Innovative biomarkers for prostate cancer early diagnosis and progression

Critical Reviews in Oncology/Hematology 73 (2010) 10–22

Innovative biomarkers for prostate cancer early diagnosis and progression Jingjing You a,b , Paul Cozzi a,c , Bradley Walsh b , Mark Willcox d,e , John Kearsley a,f , Pamela Russell a,g , Yong Li a,f,∗ a

e

Faculty of Medicine, University of New South Wales, Kensington, NSW 2052, Australia b Minomic International Ltd, Frenchs Forest, NSW 2067, Australia c Department of Surgery, St George Hospital, Gray St, Kogarah, UNSW, NSW 2217, Australia d Institute for Eye Research Ltd, Kensington, NSW 2052, Sydney, Australia School of Optometry and Vision Science, University of New South Wales, Kensington, NSW 2052, Australia f Cancer Care Centre, St George Hospital, Gray St, Kogarah 2217, NSW, Australia g Oncology Research Centre, Prince of Wales Hospital, Barker St, Randwick, NSW 2031, Australia Accepted 25 February 2009

Contents 1. 2.

3.

4.

5.

Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Conventional biomarkers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1. Serum PSA and CaP diagnosis and progression . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2. PSA derivatives and CaP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3. Human Kallikrein 2 in CaP diagnosis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.4. Prostate cancer gene 3 in urine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.5. Early prostate cancer antigen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.6. Prostate cancer-specific autoantibodies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.7. Prostate-specific membrane antigen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.8. Gene fusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Novel proteomic biomarkers for CaP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1. Current proteomic techniques for CaP biomarkers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2. Proteomic CaP biomarkers in urine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3. Proteomic CaP biomarkers in sera . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Proteomic pattern diagnosis for CaP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1. ProteinChip/SELDI technology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2. Proteomic patterns from SELDI-TOF-MS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Conclusions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Conflict of interest . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Reviewer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Biographies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

11 11 12 13 14 14 14 14 15 15 16 16 16 16 17 17 17 18 18 19 19 19 22

Abstract The marker currently used for prostate cancer (CaP) detection is an increase in serum prostate-specific antigen (PSA). However, the PSA test which may give false positive or negative information, is not reliable and does not allow the differentiation of benign prostate hyperplasia

∗

Corresponding author at: Cancer Care Centre, St George Hospital, Gray St, Kogarah, NSW 2217, Australia. Tel.: +61 2 9113 2514; fax: +61 2 9113 2514. E-mail address: [email protected] (Y. Li).

1040-8428/$ – see front matter © 2009 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.critrevonc.2009.02.007

J. You et al. / Critical Reviews in Oncology/Hematology 73 (2010) 10–22

11

(BPH), non-aggressive CaP and aggressive CaP. There is thus an urgent need to search for novel CaP biomarkers to improve the early detection and accuracy of diagnosis, determine the aggressiveness of CaP and to monitor the efficacy of treatment. Proteomic techniques allow for a high-throughput analysis of bio-fluids with the visualization and quantification of thousands of potential protein markers and represent very promising tools in the search for new, improved molecular markers of CaP. In this review, we will summarize conventional CaP biomarkers and focus on novel identified biomarkers for CaP early diagnosis and progression that might be used in the future. © 2009 Elsevier Ireland Ltd. All rights reserved. Keywords: Prostate cancer; Biomarkers; Proteomics; SELDI-TOF-MS; Diagnosis; Prognosis

1. Introduction Prostate cancer (CaP) is the second most prevalent type of cancer in males particularly in Northern America and Australia/New Zealand, and results in the sixth highest mortality rate in men worldwide in 2002 [1]. In New South Wales, Australia, CaP has the highest incidence rate in all persons and in men in 2004, resulting in 5477 new cases (28.6% of all cancers in men) and 905 deaths which accounts for 12.7% of all cancer deaths in men [2]. It has been reported that approximately 1 in 8 men will develop CaP by 75 years of age and 1 in 5 by the age of 85 years [2]. In order to cure CaP patients successfully, it is important to detect the disease at an early stage as well as to monitor its progress accurately. Currently available diagnostic techniques include pathohistology of prostate biopsies, digital rectal examination (DRE), transrectal ultrasonography (TRUS), and assaying prostate-specific antigen (PSA). DRE and TRUS are widely employed by diagnosticians but are very limited in their ability to diagnose CaP and do not provide the ability to distinguish between benign prostate hyperplasia (BPH) and CaP. Pathohistology of prostate tissue can definitively identify CaP in most cases. This method is the most commonly used prognostic indicator for CaP and results in a grading called the Gleason score which is based on the architecture of cancer tissue observed under a microscope [3]. The lower the Gleason score is, the better the prognostic outcome [3]. However, there are limitations to this method of screening. First of all, a biopsy or similar operation must be performed in order to obtain the cancer tissue for testing. Second, the Gleason’s grading scale used by pathologists is at least semiquantitive since it may be difficult to search every cell of every tissue slice. Third, there is a lack of concordance between the threshold of scoring by different pathologists [4]. For these reasons Gleason scores themselves have limited quantitative value. Using biomarkers overcomes the problem of quantification, and thus can provide a more accurate way for early diagnosis of CaP and for monitoring its progression. Gleason score is the most used prognostic factor for CaP, with high scores particularly from 7 to 10 presenting a higher risk of death from CaP than low Gleason score (Gleason score < 4) cancers when patients aged 74 were treated conservatively [5]. However patients aged from 55 to 74 with Gleason score between 5 and 6 subjected to treatments are

likely die from competing medical conditions and patients with Gleason score greater than 6 are likely to die from CaP despite treatment [6]. After age 75 years average life expectancy in men is less than 10 years and there is general agreement that men older than 75 years are unlikely to benefit from CaP screening [7,8]. However, despite the apparent lack of benefit from screening for CaP in men older than 75 years indirect evidence suggests that PSA testing in elderly men is a fairly common occurrence [9,10]. Selection of CaP treatment is difficult with CaP identified by PSA test as it does not differentiate the clinical significant CaPs [11]. A recent investigation of this question using patient reported data documented a CaP screening rate of over 30% in men 75 years or older [12]. Prognostic biomarkers that can identify or predict clinically significant CaP in patients are important in management of the disease. Novel biomarkers could be useful to determine the benefit of such screening in these patients. Ideally these prognostic/predictive biomarkers would be less invasive to obtain, are useful to screen CaP patients particularly for older ones, and guide their management to provide maximum benefit whilst minimizing the risks from the side-effects of treatments. In this review biomarkers are grouped into three categories: the conventional biomarkers found prior to the proteomic era, the innovative proteomic biomarkers and the diagnostic proteomic patterns. Conventional biomarkers are molecules found in tissues or human body fluids; they can be proteins, DNA or RNA. The second type particularly refers to biomarkers found using proteomic techniques. The proteomic pattern type is a new approach which involves looking at mass spectrometric patterns rather than individual molecules. These patterns are produced using mass spectrometry (MS) particularly the surface-enhanced laser desorption and ionization time-of-flight (SELDI-TOF) or matrix-assisted laser desorption ionization time-of-flight (MALDI-TOF).

2. Conventional biomarkers For CaP diagnosis, PSA is the only conventional biomarker accepted by the U.S. Food and Drug Administration (FDA). Whilst not an ideal biomarker, its use as the main screening biomarker for CaP in many countries has resulted in the apparent increase of the disease’s incidence rates and a decrease in mortality rates.

12

J. You et al. / Critical Reviews in Oncology/Hematology 73 (2010) 10–22

Table 1 Biomarkers that are prostate cancer specific. Marker

Mr

Type of marker

References

A2M Akt-1 AMACR DD3/PCA3 prostate cancer antigen 3 EPCA FADD GRN-A GSTP-1 HSP27

163 56 42 n/a n/a 23 50 23 23

[110] [111,112] [113] Elucidated in the review Elucidated in the review [114] [115] [116] [117]

Ki67 KLK2 p53 PAP Prostate cancer-specific autoantibodies PIM-1 PSA PSCA PSGR PSMA PSP94 PTGS2:Reprimo DNA fragment TMPRSS2 Trp-p8

358 29 44 45 n/a 36 71 13 35 84 13 n/a 54 120

Diagnostic Prognostic Diagnostic and prognostic Diagnostic Diagnostic Prognostic Could be diagnostic and prognostic biomarkers Diagnostic Surrogate biomarker for estrogen-regulated growth in androgen-insensitive prostate cancer Prognostic Prognostic Prognostic for the outcome after radical prostatectomy Diagnostic Diagnostic Diagnostic and prognostic Diagnostic and prognostic Progress biomarker Diagnostic biomarker Diagnostic biomarker Progress biomarker Diagnosis and prognosis Diagnostic biomarker Diagnostic biomarker for higher grade CaP

[118] Elucidated in the review [119] [120] Elucidated in the review [121] Elucidated in the review [122] [123] Elucidated in the review [124] [125] Elucidated in the review [126,127]

Note: Mr is the relative molecular mass.

2.1. Serum PSA and CaP diagnosis and progression PSA is a 33 kDa glycoprotein, which belongs to the family of human kallikerin proteins and is a neutral serine protease. It has several isoforms with its isoelectric point ranging from 6.8 to 7.2 [13]. Formation of PSA depends on the secretion of androgen and it is mostly found in prostatic tissue, although low concentrations of the protein can be found in other tissues such as kidney and endometrium [14]. PSA is secreted by the prostatic epithelium and the epithelial lining of the acini and ducts of the prostatic tissue. It occurs in sperm and functions in liquefaction of the seminal fluids. PSA has the highest concentration in the prostatic lumen. In order to enter the blood circulation, PSA has to move through the prostatic basal membrane, stroma, capillary basal membrane and capillary endothelial cells [13]. Two forms of PSA are found in serum, free and bound to ␣1-antichymotrypsin or to ␣2-macroglobulin [13,15]. The PSA test measures the total amount of PSA in the blood. The generally accepted upper cut-off level of PSA to be considered normal is 4.0 ng/mL [16], although age-specific levels have also been determined, being higher in older than in younger men [17]. Levels above the point suggest that a biopsy is required for CaP detection. Recent studies have found evidence which challenges the sensitivity and specificity of this threshold. One such study showed that the PSA test is only considered reliable for serum PSA levels in men below 2.6 ng/mL (for normal) or above 10.0 ng/mL (for cancer) [18]. The specificity of PSA is only 25–35% between the range of 2.6 ng/mL and 10.0 ng/mL; this means that 70–80%

of men have to undergo unnecessary biopsies [18]. Another study started in 1993 in the USA, investigated the prevalence of CaP in men with PSA levels less than 5 ng/mL and found that 33.7% of men actually had CaP with PSA levels ≤2 ng/mL, and 50.8% of men had the disease with levels ranging from 2.1 to 4 ng/mL [19]. Among those men who would remain undiagnosed using currently accepted PSA levels, only 37.5% of them had high-grade cancer. This finding is corroborated by a more recent study conducted from 1997 to 2005 in Europe, in which 855 men whose PSA was less than 4.0 ng/mL and who had normal DRE results were examined, with the discovery that 20.5% of them had high-grade cancer [20]. Moreover, the linkage between serum PSA levels and CaP has been reported to have weakened over the past 20 years. A study by Stamey showed that serum PSA level was only well correlated with BPH [21]. Preoperative serum PSA levels and prostate weights from 1300 CaP specimens were found to be significantly correlated with each other using several regression methods [21]. A reason for the unreliability of diagnosing CaP using PSA is that the levels of PSA in serum can also be affected by factors other than CaP. Other diseases such as acute urinary retention and prostatitis can elevate the level of serum PSA [22]. The age of a person can also have an affect on PSA levels. The Mayo clinic has reported that the level of serum PSA is not proportional to positive biopsy rates but rather to the age of patients [23]. Stamey reported that 8% of men in their twenties have CaP, and the incidence of the disease rises steadily and linearly with each decade of life until 80% of men in their

J. You et al. / Critical Reviews in Oncology/Hematology 73 (2010) 10–22

seventies are afflicted [21]. Therefore it is probable that one universal cut-off point to indicate normal levels may not be sufficient when considering age which complicates the test. Moul proposed that different PSA cut-off points should be employed for different age groups [17]. An investigation of 11,861 men who all had their PSA levels measured within 2 years reported that PSA levels varied with age and the recommendation was to use the threshold value of 2 ng/ml for men aged 50–59 [17]. PSA has also been found to be linked to body mass index (BMI) in Korea. After an examination of 8640 Korean men without CaP aged between 40 and 79, PSA levels were found to be inversely correlated to BMI among men aged 40–59 [24]. It was found that obesity could decrease the levels of PSA by approximately 0.13–0.18 ng/mL. All these findings highlight the need for an improved method of testing using PSA as discussed below. 2.2. PSA derivatives and CaP As a result of the specificity of the PSA test being challenged, various methods have been proposed to improve the test, which can be classified into two groups: PSA isoforms and PSA parameters. PSA isoforms consist of free PSA (fPSA), proPSA, complexed PSA (cPSA) and benign PSA (bPSA). PSA parameters on the other hand involve looking at the percent free PSA (%fPSA), PSA density (PSAD), agespecific PSA ranges, PSA velocity (PSAV) and PSA doubling time (PSA-DT). fPSA, which refers to the PSA not bound to plasma proteins, is the most studied PSA isoform so far. The percentage of fPSA over total PSA (tPSA) has been found to enhance the specificity of cancer detection in men with tPSA between 4 and 10 ng/mL and a negative prostate biopsy [25]. This finding was supported by a study conducted in 2000 which focused on a comparison of the fPSA, PSAD and age in a total of 773 samples (379 with CaP, 394 with BPH) from patients recruited from seven medical centers, aged between 50 and 75 with PSA from 4 to 10 ng/mL [26]. This study demonstrated that only fPSA (sensitivity 95%, specificity 27%) has shown consistent and significant improvements over the conventional PSA test (sensitivity 95%, specificity 15%). This increase in specificity means that there is a slight increase in correctly identifying those people who do not have CaP. Analysis of the structural composition of fPSA revealed three subfractions: proPSA, bPSA and intact PSA (iPSA). ProPSA is the precursor protein for PSA, and it consists of four truncated forms: (−2), (−4), (−5), and (−7) proPSA. Studies have shown that proPSA defined by the sum of the four truncated proPSAs improved the specificity over tPSA and %fPSA in the PSA range of 4–10 ng/mL [27,28]; whereas a European multi-institutional trial including 2055 men showed no significant improvement comparing (−5, −7) proPSA over tPSA and %fPSA in the same range [29]. A study of men with PSA levels between 4 and 10 ng/mL

13

has also shown that a combination of proPSA, fPSA and tPSA had the highest specificity compared to any of the individual tests [30], however whether it is better than the %fPSA test has not yet been determined. More studies are needed to identify its specificity and sensitivity on CaP detection. Two multi-center studies have shown that cPSA improved the specificity of total PSA, but this was not better than f/tPSA [31]. The density of cPSA (cPSAD) had the highest specificity over tPSA and f/tPSA [32]. These studies have suggested that combining the measurement of cPSA and cPSAD could replace the PSA and f/tPSA tests for CaP diagnosis. BPSA was identified in 2000, and is an altered form of fPSA found enriched in the nodular transition zone (TZ) tissue of BPH which suggests that it could be a biomarker for BPH [33]. When compared with PSA and fPSA, serum bPSA has been found to have a better predictive power for prostatic enlargement [34]. Other PSA parameters such as PSA velocity (PSA-V) and PSA doubling time (PSA-DT) have also been assessed to improve the sensitivity and specificity. Although the diagnostic value of PSA-V is controversial, it has great potential as a prognostic marker for CaP. High PSA-V is associated with a higher risk of CaP recurrences [35] and shorter recurrence intervals [36], therefore, it could be used as a potential prognostic biomarker for CaP recurrence after radical prostatectomy (RP). Studies have also shown that high PSA-V levels before diagnosis were associated with the outcomes of CaP particularly those that are lethal. D’Amico et al. investigated the association between preoperative PSAV and the risk of death from CaP in 1054 patients. Those with an annual preoperative PSA-V > 2.0 ng/mL/year 7 years before radical prostatectomy were shown to have a higher risk of death from CaP [36]. The linkage between PSA-V level before diagnosis and the life-threatening CaP when a cure is still possible was examined by Carter et al. who proposed a cut-off level of PSA-V of 0.35 ng/ml/year to screen potentially lethal CaP in men with low PSA levels [37]. These findings suggested that the level of PSA-V could be used to guide the management of CaP. The combination of preoperative PSA-V and PSA-DT was also shown to have significant predictive value concerning the progression of CaP [38], although this finding needs to be verified by others. In summary, using combinations of PSA derivatives has shown improved specificity within the PSA range from 4 to 10 ng/mL. However, only the %fPSA test has shown consistent and significant improvement, the other abovementioned tests either showed no significant clinical use or controversial outcomes. Therefore it is important to find other possible biomarkers to diagnose this disease. There are at least 91 biomarkers for CaP found so far [39], of which only a few are CaP specific and found in human body fluids; these are summarized in Table 1. This review only focuses on the biomarkers that are well documented and/or are more statistically significant than PSA.

14

J. You et al. / Critical Reviews in Oncology/Hematology 73 (2010) 10–22

2.3. Human Kallikrein 2 in CaP diagnosis Human Kallikrein 2 (hK2 or KLK2), like PSA, belongs to the serine protease family. It also has 80% amino acid sequence homology with PSA. hK2 was initially found to be over expressed in CaP epithelial cells by immunohistochemical staining, which also suggested it could be a potential biomarker for CaP [40]. Following these initial findings, its level in serum was investigated. Several groups have successfully detected hK2 in blood serum, where it also presents in free and bound forms, like PSA [41,42]. Partin et al. examined the association between the hK2 serum levels and CaP and found using the total hK2 (thk2)/fPSA ratio, that 40% of CaP was detected in men with PSA levels between 2 and 4 ng/mL [43]. However, they also examined the use of fPSA on the same patients and found the hK2/fPSA ratio was not significantly better than the fPSA test. Nevertheless, other studies have identified potential prognostic value when using hK2. Both Haese et al. [44] and Steuber et al. [45] found that the level of hK2 in blood was significantly associated with the risk of biochemical recurrence (BCR) after RP in men with a serum tPSA ≤ 10 ng/mL and it can be an independent prognostic biomarker for CaP. When 867 men were examined, the predictive accuracy of hK2 for BCR was 73.1% whereas PSA was 69.1%, in men with tPSA ≤ 10 ng/mL and hK2 had greater accuracy (73.9%) than PSA (59.9%) [45,46]. Although hK2 showed improved accuracy over PSA for predicting BCR, further validation is needed to confirm its prognostic ability because the report by Steuber et al. [46] is the only one that presented the statistical results. 2.4. Prostate cancer gene 3 in urine Prostate cancer gene 3 (PCA3), also known as PCA3DD3 or DD3PCA3 , is a new prostate-specific gene that is highly overexpressed in CaP tissue. It was first identified using differential display analysis comparing mRNA expression in normal and tumor-bearing prostate tissues [47]. PCA3 is a prostate-specific noncoding mRNA that is highly overexpressed in more than 95% of primary CaP specimens and CaP metastases [47,48]. PCA3 encodes a prostate-specific mRNA that has shown promise as a CaP diagnostic tool. Measurement of PCA3 mRNA normalized to PSA mRNA in urine has been proposed as a marker for CaP. The possible usefulness of the urinary PCA3 assay as a CaP marker was suggested by de Kok et al. in 2002 [48]. Hessels et al. found that the median up-regulation of PCA3 in CaP tissue compared with normal prostate tissue was 66-fold and the sensitivity for the detection of CaP by the timeresolved fluorescence (TRF)-based PCA3 test in urine was 67%, the specificity was 83%, and the negative predictive value was 90% [49]. On the basis of these results, Groskopf et al. developed a commercially available highly sensitive transcription-mediated amplification (TMA) method (PCA3, Gen-Probe Incorporated) for urinary assay for general clin-

ical use [50]. The method measures both PCA3 messenger RNA (mRNA) and PSA mRNA in first-catch urines collected following a DRE. The function of the DRE is to enhance the release of prostate cells through the prostate duct system into the urinary tract and thus into the urine. Recent studies have shown that the PCA3 urine assay has promise in improving the diagnostic accuracy of CaP detection, especially in the PSA gray zone [50–52]. van Gils et al. showed that the TRF-based PCA3 urine test in a total of 534 men (the largest numbers until now) for five different institutions, when used as a reflex test, can improve the specificity in CaP diagnosis and could prevent many unnecessary prostate biopsies [52]. Clinical studies of the new test urinary PCA3 levels have reflected early CaP more accurately than serum PSA levels (p < 0.01) [51]. Nakanishi have recently demonstrated that the PCA3 score is significantly associated with tumor volume and Gleason score in prostatectomy specimens, suggesting that the urinary PCA3 score may be a novel molecular marker not only for CaP detection, but also for the classification of men diagnosed with CaP [53]. Deras also recently reported the quantitative urinary PCA3 score was directly related to the probability of positive biopsy and could improve diagnostic accuracy for CaP detection [54]. Although the results from some institutes are promising, the diagnostic value needs to be further validated in a multicenter setting and to be followed up to show if indeed the PCA3 urine test is able to “predict” the presence of CaP. 2.5. Early prostate cancer antigen Early prostate cancer antigen (EPCA) is a nuclear matrix protein which is association with CaP and so could be used as a good candidate as a biomarker. The association was initially established using CaP tissue through immunohistochemical assays which showed 84% sensitivity and 85% specificity for CaP detection [55,56]. Paul et al. developed an enzyme-linked immunosorbent assay (ELISA) to detect the EPCA level in blood and showed 92% sensitivity and 94% specificity in CaP detection, however only 46 plasma samples were examined [57]. A drawback of using EPCA alone is that not all CaP tissues express EPCA [57]. Although its potential as a complement to PSA testing is promising, further researcher is needed to confirm this test and possibly investigate its complement with the PSA test. 2.6. Prostate cancer-specific autoantibodies Cancer-specific autoantibodies are generated when the immune system responds to over-expressed tumor associated antigens (TAAs). Prostate cancer-specific antoantibodies that have been detected in blood include autoantibodies against ␣-methyl-acyl-coenzymA-racemase (AMACR) which is over-expressed in CaP tissue [58] and autoantibodies against Huntington-interacting protein 1 [59]. Within the

J. You et al. / Critical Reviews in Oncology/Hematology 73 (2010) 10–22

PSA range from 4 to 10 ng/ml, the immunoreactivity against AMACR showed higher sensitivity and specificity than PSA [58]. Bradley et al. [59] has also shown improved specificity (97%) for CaP detection when combining the two autoantibodies together. More tests are needed to further confirm these results. 2.7. Prostate-specific membrane antigen Prostate-specific membrane antigen (PSMA) was originally found only on the membrane of prostate epithelium in 1987 [60], and its expression was increased in CaP tissue [61]. As a result, PSMA is proposed to be a potential biomarker for CaP and has been extensively studied. Its role as a serum prognostic CaP biomarker, however, is controversial. Studies using Western blots showed that PSMA could distinguish between late stage CaP and early stage disease [62,63], whereas other studies showed that PSMA was not more effective than PSA [64,65]. A study using a newly developed technology, the ProteinChip technique, demonstrated higher PSMA serum levels in CaP patients compare to BPH and normal groups [66]. This pilot study suggested a new approach for examining serum PSMA. However, a number of other factors such as secretion from normal tissues or other tumors and age are likely to contribute to the serum levels of PSMA and so might affect its relation to CaP [67]. Overall, PSMA is regarded as a promising biomarker for CaP and further investigations and perhaps more sensitive detection methods are needed to fully evaluate its usefulness. 2.8. Gene fusions Gene fusions are a result of chromosomal translocations and are commonly found in haematological malignancies but more rarely in solid tumors primarily due to there being less available information and the complexity of the prostate cancer karyotypes. Recurrent gene fusions in CaP were first discovered by Tomlins et al. who used a selfdeveloped algorithm named Cancer outlier Profile Analysis (COPA) to analyze DNA microarrays [68]. The recurrent gene fusions discovered involved the 5 untranslated region of the prostate-specific, androgen-regulated transmembrane protease, serine 2 gene (TMPRSS2) fused to either the erythroblastosis virus E26 transforming sequence (ETS) variant gene 1 (ETV1) or the v-ETS erythroblostosis virus E26 oncogene homolog (ERG) which are two transcription factors from the ETS family [68]. After discovery, the TMPRSS2:ETV1 and TMPRSS2:ERG gene fusions were then validated by measuring the level of RNA expression in several prostate cancer cell lines and tissue specimens using PCR-based assays, single nucleotide polymorphism arrays and fluorescent in situ hybridization (FISH)-based assays in many independent studies [68–71]. Following their discoveries, additional gene fusions have been reported involving

15

another two ETS members, ETV4 [72] and ETV5 [73]. Other 5 partners such as, HERVK22q11.3, HNRPA2B1 and C15ORF21 have also been identified in gene fusions with ETV1 whereas SLC45A3 was capable of fusing with both ETV1 and ETV5 [74,73]. Another two 5 partners for ETV4 identified were kallikerein 2 (KLK2) and calciumactivated nucleotidase 1 (CANT1) [75]. Among the gene fusions, TMRSS2:ERG is so far the most common one found [76]. The effect of gene fusions on CaP is still subject to investigation, however the fact that TMPSS2:ETS gene fusions can only be found in prostatic intraepithelial neoplasia (PIN) (an intermediate stage from benign epithelium to carcinoma), carcinoma and metastases but not in benign prostatic hyperplasia (BPH) or proliferative inflammatory atrophy (PIA) suggested that the gene fusions were likely to be the genetic trigger for the development of PIN and CaP invasion [70]. The potential of gene fusions as CaP biomarkers has also been investigated with TMRSS2:ERG being the main focus because it is the most common gene fusion in prostate cancer. The association between TMRSS2:ERG and Gleason score is controversial, probably due to the limitation of technology sensitivity, relatively small sample size, varied population and different experimental design. Although Winnes et al. found a significant association between positive TMRSS2:ERG fusion and lower Gleason score and better survival rates using 50 patient tissue samples [71], no association was found between TMRSS2:ERG fusion status and Gleason score on samples from 300 patients by Perner et al. [70]. Several studies suggested that TMRSS2:ERG gene fusion is associated with aggressiveness of prostate cancer and death from CaP. Mosquera et al. found no association between the gene fusion and Gleason score but other morphological features linked to aggressive CaP were significantly related to the gene fusion status [77]. Perner et al. has shown a significant association between positive TMRSS2:ERG and higher tumour stage using tissue samples from 211 patients [78]. Nam et al. [79] and Wang et al. (2006) [69] both reported that TMRSS2:ERG fusion was strongly correlated to the recurrence of CaP. Demichelis et al. reported TMRSS2:ERG was significantly associated with CaP specific death [80]. In addition to its prognostic value, the presence of TMRSS2:ERG fusion has helped to improve CaP detection when combined with other biomarkers possibly due to its specificity for CaP. One study used a multiplexed q-PCR based test to analyze a panel of 7 potential urinary CaP biomarkers including TMRSS2:ERG. This study showed improved specificity compared to the serum PSA test [81]. Combining TMRSS2:ERG and PCA3 also improved the predictive value of CaP compared to PCA3 test alone [82]. Overall, most of the studies have suggested that TMRSS2:ERG could be a prognostic biomarker for aggressive prostate cancer. Its presence has also impacted on the ability to improve CaP detection when coupled with other biomarkers. However, studies on the clinical implications of TMRSS2:ERG have ao far involved relatively small sam-

16

J. You et al. / Critical Reviews in Oncology/Hematology 73 (2010) 10–22

ple sizes and different experimental designs. Therefore, more studies are needed to validate its clinical significance in CaP.

3. Novel proteomic biomarkers for CaP In the genomic era, biomarkers are generally found by studying one or a few genes. This is time consuming. With more proteins than genes, this methodology may miss out important biomarkers. Nowadays, a proteomic study which attempts to investigate all the proteins expressed by a genome has become a direct and fast way for detecting biomarkers in a variety of sample types, but most especially in biological fluids. 3.1. Current proteomic techniques for CaP biomarkers Common proteomic techniques used can be divided into gel-based and gel-free techniques. The typical and most commonly used gel-based technique is the two-dimensional gel electrophoresis (2DGE) which can simultaneously separate and visualize thousands of proteins in a gel. The principle of applying this technique in biomarker hunting is to compare the final proteomic pattern shown on the gels of the normal and disease samples and to search for statistically significant differentially expressed protein spots with the help of image software. This is then followed by identification of spots of interest using mass spectrometry (MS). MS is a gel-free technique which is coupled with 2DGE for protein identification. It is also widely used for detection of proteins or peptides that are smaller than 20 kDa and cannot be found in 2D gels. MS has two major components: ionization and mass analysers. Electrospray ionization (ESI) and MALDI are the two major ionization techniques used in MS. A recently invented modified version of MALDI named surface-enhanced laser desorption and ionization (SELDI) has also been widely used. Four major mass analysers are TOF, quadrupole mass analyser, quadrupole ion trap (QIT) and Fourier transform ion cyclotron resonance (FTICR). In biomarker research, combinations of gel-based and gel-free techniques are commonly used. Many publications using proteomic approaches to search for CaP biomarkers have used either tissues or cancer cell lines. This has resulted in several potential biomarkers being identified. Lexander et al. examined cell suspensions from 35 prostatectomy specimens, 29 cancer samples and 10 benign samples using 2DGE coupled with MS [83]. A total of 39 spots were found to be expressed differentially and 30 were identified. Among these, 26 were identified by MS (20 over-expressed and 6 under-expressed in group with Gleason score 8–9 and/or aneuploid cancer), and 4 (which were also over-expressed protein spots) were identified by finding the corresponding spot position on a matched set derived from colon cancer. These potential biomarkers for highgrade CaP were identified as nucleoside diphosphate kinase 1, chromobox protein, 39S ribosomal protein L12, cytosol

aminopeptidase, endopeptidase C1p, inorganic pyrophosphatase, metaxin 2, GST-pi, acyl-CpA dehydrogenase, lysophospholipase, NADH-ubiquinone oxidoreductase, 60 and 70 kDa GRP-78, b-actin, cytokeratins 7, 8, and 18 and stomatin-like protein 2, a-actin and mutant desmin. Sardana et al. examined conditioned media from three human CaP cell lines, PC3 (bone metastasis), LNCaP (lymph node metastasis), and 22Rv1 (localized to prostate), using strong cation exchange high performance liquid chromatography (HPLC) and HPLC-tandem MS. Several previously identified CaP biomarkers including PSA, hK2 and PSMA were also found in this study [84]. Four novel potential biomarkers discovered were identified as follistatin, chemokine (C-X-C motif) ligand 16, pentraxin 3 and spondin 2. ELISA tests of these proteins using serum samples from 42 patients with and without CaP found significant correlation with PSA levels. However, the search for CaP biomarkers using proteomic studies on various biological fluids has achieved little success to date. This is probably due to two main reasons. The concentration of potential biomarkers varies in different human body fluids with the highest in tissue interstitial fluid as it directly contacts tumour tissues and lower in circulating body fluids, hence it is more difficult to detect biomarkers using the easily accessible circulating fluids such as serum/plasma [85]. Secondly, proteomic analysis of biological fluids is challenging. For instance, the huge dynamic range of serum proteins makes identification of proteins of low abundant extremely difficult [86,87]. In urine, the low concentration of proteins and high salt concentrations create problems for proteomic analysis [88]. 3.2. Proteomic CaP biomarkers in urine By comparing the high-resolution 2DGE of urinary proteomic pattern of patients with CaP and with BPH and normal controls, Grover and Resnick have isolated two proteins which they called A (36 kDa) and B (23 kDa) that were absent from urine from CaP and BPH patients, and a protein F (18–28 kDa) that was highly expressed in BPH but not in CaP and normal controls [89]. However, these proteins were not identified. Using a 2DGE and MALDI-TOF-MS approach on 12 urine samples (6 from CaP and 6 from BPH), Rehman et al. identified a novel potential biomarker, calgranulin B/MRP-14, which was only presented in CaP but not in BPH [90]. Calgranulin B was also discovered as a potential biomarker for CaP in pooled serum samples (a CaP serum sample pooled from 10 different CaP patients and a BPH serum sample pooled from 10 different BPH patients) which is discussed later [91]. However, the samples used in both studies were not enough to prove its significance; further validation studies are required for this marker. 3.3. Proteomic CaP biomarkers in sera In search of serum CaP biomarkers, Lam et al. performed MS based mass profiling (MP) combined with multivariate

J. You et al. / Critical Reviews in Oncology/Hematology 73 (2010) 10–22

analysis on 16 CaP patients and 15 healthy individuals, and found 17 serum proteins specific to metastatic CaP [92]. Among them, a protein detected at m/z 7771 was found to be significantly decreased in the sera of all the patients with metastatic CaP but not in those with localized CaP or healthy individuals, and was identified as platelet factor 4, a chemokine involved in prothrombolytic and antiangiogenic activities. Western blots and ELISA confirmed this result. Confirmations of these results are needed on a larger patient population. Malik et al. identified Apolipoprotein A-II (m/z at 8946) as a potential CaP biomarker using HPLC, reverse-phase chromatography, SDS-PAGE, LC–MS/MS and SELDI-TOF-MS [93]. However, the same peak of m/z at 8946 on IMAC 30 chips has been identified as various different proteins by research groups investigating different cancer types [94,95]. Ward et al. suggested that independent assays were needed to validate the identifications [94]. Other potential CaP serum protein biomarkers that have been found using SELDI-TOF including collective variant forms of Serum Amyloid A (SAAs) for bone lesions (the metastatic stage of CaP) patients [96] and three unidentified putative protein markers with relative molecular weights of 15.2, 15.9 and 17.5 kDa [97]. Qin et al. applied a different approach, ie., using anion displacement liquid chromatofocusing chromatography for serum fractionation, 2D differential in-gel electrophoresis (DIGE) which allows for analysis of proteins from different groups on the same gel coupled with MS on sera from CaP and BPH patients [91]. Three potential biomarkers have been identified as squamous cell carcinoma antigen 1 (SCCA1), calgranulin B and haptoglobin-related protein. Although proteomic techniques have provided a faster way for screening biomarkers, the biomarkers found are all subject to validation. The sample sizes used were not big enough, and independent confirmation from other research institutions is needed.

4. Proteomic pattern diagnosis for CaP The concept of using a specific proteomic pattern for disease diagnosis was proposed by Petricoin et al. in 2002 with the invention of the SELDI technique [98]. 4.1. ProteinChip/SELDI technology The SELDI-TOF-MS ProteinChip has been introduced by Hutchens and Yip [99]. This technology utilizes affinity surfaces on chips to retain certain adherent proteins based on their physical or chemical characteristics, which is then followed by direct analysis using TOF-MS. Such surfaces include normal phase for general protein binding; hydrophobic surfaces for reversed-phase capture; cationand anion-exchange surfaces; and immobilized metal affinity capture (IMAC) for metal-binding proteins chips. Other surfaces can be made that specifically bind for example,

17

Fig. 1. Illustration of different types of SELDI chips (adopted from http:// www.bio-infra.com/korean/product/seldi.asp).

antibody–antigen complexes, DNA–protein complexes, various receptors or drugs (see Fig. 1). An energy absorbing matrix (EAM) such as sinnapinic acid which crystallizes with the proteins when it dries, absorbs energy when hit by a laser and allows the proteins to desorb and ionize for MS analysis is overlayed over the bound proteins and the chips are then analysed using MS. SELDI-TOF-MS is more sensitive and requires only small amounts of sample (2–3 ␮L) compared to 2DGE. An additional advantage of this technique compared to 2DGE is its ability to screen small peptides (∼500 Da). 4.2. Proteomic patterns from SELDI-TOF-MS The first study on a proteomic pattern using SELDI was performed by Petricoin et al. for detection of ovarian cancer [98]. In that study, 100 normal sera, 100 cancer sera and 16 benign sera samples were analyzed by SELDI. The generated spectra were equally and randomly divided into a training set and a masked set with each including 50 normal and 50 cancer spectra. The training set was used to train a pattern algorithm called a genetic algorithm which selected the data set that could differentiate normal and cancer groups most efficiently. The chosen data set was then tested on the masked set and was able to identify 50/50 of the cancer samples, 47/50 of the normal samples and recognized 16/16 of the benign samples as different from either normal or cancer samples [98]. After this initial study, Petricoin et al. identified a serum proteomic pattern which successfully detected 36/38 patients with CaP, and 177/228 with BPH [100]. Other researchers have also used the proteomic pattern diagnostics approach with CaP and shown promising results. By training two different bioinformatic learning algorithms, the AdaBoost classifier and the Boost Decision Stump Feature Selection classifier, coupled with SELDI on 197 CaP patients, 92 BPH and 96 controls, Qu et al. achieved a 100% and 97% of sensitivity and specificity respectively for the test set including 30 CaP patients, 15 BPH and 15 controls

18

J. You et al. / Critical Reviews in Oncology/Hematology 73 (2010) 10–22

[101]. Ornstein et al. used a pattern recognition bioinformatics tool, the Proteome Quest beta version 1.0, coupled with SELDI on a QSTAR Pulsar on sera samples with PSA levels between 2.5 and 15 ng/mL [102]. The generated proteomic pattern achieved 100% sensitivity and 67% of specificity on the test samples, which means 67% of the patients could have avoided unnecessary biopsies with no cancers being missed. Kohli et al. found a specific biomarker profile significantly (p < 0.05) associated with the prognosis of advanced-stage CaP treated with androgen deprivation therapy (ADT) using SELDI-TOF-MS with Kruskal–Wallis tests for determining group differences and the covariate-adjusted Cox regressions for assessing the PSAindependent associations of each individual peak with overall survival [103]. Overall, these studies have presented exciting results, which clearly indicate that the proteomic pattern diagnosis has great potential for clinical application. It has several advantages such as higher sensitivities and specificities by using a pattern that consists of a combination of peaks rather than a single biomarker; it is independent from protein identification which significantly reduces labour intensity; has high-throughput ability; it is able to detect truncated, modified or fragmented proteins (techniques such as ELISA cannot differentiate between different forms of the same protein) (reviewed in Petricoin and Loitta [104]). The main advantage of SELDI-TOF-MS is its ability to detect low molecular weight (LMW) proteins and peptides, thus it could detect the low abundant proteins or peptides bound to the highly abundant proteins such as albumin in serum [104]. However, this claim was recently challenged by Ekblad et al. as they found that a number of peaks in the 1500–4000 m/z range of the spectra from clinical samples actually originate from abundant proteins such as the human serum albumin due to in-source decay and that these peaks can mask the real peaks generated from LMW peptides [105]. This finding raised the question of the reliability of SELDI data. Another problem associated with SELDI-TOF-MS is its reproducibility. The pattern diagnosis has been described as a ‘black box’ approach with samples entering at one end and data coming out at the other end, and everything else is hidden in the box [106]. The review by Baggerly et al. of the work done by Petricoin et al. on ovarian cancer found their results were not reproducible which was likely due to the different processing of normal and cancer samples [106]. This addressed an important issue in reproducibility of the proteomic pattern and highlighted the critical need for careful experimental design. Another study assessing the reproducibility of SELDI-TOF-MS also examined the ovarian data set, the CaP data sets, and Type 1 diabetes data sets and found satisfactory concordances in both of the CaP and Type 1 diabetes data sets [107]. Both of these data sets were generated from the same platforms (chips), whilst the unsatisfactory ovarian cancer data set was generated from different chips (H4 and WCX2) [107]. This assessment study also indicates the same need for careful experimental design in order

to generate reproducible data. Data sets generated from different labs have not been tested yet. It may be difficult for different labs to achieve the concordant data sets as there are many variations involved such as sample handling and preparation, temperature, humidity, and the actual instrument used varies between the different labs [106]. Several improvements have been suggested to increase the reliability and reproducibility of SELDI-TOF-MS including a standardized experimental design and data analysis procedure [106] and a more sensitive mass analyzer such as QSTAR pulsar MS, which can completely resolve peaks differing by a m/z of 0.375 whereas the commonly used PBSII-TOF MS can only differentiate between peaks differing by an m/z of 20 [108]. Another approach is to use the more sensitive MS, MALDI-TOF-MS. M’koma et al. have found 130 verifiable signals of a mass range of 1000–5000 m/z to suggest 71.2% specificity and 67.4% sensitivity in discriminating CaP vs. BPH [109]. They further demonstrated that comparing BPH and high-grade prostatic intraepithelial neoplasia (HGPIN) resulted in 73.6% specificity and 69.2% sensitivity, and comparing CaP and HGPIN resulted in 80.8% specificity and 81.0% sensitivity. Certainly, the use of proteomic pattern approach showed great potential for the detection of cancers in the aforementioned studies, however its reliability and reproducibility is still under investigation. Unless it has been validated, this approach could not be used in a real clinical setting.

5. Conclusions Identification of novel biomarkers is a new area of development. Modern advances in the field of proteomic techniques hold the promise of providing the clinical urologist/oncologist with new tools to find novel CaP biomarkers for early diagnosis and prognosis. 2D-DIGE is a very promising technology which can be useful for CaP biomarker identification in the future. SELDI-TOF-MS has been widely used in cancer detection and shows potential. It is used in two areas: proteomic pattern diagnostics and biomarker identifications. Proteomic pattern diagnostics requires no protein identification, produces high sensitivities and specificities and is a much higher throughput method, however its reliability and reproducibility needs further investigation. As reviewed above, there is still not yet an effective biomarker approved for CaP. As a development of the technology, the use of a proteomic pattern has become a new way to detect biomarkers. The biomarkers and the proteomic patterns listed in this review all show promise but they all need further validation. The next question for scientists hunting biomarkers is to standardize criteria for biomarker validation.

Conflict of interest None.

J. You et al. / Critical Reviews in Oncology/Hematology 73 (2010) 10–22

Reviewer Professor Warren D. Heston, Ph.D., Lerner Research Institute/NB40, Department of Cancer Biology, 9500 Euclid Avenue, Cleveland, Ohio 44195, United States. Acknowledgements Supported by Australian Research Council (ARC) (LP 0774951) and Cancer Institute NSW, Australia (06-ECF-117). References [1] Parkin DM, Bray F, Ferlay J, Pisani P. Global cancer statistics, 2002. CA Cancer J Clin 2005;55:74–108. [2] Tracey E, Chen S, Baker D, et al. Cancer in New South Wales, Incidence and Mortality Report 2004. Cancer Institute NSW: NSW Central Cancer Registry; 2006. [3] Epstein JI, Allsbrook WC, Amin MB, Egevad LL. The 2005 International Society of Urological Pathology (ISUP) Consensus Conference on gleason grading of prostatic carcinoma. Am J Surg Pathol 2005;29:1228–42. [4] Bostwick DG, Grignon DJ, Hammond ME, et al. Prognostic factors in prostate cancer. College of American Pathologists Consensus Statement 1999. Arch Pathol Lab Med 2000;124:995–1000. [5] Albertsen PC, Hanley JA, Gleason DF, Barry MJ. Competing risk analysis of men aged 55–74 years at diagnosis managed conservatively for clinically localized prostate cancer. JAMA 1998;280:975–80. [6] Albertsen PC, Hanley JA, Fine J. 20-year outcomes following conservative management of clinically localized prostate cancer. JAMA 2005;293:2095–101. [7] Fowler FJ, Bin L, Collins MM, et al. Prostate cancer screening and beliefs about treatment efficacy: a national survey of primary care physicians and urologists. Am J Med 1998;104:526–32. [8] Poteat HT, Chen P, Loughlin KR, et al. Appropriateness of prostatespecific antigen testing. Am J Clin Pathol 2000;113:421–8. [9] Merrill RM. Demographics and health-related factors of men receiving prostate-specific antigen screening in Utah. Prev Med 2001;33:646–52. [10] Etzioni R, Berry KM, Legler JM, Shaw P. Prostate-specific antigen testing in black and white men: an analysis of Medicare claims from 1991–1998. Urology 2002;59:251–5. [11] Draisma G, Boer R, Otto SJ, et al. Lead times and overdetection due to prostate-specific antigen screening: estimates from the European randomized study of screening for prostate cancer. J Natl Cancer Inst 2003;95:868–78. [12] Scales CD, Curtis LH, Norris RD, et al. Prostate specific antigen testing in men older than 75 years in the United States. J Urol 2006;176:511–4. [13] Lukes M, Urban M, Zalesky M, et al. Prostate-specific antigen: current status. Folia Biol 2001;47:41–9. Praha. [14] Clements J, Mukhtar A. Glandular kallikreins and prostate-specific antigen are expressed in the human endometrium. J Clin Endocrinol Metab 1994;78:1536–9. [15] Lilja H, Christensson A, Dahlen U, et al. Prostate-specific antigen in serum occurs predominantly in complex with alpha-1antichymotrypsin. Clin Chem 1991;37:1618–25. [16] Antenor JAV, Han M, Roehl KA, et al. Relationship between initial prostate specific antigen level and subsequent prostate cancer detection in a longitudinal screening study. J Urol 2004;172: 90–3.

19

[17] Moul JW, Sun L, Hotaling JM, et al. Age adjusted prostate specific antigen and prostate specific antigen velocity cut points in prostate cancer screening. J Urol 2007;177:499–503. [18] Smith DS, Humphrey PA, Catalona WJ. The early detection of prostate carcinoma with prostate specific antigen. Cancer 1997;80:1852–6. [19] Thompson IM, Pauler DK, Goodman PJ, et al. Prevalence of prostate cancer among men with a prostate-specific antigen level
20

J. You et al. / Critical Reviews in Oncology/Hematology 73 (2010) 10–22

[36] D’Amico AV, Chen MH, Roehl KA, Catalona WJ. Preoperative PSA velocity and the risk of death from prostate cancer after radical prostatectomy. N Engl J Med 2004;351:125–35. [37] Carter HB, Ferrucci L, Kettermann A, et al. Detection of lifethreatening prostate cancer with prostate-specific antigen velocity during a window of curability. J Natl Cancer Inst 2006;98:1521–7. [38] Sengupta S, Myers RP, Slezak JM, et al. Preoperative prostate specific antigen doubling time and velocity are strong and independent predictors of outcomes following radical prostatectomy. J Urol 2005;174:2191–6. [39] Tricoli JV, Schoenfeldt M, Conley BA. Detection of prostate cancer and predicting progression: current and future diagnostic markers. Clin Cancer Res 2004;10:3943–53. [40] Darson MF, Pacelli A, Roche P, et al. Human glandular kallikrein 2 (hK2) expression in prostatic intraepithelial neoplasia and adenocarcinoma: a novel prostate cancer marker. Urology 1997;49:857–62. [41] Piironen T, Lovgren J, Karp M, et al. Immunofluorometric assay for sensitive and specific measurement of human prostatic glandular kallikrein (hK2) in serum. Clin Chem 1996;42:1034–41. [42] Grauer LS, Finlay JA, Mikolajczyk SD, et al. Detection of human glandular kallikrein, hK2, as its precursor form and in complex with protease inhibitors in prostate carcinoma serum. J Androl 1998;19:407–11. [43] Partin AW, Catalona WJ, Finlay JA, et al. Use of human glandular kallikrein 2 for the detection of prostate cancer: preliminary analysis. Urology 1999;54:839–45. [44] Haese A, Graefen M, Steuber T, et al. Human glandular kallikrein 2 levels in serum for discrimination of pathologically organ-confined from locally-advanced prostate cancer in total PSA-levels below 10 ng/ml. Prostate 2001;49:101–9. [45] Steuber T, Vickers AJ, Haese A, et al. Risk assessment for biochemical recurrence prior to radical prostatectomy: significant enhancement contributed by human glandular kallikrein 2 (hk2) and free prostate specific antigen (PSA) in men with moderate PSA-elevation in serum. Int J Cancer 2006;118:1234–40. [46] Steuber T, Vickers AJ, Serio AM, et al. Comparison of free and total forms of serum human kallikrein 2 and prostate-specific antigen for prediction of locally advanced and recurrent prostate cancer. Clin Chem 2007;53:233–40. [47] Bussemakers MJG, van Bokhoven A, Verhaegh GW, et al. DD3: a new prostate-specific gene, highly overexpressed in prostate cancer. Cancer Res 1999;59:5975–9. [48] de Kok JB, Verhaegh GW, Roelofs RW, et al. DD3(PCA3), a very sensitive and specific marker to detect prostate tumors. Cancer Res 2002;62:2695–8. [49] Hessels D, Gunnewiek J, van Oort I, et al. DD3(PCA3)-based molecular urine analysis for the diagnosis of prostate cancer. Eur Urol 2003;44:8–15. [50] Groskopf J, Aubin SMJ, Deras IL, et al. APTIMA PCA3 molecular urine test: development of a method to aid in the diagnosis of prostate cancer. Clin Chem 2006;52:1089–95. [51] Marks LS, Fradet Y, Deras IL, et al. PCA3 molecular urine assay for prostate cancer in men undergoing repeat biopsy. Urology 2007;69:532–5. [52] van Gils MP, Hessels D, van Hooj O, et al. The time resolved fluorescence-based PCA3 test on urinary sediments after extended digital rectal examination; a Dutch multicenter validation of the diagnostic performance. J Urol 2007;177:568–1568. [53] Nakanishi H, Groskopf J, Fritsche HA, et al. PCA3 molecular urine assay correlates with prostate cancer tumor volume: implication in selecting candidates for active surveillance. J Urol 2008;179:1804–10. [54] Deras IL, Aubin SMJ, Blase A, et al. PCA3: a molecular urine assay for predicting prostate biopsy outcome. J Urol 2008;179:1587–92. [55] Dhir R, Vietmeier B, Arlotti J, et al. Early identification of individuals with prostate cancer in negative biopsies. J Urol 2004;171: 1419–23.

[56] Uetsuki H, Tsunemori H, Taoka R, et al. Expression of a novel biomarker, EPCA, in adenocarcinomas and precancerous lesions in the prostate. J Urol 2005;174:514–8. [57] Paul B, Dhir R, Landsittel D, et al. Detection of prostate cancer with a blood-based assay for early prostate cancer antigen. Cancer Res 2005;65:4097–100. [58] Sreekumar A, Laxman B, Rhodes DR, et al. Humoral immune response to alpha-methylacyl-CoA racemase and prostate cancer. J Natl Cancer Inst 2004;96:834–43. [59] Bradley SV, Oravecz-Wilson KI, Bougeard G, et al. Serum antibodies to huntingtin interacting protein-1: a new blood test for prostate cancer. Cancer Res 2005;65:4126–33. [60] Horoszewicz JS, Kawinski E, Murphy GP. Monoclonal antibodies to a new antigenic marker in epithelial prostatic cells and serum of prostatic cancer patients. Anticancer Res 1987;7:927–36. [61] Silver DA, Pellicer I, Fair WR, et al. Prostate-specific membrane antigen expression in normal and malignant human tissues. Clin Cancer Res 1997;3:81–5. [62] Rochon YP, Horoszewicz JS, Boynton AL, et al. Western blot assay for prostate specific membrane antigen in serum of prostate cancer patients. Prostate 1994;25:219–23. [63] Murphy GP, Kenny GM, Ragde H, et al. Measurement of serum prostate-specific membrane antigen, a new prognostic marker for prostate cancer. Urology 1998;51:89–97. [64] Douglas TH, Morgan TO, McLeod DG, et al. Comparison of serum prostate specific membrane antigen, prostate specific antigen, and free prostate specific antigen levels in radical prostatectomy patients. Cancer 1997;80:107–14. [65] Beckett ML, Cazares LH, Vlahou A, et al. Prostate-specific membrane antigen levels in sera from healthy men and patients with benign prostate hyperplasia or prostate cancer. Clin Cancer Res 1999;5:4034–40. [66] Xiao Z, Adam BL, Cazares LH, et al. Quantitation of serum prostatespecific membrane antigen by a novel protein biochip immunoassay discriminates benign from malignant prostate disease. Cancer Res 2001;61:6029–33. [67] Kinoshita Y, Kuratsukuri K, Landas S, et al. Expression of prostatespecific membrane antigen in normal and malignant human tissues. World J Surg 2006;30:628–36. [68] Tomlins SA, Rhodes DR, Perner S, et al. Recurrent fusion of TMPRSS2 and ETS transcription factor genes in prostate cancer. Science 2005;310:644–8. [69] 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. [70] Perner S, Mosquera JM, Demichelis F, et al. TMPRSS2-ERG fusion prostate cancer: an early molecular event associated with invasion. Am J Surg Pathol 2007;31:882–8. [71] 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. [72] Tomlins SA, Mehra R, Rhodes DR, et al. TMPRSS2: ETV4 gene fusions define a third molecular subtype of prostate cancer. Cancer Res 2006;66:3396–400. [73] Helgeson BE, Tomlins SA, Shah N, et al. Characterization of TMPRSS2:ETV5 and SLC45A3:ETV5 gene fusions in prostate cancer. Cancer Res 2008;68:73–80. [74] Tomlins SA, Laxman B, Dhanasekaran SM, et al. Distinct classes of chromosomal rearrangements create oncogenic ETS gene fusions in prostate cancer. Nature 2007;448:595–9. [75] Hermans KG, Bressers AA, van der Korput HA, et al. Two unique novel prostate-specific and androgen-regulated fusion partners of ETV4 in prostate cancer. Cancer Res 2008;68:3094–8. [76] Soller MJ, Elfving P, Lundgren R, Panagopols I. Confirmation of the high frequency of the TMPRSS2/ERG fusion gene in prostate cancer. Genes Chromosomes Cancer 2006;45:717–9.

J. You et al. / Critical Reviews in Oncology/Hematology 73 (2010) 10–22 [77] Mosquera JM, Perner S, Demichelis F, et al. Morphological features of TMPRSS2-ERG gene fusion prostate cancer. J Pathol 2007;212:91–101. [78] Perner S, Demichelis F, Beroukhim R, et al. TMPRSS2: ERG fusionassociated deletions provide insight into the heterogeneity of prostate cancer. Cancer Res 2006;66:8337–41. [79] Nam RK, Sugar L, Wang ZH, 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. [80] Demichelis F, Fall K, Perner S, et al. TMPRSS2: ERG gene fusion associated with lethal prostate cancer in a watchful waiting cohort. Oncogene 2007;26:4596–9. [81] Laxman B, Morris DS, Yu JJ, et al. A first-generation multiplex biomarker analysis of urine for the early detection of prostate cancer. Cancer Res 2008;68:645–9. [82] Hessels D, Smit FP, Verhaegh GW, et al. Detection of TMPRSS2ERG fusion transcripts and prostate cancer antigen 3 in urinary sediments may improve diagnosis of prostate cancer. Clin Cancer Res 2007;13:5103–8. [83] Lexander H, Palmberg C, Hellman U, et al. Correlation of protein expression, Gleason score and DNA ploidy in prostate cancer. Proteomics 2006;6:4370–80. [84] Sardana G, Jung K, Stephan C, Diamandis EP. Proteomic analysis of conditioned media from the PC3, LNCaP, and 22Rv1 prostate cancer cell lines: discovery and validation of candidate prostate cancer biomarkers. J Proteome Res 2008. [85] Ahn SM, Simpson RJ. Body fluid proteomics: prospects for biomarker discovery. Proteom Clin Appl 2007;1:1004–15. [86] Rabilloud T. Two-dimensional gel electrophoresis in proteomics: old, old fashioned, but it still climbs up the mountains. Proteomics 2002;2:3–10. [87] Moritz RL, Ji H, Schutz F, et al. A proteome strategy for fractionating proteins and peptides using continuous free-flow electrophoresis coupled off-line to reversed-phase high-performance liquid chromatography. Anal Chem 2004;76:4811–24. [88] Norden AGW, Sharratt P, Cutillas PR, et al. Quantitative amino acid and proteomic analysis: very low excretion of polypeptides >750 Da in normal urine. Kidney Int 2004;66:1994–2003. [89] Grover PK, Resnick MI. High resolution two-dimensional electrophoretic analysis of urinary proteins of patients with prostatic cancer. Electrophoresis 1997;18:814–8. [90] Rehman I, Azzouzi AR, Catto JWF, et al. Proteomic analysis of voided urine after prostatic massage from patients with prostate cancer: a pilot study. Urology 2004;64:1238–43. [91] Qin S, Ferdinand AS, Richie JP, et al. Chromatofocusing fractionation and two-dimensional difference gel electrophoresis for low abundance serum proteins. Proteomics 2005;5:3183–92. [92] Lam YW, Mobley JA, Evans JE, et al. Mass profiling-directed isolation and identification of a stage-specific serologic protein biomarker of advanced prostate cancer. Proteomics 2005;5:2927–38. [93] Malik G, Ward MD, Gupta SK, et al. Serum levels of an isoform of apolipoprotein A-II as a potential marker for prostate cancer. Clin Cancer Res 2005;11:1073–85. [94] Ward DG, Suggett N, Cheng Y, et al. Identification of serum biomarkers for colon cancer by proteomic analysis. Br J Cancer 2006;94:1898–905. [95] Paradis V, Degos F, Dargere D, et al. Identification of a new marker of hepatocellular carcinoma by serum protein profiling of patients with chronic liver diseases. Hepatology 2005;41:40–7. [96] Le L, Chi K, Tyldesley S, et al. Identification of serum amyloid A as a biomarker to distinguish prostate cancer patients with bone lesions. Clin Chem 2005;51:695–707. [97] Lehrer S, Roboz J, Ding H, et al. Putative protein markers in the sera of men with prostatic neoplasms. BJU Int 2003;92:223–5. [98] Petricoin EF, Ardekani AM, Hitt BA, et al. Use of proteomic patterns in serum to identify ovarian cancer. Lancet 2002;359:572–7 [see comment].

21

[99] Hutchens T, Yip T. New desorption strategies for the mass spectrometric analysis of macromolecules. Rapid Commun Mass Spectrom 1993;7:576–80. [100] Petricoin III EF, Ornstein DK, Paweletz CP, et al. Serum proteomic patterns for detection of prostate cancer. J Natl Cancer Inst 2002;94:1576–8 [see comment]. [101] Qu Y, Adam B-L, Yasui Y, et al. Boosted decision tree analysis of surface-enhanced laser desorption/ionization mass spectral serum profiles discriminates prostate cancer from noncancer patients. Clin Chem 2002;48:1835–43 [see comment]. [102] Ornstein DK, Rayford W, Fusaro VA, et al. Serum proteomic profiling can discriminate prostate cancer from benign prostates in men with total prostate specific antigen levels between 2.5 and 15.0 ng/ml. J Urol 2004;172:1302–5. [103] Kohli M, Siegel E, Bhattacharya S, et al. Surface-enhanced laser desorption/ionization time-of-flight mass spectrometry (SELDI-TOF MS) for determining prognosis in advanced stage hormone relapsing prostate cancer. Cancer Biomark 2006;2:249–58. [104] Petricoin EF, Liotta LA. SELDI-TOF-based serum proteomic pattern diagnostics for early detection of cancer. Curr Opin Biotechnol 2004;15:24–30. [105] Ekblad L, Baldetorp B, Ferno M, et al. In-source decay causes artifacts in SELDI-TOF MS spectra. J Proteome Res 2007;6:1609–14. [106] Baggerly KA, Morris JS, Coombes KR. Reproducibility of SELDITOF protein patterns in serum: comparing datasets from different experiments. Bioinformatics 2004;20:777–85. [107] Lai Y, Adam B-l, Podolsky R, She J-X. A mixture model approach to the tests of concordance and discordance between two large-scale experiments with two-sample groups. Bioinformatics 2007;23:1243–50. [108] Petricoin EF, Ornstein DK, Liotta LA. Clinical proteomics: applications for prostate cancer biomarker discovery and detection. Urology 2004;22:322–8. [109] M’Koma AE, Blum DL, Norris JL, et al. Detection of pre-neoplastic and neoplastic prostate disease by MADI profiling of urine. Biochem Biophys Res Commun 2007;353:829–34. [110] Zhang WM, Finne P, Leinonen J, et al. Determination of prostatespecific antigen complexed to alpha(2)-macroclobulin in serum increases the specificity of free to total PSA for prostate cancer. Urology 2000;56:267–72. [111] Ayala G, Thompson T, Yang G, et al. High levels of phosphorylated form of Akt-1 in prostate cancer and non-neoplastic prostate tissues are strong predictors of biochemical recurrence. Clin Cancer Res 2004;10:6572–8. [112] Le Page C, Koumakpayi IH, Alam-Fahmy M, et al. Expression and localisation of Akt-1, Akt-2 and Akt-3 correlate with clinical outcome of prostate cancer patients. Br J Cancer 2006;94:1906–12. [113] Zehentner BK, Secrist H, Zhang XQ, et al. Detection of alpha-methylacyl-coenzyme-A racemase transcripts in blood and urine samples of prostate cancer patients. Mol Diagn Ther 2006;10:397–403. [114] Shimada K, Matsuyoshi S, Nakamura M, et al. Phosphorylation status of Fas-associated death domain-containing protein (FADD) is associated with prostate cancer progression. J Pathol 2005;206: 423–32. [115] Deftos LJ, Abrahamsson PA. Granins and prostate cancer. Urology 1998;51:141–5. [116] Goessl C, Muller M, Heicappell R, et al. DNA-based detection of prostate cancer in blood, urine, and ejaculates. In: Circulating nucleic acids in plasma or serum Ii. New York: New York Acad. Sciences; 2001. p. 51–58. [117] Bonkhoff H, Fixemer T, Hunsicker I, Remberger K. Estrogen receptor gene expression and its relation to the estrogen-inducible HSP27 heat shock protein in hormone refractory prostate cancer. Prostate 2000;45:36–41. [118] Halvorsen OJ, Haukaas S, Hoisaeter PA, Akslen LA. Maximum Ki-67 staining in prostate cancer provides independent prognostic

22

[119]

[120]

[121] [122] [123]

[124]

[125]

[126]

[127]

J. You et al. / Critical Reviews in Oncology/Hematology 73 (2010) 10–22 information after radical prostatectomy. Anticancer Res 2001;21: 4071–6. Brewster SF, Oxley JD, Trivella M, et al. Preoperative p53, bcl-2, CD44 and E-cadherin immunohistochemistry as predictors of biochemical relapse after radical prostatectomy. J Urol 1999;161:1238–43. Huang CL, Brassil D, Rozzell M, et al. Comparison of prostate secretory protein with prostate specific antigen and prostatic acid phosphatase as aserum biomarker for diagnosis and monitoring patients with prostate carcinoma. Prostate 1993;23:201–12. Xu Y, Zhang T, Tang H, et al. Overexpression of PIM-1 is a potential biomarker in prostate carcinoma. J Surg Oncol 2005;92:326–30. Said J. Biomarker discovery in urogenital cancer. Biomarkers 2005;10:S83–6. Wang JH, Weng JS, Cai Y, et al. The prostate-specific G-protein coupled receptors PSGR and PSGR2 are prostate cancer biomarkers that are complementary to alpha-methylacyl-CoA racemase. Prostate 2006;66:847–57. Van Huizen I, Wu GJ, Moussa M, et al. Establishment of a serum tumor marker for preclinical trials of mouse prostate cancer models. Clin Cancer Res 2005;11:7911–9. Ellinger J, Bastian PJ, Jurgan T, et al. CpG island hypermethylation at multiple gene sites in diagnosis and prognosis of prostate cancer. Urology 2008;71:161–7. Fuessel S, Sickert D, Meye A, et al. Multiple tumor marker analyses (PSA, hK2, PSCA, trp-p8) in primary prostate cancers using quantitative RT-PCR. Int J Oncol 2003;23:221–8. Henshall SM, Afar DEH, Hiller J, et al. Survival analysis of genomewide gene expression profiles of prostate cancers identifies new prognostic targets of disease relapse. Cancer Res 2003;63:4196–203.

Biographies Jing Jing You is a (Ph.D.) candidate studying at the St George Clinical School, Faculty of Medicine of University of New South Wales (UNSW), Sydney, Australia. The objective of her research is to identify novel biomarkers from tears and sera for prostate cancer diagnosis and prognosis. Paul Cozzi is Senior Urologist at the Department of Surgery, Urology Sydney and Senior Lecturer at UNSW. He has pioneered new surgical treatments for prostate cancer in the Australian setting and established a database of patients with urological cancers, including prostate cancer patients at St George Hospital. His interests at present include clinical and laboratory research into prostate and bladder cancer with a special interest in experimental therapeutics for the treatment of these malignancies and early diagnosis of cancer. Bradley Walsh is the (CEO) of Minomic International Ltd. who has ∼20 years experience in protein science, and has been associated with proteome technology since its inception. He has an outstanding perspective of the power of the technology and the means by which it can be applied in a broad variety of projects. In recent years he has searched for biomarkers in various body fluids from diabetic patients and

prostate cancer patients using proteomics technologies and has had some success in elucidating potential biomarkers. Mark Willcox is Professor at the School of Optometry and Vision Science, UNSW and Chief Scientific Officer of the Institute for Eye Researcgh Ltd., Sydney. He has been working on tear protein analysis for many years, and identified a tear protein, lacryglobin, as a potential biomarker for cancers. Prof. Willcox is a world-renowned expert in the field of tear film study. John Kearsley is Director of Radiation Oncology at St George Cancer Care Centre, Head of Prostate Cancer Institute in St George Hospital and Conjoint Professor, Faculty of Medicine, UNSW. He is a prostate cancer clinician. Pamela J Russell completed her (M.Sc.) and (Ph.D.) at Walter and Eliza Hall Institute of Medical Research, Melbourne, Victoria under the direction of Sir Macfarlane Burnet (Nobel Prize winner) and Sir Gustav Nossal, respectively, in immunology. She subsequently worked on genitourinary cancer with Dr. Derek Raghavan, and is now the Director of the Oncology Research Centre at Prince of Wales Hospital UNSW, Conjoint Professor of Faculty of Medicine UNSW, the secretary of the Australian and New Zealand Prostate and Urological Cancer Clinical Trials Groups, on the Scientific Advisory Committee of Minomic International Ltd. and an executive of the Australian Prostate Cancer Collaboration. She has performed studies on antibody targeted therapy and gene therapy for prostate cancer and is a renowned expert in prostate cancer research. Yong Li completed his (Ph.D.) degree in Faculty of Medicine in 2000 from the UNSW, Sydney, Australia, and postdoctoral training at Cancer Care Center, St George Hospital, Sydney, UNSW, Australia, and then became an independent investigator. Currently, Dr. Li is a Senior Cancer Institute NSW Research Fellow and the Head of Cancer Research Program, St George Hospital and Conjoint Senior Lecturer, Faculty of Medicine at UNSW. Dr. Li’ research program is aimed at (a) to investigate novel biomarkers or tumor-associated antigens for cancer diagnosis, monitoring and therapy; (b) to use targeted cancer therapy and combination therapy to control micrometastatic prostate cancer and ovarian cancer; (c) to investigate the mechanism of cancer metastasis. These studies will provide new insights into the mechanisms of cancer metastasis, control of micrometastatic development, which may lead to the development of novel therapies for the treatment of cancer progression. The success in identifying proteins of interest in cancer patients may be used to stage the disease, monitor cancer progression and recurrence, and to guide clinicians in choosing the best treatment methods for an individual patient.