Prognostic Factors in Localised Prostate Cancer with Emphasis on the Application of Molecular Techniques

European Urology

European Urology 41 (2002) 363±371

Prognostic Factors in Localised Prostate Cancer with Emphasis on the Application of Molecular Techniques P.C.M.S. Verhagena,*, M.G.J. Tilanusb, R.A. de Wegerb, R.J.A. van Moorselaarc, J.G. van den Tweelb, T.A. Boonc a

Department of Urology, University Hospital Rotterdam, P.O. Box 2040, 3000 CA, Rotterdam, The Netherlands Department of Pathology, University Medical Centre Utrecht, P.O. Box 85500, 3508 GA, Utrecht, The Netherlands c Department of Urology, University Medical Centre Utrecht, P.O. Box 85500, 3508 GA, Utrecht, The Netherlands b

Accepted 31 January 2002

Abstract Prostate cancer is the most prevalent malignancy in males in the Western world and the second leading cause of male cancer death. Prostate speci®c antigen (PSA) based screening and case ®nding leads to identi®cation of early stage prostate cancer. It is often dif®cult to discriminate between patients that need curative treatment and those that can be managed conservatively. Prognostic factors are used to make this clinical decision. Based on the classi®cation proposed by the American College of Pathologists and the World Health Organisation, selected prognostic factors in prostate cancer are described. Clinical applicable factors are stage, grade and serum PSA. Prognostic factors that are not routinely used (for various reasons) are ploidy, histological type and cancer volume in needle biopsies. All other factors (including circulating tumour cells, angiogenesis, growth factors, proliferation rate, apoptosis, nuclear morphometry, neuroendocrine differentiation, loss of chromosomal regions, tumour suppresser genes and adhesion molecules) are promising as prognostic factor although currently their use in clinical decisions is not recommended. The role of these factors in prostate cancer growth and their predictive value are discussed. The rapid developments in molecular techniques allow assessment of structure or function of thousands of genes in a prostate biopsy sample. We expect that molecular characterisation of tumour material will become a clinically important tool to predict prognosis in patients with localised prostate cancer. # 2002 Elsevier Science B.V. All rights reserved. Keywords: Prostate cancer; Prognostic factors; Molecular techniques 1. Introduction Prostate cancer is the most commonly diagnosed cancer in men in the Western world. It is the second leading cause of male cancer deaths [1]. Below the age of 50, the incidence of prostate cancer is very low. Above the age of 55, there is a steep rise with age, to over 900 cases per 100,000 per year at the age of 80 [2]. In the past decade, the incidence of prostate cancer has shown marked changes. In the United States, the incidence rate peaked in 1992 at 240 cases per *

Corresponding author. Tel.: ‡31-104633132/094; Fax: ‡31-104635838. E-mail address: [email protected] (P.C.M.S. Verhagen).

100,000 per year, and has been declining since [1]. This is mainly due to the introduction of prostate speci®c antigen (PSA) serum tests and the increasing awareness of prostate cancer in the population. The prostate cancer mortality rates have shown only small changes in the past decade. From 1992 to 1995, the age adjusted mortality rate in the United States has shown a slight decline [3]. The natural history of prostate cancer is variable and dif®cult to predict. In autopsy studies of asymptomatic men small foci of prostate cancer have been found in up to one-third of men in their ®fth decade [4]. In 1993, Coffey estimated that one out of four prostate cancers becomes clinically apparent and that one out of three of these apparent cancers will lead to death [5].

0302-2838/02/$ ± see front matter # 2002 Elsevier Science B.V. All rights reserved. PII: S 0 3 0 2 - 2 8 3 8 ( 0 2 ) 0 0 0 4 8 - 9

364

P.C.M.S. Verhagen et al. / European Urology 41 (2002) 363±371

Long-time survival has been reported in selected early stage prostate cancer patients after conservative management (disease speci®c survival over 80% after 10±15 years) [6]. Because of a selection-bias (patients with favourable characteristics are more likely to be managed conservatively), prostate cancer death risk in unselected early stage cases is higher. Aus et al. who traced all conservatively managed prostate cancer patients who died within a certain period concluded that death was attributable to prostate cancer in 50% of originally M0 cases [7]. Two conclusions can be drawn from these studies: (1) a subset of patients will die from their cancer if managed conservatively and (2) it takes many years before a localised cancer leads to death. In order to plan an appropriate therapy in early prostate cancer prognostic factors are necessary. In deciding on a treatment modality in clinically localised disease (radical prostatectomy, external beam radiotherapy, brachytherapy or watchfull waiting) three aspects must be taken into account: (1) the expected biological behaviour of the tumour, (2) the life expectancy of the patient, and (3) quantity versus quality of life trade-offs. This review will focus on the dif®culties in predicting tumour behaviour. The appli-

cation of molecular techniques to prostate cancer samples has resulted in many papers describing a correlation between a certain factor and the classical prognostic factors (grade, stage, PSA). It is not surprising that more genetic abnormalities are found in more advanced lesions. This does not imply that the presence of these abnormalities add something to what we already knew (independent prognostic value in multivariate analysis). The predictive value of factors assessed on needle biopsies is hampered by the multifocal and heterogeneous nature of prostate cancer. Due to this sampling error some factors may have prognostic value when assessed in radical prostatectomy specimens, but not when assessed in corresponding needle biopsies. Stackhouse et al. showed this for p53 and bcl-2 immunostaining [8]. Lack of standardisation is another source of variability in results. These problems contribute to the fact that in spite of many promising reports, currently no molecular factors are available that are recommended in clinical decisions. To classify factors the College of American Pathologists (CAP) de®ned three categories, based on clinical applicability. Table 1 shows the classi®cations proposed by the 1999 CAP Conference on Solid Tumor

Table 1 Classi®cation of prognostic factors by CAP and WHO in year 1999 CAP

WHO

Category I: factors that have been proven to be prognostic or predictive based on evidence from multiple published trials and are recommended for routine reporting TNM stage TNM stage Histological grade (Gleason) Histological grade (Gleason and WHO nuclear grade) Surgical margin status Surgical margin status Perioperative PSA Perioperative PSA Pathological effects of treatment Location of cancer within prostate Category II: factors that show promise as predictive factors based on evidence from multiple published studies but that require further evaluation before recommendation or are recommended despite incomplete data as diagnostic or prognostic markers DNA ploidy DNA ploidy Histological type Histological type Cancer volume in biopsy Cancer volume in biopsy Cancer volume in RP Cancer volume in RP Category III: factors that have some scienti®c evidence to support their adoption as diagnostic or prognostic agents but are not currently recommended; also, factors of uncertain signi®cance Prostate-speci®c membrane antigen Prostate-speci®c membrane antigen Other serum tests (RT-PCR, PSM, hK2, IGF) Other serum tests (RT-PCR, PSM, hK2, IGF) Perineural invasion Perineural invasion Vascular/lymphatic invasion Vascular/lymphatic invasion Microvessel density Microvessel density Stromal factors (TGF-b, integrins) Stromal factors (TGF-b, integrins) Proliferation markers and apoptosis Proliferation markers and apoptosis Nuclear morphometry and karyometric analysis Nuclear morphometry and karyometric analysis Androgen receptor Androgen receptor Neuroendocrine markers Neuroendocrine markers Genetic markers Genetic markers All other factors All other factors

P.C.M.S. Verhagen et al. / European Urology 41 (2002) 363±371

Prognostic Factors and the 1999 World Health Organisation (WHO) Second International Consultation on Prostate Cancer [9]. 2. Factors that have been proven to be prognostic or predictive based on evidence from multiple published trials and are recommended for routine reporting 2.1. Stage, grade, PSA Stage (TNM), grade (Gleason score or Gleason score and WHO nuclear grade) and serum PSA are well established prognostic factors that are routinely used in clinical decisions. These factors are extensively discussed elsewhere (e.g. [9,10]). Surgical margin status and location of cancer within the prostate can only be assessed after radical prostatectomy and therefore have no in¯uence on treatment decisions in early disease. 3. Factors that show promise as predictive factors based on evidence from multiple published studies but that require further evaluation before recommendation or are recommended despite incomplete data as diagnostic or prognostic markers 3.1. Ploidy Normal cells have a diploid DNA content, which can be determined by static or ¯ow cytometry. Prostate tumours of low grade and stage are usually also diploid, whereas tumours of higher grade and stage are more often aneuploid. Since 1966, this topic has received much attention [11,12]. The majority of papers reported a decrease in survival in patients with aneuploid tumours. In several reports ploidy was found to be an independent prognostic factor in multivariate analysis. One study reported a high concordance between ploidy of biopsies and surgical specimens [13]. In individual patients with clinical localised disease the prognostic value of ploidy measurements remains limited. 3.2. Histological type Over 95% of prostatic tumours are adenocarcinomas and studies on prognostic factors in general apply to this category. Other histological types include transitional cell carcinoma, squamous cell carcinoma, undifferentiated carcinoma, rhabdomyosarcoma and leiomyosarcoma. Neuroendocrine cells can be identi®ed in benign and malignant prostatic tissues. Neuroendocrine differentiation in prostate tumours refers

365

to three different entities: (1) small cell prostate cancer (SCPC), (2) prostatic carcinoid (very rare), and (3) adenocarcinoma with scattered neuroendocrine cells [14]. SCPC accounts for <5% of prostate tumours and is reported to follow an aggressive course [15]. 3.3. Cancer volume in needle biopsies Prostate cancer volume is dif®cult to assess because macroscopically the tumour is hard to recognise and often multifocal. In radical prostatectomy specimens volume can be measured and is reported to correlate with grade, stage and outcome [16]. Several studies report a good correlation between sextant biopsy involvement and cancer volume in radical prostatectomy specimens (e.g. [17]) or PSA recurrence after radical prostatectomy [18]. Others have clearly shown that the presence of signi®cant disease cannot be ruled out on biopsy parameters [19,20]. While one small focus of tumour in sextant biopsies does not imply the presence of a limited cancer, extensive involvement of multiple biopsies points towards extensive disease and in this respect it can contribute to treatment decisions. 4. Factors that have some scientific evidence to support their adoption as diagnostic or prognostic agents but are not currently recommended; also, factors of uncertain significance 4.1. Circulating prostate cells Several studies tried to correlate preoperative PSA reverse transciptase polymerase chain reaction (RT-PCR) positivity of blood (which is assumed to indicate the presence of intact prostate cells) with biochemical recurrence after radical prostatectomy. Some found such a correlation [21] and others did not [22]. Some found superior results using prostate speci®c membrane antigen (PSMA) mRNA [23]. Some studies mention false positive results with the nested PCR ampli®cation which is the method applied in most reports. 4.2. PSA density, PSA free to total ratio, PSMA PSA density (PSA level divided by prostatic volume) and PSA free to total ratio (free PSA divided by free PSA ‡ a1-chymotrypsin bound PSA) are methods applied to enhance the use of PSA for tumour detection. Application of these methods may improve prognostic value although de®nite conclusions cannot be drawn [18,24,25]. Measurement of serum PSMA appeared to have prognostic value in one study [26].

366

P.C.M.S. Verhagen et al. / European Urology 41 (2002) 363±371

4.3. Perineural invasion, vascular/lymphatic invasion Perineural invasion in needle biopsies was an independent predictor of PSA recurrence in some series [27]. However, Egan and Bostwick did not ®nd perineural invasion to independently predict extra prostatic extension and recommended not to routinely evaluate this ®nding [28]. Microvascular or lymphatic invasion in radical prostatectomy specimens correlated with outcome and was an independent predictor of PSA recurrence in a recent series [29] but it was dependent on grade in another study [30]. 4.4. Angiogenesis, microvessel density Microvessel density can be studied by applying antibodies to factor VIII, CD31 or CD34 to prostatic tissues. Microvessel density was higher in tumours compared to adjacent normal tissue [31]. Prognostic value has been reported after radiotherapy [32] and after radical prostatectomy [33,34]. Microvessel density assessed in needle biopsies from 186 patients who subsequently underwent radical prostatectomy improved prediction of extraprostatic disease when combined with Gleason score and serum PSA [35]. 4.5. Growth factors Epidermal growth factor (EGF) and transforming growth factor-a (TGF-a) both activate the same receptor (EGFR) which leads to proliferation of normal epithelial cells. Transforming growth factor-b (TGFb) enhances cancer growth and metastasis [36]. EGF receptor expression was reported to correlate with progression free, cancer speci®c and overall survival in 147 prostatic cancers, but had no independent prognostic value in multivariate analysis [37]. Loss of expression of TGF-b receptor R1 correlated with tumour stage and survival [38]. TGF-b1 level in blood was associated with advanced disease and metastasis [39]. C-erbB-2 (HER-2/neu) is an oncoprotein related to EGFR, with prognostic value in breast and ovarian cancer. Ross et al. found its presence to be associated with high tumour grade and stage, but it was not an independent prognostic factor [40]. 4.6. Antigens associated with proliferation Several markers are applied to measure cell proliferation: Ki-67, MIB-1 (both directed to the same antigen) and proliferating cell nuclear antigen (PCNA). Ki-67 is a nuclear antigen present throughout the cell cycle, but not present at rest (G0 or early G1 phase). The proliferation rate of prostate cancer in general is low. Several studies reported (independent) prognostic value of Ki-67/MIB-1 [41±44] and PCNA [45,46]. Naito et al. reported independent prognostic signi®-

cance of PCNA index assessed in needle biopsies of 54 prostatic carcinomas [47]. Coetzee et al. found that the proliferative index by Ki-67/MIB-1 in 244 radical prostatectomy specimens added little above GSS, pathological stage and ploidy [48]. 4.7. Apoptosis, bcl-2 Programmed cell death (apoptosis) is essential in normal development and can occur in cancer cells following hormonal treatment. The bcl-2 oncoprotein inhibits apoptosis. In primary prostate cancer, overexpression has been reported in 32±41%. In a series of 175 radical prostatectomy specimens, Bauer et al. showed that bcl-2 expression was an independent predictor of PSA recurrence [49]. A total of 47/175 patients had bcl-2 over-expression. Of these men, 67% recurred at 5 years, compared to 30.5% in bcl-2 negative patients. The bcl-2 expression in needle biopsies from this same cohort of patients was not predictive of recurrence [8]. Huang et al. found pretreatment bcl-2 expression in biopsies to be predictive of recurrence in patients treated by radiotherapy [50]. 4.8. Morphometric studies Nuclear morphometry uses automated techniques to quantify abnormalities in nuclear shape. Nuclear roundness was found to predict progression in patients after radical prostatectomy [51] and after TURP [52]. Blom et al. found variation in nuclear size to be predictive of outcome after radical prostatectomy, but only within the group of (Mosto®) grade 2 tumours [53]. 4.9. Genetic markers 4.9.1. Chromosomal regions Chromosome aberrations found by classical cytogenetics include structural changes in chromosome 1, 2, 7, 3p, 6p, 8p, 10q, 13q, 15q and 16q. Whole chromosome gain of 7, 14, 20 and 22 and loss of 1, 2, 4, 5 and Y are reported. These changes seem independent of stage and grade [54,55]. Comparative genomic hybridization (CGH) of primary prostate cancers revealed frequent loss of certain chromosomal regions (6q, 8p, 13q, and 16q) and gain of 8q and 7 [56]. Loss of heterozygosity (assessed by allelotyping) is reported in 6q13±21, 7q31.1, 8p12±p22, 10p, 10q23±q25, 13q14, 16q, 17q21 and 18q [57±59]. Loss at 7q31.1 was present in 30% of primary prostate cancers which correlated with tumour grade and lymph node metastasis [60]. 4.9.2. Tumour suppressor genes 4.9.2.1. The p53 tumour suppressor gene and WAF1/ CIP1. The p53 tumour suppressor gene has a function in

P.C.M.S. Verhagen et al. / European Urology 41 (2002) 363±371

cell cycle control and regulates the WAF1/CIP1 gene. Aberrations in the p53 gene can be assessed by direct sequencing or immunohistochemistry (nuclear accumulation of the aberrant p53 protein). False positive results in immunohistochemical staining are reported [61]. The frequency of p53 mutations in primary prostate cancers is reported to be low (up to 14%) [62,63]. Others found 61±80% of clinically localised cancers to be immunohistochemically positive for p53 [49,64]. Nuclear accumulation of p53 was reported to be of limited prognostic value [46]. Others found it to be an independent predictor of PSA recurrence after radical prostatectomy [49,64]. Corresponding needle biopsies, available from one of these series, did not show prognostic significance when analysed for p53 accumulation [8]. Taken together, assessment of prognostic value of p53 is hampered by variability in techniques and criteria for positivity, heterogeneous expression within a tumour and limited number of studies performed. WAF1/CIP1 is activated by p53. The gene product, p21, inhibits proliferation and directs the cell into apoptosis. Mutations have been reported in 17% of primary prostate cancers [65]. The p21 expression is increased in cancers compared to benign prostatic tissue [66]. Several studies found increased p21 expression to be associated with other prognostic factors (stage, grade) [67] or survival [68,69]. 4.9.2.2. PTEN/MMAC1. Germ line mutations in the PTEN gene cause Cowden disease, characterised by hamartomas and a predisposition to various tumours. The biological target of PTEN are inositol phospholipids. PTEN dephosphorylates these lipids, which play a central role in various cellular processes, including cell cycle regulation and cell survival [70]. PTEN is frequently inactivated in prostate cancer [71,72]. Loss of PTEN expression in 109 primary prostate cancers was shown to correlate with Gleason score and advanced stage [73]. 4.9.2.3. P27 (Kip1). P27 (Kip1) is a cyclin-dependent kinase inhibitor with a role in cell cycle arrest and apoptosis [74]. Decreased expression in prostate cancer is reported to correlate with grade, stage and prognosis [75±78]. Expression of p27 in needle biopsies is reported to predict the expression level in radical prostatectomy specimens [79]. 4.9.2.4. Other tumour suppressor genes. The RB gene (responsible for hereditary retino blastoma) has drawn attention because it is located on 13q14.1 which is frequently deleted in prostate cancer. Bookstein found abnormal RB expression in two out of seven advanced

367

cancers [80]. Although allelic loss is reported in up to 60% at 13q14, no correlation could be established between LOH in the RB region and absence of the RB protein indicating the presence of another tumour suppressor gene in this region [81]. MXI1 is located on 10q24±25. The protein product negatively regulates the oncogene myc. Mutations were shown in primary tumours with concomitant 10q24±25 deletions [82]. 4.9.3. Adhesion molecules 4.9.3.1. E-cadherine/a-catenin. E-cadherin is a cell adhesion molecule, mapped to chromosome 16q22.1. The protein is located at the cell membrane and forms a complex with the cytoplasmatic catenins. In normal prostate, membranous staining of E-cadherin in the epithelial cells can be demonstrated. In tumours this staining pattern is often aberrant or absent [83,84]. There is frequent loss of 16q22 in prostate cancer, but concomitant mutations of E-cadherin are not often found. Decreased expression appears to be due to hypermethylation of CpG islands in the E-cadherin promotor region [85]. All studies describe an inverse correlation between E-cadherin expression and grade and stage. Because E-cadherin depends on the presence of a-catenin to form a functional complex, some studies analysed these molecules together and report improved prognostic value if they are both taken into account [86,87]. 4.9.3.2. CD44. CD44, located on 11p13 encodes a transmembrane glycoprotein with a function in intercellular interactions. The gene has 19 exons. Standard CD44 (CD44s) is composed of exons 1±5 and 15±19; several splice variants contain one or more of the exons 6±14. In gastrointestinal epithelia, CD44 is not expressed. In tumours of the gastrointestinal tract, CD44 expression is correlated with adverse outcome. In contrast, CD44 is expressed in normal prostatic tissue. Loss of CD44s expression in radical prostatectomy specimens was an independent predictor for recurrence [78,88]. Downregulation of CD44s and CD44v6 assessed on archival needle biopsies was predictive of poor outcome [89]. 4.9.3.3. KAI1. The KAI1 gene, located on 11p11.2 is another transmembrane protein with a putative role in regulation of cell development, activation, growth and motility [90]. Mashimo et al. identified a region with strong homology to the p53 responsive consensus sequence in the KAI1 promotor. They found evidence that KAI1 expression is directly activated by p53 [91]. Several studies showed an inverse

368

P.C.M.S. Verhagen et al. / European Urology 41 (2002) 363±371

correlation between KAI1 expression and tumour grade and/or stage [92]. 4.10. Other factors Studies on the ras oncogene report variable frequency of mutations in primary cancers (5±33%) [93]. The c-myc oncogene is located on chromosome 8q24, a region that shows frequent gain in prostate cancer. Sato et al. reported ampli®cation of c-myc in high grade advanced cancers to be predictive of outcome [94]. PSA is expressed by normal prostatic epithelial cells and also by the majority of the malignant prostatic cells. There is a strong inverse correlation between Gleason grade and the PSA content of prostate cancer. The prognostic value of PSA expression in biopsies or TURP specimens was reported to be limited [52,95] or absent [96]. Decreased androgen receptor content was found to predict worse outcome in patients with advanced cancers [97]. During normal replication chromosomes shorten and this limits the life span of the cell. Telomerase is an enzyme that reverses this chromosome shortening. Telomerase activity was found to be absent in benign prostatic tissues and present in prostate cancer [98]. 5. Conclusion The factors grade, stage, serum PSA, ploidy and cancer volume in needle biopsies have been studied extensively in clinical settings and can to some extent predict the course of localised prostate cancer. For various reasons these prognostic factors have limited power in discriminating patients with early prostate cancer in those who need a cure and those who do not. Given the fact that the majority of patients in a screening population have a cancer that will not change their life expectancy, there is a great need for more precise prognostic factors. In clinical decisions, physicians tend to give the patient the bene®t of the doubt. This implies that if prognostic factors leave uncertainty about the outcome, there effect on treatment decisions will be limited. An example of this is the infrequent application of DNA ploidy measurements of biopsies in clinical decisions, in spite of the extensive literature that supports its use.

The multifocal and heterogeneous nature of prostate cancer makes it dif®cult to obtain a representative biopsy sample. Improvements in biopsy procedures will be mandatory in order to make progress on this issue. Molecular techniques have made it possible to study genetic abnormalities in detail. This contributes to the understanding of the multiple steps of prostate cancer initiation and progression. In addition, these steps can serve as prognostic factors that provide us with information on the nature of the cancer. More advanced tumours have more genetic abnormalities. This is re¯ected in numerous reports that show progressive abnormalities of a marker with advancing tumour grade and stage. This does not imply that these markers are prognostically important. Prognostic signi®cance can only be shown in multivariate analysis of prospective trials, where new markers are compared with established prognosticators and outcome. Most trials currently available used radical prostatectomy specimens to investigate a marker. We need markers that can be assessed before the treatment decision is made. For tissue factors this implies that they have to be tested on biopsy materials. We need a standardised approach. For example the wide variability of reported p53 staining in primary prostate cancer (0±80%, see above) must be attributed to differences in technique and scoring. We need markers in patients with localised cancers which means that a very long follow-up is necessary before a correlation with survival can be established. These aspects make this kind of research dif®cult. It is not surprising therefore, that the use of molecular factors cannot be advocated in clinical decisions at present. It is our conviction that this will change. It is unlikely that a single marker will emerge, which can provide all the information necessary. cDNA micro-arrays make it possible to obtain information on expression of thousands of genes from one biopsy sample in a single experiment (e.g. [99]). It is likely that this extensive molecular characterisation of a tumour will lead to a better prediction of its biological behaviour. A major improvement in prostate cancer therapy will be possible if patients with localised disease can be adequately selected for curative or expectant management. The life expectancy of the patient and quality versus quantity of life trade-offs are the complementary issues that still may form a dif®cult subject for debate.

References [1] Greenlee RT, Hill-Harmon MB, Murray T, Thun M. Cancer statistics, 2001 CA Cancer J Clin 2001;51:15±36.

[2] Pienta KJ, Esper PS. Risk factors for prostate cancer. Ann Intern Med 1993;118:793±803.

P.C.M.S. Verhagen et al. / European Urology 41 (2002) 363±371 [3] Merrill RM, Stephenson RA. Trends in mortality rates in patients with prostate cancer during the era of prostate speci®c antigen screening. J Urol 2000;163:503±10. [4] Sakr WA, Haas GP, Cassin BF, Pontes JE, Crissman JD. The frequency of carcinoma and intraepithelial neoplasia of the prostate in young male patients. J Urol 1993;150:379±85. [5] Coffey DS. Prostate cancer. An overview of an increasing dilemma. Cancer 1993;71:880±6. [6] Johansson JE, Holmberg L, Johansson S, Bergstrom R, Adami HO. Fifteen years survival in prostate cancer. A prospective, populationbased study in Sweden. JAMA 1997;277:467±71. [7] Aus G, Hugosson J, Norlen L. Long-term survival and mortality in prostate cancer treated with noncurative intent. J Urol 1995;154:460±5. [8] Stackhouse GB, Sesterhenn IA, Bauer JJ, Mosto® FK, Connelly RR, Srivastava SK, Moul JW. The p53 and bcl-2 immunohistochemistry in pretreatment prostate needle biopsies to predict recurrence of prostate cancer after radical prostatectomy. J Urol 1999;162:2040±5. [9] Bostwick DG, Foster CS. Predictive factors in prostate cancer: Current concepts from the 1999 College of American Pathologists Conference on Solid Tumor Prognostic Factors and the 1999 World Health Organization Second International Consultation on Prostate Cancer. Semin Urol Oncol 1999;17:222±72. [10] Partin AW, Kattan MW, Subong EN, Walsh PC, Wojno KJ, Oesterling JE, et al. Combination of prostate-speci®c antigen, clinical stage, and Gleason score to predict pathological stage of localized prostate cancer. A multi-institutional update. JAMA 1997;277:1445±51. [11] Adolfsson J. Prognostic value of deoxyribonucleic acid content in prostate cancer: a review of current results. Int J Cancer 1994;58: 211±6. [12] Shankey TV, Kallioniemi OP, Koslowski JM, Lieber ML, Mayall BH, Miller G, Smith GJ. Consensus review of the clinical utility of DNA content cytometry in prostate cancer. Cytometry 1993;14:497± 500. [13] Ross JS, Figge H, Bui HX, del Rosario AD, Jennings TA, Rifkin MD, Fisher HA. Prediction of pathologic stage and postprostatectomy disease recurrence by DNA ploidy analysis of initial needle biopsy specimens of prostate cancer. Cancer 1994;74:2811±8. [14] di Sant'Agnese PA. Neuroendocrine differentiation in carcinoma of the prostate. Diagnostic, prognostic, and therapeutic implications. Cancer 1992;70:254±68. [15] Oesterling JE, Hauzeur CG, Farrow GM. Small cell anaplastic carcinoma of the prostate: A clinical, pathological and immunohistological study of 27 patients. J Urol 1992;147:804±7. [16] Epstein JI, Carmichael M, Partin AW, Walsh PC. Is tumor volume an independent predictor of progression following radical prostatectomy? A multivariate analysis of 185 clinical stage B adenocarcinomas of the prostate with 5 years of followup. J Urol 1993;149:1478± 81. [17] Peller PA, Young DC, Marmaduke DP, Marsh WL, Badalament RA. Sextant prostate biopsies. A histopathologic correlation with radical prostatectomy specimens. Cancer 1995;75:530±8. [18] Ravery V, Chastang C, Toublanc M, Boccon-Gibod L, Delmas V. Percentage of cancer on biopsy cores accurately predicts extracapsular extension and biochemical relapse after radical prostatectomy for T1-T2 prostate cancer. Eur Urol 2000;37:449±55. [19] Cupp MR, Bostwick DG, Myers RP, Oesterling JE. The volume of prostate cancer in the biopsy specimen cannot reliably predict the quantity of cancer in the radical prostatectomy specimen on an individual basis. J Urol 1995;153:1543±8. [20] Bruce RG, Rankin WR, Cibull ML, Rayens MK, Banks ER, Wood Jr DP. Single focus of adenocarcinoma in the prostate biopsy specimen is not predictive of the pathologic stage of disease. Urology 1996;48:75±9. [21] Olsson CA, De Vries GM, Benson MC, Raffo A, Buttyan R, Cama C, et al. The use of RT-PCR for prostate-speci®c antigen assay to

[22]

[23]

[24] [25] [26] [27]

[28]

[29]

[30] [31] [32]

[33]

[34]

[35] [36] [37] [38]

[39]

369

predict potential surgical failures before radical prostatectomy: Molecular staging of prostate cancer. Br J Urol 1996;77:411±7. Ellis WJ, Vessella RL, Corey E, Arfman EW, Oswin MM, Melchior S, Lange PH. The value of a reverse transcriptase polymerase chain reaction assay in preoperative staging and followup of patients with prostate cancer. J Urol 1998;159:1134±8. Loric S, Dumas F, Eschwege P, Blanchet P, Benoit G, Jardin A, Lacour B. Enhanced detection of hematogenous circulating prostatic cells in patients with prostate adenocarcinoma by using nested reverse transcription polymerase chain reaction assay based on prostate-speci®c membrane antigen. Clin Chem 1995;41:1698±704. Epstein JI, Walsh PC, Carmichael M, Brendler CB. Pathologic and clinical ®ndings to predict tumor extent of nonpalpable (stage T1c) prostate cancer. JAMA 1994;271:368±74. Pannek J, Rittenhouse HG, Chan DW, Epstein JI, Walsh PC, Partin AW. The use of percent free prostate speci®c antigen for staging clinically localized prostate cancer. J Urol 1998;159:1238±42. Murphy GP, et al. Measurement of serum prostate-speci®c membrane antigen, a new prognostic marker for prostate cancer. Urology 1998;51:89±97. de la Taille A, Rubin MA, Bagiella E, Olsson CA, Buttyan R, Burchardt T, et al. Can perineural invasion on prostate needle biopsy predict prostate speci®c antigen recurrence after radical prostatectomy? J Urol 1999;162:103±6. Egan AJ, Bostwick DG. Prediction of extraprostatic extension of prostate cancer based on needle biopsy ®ndings: Perineural invasion lacks signi®cance on multivariate analysis. Am J Surg Pathol 1997;21:1496±500. de la Taille A, Rubin MA, Buttyan R, Olsson CA, Bagiella E, Burchardt M, et al. Is microvascular invasion on radical prostatectomy specimens a useful predictor of PSA recurrence for prostate cancer patients? Eur Urol 2000;38:79±84. Bahnson RR, Dresner SM, Gooding W, Becich MJ. Incidence and prognostic signi®cance of lymphatic and vascular invasion in radical prostatectomy specimens. Prostate 1989;15:149±55. Brawer MK, Deering RE, Brown M, Preston SD, Bigler SA. Predictors of pathologic stage in prostatic carcinoma. The role of neovascularity. Cancer 1994;73:678±87. Hall MC, Troncoso P, Pollack A, Zhau HY, Zagars GK, Chung LW, von Eschenbach AC. Signi®cance of tumor angiogenesis in clinically localized prostate carcinoma treated with external beam radiotherapy. Urology 1994;44:869±75. Bettencourt MC, Bauer JJ, Sesterhenn IA, Connelly RR, Moul JW. CD34 immunohistochemical assessment of angiogenesis as a prognostic marker for prostate cancer recurrence after radical prostatectomy. J Urol 1998;160:459±65. de la Taille A, Katz AE, Bagiella E, Buttyan R, Sharir S, Olsson CA, et al. Microvessel density as a predictor of PSA recurrence after radical prostatectomy. A comparison of CD34 and CD31. Am J Clin Pathol 2000;113:555±62. Bostwick DG, et al. Optimized microvessel density analysis improves prediction of cancer stage from prostate needle biopsies. Urology 1996;48:47±57. Barrack ER. TGF beta in prostate cancer: A growth inhibitor that can enhance tumorigenicity. Prostate 1997;31:61±70. Visakorpi T, Kallioniemi OP, Koivula T, Harvey J, Isola J. Expression of epidermal growth factor receptor and ERBB2 (HER-2/Neu) oncoprotein in prostatic carcinomas. Mod Pathol 1992;5:643±8. Kim IY, Ahn HJ, Lang S, Oefelein MG, Oyasu R, Kozlowski JM, Lee C. Loss of expression of transforming growth factor-beta receptors is associated with poor prognosis in prostate cancer patients. Clin Cancer Res 1998;4:1625±30. Adler HL, McCurdy MA, Kattan MW, Timme TL, Scardino PT, Thompson TC. Elevated levels of circulating interleukin-6 and transforming growth factor-beta1 in patients with metastatic prostatic carcinoma. J Urol 1999;161:182±7.

370

P.C.M.S. Verhagen et al. / European Urology 41 (2002) 363±371

[40] Ross JS, Nazeer T, Church K, Amato C, Figge H, Rifkin MD, Fisher HA. Contribution of HER-2/neu oncogene expression to tumor grade and DNA content analysis in the prediction of prostatic carcinoma metastasis. Cancer 1993;72:3020±8. [41] Bettencourt MC, Bauer JJ, Sesterhenn IA, Mosto® FK, McLeod DG, Moul JW. Ki-67 expression is a prognostic marker of prostate cancer recurrence after radical prostatectomy. J Urol 1996;156:1064±8. [42] Bubendorf L, Sauter G, Moch H, Schmid HP, Gasser TC, Jordan P, Mihatsch MJ. Ki67 labelling index: An independent predictor of progression in prostate cancer treated by radical prostatectomy. J Pathol 1996;178:437±41. [43] Moul JW, Bettencourt MC, Sesterhenn IA, Mosto® FK, McLeod DG, Srivastava S, Bauer JJ. Protein expression of p53, bcl-2, and KI-67 (MIB-1) as prognostic biomarkers in patients with surgically treated, clinically localized prostate cancer. Surgery 1996;120:159±166;. discussion 166-7. [44] Aaltomaa S, Lipponen P, Vesalainen S, Ala-Opas M, Eskelinen M, Syrjanen K. Value of Ki-67 immunolabelling as a prognostic factor in prostate cancer. Eur Urol 1997;32:410±5. [45] Botticelli AR, Criscuolo M, Martinelli AM, Botticelli L, Filoni A, Migaldi M. Proliferating cell nuclear antigen/cyclin in incidental carcinoma of the prostate. Virchows Arch A Pathol Anat Histopathol 1993;423:365±8. [46] Vesalainen SL, Lipponen PK, Talja MT, Alhava EM, Syrjanen KJ. Proliferating cell nuclear antigen and p53 expression as prognostic factors in T1-2M0 prostatic adenocarcinoma. Int J Cancer 1994;58:303±8. [47] Naito S, Sakamoto N, Kotoh S, Goto K, Koga H, Hasegawa S, et al. Proliferating cell nuclear antigen in needle biopsy specimens of prostatic carcinoma. Eur Urol 1994;26:164±9. [48] Coetzee LJ, Lay®eld LJ, Hars V, Paulson DF. Proliferative index determination in prostatic carcinoma tissue: Is there any additional prognostic value greater than that of Gleason score, ploidy and pathological stage? J Urol 1997;157:214±8. [49] Bauer JJ, Sesterhenn IA, Mosto® FK, McLeod DG, Srivastava S, Moul JW. Elevated levels of apoptosis regulator proteins p53 and bcl2 are independent prognostic biomarkers in surgically treated clinically localized prostate cancer. J Urol 1996;156:1511±6. [50] Huang A, Gandour-Edwards R, Rosenthal SA, Siders DB, Deitch AD, White RW. The p53 and bcl-2 immunohistochemical alterations in prostate cancer treated with radiation therapy. Urology 1998;51:346±51. [51] Diamond DA, Berry SJ, Jewett HJ, Eggleston JC, Coffey DS. A new method to assess metastatic potential of human prostate cancer: relative nuclear roundness. J Urol 1982;128:729±34. [52] Mohler JL, et al. Prediction of prognosis in untreated stage A2 prostatic carcinoma. Cancer 1992;69:511±9. [53] Blom JH, Ten Kate FJ, Schroeder FH, van der Heul RO. Morphometrically estimated variation in nuclear size. A useful tool in grading prostatic cancer. Urol Res 1990;18:93±9. [54] Konig JJ, Teubel W, Kamst E, Romijn JC, Schroder FH, Hagemeijer A. Cytogenetic analysis of 39 prostate carcinomas and evaluation of short-term tissue culture techniques. Cancer Genet Cytogenet 1998;101:116±22. [55] Arps S, Rodewald A, Schmalenberger B, Carl P, Bressel M, Kastendieck H. Cytogenetic survey of 32 cancers of the prostate. Cancer Genet Cytogenet 1993;66:93±9. [56] Visakorpi T, Kallioniemi AH, Syvanen AC, Hyytinen ER, Karhu R, Tammela T, et al. Genetic changes in primary and recurrent prostate cancer by comparative genomic hybridization. Cancer Res 1995;55:342±7. [57] Isaacs WB, Bova GS, Morton RA, Bussemakers MJ, Brooks JD, Ewing CM. Molecular biology of prostate cancer progression. Cancer Surv 1995;23:19±32. [58] Trapman J, Sleddens HF, van der Weiden MM, Dinjens WN, Konig JJ, Schroder FH, et al. Loss of heterozygosity of chromosome 8

[59] [60]

[61] [62] [63]

[64] [65] [66] [67]

[68]

[69] [70] [71] [72] [73]

[74]

[75]

[76] [77]

microsatellite loci implicates a candidate tumor suppressor gene between the loci D8S87 and D8S133 in human prostate cancer. Cancer Res 1994;54:6061±4. Cunningham JM, et al. Allelic imbalance and microsatellite instability in prostatic adenocarcinoma. Cancer Res 1996;56:4475± 82. Oakahashi S, Shan AL, Ritland SR, Delacey KA, Bostwick DG, Lieber MM, et al. Frequent loss of heterozygosity at 7q31.1 in primary prostate cancer is associated with tumor aggressiveness and progression. Cancer Res 1995;55:4114±9. Battifora H. The p53 immunohistochemistry: A word of caution. Hum Pathol 1994;25:435±7. Bookstein R, MacGrogan D, Hilsenbeck SG, Sharkey F, Allred DC. The p53 is mutated in a subset of advanced-stage prostate cancers. Cancer Res 1993;53:3369±73. Navone NM, Troncoso P, Pisters LL, Goodrow TL, Palmer JL, Nichols WW, et al. The p53 protein accumulation and gene mutation in the progression of human prostate carcinoma. J Natl Cancer Inst 1993;85:1657±69. Stricker HJ, Jay JK, Linden MD, Tamboli P, Amin MB. Determining prognosis of clinically localized prostate cancer by immunohistochemical detection of mutant p53. Urology 1996;47:366±9. Gao X, Chen YQ, Wu N, Grignon DJ, Sakr W, Porter AT, Honn KV. Somatic mutations of the WAF1/CIP1 gene in primary prostate cancer. Oncogene 1995;11:1395±8. Bushman EC, Nayak RN, Bushman W. Immunohistochemical staining of ras p21: staining in benign and malignant prostate tissue. J Urol 1995;153:233±7. Sarkar FH, Li Y, Sakr WA, Grignon DJ, Madan SS, Wood Jr DP, Adsay V. Relationship of p21(WAF1) expression with disease-free survival and biochemical recurrence in prostate adenocarcinomas (PCa). Prostate 1999;40:256±60. Baretton GB, Klenk U, Diebold J, Schmeller N, Lohrs U. Proliferation- and apoptosis-associated factors in advanced prostatic carcinomas before and after androgen deprivation therapy: prognostic signi®cance of p21/WAF1/CIP1 expression. Br J Cancer 1999;80:546±55. Aaltomaa S, Lipponen P, Eskelinen M, Ala-Opas M, Kosma VM. Prognostic value and expression of p21(Waf1/Cip1) protein in prostate cancer. Prostate 1999;39:8±15. Cantley LC, Neel BG. New insights into tumor suppression: PTEN suppresses tumor formation by restraining the phosphoinositide 3kinase/AKT pathway. Proc Natl Acad Sci USA 1999;96:4240±5. Cairns P, et al. Frequent inactivation of PTEN/MMAC1 in primary prostate cancer. Cancer Res 1997;57:4997±5000. Vlietstra RJ, van Alewijk DC, Hermans KG, van Steenbrugge GJ, Trapman J. Frequent inactivation of PTEN in prostate cancer cell lines and xenografts. Cancer Res 1998;58:2720±3. McMenamin ME, Soung P, Perera S, Kaplan I, Loda M, Sellers WR. Loss of PTEN expression in paraf®n-embedded primary prostate cancer correlates with high Gleason score and advanced stage. Cancer Res 1999;59:4291±6. Lloyd RV, Erickson LA, Jin L, Kulig E, Qian X, Cheville JC, Scheithauer BW. p27Kip1: A multifunctional cyclin-dependent kinase inhibitor with prognostic signi®cance in human cancers. Am J Pathol 1999;154:313±23. Cote RJ, Shi Y, Groshen S, Feng AC, Cordon-Cardo C, Skinner D, Lieskovosky G. Association of p27Kip1 levels with recurrence and survival in patients with stage C prostate carcinoma. J Natl Cancer Inst 1998;90:916±20. Yang RM, Naitoh J, Murphy M, Wang HJ, Phillipson J, deKernion JB, et al. Low p27 expression predicts poor disease-free survival in patients with prostate cancer. J Urol 1998;159:941±5. Fernandez PL, Arce Y, Farr X, Martnez A, Nadal A, Rey MJ, et al. Expression of p27/Kip1 is down-regulated in human prostate carcinoma progression. J Pathol 1999;187:563±6.

P.C.M.S. Verhagen et al. / European Urology 41 (2002) 363±371 [78] Vis AN, Noordzij MA, Fitoz K, Wildhagen MF, Schroder FH, van der Kwast TH. Prognostic value of cell cycle proteins p27Kip1 and MIB-1, and the cell adhesion protein CD44s in surgically treated patients with prostate cancer. J Urol 2000;164:2156±61. [79] Thomas GV, Schrage MI, Rosenfelt L, Kim JH, Salur G, deKernion JB, et al. Preoperative prostate needle biopsy p27 correlates with subsequent radical prostatetcomy p27, gleason grade and pathological stage. J Urol 2000;164:1987±91. [80] Bookstein R, Rio P, Madreperla SA, Hong F, Allred C, Grizzle WE, Lee WH. Promoter deletion and loss of retinoblastoma gene expression in human prostate carcinoma. Proc Natl Acad Sci USA 1990;87:7762±6. [81] Vesalainen S, Lipponen P. Expression of retinoblastoma gene (Rb) protein in T12M0 prostatic adenocarcinoma. J Cancer Res Clin Oncol 1995;121:429±33. [82] Prochownik EV, Eagle Grove L, Deubler D, Zhu XL, Stephenson RA, Rohr LR, et al. Commonly occurring loss and mutation of the MXI1 gene in prostate cancer. Genes Chromosomes Cancer 1998;22:295±304. [83] Umbas R, Isaacs WB, Bringuier PP, Schaafsma HE, Karthaus HF, Oosterhof GO, et al. Decreased E-cadherin expression is associated with poor prognosis in patients with prostate cancer. Cancer Res 1994;54:3929±33. [84] Cheng L, Nagabhushan M, Pretlow TP, Amini SB, Pretlow TG. Expression of E-cadherin in primary and metastatic prostate cancer. Am J Pathol 1996;148:1375±80. [85] Graff JR, et al. E-cadherin expression is silenced by DNA hypermethylation in human breast and prostate carcinomas. Cancer Res 1995;55:5195±9. [86] Morita N, Uemura H, Tsumatani K, Cho M, Hirao Y, Okajima E, et al. E-cadherin and alpha-, beta- and gamma-catenin expression in prostate cancers: Correlation with tumour invasion. Br J Cancer 1999;79:1879±83. [87] Richmond PJ, Karayiannakis AJ, Nagafuchi A, Kaisary AV, Pignatelli M. Aberrant E-cadherin and alpha-catenin expression in prostate cancer: correlation with patient survival. Cancer Res 1997; 57:3189±93. [88] Noordzij MA, van Steenbrugge GJ, Verkaik NS, Schroder FH, van der Kwast TH. The prognostic value of CD44 isoforms in prostate

[89] [90]

[91]

[92] [93] [94]

[95]

[96] [97]

[98] [99]

371

cancer patients treated by radical prostatectomy. Clin Cancer Res 1997;3:805±15. Aaltomaa S, Lipponen P, Ala-Opas M, Kosma VM. Expression and prognostic value of CD44 standard and variant v3 and v6 isoforms in prostate cancer. Eur Urol 2001;39:138±44. Dong JT, Lamb PW, Rinker-Schaeffer CW, Vukanovic J, Ichikawa T, Isaacs JT, Barrett JC. KAI1, a metastasis suppressor gene for prostate cancer on human chromosome 11p11.2. Science 1995;268: 884±6. Mashimo T, Watabe M, Hirota S, Hosobe S, Miura K, Tegtmeyer PJ, Watabe K. The expression of the KAI1 gene, a tumor metastasis suppressor, is directly activated by p53. Proc Natl Acad Sci USA 1998;95:11307±11. Ueda T, et al. Expression of the KAI1 protein in benign prostatic hyperplasia and prostate cancer. Am J Pathol 1996;149:1435±40. Konishi N, et al. Intratumor cellular heterogeneity and alterations in ras oncogene and p53 tumor suppressor gene in human prostate carcinoma. Am J Pathol 1995;147:1112±22. Sato K, Qian J, Slezak JM, Lieber MM, Bostwick DG, Bergstralh EJ, Jenkins RB. Clinical signi®cance of alterations of chromosome 8 in high-grade, advanced, nonmetastatic prostate carcinoma. J Natl Cancer Inst 1999;91:1574±80. Sakai H, Yogi Y, Minami Y, Yushita Y, Kanetake H, Saito Y. Prostate speci®c antigen and prostatic acid phosphatase immunoreactivity as prognostic indicators of advanced prostatic carcinoma. J Urol 1993;149:1020±3. Berner A, Harvei S, Tretli S, Fossa SD, Nesland JM. Prostatic carcinoma: A multivariate analysis of prognostic factors. Br J Cancer 1994;69:924±30. Takeda H, Akakura K, Masai M, Akimoto S, Yatani R, Shimazaki J. Androgen receptor content of prostate carcinoma cells estimated by immunohistochemistry is related to prognosis of patients with stage D2 prostate carcinoma. Cancer 1996;77:934±40. Sommerfeld HJ, Meeker AK, Piatyszek MA, Bova GS, Shay JW, Coffey DS. Telomerase activity: A prevalent marker of malignant human prostate tissue. Cancer Res 1996;56:218±22. Carlisle AJ, et al. Development of a prostate cDNA micro-array and statistical gene expression analysis package. Mol Carcinog 2000;28: 12±22.