Centres, Institute of Cancer, ³Department of Histopathology and Morbid Anatomy, Barts and London School of Medicine and Dentistry, Queen Mary, University of London, Charterhouse Square, London EC1M 6BQ, UK

Abstract

Aim: To investigate the existence of TMPRSS2:ERG fusion gene in circulating tumor cells (CTC) from prostate cancer patients and its potential in monitoring tumor metastasis.  Methods: We analyzed the frequency of TMPRSS2:ERG and TMPRSS2:ETV1 transcripts in 27 prostate cancer biopsies from prostatectomies, and TMPRSS2:ERG transcripts in CTC isolated from 15 patients with advanced androgen independent disease using reverse transcription polymerase chain reaction (RT-PCR). Fluorescence in situ hybridization (FISH) was applied to analyze the genomic truncation of ERG, which is the result of TMPRSS2:ERG fusion in 10 of the 15 CTC samples. Results: TMPRSS2:ERG transcripts were found in 44% of our samples, but we did not detect expression of TMPRSS2:ETV1. Using FISH analysis we detected chromosomal rearrangements affecting the ERG gene in 6 of 10 CTC samples, including 1 case with associated TMPRSS2:ERG fusion at the primary site. However, TMPRSS2:ERG transcripts were not detected in any of the 15 CTC samples, including the 10 cases analyzed by FISH.  Conclusion: Although further study is required to address the association between TMPRSS2:ERG fusion and prostate cancer metastasis, detection of genomic truncation of the ERG gene by FISH analysis could be useful for monitoring the appearance of CTC and the potential for prostate cancer metastasis. (Asian J Androl 2008 May; 10: 467_473)

Keywords: TMPRSS2:ERG; fusion gene; prostate cancer; metastasis; circulating tumor cells; fluorescence in situ hybridization; polymerase chain reaction

Correspondence to: Dr Yong-Jie Lu, Medical Oncology Centre, Cancer Institute, Barts and London School of Medicine and Dentistry, Queen Mary, University of London, Charterhouse Square, London EC1M 6BQ, UK.
Tel: +44-20-7882-6140 Fax: +44-20-7882-6004
E-mail: yong.lu@cancer.org.uk
Received 2007-11-09 Accepted 2008-02-20

DOI: 10.1111/j.1745-7262.2008.00401.x

1 Introduction

Prostate cancer is the most common non-cutaneous malignancy in men living in developed nations [1]. Most prostate cancers diagnosed at an early stage are latent, but once having progressed they become life threatening, and, despite recent advances, late stage prostate cancer is still incurable [2]. With the general application of prostate specific antigen (PSA) testing for prostate cancer screening and diagnosis, there is also a dilemma in the clinical treatment of early stage cancers, because it is difficult to predict their progression potential [2]. Current methods either over-treat the majority of early prostate cancer patients who will not die from prostate cancer even without treatment, or miss the opportunity to cure early stage aggressive disease. As it is the metastatic and androgen independent disease that is responsible for cancer deaths, detecting signs of aggressive cancer in clinically early stages of the disease would be invaluable [3, 4]. However, there are currently no reliable prognostic biomarkers to predict prostate cancer progression and metastastic potential [2, 4].

Fusion genes have been studied a great deal in hematological malignancies and soft tissue sarcomas, where they frequently define a tumor subtype and are associated with prognosis. The recurrent fusion of the TMPRSS2 and ETS family transcription factor genes was recently identified in prostate cancer [5, 6]. Fusion genes are de novo genes generated in cancer cells. They can be used as a marker to detect cancer cells in minimum residual disease [7], and also as targets for cancer cell specific treatment [5, 8]. If these fusion genes, particularly the high frequency of TMPRSS2:ERG fusion gene, play a role in tumor progression and, most importantly, tumor metastasis, they will be invaluable markers for treatment stratification and targets for novel forms of therapy. However, studies on the association between TMPRSS2:ETS fusion gene and tumor progression have generated inconsistent results, and few studies have attempted to address the role of the fusion gene in advanced metastastic prostate cancer [9_17].

Long distance tumor metastasis is caused by tumor cell migration from the primary site into the blood stream, culminating in survival and re-establishment at a new site. It has been known for 150 years that tumor cells are detectable in the blood of patients with advanced cancer [18]. Sampling circulating tumor cells (CTC) may be used to study genetic and gene expression alterations in a continual, relatively non-invasive fashion, and to monitor treatment. Using genetic and/or biomarkers to detect the CTC might also provide informed treatment planning. With the recent advances in use of specific antibodies and various sorting techniques following antibody labeling, the ratio of epithelial to blood cells can be augmented so that analysis of these cells is more feasible [19_21]. However, many genes associated with prostate carcinogenesis are also expressed in hematopoietic cells, and the existence of non-tumor epithelial cells in the peripheral blood complicates CTC analysis [22, 23]. For prostate cancer, the detection of micrometastases is also limited by the lack of specific cancer cell markers [22, 24, 25]. As the TMPRSS2:ERG fusion gene occurs at a high frequency in prostate cancer and is a specific genetic marker of abnormal prostate cells, we have attempted to detect its existence in CTC to monitor tumor progression and metastatic potential.

2 Materials and methods

2.1 Materials

Twenty-seven primary prostate cancer biopsies from radical prostatectomy and fifteen peripheral blood samples from patients with advanced prostate cancer (with metastasis or PSA > 40 ng/mL) were obtained from local hospitals with patients' consent and local research ethical committee's approval. A cancer biopsy and a blood sample were available from 1 patient (primary biopsy PC127 and blood sample PCB34). The clinico-pathological data are summarized in Tables 1 and 2. All blood samples were taken from patients who received second line hormone therapy, except for PCB32, who was on first line treatment. Tissue biopsies were kept snap frozen in liquid nitrogen and all samples in the present study were confirmed with cancer by a consultant histopathologist (Dan Berney), who reviewed fresh frozen sections from the collected tissue.

2.2 Cell separation and circulating tumor cells purification from blood samples

Peripheral blood samples were collected in EDTA-coated tubes and processed as follows. From a small volume (5 mL) blood collection, lymphocytes and tumor cells were separated from red cells by spinning in Ficoll gradient buffer and then used directly for RNA extraction. From a large volume (20 mL) blood collection, CTC were purified by magnetic cell sorting. Initially, the red blood cells and the majority of the white blood cells were removed from the 20 mL blood sample by spinning in an Oncoquick tube pre-filled with separation buffer (Greiner; Kremsmuenster, Austria). The aspirate was then washed twice with Oncoquick washing buffer and resuspended in MACS running buffer (Miltenyi Biotech; Surrey, UK) for CTC selection. A negative selection for cells which were labeled with CD45 (a lymphocyte marker) was performed using AutoMACS (Miltenyi Biotech). This was followed by a positive selection for cells labeled with EpCAM (Epithelial Cell Adhesion Molecule) microbeads.

2.3 Immunostaining

EpCam (Abcam, Cambridge, UK) antibodies were used at concentrations recommended by the manufacturers. CTC were air dried onto slides, fixed, incubated overnight with primary antibodies and detected using appropriate secondary antibodies. Nuclear counterstaining was performed using 4,6-diamidino-2-phenylindole dihydrochloride (DAPI). Stained preparations were analyzed on a Zeiss 510 confocal microscope (Carl Zeiss Ltd.; Jena, Germany).

2.4 RNA extraction and reverse transcription polymerase chain reaction (RT-PCR) analysis

RNA was extracted from snap frozen tissues and Ficoll separated cells from the small blood volume samples using the Trizol method following the manufacturer's instructions (Invitrogen, Carlsbad, CA, USA). The RNeasy Mini kit (Qiagen, Crawley, UK) was also used according to the manufacturer's instructions to extract RNA from the CTC.

Reverse transcription of RNA was performed using the Superscript II enzyme (Invitrogen). Random primers (6pN) were used for RNA extracted from the cancer biopsies and the Poly-A primer was used for RNA from the circulating cells.

The primers (for TMPRSS2:ERG forward primer CAGGAGGCGGAGGCGGA and reverse primer GGCGTTGTAGCTGGGGGTGAG and for TMPRSS2:ETV1 forward primer CAGGAGGCGGAGGCGGA and reverse primer TTGTGGTGGGAAGGGGATGTTT) described by Tomlins et al. [5, 8] were used for PCR amplification with an annealing temperature of 65ºC. For standard RT-PCR, 35 cycles were used, and 40 cycles were used for quantitative RT-PCR (Q-RT-PCR). β-actin with the forward primer gatgagatt ggcatggcttt and reverse primer caccttcac cgttccagttt was used as a positive control. The Opticon DNA Engine 2 (MJ Research, Waltham, USA) was used to perform the Q-RT-PCR thermal cycling and Opticon monitor software (MJ Research) was used for data analysis.

2.5 Sequence analysis

PCR products were cloned into the pCR 2.1-TOPO Vector using the TOPO TA Cloning Kit (Invitrogen). Single bacterial clones were picked and directly lysed in PCR buffer for insert amplification using the M13F and M13R primers and then sequenced using the ABI Prism 3700 DNA Analyser (Applied Biosystem). DNA sequences were analyzed using the 4 Peaks software (Netherlands Cancer Institute, Amsterdam, the Netherlands).

2.6 Fluorescence in situ hybridization

The BAC clones (RP11-95I21 and RP11-476D17) on either side of the ERG gene were obtained from The Sanger Institute (Cambridge, UK) and differentially labeled in red and green fluorescent dyes, as previously described [26]. CTC cells were dropped onto glass slides and fixed in situ with ethanol. Before hybridization, they were pretreated in 70% acetic acid for 10 min to remove the cytoplasm. Then slides and probes were denatured separately and hybridized overnight at 37ºC following standard fluorescence in situ hybridization (FISH) methodology. Hybridized cells were counterstained by DAPI and images were captured using an Olympus fluorescence microscope (Olympus Optical Co. Ltd., Tokyo, Japan) mounted with a cooled coupled device camera controlled by the computer software Mac Probe 4.3 (Applied Imaging, Newcastle, UK).

2.7 Statistics

The Fisher exact test was performed to compare the frequency of TMPRSS2:ERG fusion gene between the primary sample and the CTC.

3 Results

We analyzed 27 primary prostate cancer biopsies from radical prostatectomy using standard RT-PCR for both the TMPRSS2:ERG and TMPRSS2:ETV1 fusion genes. Of the samples, 12 (44%) were TMPRSS2:ERG fusion positive and in 7 samples more than one form of fusion transcripts were detected (Table 1). The most common fusion form detected in 9 of the 12 positive cases was between exon 1 of TMPRSS2 and exon 4 of ERG (T1/E4), as originally reported by Tomlins et al. [5, 8]. Other fusion forms, such as T1/E2, T1/E5 and T2/E5, as previously reported by Clark et al. [14, 27_30], were also detected. The TMPRSS2:ETV1 fusion gene was not detected in any samples.

As the TMPRSS2:ERG fusion gene occurred at a very high frequency, we further investigated the possibility of detecting its fusion transcripts by blood testing. We analyzed four peripheral blood samples (one of them, PCB34, taken from a fusion positive case, PC127) using standard RT-PCR. We took 5 mL of blood from each sample, and failed to detect any fusion products.

Because both lack of CTC in the small blood volume and a large amount of lymphocyte contamination could result in the failure to detect the fusion gene, we further analyzed the separated CTC from 20 mL blood samples from patients with advanced androgen independent prostate cancer receiving second line hormone therapy. The purity of cancer cells was enriched and each sample contained more than 100 nucleated CD45 negative, cytokeratin/EpCAM positive cells (Figure 1). However, of the 15 samples analyzed using both standard RT-PCR and Q-RT-PCR, including 1 case in which the primary cancer biopsy was fusion positive, no fusion transcripts were detected in any CTC (Figure 2A, B). We confirmed the sensitivity of our RT-PCR protocol in detecting TMPRSS2:ERG fusion transcripts by mixing different dilutions of fusion positive VCaP cells with fusion negative LNCaP cells. We detected the fusion product in approximately 20 (100 pg RNA) but not 10 (50 pg RNA) VCaP cells mixed with 200 LNCaP cells (Figure 2C).

As RT-PCR is limited to detecting the fusion transcript in single cells, we applied FISH to analyze the genomic truncation of ERG (which is the result of TMPRSS2:ERG fusion) in individual cells. Isolated CTC from 10 of the 15 blood samples were available for this type of analysis, and we found ERG truncation in 6 of them, including the sample accompanied by a TMPRSS2:ERG fusion positive primary tumor (Table 2, Figure 3). Although the frequency of TMPRSS2:ERG fusion was higher than that detected in our primary samples, it was not statistically significant (P = 0.319)

4 Discussion

There are currently no reliable markers for predicting the behavior of prostate cancer diagnosed at an early stage. Subsequently, there is a dilemma as to how to treat localized tumors, which are increasingly being detected by PSA screening of tumors. The high frequency of TMPRSS2:ERG fusion in prostate cancer and the detection of the fusion transcript in early stage cancers and even pre-cancerous lesions [27, 31] has promted investigations into the potential of using the fusion product to predict cancer progression [9_12, 14_17].

In the present study, we detected a high frequency of the TMPRSS2:ERG fusion gene in primary prostate cancer biopsies from radical prostatectomy using RT-PCR, which is comparable to previous studies [5, 14, 27_33]. We also found the co-existence of multiple forms of fusion variants of TMPRSS2:ERG, which correlates with several previous studies [14, 27_30]. Most importantly, we detected that ERG truncation generally resulted from the gene fusion at a high frequency (6/10) in CTC samples prepared from the peripheral blood, suggesting that cancer cells with TMPRSS2:ERG fusion frequently migrate into the blood vessel for long distance seeding. Our observation of TMPRSS2:ERG fusion gene positive CTC is consistent with a previous observation where fusion genes were passed on from the primary tumors to lymph node metastatic cells [34]. The frequency of TMPRSS2:ERG fusion was not significantly higher in the CTC samples when compared to that detected in our primary samples. However, the number of CTC samples analyzed in the present study is small, and further investigation of this fusion gene in a larger series of CTC samples will be required.

CTC analysis can be used to monitor tumor progression and response to therapies [35]. However, no markers can currently be used to separate cancer from normal prostate cells. Truncation of ERG or fusion of TMPRSS2:ERG is a specific marker for cancer cells. The TMPRSS2:ERG fusion may be used to detect tumor cells circulating in the blood in a large proportion of cases, although not all cases of CTC may be identified. This makes the detection of ERG truncation or TMPRSS2:ERG fusion in CTC a specific tool for monitoring tumor metastasis before apparent long distance metastasis occurs. In our study, we detected ERG alteration positive cells in two cases of prostate cancers without detected metastatic tumors. The present study has mainly established the principle in detecting TMPRSS2:ERG fusion in CTC using advanced prostate cancers. Further investigation is required to evaluate its application in monitoring early stage disease.

Although the TMPRSS2:ERG fusion was detected at the genomic level, we did not detect the fusion transcript in any of the CTC isolated from patient blood samples. RT-PCR is a sensitive method for detecting fusion gene transcripts in a small amount of cells. The protocol we used can detect a sample containing 20 TMPRSS2:ERG fusion positive cells from the VCaP cell line. There are more than 100 prostate epithelial cells in each of the selected populations from blood samples determined by immunostaining using EpCAM. Although we still cannot absolutely exclude the limitation of technical sensitivity as a reason for the failure to detect the fusion transcripts in the small number of CTC, it is most likely that the expression of TMPRSS2:ERG is switched off/down in the CTC. First, as all the patients are androgen independent, it is most likely that fusion gene expression in CTC from these patients is switched-off/down as a result of androgen ablation therapy [36]. In a previous report, the expression of TMPRSS2:ERG was not detectable in androgen independent cancers, including some metastatic samples, although the genomic fusion existed in some cases [14]. Second, most of the CTC in the blood stream fail to proliferate in culture and appear terminally differentiated, which might also affect fusion gene expression.

In summary, we have detected a high frequency of TMPRSS2:ERG fusion not just in the primary tumors but also in CTC. Although the expression of TMPRSS2:ERG fusion gene was not detected in CTC, genomic detection of the fusion gene by FISH in CTC could be used clinically to monitor the early signs of prostate cancer metastasis.

Acknowledgment

We thank the Orchid Cancer Appeal and Prostate Research Campaign UK for funding support. David M. Prowse is a Research Council UK Academic fellow.

References

1 Jemal A, Murray T, Ward E, Samuels A, Tiwari RC, Ghafoor A, et al. Cancer statistics, 2005. CA Cancer J Clin 2005; 55: 10_30.

2 Ulbright TM. Male genital tract. In: Alison MR, editor. The Cancer Handbook. London: Nature publishing group; 2002: p665_87.

3 Xu XF, Zhou SW, Zhang X, Ye ZQ, Zhang JH, Ma X, et al. Prostate androgen-regulated gene: a novel potential target for androgen-independent prostate cancer therapy. Asian J Androl 2006; 8: 455_62.

4 Yang J, Wu HF, Qian LX, Zhang W, Hua LX, Yu ML, et al. Increased expressions of vascular endothelial growth factor (VEGF), VEGF-C and VEGF receptor-3 in prostate cancer tissue are associated with tumor progression. Asian J Androl 2006; 8: 169_75.

5 Tomlins SA, Rhodes DR, Perner S, Dhanasekaran SM, Mehra R, Sun XW, et al. Recurrent fusion of TMPRSS2 and ETS transcription factor genes in prostate cancer. Science 2005; 310: 644_8.

6 Tomlins SA, Mehra R, Rhodes DR, Smith LR, Roulston D, Helgeson BE, et al. TMPRSS2:ETV4 gene fusions define a third molecular subtype of prostate cancer. Cancer Res 2006; 66: 3396_400.

7 Yin JA, Grimwade D. Minimal residual disease evaluation in acute myeloid leukaemia. Lancet 2002; 360: 160_62.

8 Rowley JD. Chromosome translocations: dangerous liaisons revisited. Nat Rev Cancer 2001; 1: 245_50.

9 Winnes M, Lissbrant E, Damber JE, Stenman G. Molecular genetic analyses of the TMPRSS2-ERG and TMPRSS2-ETV1 gene fusions in 50 cases of prostate cancer. Oncol Rep 2007; 17: 1033_6.

10 Rajput AB, Miller MA, De Luca A, Boyd N, Leung S, Hurtado-Coll A, et al. Frequency of the TMPRSS2:ERG gene fusion is increased in moderate to poorly differentiated prostate cancers. J Clin Pathol 2007; 60: 1238_43.

11 Nam RK, Sugar L, Wang Z, Yang W, Kitching R, Klotz LH, 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_5.

12 Mehra R, Tomlins SA, Shen R, Nadeem O, Wang L, Wei JT, et al. Comprehensive assessment of TMPRSS2 and ETS family gene aberrations in clinically localized prostate cancer. Mod Pathol 2007; 20: 538_44.

13 Lapointe J, Kim YH, Miller MA, Li C, Kaygusuz G, van de Rijn M, et al. A variant TMPRSS2 isoform and ERG fusion product in prostate cancer with implications for molecular diagnosis. Mod Pathol 2007; 20: 467_73.

14 Hermans KG, van Marion R, van Dekken H, Jenster G, van Weerden WM, Trapman J. TMPRSS2:ERG fusion by translocation or interstitial deletion is highly relevant in androgen-dependent prostate cancer, but is bypassed in late-stage androgen receptor-negative prostate cancer. Cancer Res 2006; 66: 10658_63.

15 Demichelis F, Fall K, Perner S, Andren O, Schmidt F, Setlur SR, et al. TMPRSS2:ERG gene fusion associated with lethal prostate cancer in a watchful waiting cohort. Oncogene 2007; 26: 4596_9.

16 Attard G, Clark J, Ambroisine L, Fisher G, Kovacs G, Flohr P, et al. Duplication of the fusion of TMPRSS2 to ERG sequences identifies fatal human prostate cancer. Oncogene 2008;27: 253_63.

17 Perner S, Mosquera JM, Demichelis F, Hofer MD, Paris PL, Simko J, et al. TMPRSS2-ERG fusion prostate cancer: an early molecular event associated with invasion. Am J Surg Pathol 2007; 31: 882_8.

18 Fleming JA, Stewart JW. A critical and comparative study of methods of isolating tumor cells from the blood. J Clin Pathol 1967; 20: 145_51.

19 Pelkey TJ, Frierson HF Jr, Bruns DE. Molecular and immunological detection of circulating tumor cells and micrometastases from solid tumors. Clin Chem 1996; 42: 1369_81.

20 Smirnov DA, Zweitzig DR, Foulk BW, Miller MC, Doyle GV, Pienta KJ, et al. Global gene expression profiling of circulating tumor cells. Cancer Res 2005; 65: 4993_7.

21 Ady N, Morat L, Fizazi K, Soria JC, Mathieu MC, Prapotnich D, et al. Detection of HER-2/neu-positive circulating epithelial cells in prostate cancer patients. Br J Cancer 2004; 90: 443_8.

22 Li X, Wong C, Mysel R, Slobodov G, Metwalli A, Kruska J, et al. Screening and identification of differentially expressed transcripts in circulating cells of prostate cancer patients using suppression subtractive hybridization. Mol Cancer 2005; 4: 30.

23 McIntyre IG, Spreckley K, Clarke RB, Anderson E, Clarke NW, George NJ. Optimization of the reverse transcriptase polymerase chain reaction for the detection of circulating prostate cells. Br J Cancer 2000; 83: 992_7.

24 Llanes L, Ferruelo A, Paez A, Gomez JM, Moreno A, Berenguer A. The clinical utility of the prostate specific membrane antigen reverse-transcription/polymerase chain reaction to detect circulating prostate cells: an analysis in healthy men and women. BJU Int 2002; 89: 882_5.

25 Llanes L, Paez A, Ferruelo A, Lujan M, Romero I, Berenguer A. Detecting circulating prostate cells in patients with clinically localized prostate cancer: clinical implications for molecular staging. BJU Int 2000; 86: 1023_7.

26 Lu YJ, Birdsall S, Summersgill B, Smedley D, Osin P, Fisher C, et al. Dual colour fluorescence in situ hybridization to paraffin-embedded samples to deduce the presence of the der(X)t(X;18)(p11.2;q11.2) and involvement of either the SSX1 or SSX2 gene: a diagnostic and prognostic aid for synovial sarcoma. J Pathol 1999; 187: 490_6.

27 Clark J, Merson S, Jhavar S, Flohr P, Edwards S, Foster CS, et al. Diversity of TMPRSS2-ERG fusion transcripts in the human prostate. Oncogene 2007; 26: 2667_73.

28 Iljin K, Wolf M, Edgren H, Gupta S, Kilpinen S, Skotheim RI, et al. TMPRSS2 fusions with oncogenic ETS factors in prostate cancer involve unbalanced genomic rearrangements and are associated with HDAC1 and epigenetic reprogramming. Cancer Res 2006; 66: 10242_6.

29 Wang J, Cai Y, Ren C, Ittmann M. Expression of Variant TMPRSS2/ERG Fusion Messenger RNAs Is Associated with Aggressive Prostate Cancer. Cancer Res 2006; 66: 8347_51.

30 Soller MJ, Isaksson M, Elfving P, Soller W, Lundgren R, Panagopoulos I. Confirmation of the high frequency of the TMPRSS2/ERG fusion gene in prostate cancer. Genes Chromosomes Cancer 2006; 45: 717_9.

31 Cerveira N, Ribeiro FR, Peixoto A, Costa V, Henrique R, Jeronimo C, et al. TMPRSS2-ERG gene fusion causing ERG overexpression precedes chromosome copy number changes in prostate carcinomas and paired HGPIN lesions. Neoplasia 2006; 8: 826_32.

32 Yoshimoto M, Joshua AM, Chilton-Macneill S, Bayani J, Selvarajah S, Evans AJ, et al. Three-color FISH analysis of TMPRSS2/ERG fusions in prostate cancer indicates that genomic microdeletion of chromosome 21 is associated with rearrangement. Neoplasia 2006; 8: 465_9.

33 Laxman B, Tomlins SA, Mehra R, Morris DS, Wang L, Helgeson BE, et al. Noninvasive detection of TMPRSS2:ERG fusion transcripts in the urine of men with prostate cancer. Neoplasia 2006; 8: 885_8.

34 Perner S, Demichelis F, Beroukhim R, Schmidt FH, Mosquera JM, Setlur S, et al. TMPRSS2:ERG fusion-associated deletions provide insight into the heterogeneity of prostate cancer. Cancer Res 2006; 66: 8337_41.

35 Hayes DF, Cristofanilli M, Budd GT, Ellis MJ, Stopeck A, Miller MC, et al. Circulating tumor cells at each follow-up time point during therapy of metastatic breast cancer patients predict progression-free and overall survival. Clin Cancer Res 2006; 12: 4218_24.

36 Lin B, Ferguson C, White JT, Wang S, Vessella R, True LD, et al. Prostate-localized and and rogen-regulated expression of the membrane-bound serine protease TMPRSS2. Cancer Res 1999; 59: 4180_4.