Pathological and clinical characteristics of large prostate cancers predominantly located in the transition zone

Transcription

1 Pathological and clinical characteristics of large prostate cancers predominantly located in the transition zone (2002) 5, ß 2002 Nature Publishing Group All rights reserved /02 $ A Erbersdobler 1 *, S Huhle 1, J Palisaar 2, M Graefen 2, P Hammerer 2, J Noldus 2 & H Huland 2 1 Institute of Pathology, University of Hamburg, Germany; and 2 Department of Urology, University of Hamburg, Germany Prostate carcinomas located in the transition zone are suspected to behave differently from the more frequent peripheral zone cancers. In this study, large transition zone prostate cancers were investigated for pathological and clinical features. From 365 consecutive radical prostatectomy specimens, 73 cases were disclosed with tumours larger than 10 cm 3. Of these, 14 were predominantly (>70% tumour area) located in the transition zone. Pathological investigations included a complete histological work-up, immunohistochemistry for p53 and bcl-2, and interphase cytogenetics for chromosomes 7, 8, 17, and X. Despite large tumour volumes and high preoperative prostate specific antigen (PSA)-values, most tumours showed quite favourable pathological features. Only two of these patients suffered from a postoperative PSA-recurrence during a median followup of 50 months. For comparison, 36 cases that contained tumours predominantly located in the peripheral zone mostly displayed adverse prognostic signs and 68.8% of these patients suffered from postoperative PSA-recurrence. We conclude that the peculiar pathological and clinical characteristics of large prostate cancers in the transition zone might be important for prognostic considerations. (2002) 5, doi: /sj.pcan Keywords: prostatic neoplasm; tumour volume; zonal location; interphase cytogenetics Introduction Tumour size is regarded as an important factor in the development and progression of clinically manifest prostate cancer. 1,2 However, some authors do not believe tumour volume to be an independent prognostic parameter, because it is related to Gleason grade, surgical margins, and capsular penetration and does not provide additional information over these established prognostic factors. 3 In 1993, Stamey and co-workers reported three cases of prostate cancers with large tumour volumes and excessive prostate specific antigen (PSA)-values, which nevertheless were found to be organ-confined after pathological investigation of the prostatectomy specimens. These tumours were predominantly located in the transition zone and the three patients remained free of tumour recurrence up to 650 days following prostatectomy. 4 A higher cure rate of transition zone cancers as compared to cancers in the peripheral zone of the prostate could also been shown in larger studies. 5,6 The predictive value of cancer volume might be modified by the zonal location of the tumour and large cancers in the transition zone could be a challenge for any prognostic algorithm that uses established preoperative factors. The aim of the present study was to investigate some clinical and pathological parameters in a group of 14 prostate carcinomas larger than 10 cm 3 that were predominantly located in the transition zone. *Correspondence: Dr A Erbersdobler, Institute of Pathology, University Hospital Hamburg-Eppendorf, Martinistr. 52, D Hamburg, Germany. erbersdobler@uke.uni-hamburg.de Received 12 February 2002; revised 17 April 2002; accepted 25 April 2002 Materials and methods Patients and tissues The pathologic reports of 365 radical prostatectomies, performed between 1994 and 1997 at the Department of

2 280 Urology, University Hospital Hamburg-Eppendorf, were reviewed. All prostatectomy specimens had been completely processed in 3 5 mm steps according to the Stanford protocol. 7 The outlines of tumour foci had been marked on whole mount histological slides, allowing for the determination of zonal location and the calculation of tumour volume with computer-assisted planimetry. 1,7 From the 365 reports, cases were selected that contained prostate cancers with total tumour volumes larger than 10 cm 3. The presumed zonal origin was determined according to the criteria defined by McNeal et al. 7 Briefly, a tumour was assigned to a specific zone if at least 70% of its volume was estimated to be located in that zone. In this way, 14 cases were identified that contained tumours predominantly located in the transition zone. No patient had received preoperative anti-androgen therapy. Intraoperative frozen and permanent histological sections of regional lymph nodes had not disclosed any metastases. All histological slides of these cases were carefully reviewed and the relationship of the tumours to the boundary of the transition zone, a stromal band extending from posterior to the urethra at the midline to the anterior border of the prostate, was assessed. Foci of prostate cancer were considered separate when divided by at least 3-mm non-neoplastic tissue and when no continuity was observed in sequential slides. 7 The TNM-classification (1997), Gleason grades, and margin status were determined. Clinical data were obtained from the referring urologists and included preoperative PSAvalues and the results of digital rectal examination and transrectal ultrasound. Postoperative follow-up examinations were routinely performed after 2, 6, 12, 18, and 24 months and yearly thereafter. A postoperative PSA value was considered positive if it was above 0.1 ng/ml in two consecutive samples. The follow-up period was defined as the time between the operation and the latest negative PSA-value, or until the first follow-up visit that disclosed PSA-recurrence. Postoperative antiandrogen or radiation therapy was usually not given until tumour recurrence was confirmed. Immunohistochemistry The avidin biotin complex (ABC) peroxidase technique was performed on selected paraffin sections, 4 mm thick. After deparaffinization and hydration, the specimens were placed in a plastic coplin jar, immersed in 10 mm citric acid monohydrate (ph 6) and pretreated in a microwave oven at 750 W for min. Endogenous peroxidase was blocked by immersion in 3% hydrogen peroxide in phosphate buffered saline solution (PBS; 10.4 mm Na 2 HPO 4, 3.16 mm KH 2 PO 4, 150 mm NaCl, ph 7.6). The sections were treated with 1.5% normal horse serum in PBS for 30 min at room temperature and then incubated with the following primary antibodies for 60 min at 37 C: p53 (clone AB-6; Calbiochem, Bad Soden, Germany; dilution 1:100), and bcl-2 (clone 124; Dako, Glostrup, Denmark; dilution 1:40). Bound primary antibody was detected by a commercially available reagent kit (Vectastain 1 Elite, Vector Laboratories, Burlingame, CA). Diaminobenzidine (Pierce, Rockford, IL) was used as chromogen. The sections were counterstained with hematoxylin. For p53, a colorectal carcinoma with a known p53 mutation was used as positive control. For bcl-2, basal cells in non-neoplastic prostate glands served as an internal positive control. Negative controls were performed by omitting the primary antibody. A case was considered positive for p53 or bcl-2 expression when at least 5% of tumour cells were marked by the corresponding antibody. 8,9 No further quantification according to the number of positive cells or the intensity of staining was performed. Interphase cytogenetics Chromogenic in situ hybridisations (CISH) were performed on selected 5 6 mm thick paraffin sections as described previously, 10 using digoxigenated nucleic acid probes specific for chromosomes 7, 8, 17, and X (Oncor, Gaithersburg, MD). Briefly, pretreatment of the slides consisted of boiling in 10 mm citric acid monohydrate in a microwave oven for 30 s, followed by an incubation with 1 M sodium thiocyanate for 10 min at 80 C. The slides were then digested with pepsin (2 4 mg/ml in 0.2 M hydrochloric acid) at 37 C for 2 8 min. Pepsin concentration and digestion time had to be optimised for each case. After washing and heating at 80 C for 30 min, the sections were covered with 30 ml hybridisation solution containing 65% deionized formamide, 26SSC (0.3 M sodium chloride, 0.03 M sodium citrate), 20% dextran sulfate, 1 mg/ml salmon sperm DNA, and 0.1 ng/ml DNA probe. The sections were denatured in a 78 C water bath for 10 min, followed by hybridisation at 37 C overnight. After washings in a solution containing 26SSC and formamide for 20 min at 40 C, hybridisation signals were visualised by the ABC method as described earlier. Slides were counterstained with haematoxylin and permanently mounted. For each tumour, two separate areas were evaluated by counting the hybridisation signals in (median, 253) non-overlapping nuclei. Signals in adjacent non-neoplastic mucosa or in stromal cells served as an internal control. A normal (disomic) status of the autosomes (chromosomes 7, 8, 17) was characterised by two signals in about 60% 70% of the nuclei and one signal in another 30% 40%, accounting for the fact that paraffin sections, 5 6 mm thick, do not contain whole nuclei. A chromosomal gain was diagnosed if more than 6% of nuclei contained more than two signals for the autosomes, or more than one signal for the X-chromosome. Tetrasomy for a given chromosome was assumed when the gain of signal numbers did not exceed four per nucleus, and aneusomy was assessed when more than 5% of nuclei contained five or more hybridisation signals. A diagnosis of monosomy required signal losses in more than 50% of nuclei. These cut-off values were adopted from the literature and from our own previous investigations on prostatic tissue sections. 11,12 Results Of the 365 radical prostatectomy specimens, 73 contained tumours larger than 10 cm 3. Of these, 21 could not be assigned to a specific zone. Two tumours were mainly located in the central zone and 36 in the peripheral zone.

3 Table 1 Clinical characteristics of 14 prostate cancers with tumour volumes larger than 10 cm 3 predominantly located in the transition zone 281 Age Clinical Preoperative Postoperative Follow-up Case (y) T-class PSA (ng/ml) PSA-recurrence (months) 1 71 T negative T2b 22.4 negative T1c 16.4 negative T2b 31.1 negative T1c 34.0 negative T1c 16.4 no data no data 7 73 T2a 23.9 negative T1c 14.4 negative T2b 5.9 negative T1c 21.7 negative T2b 35.5 negative T2a 25.6 after 38 months T1c 8.2 negative T2b 27.0 after 24 months 36 The remaining 14 cases contained tumours that were predominantly ( 70% of tumour area) located in the transition zone. The clinical and pathological data of these 14 patients are given in Tables 1 and 2. The median preoperative PSA-value was 22.1 ng/ml (range ng/ml). Seven tumours were confined to the prostate (pt2a or pt2b), five tumours penetrated the prostatic capsule (pt3a), and two cancers invaded the seminal vesicles (pt3b). A positive surgical margin was recorded in half of the cases. The median tumour volume was 17.2 cm 3 (range cm 3 ). Two prostate cancers (cases 7 and 9) were exclusively located in the transition zone. Four prostates (cases 3, 6, 12, and 13) had separate small cancer foci in the peripheral zone with a distance of non-neoplastic prostatic tissue of at least 3 mm in between. The remaining eight cases had tumour foci in the peripheral zone that were considered contiguous to the larger transition zone cancers. However, in two of these eight cases (cases 1 and 4), histology of the small tumours in the peripheral zone differed substantially from the transition zone cancers. Gleason patterns 2 and 3 were the dominant tumour grades in the transition zone. A Gleason pattern 4 was present in nine cases and was more frequent in peripheral zone tumour foci. Overexpression of the p53-protein was recorded in one tumour (case 14) only and no cancer exhibited any bcl-2-expression. Interphase cytogenetics for chromosomes 7, 8, 17, and X revealed a normal (disomic) chromosomal status in four cases. The remaining 10 tumours exhibited focal gains of signal counts at least for some of the chromosomes studied. In most cases, these gains of signal counts were compatible with a tetrasomic status. In two tumours (cases 2 and 5), however, more than 5% of nuclei displayed more than four hybridisation signals for chromosomes 7 and 8, thus qualifying for the diagnosis of aneusomy. Both cases contained high-grade (Gleason pattern 4) tumours. One case (case 6) failed to give satisfactory hybridisation results for chromosome 8 even after repeated experiments. No monosomies of chromosomes 7, 8, or 17 were detected. Figures 1 3 are derived from case 1 and illustrate a whole mount transversal section, a Gleason pattern 2 cancer area, and an in situ hybridisation for chromosome 8, respectively. Clinical follow-up data were available from 13 patients and ranged from 21 months to 71 months (median, 50 months). During that follow-up period, 11 of the Table 2 zone Pathological characteristics of 14 prostate cancers with tumour volumes larger than 10cm 3 predominantly located in the transition Case Pathol. TN-class Volume (cm 3 ) TZ (%) PZ (%) Margin Gleason TZ Gleason PZ p53 bcl-2 Interphase cytogenetics (chromosomes) 1 pt3b, pn neg 3 þ 2 3þ 4 neg neg tet(7), tet(8), tet(17), di(x) 2 pt3a, pn anterior pos 3 þ 4 3þ 4 neg neg an(7), an(8), tet(17), di(x) 3 pt2b, pn neg 2 þ 3 3þ 4 neg neg di(7), tet(8), tet(17), di(x) 4 pt3a, pn neg 2 þ 3 3þ 4 neg neg tet(7), tet(8), tet(17), di(x) 5 pt3a, pn neg 3 þ 4 3þ 4 neg neg an(7), an(8), tet(17), di(x) 6 pt2b, pn anterior pos 2 þ 3 3þ 3 neg neg di(7), no data (8), di(17), mon(x) 7 pt2a, pn neg 2 þ 3 neg neg di(7), di(8), di(17), mon(x) 8 pt2b, pn neg 2 þ 3 3þ 2 neg neg di(7), di(8), di(17), mon(x) 9 pt3a, pn neg 2 þ 3 neg neg tet(7), tet(8), tet(17), di(x) 10 pt2b, pn anterior pos 2 þ 3 3þ 2 neg neg di(7), di(8), tet(17), di(x) 11 pt3a, pn basal pos 3 þ 2 3þ 4 neg neg tet(7), di(8), di(17), mon(x) 12 pt2b, pn apical pos 2 þ 3 3þ 4 neg neg di(7), di(8), tet(17), di(x) 13 pt2b, pn apical pos 3 þ 4 3þ 3 neg neg di(7), di(8), di(17), mon(x) 14 pt3b, pn anterior pos 3 þ 4 3þ 4 pos neg tet(7), tet(8), tet(17), di(x) TZ: transition zone; PZ: peripheral zone; neg: negative; pos: positive; di: disomy; tet: tetrasomy; an: aneusomy; mon: monosomy.

4 282 Figure 1 Transversal whole-mount histological section of the prostate (case 1) at the level of the verumontaneum (U ¼ Urethra), showing two large cancer foci in the transition zone and a smaller focus in the peripheral zone. Hematoxylin and eosin; original magnification62. Figure 3 Same tumour area as in Figure 2. In situ hybridisation with a DNA-probe specific for chromosome 8. Tetrasomic signal distribution with three or four (arrowheads) hybridisation signals in many tumour nuclei. Counterstain with hematoxylin; original magnification Figure 2 Same case as Figure 1. Tumour focus in the transition zone with a Gleason pattern 2. Hematoxylin and eosin; original magnification patients remained free of tumour recurrence as defined by postoperative PSA-values below 0.1 ng/ml in two consecutive samples. The two remaining patients developed a postoperative rise in PSA after 38 months (case 12) and 24 months (case 14). For comparison, clinical and standard pathological data were also obtained from the 36 cases with large ( > 10 cm 3 ) tumours predominantly located in the peripheral zone. The median tumour volume of these cases was 15.3 cm 3 and the median preoperative PSA value was 17.4 ng/ml. Only two of these cancers were organ confined (pt2b), whereas 10 showed capsular penetration (pt3a) and the remaining tumours invaded the seminal vesicles (pt3b; 22 cases) or the bladder neck (pt4; 2 cases). All but two cases contained high-grade (Gleason 4/5 pattern) tumours. Follow-up data were available in 32 patients. Of these, 22 (68.8%) suffered from PSArecurrence after a median time of 12 months. Discussion We are reporting on 14 patients with prostate cancers treated by radical prostatectomy, whose tumours measured more than 10 cm 3 and were predominantly ( 70%) located in the transition zone. Despite the large volumes, these tumours had quite favourable pathological features and a relatively low rate of biochemical relapse, especially when compared to a group of tumours with similar size that were predominantly located in the peripheral zone. Since the description of different anatomic zones in the prostate by McNeal et al, it has become evident that prostate cancer arises about three times more frequently in the peripheral zone than in the transition zone. 7 Further differences between tumours in both locations include the mode of detection, morphology, and clinical outcome. 7,13,14 Stamey and coworkers reported three patients with large prostate cancers in the transition zone and excessive PSA-values between 150 and 456 ng/ml. After radical prostatectomy, however, these tumours were found to be organ-confined and the patients remained free of PSA recurrence up to 650 days. 4 This prompted us to review a series of 365 consecutive prostatectomy specimens in order to elaborate some pathological and clinical characteristics of large prostate cancers in the transition zone. We only found 14 cases (3.8%) that fulfilled the criteria mentioned earlier, indicating that this tumour-type is an infrequent subset of prostate cancers. Interestingly, in six patients the tumour was not palpable by digital rectal examination. Preoperative PSA-values were rather high in most cases (median, 22.1 ng/ml), although not as extremely elevated as in the patients described by Stamey. Only two tumours were entirely located in the transition zone, whereas the others also harboured smaller tumour foci in the peripheral zone, either separate or contiguous with the large transition zone cancer. While it can not be totally excluded that the latter tumours primarily originated in the peripheral zone and subsequently invaded the transition zone, a probable transition zone origin had been proposed by McNeal et al if more than 70% of the

5 tumour-area lies within that zone. 7 In half of the cases, the tumours were confined to the prostate. Interestingly, one large transition-zone cancer (case 7) was exclusively located in the right lobe, thus qualifying for a pt2aclassification. Only two cancers invaded the seminal vesicles and in no cases were there any metastases in frozen or permanent histological sections from regional lymph nodes. A low frequency of tumour spread by these routes was also found in previous reports. 5,7,13 Positive surgical margins were most often located anteriorly, which is in keeping with the findings by other groups. 5 Gleason patterns 2 or 3 were the dominant grades in the transition zone, but most cases also harboured poorly differentiated cancer, either as a minor component in the transition zone, or in peripheral zone tumour areas. A tendency towards lower Gleason grades in transition zone cancers was found in most studies, 1,7,12 14 but others reported that the amount of poorly differentiated tumour was nearly equal in both zones. 5 An abnormal expression of p53 or bcl-2 is regarded an adverse prognostic sign in prostate cancer High rates of abnormal bcl-2-expression have also been found in high-grade prostatic intraepithelial neoplasia, 15,16 which is regarded a putative precursor of prostatic carcinoma in the peripheral zone, but not in the transition zone. 13 We observed a quite infrequent expression of both markers in our group of 14 large transition zone-cancers. Only one case was positive for p53 and no case at all was positive for bcl-2. Numerical chromosomal aberrations are a frequent genetic event in prostatic neoplasms and the extent of alterations is associated with disease progression Interphase cytogenetics is an in situ hybridisation method with chromosome-specific probes, that works well on paraffin-embedded archival tissue. It allows for the detection of numerical changes of single chromosomes together with a well preserved histology. We analysed chromosomes 7, 8, 17, and X, because numerical gains of these chromosomes are among the most frequent alterations reported. 10,11 Four cases had a normal copy number for all chromosomes tested, whereas gains of at least one chromosome were found in the remaining 10 tumours. In most cases, signal counts were compatible with a tetrasomic chromosomal status. Only two tumours exhibited an aneusomy for chromosomes 7 and 8. The low rate of chromosomal aneusomies in prostatic carcinomas lying in the transition zone is in keeping with previous studies using DNA image analysis or interphase cytogenetics, 12,14 and contrasts with the findings in cancers from the peripheral zone. 10 On the other hand, in a study that investigated allelic losses by microsatellite analyses, no significant differences could be observed between tumour foci in the transition zone and in the peripheral zone. 17 Thus, the question of genetic differences between cancers in both prostatic zones has not been satisfactorily answered, yet. There are few studies that investigated the impact of zonal tumour location on cure rates after radical prostatectomy. Noguchi and coworkers matched 79 cases of transition zone cancers by total tumour volume to an equal number of cancers in the peripheral zone. They found a significant difference in the 5-year postoperative biochemical disease-free status of 72% and 49%, respectively. 5 In another study that investigated prognostic parameters in a large group of patients treated by radical prostatectomy, cure rates of cancers in the peripheral zone were only 32.5% for patients with a preoperative PSA greater than 10 ng/ml and only 6.7% for patients with a preoperative PSA greater than 20 ng/ml. 6 These observations fit well to the 31.2% freedom of PSA-recurrence in our control group of 36 patients with large cancers in the peripheral zone and a median preoperative PSA of 17.4 ng/ml. On the other hand, with comparable follow-up periods, 11 of 13 patients (85%) with large cancers in the transition zone remained disease-free in spite of a median preoperative PSA-value greater than 20 ng/ml. Therefore, although the number of patients in our study is rather small and the follow-up period is still too short to draw definitive conclusions, it seems as if transition zone cancers harbour a better prognosis even with very large tumour volumes. Hitherto, it is not quite clear why there exist differences in behaviour between peripheral zone and transition zone cancers, but it is tempting to speculate that these differences can not be put down to one single reason. Rather, a combination of anatomical peculiarities and pathological features, especially the frequently lower Gleason grades, may account for a better prognosis associated with prostate cancers in the transition zone. The present study confirms that tumour volume alone is not a reliable prognostic parameter. Furthermore, although being quite rare, large prostate cancers in the transition zone might be a source of error with any prognostic algorithm that uses preoperative PSA-values. While it is currently impossible to determine the exact location of the tumour preoperatively, it should at least be mentioned in the pathological report of the prostatectomy specimen if the largest cancer lies in the transition zone. Acknowledgements This work was supported by a grant from the Deutsche Forschungsgemeinschaft to H Huland. The authors thank Cornelia Ebisch for technical assistance. References 1 McNeal JE et al. Histologic differentiation, cancer volume, and pelvic lymph node metastasis in adenocarcinoma of the prostate. Cancer 1990; 66: Stamey TA et al. Biological determinants of cancer progression in men with prostate cancer. JAMA 1999; 281: 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 follow-up. J Urol 1993; 149: Stamey TA, Dietrick DD, Issa MM. Large, organ confined, impalpable transition zone prostate cancer: Association with metastatic levels of prostate specific antigen. J Urol 1993; 149: Noguchi M, Stamey TA, McNeal JE, Yemoto CEM. An analysis of 148 consecutive transition zone cancers: Clinical and histological characteristics. J Urol 2000; 163: Stamey TA et al. Prostate cancer is highly predictable: a prognostic equation based on all morphological variables in radical prostatectomy specimens. J Urol 2000; 163:

6 284 7 McNeal JE, Redwine EA, Freiha FS, Stamey TA. Zonal distribution of prostatic adenocarcinoma. Correlation with histologic pattern and direction of spread. Am J Surg Pathol 1988; 12: Grossfeld GD et al. Locally recurrent prostate tumors following either radiation therapy or radical prostatectomy have changes in KI-67 labeling index, p53 and bcl-2 immunoreactivity. JUrol1998; 159: Stapleton AMF et al. Assessment of the biological markers p53, Ki-67, and apoptotic index as predictive indicators of prostate carcinoma recurrence after surgery. Cancer 1998; 82: Henke R-P et al. Frequency and distribution of numerical chromosomal aberrations in prostatic cancer. Hum Pathol 1994; 25: Qian J et al. Chromosomal anomalies in prostatic intraepithelial neoplasia and carcinoma detected by fluorescence in situ hybridization. Cancer Res 1995; 55: Erbersdobler A, Hammerer P, Huland H, Henke R-P. Numerical chromosomal aberrations in transition-zone carcinomas of the prostate. J Urol 1997; 158: Greene DR et al. A comparison of the morphological features of cancer arising in the transition zone and in the peripheral zone of the prostate. J Urol 1991; 146: Grignon DJ, Sakr WA. Zonal origin of prostatic adenocarcinoma: Are there biologic differences between transition zone and peripheral zone adenocarcinomas of the prostate gland? J Cell Biochem 1994; 19: Bauer JJ et al. Elevated levels of apoptosis regulator proteins p53 and bcl-2 are independent prognostic biomarkers in surgically treated clinically localized prostate cancer. JUrol1996; 156: Häussler O et al. Cell proliferation, apoptosis, oncogene, and tumor suppressor gene status in adenosis with comparison to benign prostatic hyperplasia, prostatic intraepithelial neoplasia, and cancer. Hum Pathol 1999; 30: Erbersdobler A et al. Allelic losses at 8p, 10q, 11p, 13q, 16q, l7p, and 18q in prostatic carcinomas: the impact of zonal location, Gleason grade, and tumour multifocality. Prost Cancer Prostat Dis 1999; 2: