Biologic markers in prostatic intraepithelial neoplasia: immunohistochemical and cytogenetic analyses

Biologic markers in prostatic intraepithelial neoplasia: immunohistochemical and cytogenetic analyses

Masahito Tsuji^a, Kazuya Kanda^a, Yoshihide Murakami^a, Yasushi Kurokawa^a,
Hiro-omi Kanayama^a, Toshiaki Sano^b, and Susumu Kagawa^a

^aDepartment of Urology, and ^bFirst Department of Pathology, The University of Tokushima School of Medicine, Tokushima, Japan

Abstract:Objective:We evaluated the biological properties of High-grade prostatic intraepithelial neoplasia (PIN) by immunohistochemistry and fluorescence in situ hybridization (FISH) analysis in relation to normal tissue and carcinoma lesions.
Materials and Methods:Immunohistochemical staining and FISH were performed on23formalin-fixed radical prostatectomy specimens taken from patients with PIN. Assays were performed using MIB-1, chromogranin A (CGA) and an anti-androgen receptor antibody (AR). A centromere probe for chromosome8was used to test for aneuploidy.
Results:The MIB-1index of cancerous specimens (16.2±10.5%) was significantly higher than that of benign (1.9±1.6%, p<0.0001) or PIN (4.0±4.5%, p<0.0001) specimens. The percentage of CGA positive cells was significantly lower in normal tissue (1.2±1.8%) than in PIN (3.5±2.9%, p=0.012) or carcinoma (5.4±4.9%, p=0.005) lesions. Positive staining for AR was consistently observed in the nuclei of both benign and malignant epithelial cells, but positive cytoplasmic staining was also seen in PIN epithelial cells. No significant difference in FISH detected anomalies were found between PIN and carcinoma specimens.
Conclusions:Our studies concerning proliferative activity, NE differentiation and chromosomal anomalies of prostatic specimens support the hypothesis that PIN is a biologically intermediate stage in the pathogenesis of prostatic carcinoma. The cellular distribution of AR was altered in PIN cells, but the role of AR in PIN is not yet clear. J. Med. Invest. 46:35-41, 1999

Keywords:prostatic intraepithelial neoplasia, immunohistochemistry, fluorescence in situ hybridization

INTRODUCTION
Prostatic intraepithelial neoplasia (PIN) is thought to be a precursor of prostatic carcinoma due to their similar histopathology. Indeed, the presence of PIN in a diagnostic needle biopsy, without any accompany-ing malignancy, is highly predictive of the presence of carcinoma in subsequent biopsies [1, 2]. However, although some reports have appeared on immuno-histochemical or cytogenetic analyses of PIN, its biological properties and clinical significance remain unclear [3, 4].
Ki-67 is a nuclear protein expressed during the G1, S, G2, and M phases of continuously cycling cells, but not G0 cells. The genetic locus of Ki-67is not well characterized, and it has been assigned to chromosome10 [5]. Several studies have shown that cell proliferative activity, as defined by the Ki-67 index, correlates with cell growth fraction. MIB-1antibody, which detects a formalin-resistant epitope of the Ki-67 antigen, has been a useful index in several studies [6]. Hepburn et al. [7] compared the reactivity of MIB-1antibody in paraffin-embedded sections with that of Ki-67 antibody on frozen sec-tions of prostatic carcinoma, and they reported a significant correlation between the two. Differen-tiation of prostatic neuroendocrine (NE) cells, which are distributed throughout the prostate gland as well as various ducts and the urethra [8], has been found to be a good prognostic indicator for cases of prostatic adenocarcinoma [9, 10]. However, the relationship between PIN and NE markers has not been fully elucidated. In addition, the androgen receptor (AR) is widely distributed in tissue and several mutations have been detected in prostate carcinoma [11, 12] suggesting that AR may also play an important role in prostatic carcinogenesis [13, 14].
Recently, fluorescence in situ hybridization (FISH) has been performed on interphase chromosomes in a variety of solid tumors in order to detect genetic aberrations [15, 16]. Brown et al. [17] evaluated numeric chromosomal anomalies in localized pros-tate carcinomas using FISH with centromere probes for chromosomes 4, 6-12, 17, 18, X and Y. They reported that chromosome 8 polysomy was the most frequent aneuploidy in prostate carcinoma. Konig et al. [18] also investigated numerical aber-rations of chromosomes 1, 7, 8, 10, 18 and Y in both carcinomas and benign hyperplasias. Their results suggested that polysomies of chromosomes 7, 8and10were seen in cancerous tissue, although no evidence of aneuploidy was seen in normal tissue.
The present study attempted to determine the expression of these biomarkers in PIN taken from prostatectomy specimens of untreated patients, and to evaluate the biological relationship between normal epithelium, PIN and carcinoma tissue.

MATERIALS AND METHODS
Tissue samples
Twenty-three patients with previously untreated-localized prostate cancer who underwent radical prostatectomy were studied. All tissue specimens selected for this study contained both carcinoma and PIN. Formalin-fixed, paraffin-embedded pros-tate tissue was obtained from the Department of Urology, The University of Tokushima School of Medicine and its affiliated hospitals. Patients partici-pating in this study ranged in age from 60 to 79 years (mean age 68.3). Tumors were staged according to the UICC classification, and tumor grade was deter-mined based on the Gleason sum score. High-grade PIN samples were further classified by hematoxylin-eosin staining as described by McNeal and Bostwick [19]. Additionally, for cases in which a diagnosis of PIN was not clearly evident, immunostaining was per-formed for high-molecular-weight cytokeratin using 34βE12 (Enzo Biochemicals, USA, dilution1:50).This antibody acts as a specific marker for basal cells. In the present study, the term PIN is used to indicate high-grade PIN. All slides were reviewed and a representative paraffin-embedded tissue block containing PIN, prostatic adenocarcinoma, and non-cancerous normal prostatic foci, was selected from each case for immunohistochemical exami-nation and FISH analysis. Atypical adenomatous hyperplasia (AAH) was not observed in any of these cases.

Immunohistochemical staining
Immunohistochemical analysis was performed with only slight modifications of previously described methods [20]. Paraffin-embedded 5-μm sections were prepared and deparaffinized in xylene and rehydrated in graded alcohols. Endogenous peroxidase activity was blocked by immersing sections in 0.3% hydrogen peroxide (H2O2) in methanol for 30 minutes. Sections were placed in an autoclave for 10minutes at 120°Cand were then allowed to cool at room temperature for 40 min before immunostaining. Pretreatment with normal rabbit serum for 30min at room tem-perature was followed by incubation overnight at4°C with the primary antibodies in a humidified chamber. Subsequently, biotinylated anti-mouse antibody was applied to the sections for 30min. Streptavidin-biotinylated peroxidase complex (LSAB kit;DAKO, California, USA) was then applied, and the sections were incubated with diaminobenzidine (DAB;Wako, Osaka, Japan) as chromogen. Follow-ing immunohistochemical staining, the sections were counterstained with hematoxylin to enhance nuclear detection. For negative controls, the primary antibody was substituted with phosphate-buffered saline in duplicate sections. The primary antibodies were as follows:anti-MIB-1 (Immunotech S.A., France, dilution1:50), anti-ChromograninA (CGA) (DAKO, Denmark, dilution 1 : 50), and anti-androgen receptor (AR) (Novocastra, UK, dilution1:50).

Labeled cell counts
MIB-1, CGA and AR expression were evaluated under light microscopic observation at X400 mag-nification. Immunostaining results were interpreted according to the positivity of the immunoreactive product seen in epithelial cells. The percentage of positive cells for each marker was determined by counting at least 800tumor cells in four selected fields that displayed the highest immunoreactivity.

Fluorescence in situ hybridization
FISH analysis was performed using a slight modi-fication of the procedure described previously [21]. Serial 5-μm sections were prepared from each speci-men for FISH analysis. The slides were deparaffinized, dehydrated, and treated in a 30%concentration of a pretreatment solution (Oncor, Inc. Gaithersburg, USA) at45°Cfor15minutes. The slides were then washed three times in 2x standard saline citrate (SSC), followed by digestion in pepsin solution (4mg/ml in 0.9% sodium chloride, PH1.5) at 45°Cfor 18min. This was again followed by three washes in 2x SSC. After dehydration in ethanol, air drying, and denaturing of DNA probes and target DNA in an 82°C oven for 5minutes, 10μl of a chromosome 8 centromeric probe (Oncor, inc. Gaithersburg, USA) was applied directly to each treated slide. The slides were coverslipped, sealed with rubber cement, and incubated overnight in a 37°C humidified chamber. Coverslips were then removed and the slides were washed in 2X SSC at 72°C for 5minutes. Hybridization was detected using fluorescein-isothiocyanate-conjugated (FITC) avidin, and nuclei were counter-stained with propidium iodide.

Criteria for FISH anomalies
The following criteria were applied for interpre-tation of the results [22]. 1) Overlapping nuclei were not counted. 2) Multiple stains within one nucleus have more or less the same size and intensity. Non-specific, smaller stains were not counted. 3)Paired spots were counted as one stain. Stains were enumerated in at least 200 epithelial nuclei from histologically benign regions of each of the 23pros-tates. The mean ± SD percentage of nuclei with three or more stains was found to be 4.1±3.3% (ranging from 1.0 to 15.0%), and a cut-off value of 17.3% (mean+4SD) was used to define FISH detected anomalies [23, 24]. If the percentage of cells with3or more stains exceeded this number, the foci of PIN or carcinoma were considered to have nuclei with polysomies for chromosome8.

Statistical analysis
A Student's t test and Chi-squared test were used for statistical analysis of the results. P values below0.05 were considered statistically significant.

RESULTS
Of the 23 radical prostatectomy specimens, 9 were stage pT2 and 14were stage pT3. The Gleason sum score of the 23 tumors varied from 2 to 9, with14tumors classified as low grade (Gleason sum score 2 to 6) and 9as high grade (Gleason sum score 7to 9).
MIB-1 positive nuclei were randomly distributed in individual glands comprising normal epithelia, PIN, and carcinoma tissue (Fig.1). The MIB-1 index increased from a mean of 1.9±1.6% (mean±SD) in benign to 4.0±4.5% in PIN to 16.2±10.5% in carcinoma tissue. The MIB-1 index of carcinoma tissue was significantly larger than that of benign or PIN tissue (p<0.0001). In addition, there was a significant difference in the mean MIB-1 indexes between high grade (21.2±11.3) and low grade(12.6±8.3) carcinomas (p<0.05). Although there was no significant correlation, PIN tended to have a higher MIB-1 index than benign epithelium (p=0.052) (Table1).
Immunohistochemical staining for CGA was observed in the cytoplasm of NE cells in normal epithelium, PIN, and carcinoma tissue. NE cells were relatively large and located in the basal layer (Fig.2). The percentage of NE cells increased from a mean of 1.2±1.8% in benign to 3.5±2.9% in PIN to 5.4±4.9% in carcinoma tissue. The percentage of NE cells was significantly lower in benign glands than in PIN (p=0.012) and carcinoma tissue (p=0.005, Table1). However, there was no significant differ-ence in the percentage of NE cells between PIN and carcinoma tissue or between high grade and low grade carcinomas.
AR-positive staining was observed in the nuclei of benign epithelium and stromal cells. However, the basal cell layer did not express AR. In malignant tissue, AR immunoreactivity was observed in the nuclei of carcinoma cells. On the other hand, while AR expression was present in PIN, the nuclear staining intensity was weak in general, and cytoplasmic staining was occasionally observed (Fig.3). The percentage of AR positive cells did not differ signif-icantly between normal epithelia, PIN, and carcinoma tissue (65.9±14.5%, 54.2±15.7% and 67.3±17.0%,respectively) (Table1).
The FISH results for the chromosome 8 centro-metric probe, observed in the foci of normal epithelia, PIN, and carcinoma tissue, are summarized in Table 2. The mean percentage of nuclei with three or more stains4.1%, 21.3% and 31.5% for normal epithelia, PIN, and carcinoma tissue, respectively. As a result, chromosome8polysomies were found in64.3% (9of 14) of PIN foci and61.1% (11of18) of carcinoma foci. No significant difference in FISH-detected anom-alies were found between PIN and carcinoma foci (Table2).

DISCUSSION
The clinical importance of recognizing PIN stems from its strong association with carcinoma, which most studies to date have confirmed [1- 4]. The fre-quency and extent of PIN are significantly increased in cancerous prostate specimens compared with non-cancerous prostates [1, 2]. Approximately 57% of patients show carcinoma on repeat biopsy following the appearance of high-grade PIN on a previous biopsy [25]. Therefore, when PIN is encountered in prostate specimens, all tissue should be embedded and evaluated.
In PIN foci, the majority of nuclei are enlarged and prominent nucleoli are frequently observed. The morphologic continuum of cellular proliferations with cytological changes mimics cancer. Bostwick et al. [26] identified 4 common patterns of PIN:tufting, micropapillary, cribriform and flat. However, there is no prognostic difference between these morphological patterns of PIN, and this distinction appears to have only diagnostic utility. On the other hand, atypical adenomatous hyperplasia (AAH) is a localized proliferation of small glands, which may be mistaken for adenocarcinoma. Generally, adenocarcinoma glands do not have basal cells, but AAH glands are surrounded by a fragmented layer of basal cells, as with PIN. Presently, the evidence that AAH is a cancerous precursor is unconvinc-ing [27]. Therefore, PIN and AAH show some of the morphologic features of carcinoma, although they lack stromal invasion. 34βE12 is specific for high-molecular-weight cytokeratins 1, 5, 10 and14, which correspond to molecular weights of 68, 58, 56.5 and 50kd, characteristically found in complex epithelium. This antibody serves as a specific marker for basal cells in the prostate gland, and may have a potential diagnostic use in distinguishing between PIN or AAH and malignant glands of the prostate[26, 28, 29].
Tamboli et al. [30] compared the nuclear prolif-erative activities of benign prostatic tissue, PIN, and prostate cancer in non-hormone treated patients using the MIB-1 antibody. They reported that the MIB-1 index consistently increased in benign, PIN, and malignant epithelium. In the present study, the MIB-1index also increased from normal epithelium to PIN to neoplastic prostatic tissue. Additionary, the MIB-1 index also significantly increased with increasing tumor grade. However, the proliferative activity of PIN was only slightly higher than that of normal epithelium compared with carcinoma. This finding suggests that the aggressive potential of PIN is not as high as that of carcinoma.
There is now increasing evidence that NE differ-entiation occurs frequently in prostatic carcinoma. This may have prognostic and therapeutic impli-cations [8 -10, 31]. CGA is an acidic, water-soluble protein present in the secretory vesicles of NE cells [32]. Weinstein et al. [9] reported that NE differen-tiation, as measured by the presence of CGA, is very common in prostatic carcinoma, and they suggested that NE differentiation in carcinoma correlates with poor prognosis. The regulatory role played by the NE cell in the prostate is unknown, although it is highly likely that these cells regulate both growth and differentiation [33]. In the present study, CGA positive cells were more frequent in PIN and carcino-ma than in normal epithelium. This finding suggests that PIN shares some biological characteristics with carcinoma.
Several studies have shown either the existence of point mutations in the AR gene or AR gene ampli-fication occurring during cancer progression [11, 12, 34]. However, immunohistochemical staining levels of AR protein in normal epithelium, PIN and carcinoma tissue are still controversial [13, 35]. In this study, immunohistochemistry showed the pres-ence of nuclear AR in both prostatic stromal cells and glandular epithelium. Similarly, prostatic adeno-carcinoma tissue was comprised of a majority of AR positive cells. On the other hand, there was weak nuclear staining and occasionally cytoplasmic stain-ing observed in PIN tissue. Our data bears a striking resemblance to a report by Magi-Galluzzi et al. [36], who found both uniformly decreased nuclear staining and cytoplasmic localization of AR in high-grade PIN. From these studies, we speculate that the distribution of AR protein has been altered in PIN tissue.
FISH has been recognized as a reliable method of detecting aneuploidy. Persons et al. [37] evaluated the FISH ploidy patterns of prostate tumors using chromosome 8 and 12 centromere probes, and compared their results to flow cytometry (FCM) ploidy patterns. They reported that FISH was more sensitive than FCM for detection of aneuploidy. Qian et al. [24] studied numerical chromosomal anomalies and gene copy number anomalies in PIN and carcinoma tissue using DNA probes for chr-omosomes 7, 8, 10, 12, and Y, and for chromosome region 8q24 (c-myc). Their results revealed that aneuploidy, especially of chromosome8, as well as extra copies of c-myc gene were identified both in PIN and in carcinoma tissue. Our study also demon-strated that polysomy of chromosome 8 was detected in the majority of PIN and carcinoma foci. From these findings, it would seem that PIN and prostate carcinoma show similar aneuploid characteristics.
In conclusion, the present study concerning prolif-erative activity, NE differentiation and chromosomal anomaly of prostate specimens supports the hypoth-esis that PIN is a biologically intermediate stage in the pathogenesis of prostatic carcinoma. However, since the distribution of AR protein was different in PIN from that in normal epithelium or carcinoma, more studies are needed to elucidate the role of AR in PIN.

REFERENCES
1. Davidson D, Bostwick DG, Qian J, Wollan PC, Oesterling, Rudders RA, Siroky M, Stilmant M:Prostatic intraepithelial neoplasia is a risk factor for adenocarcinoma:Predictive accuracy in needle biopsies. J Urol154:1295-9, 1995
2. Weinstein MH, Epstein JI:Significance of high grade prostatic intraepithelial neoplasia on needle biopsy. Hum Pathol 24:624-9, 1993
3. Bostwick DG:High grade prostatic intraepi-thelial neoplasia:The most likely precursor of prostate cancer. Cancer75:1823-36, 1995
4. Bostwick DG:Progression of prostatic intra-epithelial neoplasia to early invasive adenocar-cinoma. Eur Urol 30:145-52, 1996
5. Gerdes J, Li L, Schlueter C, Duchrow M, Wohlenberg C, Gerlach C, Stahmer I, Kloth S, Brandt E, Flad HD:Immunobiochemical and molecular biologic characterization of the cell proliferation-associated nuclear antigen that is defined by monoclonal antibody Ki-67.Am J Pathol 138:867-73, 1991
6. Cattoretti G, Becker MHG, Key G, Duchrow M, Schluter C, Galle J, Gerdes J:Monoclonal anti-bodies against recombinant parts of the Ki-67 antigen (MIB1 and MIB3) detect proliferating cells in microwave-processed formalin-fixed paraffin sections. J Pathol 168:357-63, 1992
7. Hepburn PJ, Glynne JE, Goddard L, Gee JM, Harper ME:Cell proliferation in prostatic car-cinoma:comparative analysis of Ki-67, MIB-1and PCNA. Histochem J27 (3):196 -203,1995
8. Di Sant'Agnese P:Neuroendocrine differentiation in prostatic carcinoma:Recent findings and new concepts. Cancer75:1850-9, 1995
9. Weinstein MH, Partin AW, Veltri RW, Epstein JI:Neuroendocrine differentiation in prostate cancer:enhanced prediction of progression after radical prostatectomy. Hum Pathol 27:683-7, 1996
10. Speights VO, Cohen MK, Riggs MW, Coffield KS, Keegan G, Arber DA : Neuroendocrine stains and proliferative indices of prostatic adeno-carcinomas in transurethral resection samples. Br J Urol80:281-6, 1997
11. Gaddipati JP, McLeod DG, Heidenberg HB, Sesterhenn IA, Finger MJ, Moul JW, Srivastava S:Frequent detection of codon 877 mutation in the androgen receptor gene in advanced prostate cancer. Cancer Res 54:2861-4, 1994
12. Culig Z, Hobisch A, Cronauer MV, Cato AC, Hittmair A, Radmayr C, Eberle J, Bartsch G, Klocker H:Mutant androgen receptor detected in an advanced-stage prostatic carcinoma is activated by adrenal androgens and progester-one. Mol Endocrinol7:1541-50, 1993
13. Miyamoto KK, McSherry SA, Dent GA, Sar M, Wilson EM, French FS, Sharief Y, Mohler JL:Immunohistochemistry of the androgen receptor in human benign and malignant pros-tate tissue. J Urol149:1015-9, 1993
14. Hobisch A, Culig Z, Radmayr C, Bartsch G, Klocker H, Hittmair A:Distant metastases from prostatic carcinoma expression androgen receptor protein. Cancer Res 55:3068-72, 1995
15. Beck JMN, Hopman AHN, Feitz WFJ, Schalken J, Schaafsma HE, Van-de-Kaa CA, Remaekers FC, Hanselaar AG, De Wilde PC:Numerical aberrations of chromosomes 1 and 7 in renal cell carcinomas as detected by interphase cyto-genetics. J Pathol176:123-35,1995
16. Herbergs J, Hopman AH, De Bruine AP, Ramaekers FC, Arends JW:In situ hybridi-zation and flow cytometric analysis of colorectal tumours suggests two routes of tumourigenesis characterized by gain of chromosome 7or loss of chromosomes 17 and18. J Pathol 179:243-7, 1996
17. Brown JA, Alcaraz A, Takahashi S, Persons DL, Lieber MM, Jenkins RB:Chromosomal aneusomies detected by fluorescent in situ hybridization analysis in clinically localized prostate carcinoma. J Urol152:1157-62, 1994
18. Konig JJ, Teubel W, Romijn JC, Schroder FH, Hagemeijer A:Gain and loss of chromosomes1, 7, 8, 10, 18, and Y in 46 prostate cancers. Hum Pathol 27:720-7, 1996
19. McNeal JE, Bostwick DG : Intraductal dysplasia : A premalignant lesion of the prostate. Hum Pathol17:64-71, 1986
20. Gerdes J, Becker MH, Key G, Cattoretti G: Immunohistochemical detection of tumor growth fraction (Ki-67 antigen) in formalin fixed and routinely processed tissue (Letter to the Editor). J Pathol168:85-6, 1992
21. Wu SQ, Hafez GR, Xing W, Newton M, Chen XR, Messing E:The correlation between the loss of chromosome 14q with histologic tumor grade, pathologic stage, and outcome of patients with nonpapillary renal cell carcinoma. Cancer 77:1154-60, 1996
22. Hopman AH, van Hooren E, van de Kaa CA, Vooijs PG, Ramaekers FC:Detection of nu-merical chromosome aberrations using in situ hybridization in paraffin sections of routinely processed bladder cancers. Mod Pathol 4:503-13, 1991
23. Ichikawa D, Hashimoto N, Hoshima M, Yamaguchi T, Sawai K, Nakamura Y, Takahashi T, Abe T, Inazawa J:Analysis of numerical aberrations in specific chromosomes by fluorescent in situ hybridization as a diagnostic tool in breast cancer. Cancer77:2064-9, 1996
24. Qian J, Jenkins RB, Bostwick DG:Detection of chromosomal anomalies and c-myc gene amplification in the cribriform pattern of pros-tatic intraepithelial neoplasia and carcinoma by fluorescence in situ hybridization. Mod Pathol10 (11):1113-9, 1997
25. Brawer MK, Bigler SA, Sohlberg OE, Nagle RB, Lange PH:Significance of prostatic intra-epithelial neoplasia on prostate needle biopsy. Urology38:103-7, 1991
26. Bostwick DG, Amin MB, Dundore P, Dundore P, Marsh W, Schultz DS : Architectural patterns of high-grade prostatic intraepithelial neoplasia. Hum Pathol 24:298-310, 1993
27. Qian J, Jenkins RB, Bostwick DG : Chromosomal anomalies in atypical adenomatous hyperplasia and carcinoma of the prostate using fluorescence in situ hybridization. Urology46:837- 42, 1995
28. Wojno KJ, Epstein JI:The utility of basal cell-specific anti-cytokeratin antibody (34βE12) in the diagnosis of prostate cancer:A review of 228 cases. Am J Surg Pathol19 (3):251-260, 1995
29. Bostwick DG, Srigley J, Grignon D, Maksem J, Humphrey P, van der Kwast TH, Bose D, Harrison J, Young RH:Atypical adenomatous hyperplasia of the prostate:Morphologic cri-teria for its distinction from well-differentiated carcinoma. Hum Pathol24:819-32, 1993
30. Tamboli P, Amin MB, Schultz DS, Linden MD, Kubus J. Comparative analysis of the nuclear proliferative index (Ki-67) in benign prostate, prostatic intraepithelial neoplasia, and prostatic carcinoma. Mod Pathol 9:1015-9, 1996
31. Cohen MK, Arber DA, Coffield KS, Keegan GT, McClintock J, Speights VO Jr : Neuroendocrine differentiation in prostatic adenocarcinoma and its relationship to tumor progression. Cancer74:1899-1903, 1994
32. Wilson BS, Lloid RV : Detection of chromogranin in neuroendocrine cells with a monoclonal antibody. Am J Pathol 115:458-68, 1984
33. Bostwick DG, Pacelli A, Lopez-Beltran : Molecular biology of prostatic intraepithelial neoplasia. The Prostate29:117-134, 1996
34. Visakorpi T, Hyytinen E, Koivisto P, Tanner M, Keinanen R, Palmberg C, Palotie A, Isola J, Kallioniemi OP:In vivo amplification of the androgen receptor gene and progression of human prostate cancer. Nat Genet 9 : 401- 6, 1995
35. Leav I, Mcneal JE, Kwan PWL, Komminoth P, Merk FB:Androgen receptor expression in prostatic dysplasia (prostatic intraepithelial neo-plasia) in the human prostate:An immuno-histochemical and in situ hybridization study. Prostate29:137- 45, 1996
36. Magi-Galluzzi C, Xu X, Hlatky L, Hahnfeldt P, Kaplan I, Hsiao P, Chang C, Loda M:Hetero-geneity of androgen receptor content in advanced prostate cancer. Mod Pathol 10 (8):839-45, 1997
37. Persons DL, Gibney DJ, Katzmann JA, Lieber MM, Farrow GM, Jenkins RB : Use of fluorescent in situ hybridization for deoxyribonucleic acid ploidy analysis of prostatic adenocarcinoma. J Urol150:120 -5, 1993

Received for publication October 9, 1998;accepted November 24, 1998.

Address correspondence and reprint requests to Masahito Tsuji, M.D., Department of Urology, The University of Tokushima School of Medicine, Kuramoto-cho, Tokushima770-8503, Japan and Fax:+81-88-633-7160.