Pathologic characteristics of prostatic adenocarcinomas: a mapping analysis of Korean patients

Transcription

1 Pathologic characteristics of prostatic adenocarcinomas: a mapping analysis of Korean patients SY Song 1 *, SR Kim 1, G Ahn 1 & HY Choi 2 1 Department of Pathology, Sungkyunkwan University School of Medicine, Samsung Medical Center, Seoul, Korea; and 2 Department of Urology, Sungkyunkwan University School of Medicine, Samsung Medical Center, Seoul, Korea (2003) 6, & 2003 Nature Publishing Group All rights reserved /03 $ Pathologic characteristics of the prostatic adenocarcinoma in Koreans are not clear. We studied 132 cases of prostatectomy specimens using mapping analysis to discover the pathologic characteristics of the Korean prostatic adenocarcinoma. Mean values were as follows: serum prostate-specific antigen level (spsa), 16.4 ng/ ml; tumor volume, 27.5%; size, 2.4 cm; Gleason score, 7.7; and p53 expression, 9.8%. Rates of multifocal tumors and high-grade prostatic intraepithelial neoplasm (HPIN) were 33.3 and 65.2%. The Gleason score, tumor volume%, tumor size and spsa were correlated with each other. Korean prostatic adenocarcinomas showed higher Gleason scores, lower rates of HPIN and multifocality, and lower p53 expression in comparison to Western prostatic adenocarcinomas. These data may be a basis for pathologic characteristics of Korean prostatic adenocarcinoma that has now been increasing. (2003) 6, doi: /sj.pcan Keywords: prostatic neoplasms; pathology; Korea Introduction Prostate cancer is one of the most common male cancers and is the second most common cause of death by cancer in Western countries. 1,2 Since early diagnosis of prostate cancer is difficult, it is usually found in the progressive form, about 33 40% of total prostate cancers. 2,3 Its pathologic characteristics and prognostic factors, such as serum prostate-specific antigen level (spsa), Gleason score, tumor volume% or size, DNA ploidy, chromosome abnormality and p53 mutation, have been studied extensively in Western countries. 4 6 In spite of being highlighted in Western countries, prostate cancer occupies only a minor proportion of cancer incidence and cancer-related deaths in Korea. It is the ninth common male cancers (2.2%; year 2000) and is the tenth common cause of death by male cancer (2.3 deaths/ ; year 2000) in Korea. 7 During the past 5 years, its significance in the incidence and cancer-related death of prostate *Correspondence: Dr SY Song, Department of Pathology, Samsung Medical Center, 50 Ilwon-dong, Kangnam-gu, Seoul , Korea. yodasong@smc.samsung.co.kr Received 13 May 2002; revised 10 September 2002; accepted 23 September 2002 cancer patient has now been increasing (almost twice) in Korea, 7 like other Asian country (eg, Singapore 8 ). Thus, it is important to understand the pathologic characteristics and prognostic factors of prostate cancer in Korea. In addition to incidence, are there any more differences between the Western prostate cancer and the Asian one? A couple of studies have answered yes to this question. Data on Japanese and Singaporean patients have shown that p53 overexpression and mutation occur in prostate cancer with a considerably lower frequency than those found in Western populations. 7,9,10 In addition to data with discordant rates of p53 change in prostate tumors from Western and Japanese populations, a considerably higher frequency of ras mutation is found in Japanese patients, up to 25%. 11,12 High-grade prostatic intraepithelial neoplasm (HPIN) is the most important candidate for the premalignant condition of prostatic carcinogenesis and is frequently found near the prostatic adenocarcinoma Another pathologic characteristic of prostate cancer is multifocality, which can be demonstrated only by mapping analysis. Owing to this, pathologists should clarify that more than two carcinomas are apart by 3 5 mm or more. About 50 85% of prostate cancers are multifocal in Western countries However, it is hard to find these kinds of basic pathologic features of Asian prostate cancers in English literature.

2 144 Even in Korean literature, we were able to find only a few reports dealing with basic pathologic features of prostate cancer, which were demonstrated by partial mapping analysis Using near-total mapping analysis, we investigated basic pathologic features including accompanying rate of HPIN and multifocality, as well as several prognostic indicators including p53 overexpression of prostate adenocarcinoma in the Korean patient in order to understand the pathologic differences between Korean and Western prostate cancers. Materials and methods Materials Samples used in this study were radically resected prostates obtained from the patients admitted to the Department of Urology, Samsung Medical Center between January 1995 and August A total of 138 prostates were resected because of the preoperative diagnosis of malignancy. Three cases were excluded because of the uncertainty of clinical information and incomplete pathological examination or preservation. Three cases were mesenchymal malignancy. No initial chemotherapy or radiotherapy was performed before tumor excision. A final of 132 samples were included in this study. Ethnically, all patients were Korean. Basic clinical data of the patients were obtained from the clinical record. Pathological mapping analysis Resected prostates were processed according to the guideline described in the textbook, Urologic Surgical Pathology. 6 In brief, they were inked, serially sectioned of 3 mm thickness and carefully observed. Representative sections were embedded in giant blocks. Entire or more than two-thirds volume of prostate was embedded to make tissue blocks. Apex and base tissues were separately dissected and conization was done. Tumor area was marked on the glass slide, measured diameter, calculated vol%, graded according to the Gleason system. 4 Routine pathologic examination was performed by experienced pathologists. Other pathologic features, such as, resection margin, neural invasion, vascular invasion, seminal vesicle invasion, presence of HPIN and multifocality were also checked and recorded. Immunohistochemistry Immunohistochemical stains were performed using the paraffin-embedded tissue blocks. Thick tissue sections, 5 mm in thickness were placed on the sialine-coated glass slides, deparaffinated in xylene, rehydrated in ethanol and washed in Tris-buffered saline. Microwave-heated antigen retrieval was done before 3% hydrogen peroxide treatment. Anti-p53 antibody (1 : 80; Zymed, San Francisco, CA, USA) was incubated for an hour and the following steps were done with the streptavidin biotin peroxidase technique. 3,3 0 -diaminobenzidine was used as a chromogen and Gill s hematoxylin was used as a counterstain. A breast carcinoma tissue known to contain p53 was run in parallel for external positive control and a slide without p53 antibody was used as a negative control. Positive result was defined as cells greater than 5% with distinct nuclear staining. Statistical analysis Statistical analysis was performed using the dbstat for Windows (version 3.0, Seoul, Korea). The association between the variables was assessed by Kruskal Wallis test, Mann Whitney U (Wilcoxon rank sum) test and w 2 test with a significance level lower than Results Clinical characteristics There were 132 patients, from 51 to 76 years old, in the study with the average age being (Table 1). The age groups are: 50s, 47 people (35.6%); 60s, 73 people (55.3%); 70s, 12 people (9.1%). Age is correlated with the Gleason score (P ¼ ) and the multifocality (P ¼ ) (Table 2). The spsa was ng/ml with the average level being ng/ml (Table 1). Compared to those who did not, the patient who had an excessive rise of spsa (416 ng/ml) had a higher Gleason score ( : , P ¼ ), bigger tumor volume ( %: %, P ¼ ), and bigger tumor size ( cm: cm, P ¼ ) (Table 2). Pathologic characteristics The Gleason scores observed ranged from 2 to 10, with the average being (Table 1). The low-grade group (Gleason score, 2 6) consisted of 24 patients (18.2%); the intermediate-grade group (Gleason score, 7) consisted of 39 patients (29.6%); and the high-grade group (Gleason score, 8 10) consisted of 69 patients (52.3%). The higher the Gleason score, the bigger the tumor volume ( %: %: %, P ¼ ), the bigger the tumor size ( cm: cm: cm, P ¼ ), and the higher spsa ( ng/ml: ng/ml: ng/ml, P ¼ ) as well as the older the age ( : : , P ¼ ) (Table 2). Tumor volumes observed ranged widely from 1 to 95% with the average being % (Table 1). Compared to those that were not, tumors with over 25 vol% had higher Gleason scores Table 1 General clinicopathologic characteristics of studied cases Parameter Mean s.d. N Gleason score Tumor volume% Tumor size (cm) Age (year) Serum PSA (ng/ml) Number of tumors N: number of studied cases; s.d., standard deviation: PSA, prostate-specific antigen.

3 Table 2 Correlation between the prognostic parameters of prostatic adenocarcinoma 145 Parameter category Gleason score Tumor volume% Tumor size (cm) Age (year) Serum PSA (ng/ml) Gleason score P-value* Tumor volume o25% %B P-value** Tumor size o2.4 cm cmb P-value** Serum PSA o16 ng/ml ng/mlb P-value** Tumor extent pt pt P-value** p53 expression Negative Positive P-value** Number of tumors Unifocal Multifocal P-value** PSA: prostate-specific antigen; NS: not significant. *Kruskal Wallis test: **Mann Whitney U (Wilcoxon rank sum) test. ( : , P ¼ ), bigger tumor size ( cm: cm, P ¼ ), and higher spsa ( ng/ml: ng/ml, P ¼ ) (Table 2). Tumor size was measurable in 120 cases, ranging widely from 0.1 to 6.0 cm, and the average was cm (Table 1). Compared to tumors larger than 2.4 cm, those that were 2.4 cm and smaller had lower Gleason scores ( : , P ¼ ), lower tumor volumes ( %: %, P ¼ ), and lower spsa ( ng/ml: ng/ml, P ¼ ) (Table 2). The spsa varied, ranging from 0.02 to ng/ml, with the average being ng/ml (Table 1). Compared to those that had lower levels, the groups that had high spsa (over 16 ng/ml) had higher Gleason scores ( : , P ¼ ), higher tumor volume ( %: %, P ¼ ), and bigger tumor size ( cm: cm, P ¼ ). Tumor extent was included in pt2, 70 cases (53.0%); pt3, 58 cases (43.9%); and pt4, 4 cases (3.0%). When tumor extent was divided into a low-extent group (pt2) and a high-extent group (pt3-4), the high-extent group had a higher Gleason score ( : , P ¼ ), higher tumor volume ( %: %, P ¼ ), bigger tumor size ( cm: cm, P ¼ ), and a higher spsa ( ng/ml: ng/ml, P ¼ ) (Table 2). The following are comparisons between the high-extent group and the low extent group: apex involvement (62.9%: 38.6%, P ¼ ), base involvement (51.6%: 24.3%, P ¼ ), seminal vesicle involvement (46.8%: 0%, P ¼ ), resection margin involvement (72.6%: 0%, P ¼ ), neural involvement (82.3%: 47.1%, P ¼ ), and lymphatic invasion (22.6%: 4.3%, P ¼ ). Multifocal tumors were observed in 44 (33.3%) cases, and HPIN was observed in 86 cases (65.2%) (Table 1). The existence of multifocal tumors and presence of HPIN are not statistically correlated to prognostic parameters including tumor extent, pt (Table 3). p53 expression Out of 132 cases, 13 (9.8%) showed positive nuclear staining against anti-p53 antibody p53-positive tumors showed higher tumor volume than p53-negative group ( : , P ¼ ) and high-extent group (pt3-4) showed more frequent p53 expression (16.1%: 4.3%, P ¼ ) (Table 3). There were no statistical correlations between protein p53 expression and other parameters, such as Gleason s score, tumor size, spsa, neural involvement, lymphatic invasion, multifocality and presence of HPIN (Tables 2 and 3).

4 146 Table 3 Relation between tumor extent and various pathologic parameters Parameters pt2 pt3-4 N P-value* Apex involvement 66 (50.0%) 27 (38.6%) 39 (62.9%) Base involvement 49 (37.1%) 17 (24.3%) 32 (51.6%) SV involvement 29 (22.0%) 0 29 (46.8%) RM involvement 45 (34.1%) 0 45 (72.6%) Neural invasion 84 (63.6%) 33 (47.1%) 51 (82.3%) Lymphatic invasion 17 (12.9%) 3 (4.3%) 14 (22.6%) Multifocality 44 (33.3%) 27 (38.6%) 16 (25.8%) Presence of HPIN 86 (65.2%) 50 (71.4%) 36 (58.6%) p53 expression 13 (9.8%) 3 (4.3%) 10 (16.1%) SV: seminal vesicle; RM: resection margin; HPIN: high-grade prostatic intraepithelial neoplasm; N: number of studied cases; NS: not significant. *w 2 test. Discussion It is well known that prostatic adenocarcinoma is a disease of old Western men. One study reports a high number of multifocal tumors, often accompanied by HPIN with the average Gleason score being around The summary of recent studies done by mapping analysis of Korean men indicates that prostatic adenocarcinoma is presently different in Koreans compared to Westerners. First, Korean prostatic adenocarcinoma has a higher Gleason score than that of the West. Second, the HPIN rate is lower than that of Western men. Third, the rate of multifocal carcinomas is lower. As the Gleason score is the grading system that represents tumor differentiation, it correlates with tumor volume, spsa, lymph node metastasis, tumor recurrence, etc. 4,20,21 In this study, the biological significance of the Gleason score is similar to that of Western results, because it closely correlates with tumor volume, tumor size and extent, and it is proportionate to the spsa. Nevertheless, considering the above, the average Gleason score for this study was 7.7, which is quite different from the score of 6.1 from 195 cases reported by Qian et al. in Comparing this study and that of Lee et al. 19 who performed mapping analysis on 58 cases of Korean prostatic adenocarcinoma in 1999, we can see that the results are quite similar to each other. We assume this is the consequence of the characteristic of Korean prostatic adenocarcinoma. If we consider the significance of the Gleason score for prostatic adenocarcinoma biologically, we can assume it may be the result of late detection of prostatic adenocarcinoma in Korean patients. As a result of a lower frequency of cases in Korea, as compared to the West, there is little relative interest in prostatic adenocarcinoma, as with gastric carcinoma or carcinoma of the uterine cervix, both of which are also usually diagnosed in the advanced stages. However, the incidence of prostatic adenocarcinoma in Korea is now increasing, therefore increasing the interest in studying the disease and revitalizing the need for early diagnosis. Thus, the characteristics for Korean prostatic adenocarcinoma may change in the future. HPIN is known to show a strong association with prostatic adenocarcinoma, 6 more than 80% is observed in prostatic adenocarcinoma in Western cases. 7,9,22 HPIN is found near prostatic adenocarcinoma, usually within 2 mm, but it is very difficult to find without a mapping analysis on numerous prostatic tissues. In Korea, Lee et al. 19 were the first to report pathological studies of a mapping analysis. Their result was 79.3%, a little lower than that of Western studies; and our HPIN is 65.2%, which is much lower than that of Western results. Therefore, Lee et al s result and the result of this study indicate that the HPIN rate of the Korean prostatic adenocarcinoma seems to be lower than that of the West. What is the cause of this difference? Perhaps because many of the prostatic adenocarcinomas diagnosed in Korea are in their late detection, it is better to consider the difference in rates (Korean and Western) as being already mixed with invasive carcinoma. Qian et al reported that if the Gleason score is high, the volume of HPIN is low. 23 This is a proper explanation relating to the low HPIN rate of Korean prostatic adenocarcinoma. The frequency of multifocal prostatic adenocarcinoma is 50 85% in the west, but there is no information on frequency in Korea. It is assumed to be % in Korea. 20,22 Our result is 33.3%, which is very much lower that that of Western results. To get a more accurate frequency figure, sufficient mapping analysis on the tissues of prostatic adenocarcinomas is needed. Multifocal adenocarcinoma has different clinical significance. First, the significance changes depending on clonality. That is, if several carcinomas are of the same clone, there is the possibility that they have spread from the original carcinoma, indicating the possibility of intraorgan metastasis. If several carcinomas are polyclonal, we can guess that each have come into being independently. 10 As reported by a study on the loss of heterozygosity, the appearance of multifocal carcinomas is understood to be independent of each other. 10 If they are independent, prostatic adenocarcinoma must be resected radically for the complete removal of hidden or possible foci of synchronous or metachronous carcinomas. Second, when the sum of the size of multifocal tumors is calculated, there is a problem in that we must choose between the biggest one and the total of all tumors. However, since there are no rules as yet, in this study we used the biggest tumor to calculate the size, as other studies have done. In this case, the size seems to be proportionate to tumor volume% with significant difference. The reason for the low frequency of multifocal carcinomas in Korea is probably the same as that of HPIN. That is, to find the multifocal carcinoma, each multifocal carcinoma that exists independent of each other must be discovered in its early period of tumorigenesis. However, the frequency of detection of the multifocal carcinoma is low because their diagnostic detections are late in Korea. Third, what is the clinical difference between unifocal

5 and multifocal prostatic carcinomas? Oh et al. 20 reported that a high Gleason score and many extracapsular invasions were shown in unifocal carcinomas. Our study contains the same result concerning extracapsular invasions in the unifocal carcinoma. However, to summarize the results of the Western study, the multifocal carcinoma generally is more aggressive than the unifocal carcinoma; the Gleason score for the multifocal carcinoma is higher with higher frequency of lymphatic and capsular invasion, and the resection margin involvement is more often positive. 11,24 It can be another difference between Western and Korean prostatic adenocarcinomas. Our study researched not only the rate of existence of the multifocal carcinoma and HPIN rate, but also the invasion of the apex and the base. Apex involvement (50.0%) is affected more than base involvement (37.1%). It may be that the apex is included in the peripheral zone, a tissue that is anatomically susceptible to carcinomas. The neural invasion rate of Korean prostatic adenocarcinoma is 63.6%, and the lymphatic invasion rate is 12.9%. These results are not much different from that of Western studies. In Asian patients, p53 changes (protein overexpression or gene mutation) are not common, ranging from 3% to approximately 10%. 7,9,10 Our result, 9.8%, is very similar to the data from Singapore and Japan. There is no explanation about low p53 changes among Asian prostatic adenocarcinoma yet. In spite of its low expression in our study, pt3-4 group showed more p53 expression than pt2 group like other reports. 25 Our study is limited to clinically operable prostatic adenocarcinoma, and there is no autopsy data and minimal examples of lymph node dissection. Therefore, there may be adverse criticisms regarding the representativeness. However, the point of this study is to examine the various characteristics of the Korean prostatic adenocarcinoma, although a limited amount of samples were used. Based on this, the characteristics of the Korean prostatic adenocarcinoma can be observed in detail if the mapping analysis is done widely and frequently. The increase in the amount of data concerning Korean prostatic adenocarcinoma will play an important role in the diagnosis and treatment of patients. References 1 Catalona WJ, Scott WW. Carcinoma of the prostate: a review. J Urol 1978; 119: Landis SH, Murray T, Bolden S, Wergo PA. Cancer statistics, CA J Clin 1998; 48: Elder JS, Catalona WJ. Management of newly diagnosed metastatic carcinoma of the prostate. Urol Clin North Am 1984; 11: Gleason DF, Mellinger GT. Veterans Administration Cooperative Urological Research Group. Prediction of prognosis for prostatic adenocarcinoma by combined histological grading and clinical staging. JUrol1974; 111: Rosai J. Male reproductive system. In: Ackerman s Surgical Pathology, 8th edn. Mosby, St Louis, 1996, pp Bostwick DG. Neoplasms of the prostate. In: Bostwick DG, Eble JN (eds). Urologic Surgical Pathology. Mosby, St. Louis, 1997, pp Chia S-J et al. Prostate tumours from an Asian population: examination of bax, bcl-2, p53 and ras and identification of bax as a prognostic marker. Br J Cancer 2000; 83: Cancer Registration and Biostatics Branch, National Cancer Center. Cancer Statistics in Korea, Uchida T et al. Infrequent involvement of p53 gene mutations in the tumorigenesis of Japanese prostate cancer. Br J Cancer 1993; 68: Watanabe M et al. p53 gene mutations in human prostate cancers in Japan: different mutation spectra between Japan and Western countries. Jpn J Cancer Res 1994; 85: Anwar K et al. Presence of ras oncogene mutations and human papillomavirus DNA in human prostate carcinomas. Cancer Res 1992; 52: Shiraishi T et al. Mutations of ras gene are relatively frequent in Japanese prostate cancers: pointing to genetic differences between populations. Anticancer Res 1998; 18: Bostwick DG, Brawer MK. Prostatic intraepithelial neoplasia and early invasion in prostate cancer. Cancer 1987; 59: Amin MB, Ro JY, Ayala AG. Putative precursor lesion of prostatic adenocarcinoma: fact of fiction? Mod Pathol 1993; 6: Bostwick DG. High-grade prostatic intraepithelial neoplasia: the most likely precursor of the prostate cancer. Cancer 1995; 75: Byar DP, Mostofi FK. Carcinoma of the prostate: prognostic evaluation of certain pathologic features in 208 radical prostatectomys. Cancer 1972; 30: Villers A, McNeal JE, Frelha FS, Stamey TA. Multiple cancers in the prostate. Cancer 1992; 70: Haggman M et al. Characteristics of localized prostate cancer: distribution, grading and pt-staging in radical prostatectomy specimens. Scand J Urol Nephrol 1993; 27: Lee TJ, Choe MS, Kim CS, Ro JY. Clinicopathologic features of prostatic adenocarcinoma: a study of 58 radical prostatectomy specimens. Korean J Pathol 1999; 33: Oh W, Song SY, Choi HY. Comparison of clinical and pathological findings in unifocal and multifocal prostate cancer. Korean J Urol 1999; 40: Hong JH et al. Prostatic intraepithelial neoplasia in radical prostatectomy specimens as a prognostic factor. Korean J Urol 2000; 41: Choi YL, Kim SR, Song SY, Choi HY. Pathologic characteristics of Korean prostatic adenocarcinoma: a mapping analysis of 60 cases. Korean J Pathol 2001; 35: Qian J, Wollan P, Bostwick DG, The extent and multicentricity of high-grade prostatic intraepithelial neoplasia in clinically localized prostatic adenocarcinoma. Hum Pathol 1997; 28: Rob K et al. Predictability and significance of multifocal prostate cancer in the radical prostatectomy specimen. Eur Urol 1998; 33 (Suppl): Bookstein R, MacGrogan D, Hilsenbeck SG, Sharkey F, Allred DC. p53 is mutated in a subset of advanced-stage prostate cancers. Cancer Res 1993; 53: