International Urology and Nephrology

, Volume 42, Issue 1, pp 73–79

Characterization of prostate cancer incidentally detected in radical cystoprostatectomy specimens from Japanese men with bladder cancer

Authors

  • Toshifumi Kurahashi
    • Division of UrologyKobe University Graduate School of Medicine
    • Division of UrologyKobe University Graduate School of Medicine
  • Junya Furukawa
    • Division of UrologyKobe University Graduate School of Medicine
  • Masafumi Kumano
    • Division of UrologyKobe University Graduate School of Medicine
  • Atsushi Takenaka
    • Division of UrologyKobe University Graduate School of Medicine
  • Masato Fujisawa
    • Division of UrologyKobe University Graduate School of Medicine
Urology - Original Paper

DOI: 10.1007/s11255-009-9578-1

Cite this article as:
Kurahashi, T., Miyake, H., Furukawa, J. et al. Int Urol Nephrol (2010) 42: 73. doi:10.1007/s11255-009-9578-1

Abstract

Objectives

The objective of this study was to investigate and characterize the clinicopathological features of incidentally detected prostate cancer in radical cystoprostatectomy specimens from Japanese men with bladder cancer.

Materials and methods

We reviewed the pathological reports of 251 male patients who underwent radical cystoprostatectomy for bladder cancer at our institution and identified men with incidentally detected prostate cancer in these specimens. Clinicopathological data of patients with incidental prostate cancer in cystoprostatectomy specimens (group A) were compared with those of 193 patients with clinically detected prostate cancer who underwent radical prostatectomy (group B). Immunohistochemical staining was also performed to measure the expression levels of Ki-67, p53 and androgen receptor (AR) proteins in specimens from both groups A and B.

Results

In this series, a total of 31 patients (12.3%; group A) were incidentally diagnosed as having prostate cancer in radical cystoprostatectomy specimens. Clinically significant cancer, defined as any tumor greater than 0.5 cc according to the report by Stamy et al. (Cancer 71:933–938, 1993) was detected in 9 (29.0%) in group A and 170 (88.1%) in group B. Mean age in group A was significantly older than that in group B, while despite the lack of significant difference in the incidence of seminal vesicle invasion between these two groups, other parameters in group A were significantly more favorable than those in group B, including serum prostate-specific antigen, pathological stage, Gleason score, perineural invasion and capsular penetration. None of the patients in group A had biochemical recurrence (median observation period, 82 months); however, biochemical recurrence occurred in 41 (21.2%) in group B (median observation period, 46 months). Furthermore, immunohistochemical study demonstrated the significantly greater expression of Ki-67, p53 and AR proteins in group B than in group A.

Conclusions

Clinicopathological features of incidentally detected prostate cancer are markedly more favorable than those of clinically detected prostate cancer, which may reflect less aggressive biological phenotypes of incidental prostate cancer arising in Japanese men.

Keywords

Incidental prostate cancerRadical cystoprostatectomyRadical prostatectomyImmunohistochemical staining

Introduction

Prostate cancer is the most frequently diagnosed malignancy and second leading cause of cancer-specific deaths in men in Western-industrialized countries [1]. However, the histological prevalence of prostate cancer far exceeds that of clinically significant disease; that is, in autopsy series, incidental prostate cancer is found in approximately 30% of 50-year-old men, gradually increasing to as high as 70% in 80-year-old men [2]. These latent cancers are usually small, well or moderately differentiated, and localized within the prostate. In general, the majority of prostate cancers detected in the prostate of cystoprostatectomy specimens from patients undergoing bladder cancer surgery have also been regarded as clinically insignificant disease [320]. Therefore, analyzing cystoprostatectomy specimens, which offer a unique opportunity to study the incidence and morphological features of incidental prostate cancer, would be an interesting approach to characterize differences between incidentally and clinically detected prostate cancers.

To date, several investigators have performed comparative studies of prostate cancer arising in Japanese versus Western men, showing the evident differences in epidemiological, pathological and molecular aspects between them [1, 21]. To our knowledge, however, there have not been any studies precisely analyzing the characteristics of prostate cancer incidentally detected in Japanese men; therefore, in this study, we retrospectively reviewed data from 31 patients with incidental prostate cancer who underwent cystoprostatectomy for bladder cancer and 193 patients who underwent radical prostatectomy for clinically detected prostate cancer at our institution. Clinicopathological differences between these two groups were compared, and immunohistochemical staining of resected specimens was performed with antibodies against Ki-67, p53 and androgen receptor (AR), which are shown to be major molecular factors involved in the progression of prostate cancer [22].

Materials and methods

We retrospectively reviewed the data of 256 male patients who underwent radical cystoprostatectomy for primary bladder cancer at our institution between 1988 and 2006. Bladder cancer was histologically diagnosed by transurethral resection. The indication for radical cystectomy included muscle invasive bladder cancer, carcinoma in situ of the bladder refractory to intravesical bacillus Calmette–Guerin therapy and recurrent multifocal high-grade superficial bladder cancer uncontrollable by repeat transurethral resection. Physical examinations, laboratory studies, chest radiographies and intravenous pyelograms were performed in all patients. Computerized tomography, magnetic resonance imaging and/or abdominal ultrasonography were used for clinical staging. In addition, all the patients were examined by digital rectal examination, and 223 of them underwent assessment of the serum prostate-specific antigen (PSA) level. Based on these preoperative examinations, patients with findings suspicious of prostate cancer underwent further examinations, and 5 were diagnosed as having prostate cancer prior to radical cystoprostatectomy; therefore, these 5 were excluded from the study, and the following analyses were performed targeting the remaining 251. The surgical procedures for radical cystoprostatectomy remained unchanged during the study period; that is, bladder, prostate and seminal vesicles were totally resected, and urethrectomy was performed in cases diagnosed as having histologically proven cancer of the prostate and/or urethra before radical cystoprostatectomy. Standard pelvic lymphadenectomy including the obturator and iliac nodes was performed for all patients. This study also included 193 patients with clinically organ-confined prostate cancer who underwent radical prostatectomy and bilateral pelvic lymphadenectomy between 2001 and 2006 without any neoadjuvant therapies.

All pathological examinations were performed by a single pathologist according to the 2002 TNM classification system. Both cystoprostatectomy and prostatectomy specimens were prepared as whole mounts and fixed in 10% neutral formalin. Whole-mount sections perpendicular to the long axis of the specimen were cut transversely at 3–5-mm intervals and processed with hematoxylin and eosin on slides for microscopic evaluation. Tumor volume in resected specimens was measured with a digitizer tablet as previously described [23]. In this series, clinically significant cancer was regarded as tumor greater than 0.5 cc according to the criterion advocated by Stamey et al. [24]. Biochemical recurrence following surgery was defined as PSA persistently >0.2 ng/ml. Irrespective of pathological findings, none of the patients received any adjuvant therapies against prostate cancer until their serum PSA levels reached 0.4 ng/ml or greater.

Immunohistochemical staining of resected specimens was performed as previously described [25]. Briefly, sections from formaldehyde-fixed, paraffin-embedded tissue were deparaffinized and rehydrated. After blocking endogenous peroxidase, sections were incubated with 5% normal blocking serum for 20 min. The sections were then incubated with the following antibodies: anti- Ki-67 mouse monoclonal antibody (Dako, Carpinteria, CA), anti-p53 mouse monoclonal antibody and anti-AR mouse monoclonal antibody (Novocastra Laboratories, Newcastle, UK). The sections were then incubated with biotinylated goat anti-mouse IgG (Vector Laboratories, Burlingame, CA). After incubation in an avidin–biotin peroxidase complex for 30 min, the samples were exposed to diaminobenzidine tetrahydrochloride solution and counterstained with hematoxylin. Staining results were interpreted by two independent observers (JF and MK) who were blinded to the clinicopathological data of the patients. For Ki-67 and p53 analyses, only nuclear staining was considered, and strong expression of Ki-67 and p53 was defined as the proportion of positively stained tumor cells greater than 5% and 20%, respectively, as previously reported [22]. The highest intensity of immunohistochemical staining for AR was visually scored from several fields of each section and classified into the following four groups: negative, weak, moderate and strong. According to the previous study, either moderate or strong staining intensity in greater than 10% of tumor cells was considered strong expression [26].

All statistical analyses were performed using Statview 5.0 software (Abacus Concepts, Inc., Berkley, CA). Clinicopathological factors were analyzed using the chi-square test, unpaired t-test or Mann–Whitney U test. Probability (p) values <0.05 were considered significant.

Results

In this study, a total of 31 patients (12.3%; group A) were incidentally diagnosed as having prostate cancer in radical cystoprostatectomy specimens. Detailed characteristics of these 31 are summarized in Table 1. As shown in Table 2, 9 patients (29.0%) in group A and 170 (88.1%) of the 193 (group B) undergoing radical prostatectomy were diagnosed as having significant prostate cancer. Furthermore, the mean age at surgery in group A was significantly older than that in group B. Despite the lack of a significant difference in the incidence of seminal vesicle invasion between these two groups, other parameters examined in this study were significantly more favorable in group A than in group B, including the serum level of PSA, pathological stage, Gleason score, perineural invasion and capsular penetration. Although there were no patients demonstrating biochemical recurrence in group A (median observation period, 48 months; range, 5–195 months), biochemical recurrence occurred in 41 (21.2%) in group B (median observation period, 46 months; range, 22–86 months).
Table 1

Characteristics of 31 patients with incidentally detected prostate cancer in radical cystoprostatectomy specimens

No.

Age (years)

Stage of bladder cancer

Stage of prostate cancer

Gleason score

Significant cancer

Expression of Ki-67

Expression of p53

Expression of AR

Follow-up (months)

Biochemical recurrence

Patient outcome

1

65

pT3a

pT2a

5

No

Weak

Weak

Weak

64

Dead (lung cancer)

2

69

pT3b

pT2a

5

No

Weak

Weak

Weak

109

Alive

3

62

pT4

pT2b

7

Yes

Weak

Weak

Weak

5

Lost to follow-up

4

72

pT3a

pT2a

6

No

Weak

Weak

Weak

52

Dead (bladder cancer)

5

73

pT4

pT3a

7

Yes

Weak

Weak

Weak

46

Alive (with cancer)

6

86

pT4

pT2b

6

No

Weak

Strong

Weak

12

Dead (gastric cancer)

7

77

pT3b

pT2b

7

Yes

Weak

Strong

Strong

6

Dead (bladder cancer)

8

69

pTis

pT2a

6

No

Weak

Weak

Weak

22

Alive

9

66

pT4

pT3b

7

Yes

Strong

Strong

Weak

24

Alive

10

79

pT4

pT2b

7

Yes

Weak

Weak

Weak

16

Alive

11

70

pT2

pT2b

6

No

Weak

Weak

Weak

12

Alive

12

78

pT4

pT2a

7

No

Weak

Weak

Weak

52

Alive

13

81

pT2

pT2a

7

No

Weak

Weak

Strong

52

Alive

14

75

pTis

pT2b

7

Yes

Weak

Weak

Weak

51

Alive

15

83

pT3a

pT2a

7

No

Weak

Weak

Strong

26

Dead (bladder cancer)

16

80

pT3a

pT2a

6

No

Weak

Weak

Weak

49

Alive

17

68

pT0

pT2a

7

No

Strong

Strong

Weak

49

Alive

18

82

pT0

pT2a

6

No

Strong

Strong

Strong

48

Alive

19

67

pTis

pT2a

6

No

Weak

Weak

Strong

44

Alive

20

83

pTis

pT2a

5

No

Weak

Weak

Weak

17

Dead (cardiac cancer)

21

59

pT1

pT2a

5

No

Weak

Weak

Strong

195

Alive

22

79

pT3a

pT2a

5

No

Weak

Weak

Weak

15

Dead (bladder cancer)

23

79

pT3b

pT4

6

Yes

Weak

Weak

Strong

24

Dead (bladder cancer)

24

80

pT3b

pT2a

5

No

Weak

Weak

Weak

12

Dead (bladder cancer)

25

74

pT3b

pT2a

5

No

Weak

Weak

Weak

4

Dead (bladder cancer)

26

69

pT0

pT2b

5

Yes

Weak

Weak

Weak

107

Alive

27

69

pT4

pT2a

5

No

Weak

Strong

Strong

97

Alive

28

70

pT2

pT2b

5

No

Weak

Weak

Weak

89

Alive

29

75

pT0

pT2a

5

No

Weak

Weak

Weak

36

Dead (gastric cancer)

30

77

pT3a

pT3a

5

Yes

Weak

Weak

Weak

67

Alive

31

66

pT4

pT2a

7

No

Weak

Weak

Weak

67

Alive

AR androgen receptor

Table 2

Comparison of clinicopathological features between patients with incidentally detected and clinically detected prostate cancer

Variables

Group A (incidental cancer, n = 31)

Group B (clinical cancer, n = 193)

p value

Age (years)a

73.6 ± 6.8

69.5 ± 6.1

<0.001

Serum PSA (ng/ml)a

4.5 ± 5.3b

10.6 ± 4.2

<0.001

Pathological stage (%)

  

<0.001

    pT2

27 (87.1)

99 (51.3)

 

    pT3

3 (9.7)

89 (46.1)

 

    pT4

1 (3.2)

5 (2.6)

 

Gleason score (%)

  

0.0063

    6 or less

20 (64.5)

73 (37.8)

 

    7

11 (35.5)

90 (46.6)

 

    8 or greater

0 (0)

30 (15.6)

 

Perineural invasion (%)

  

<0.001

    Negative

26 (83.9)

92 (47.7)

 

    Positive

5 (16.1)

101 (52.3)

 

Capsular penetration (%)

  

<0.001

    Negative

27 (87.1)

99 (51.3)

 

    Positive

4 (12.9)

94 (48.7)

 

Seminal vesicle invasion (%)

  

0.25

    Negative

29 (93.5)

166 (86.0)

 

    Positive

2 (6.5)

27 (14.0)

 

Significant cancer (%)

  

<0.001

    No

22 (71.0)

23 (11.9)

 

    Yes

9 (29.0)

170 (88.1)

 

Ki-67 expression (%)

  

<0.001

    Weak

28 (90.3)

113 (58.5)

 

    Strong

3 (9.7)

80 (41.5)

 

p53 expression (%)

  

0.0060

    Weak

25 (80.6)

105 (54.4)

 

    Strong

6 (19.4)

88 (45.6)

 

AR expression (%)

  

0.015

    Weak

23 (74.2)

98 (50.8)

 

    Strong

8 (25.8)

95 (49.2)

 

PSA prostate-specific antigen; AR androgen receptor

aValues are expressed as mean ± standard deviation

bSerum PSA values were available in 28 of the 31 patients in group A

We then evaluated the outcomes of immunohistochemical staining for Ki-67, p53 and AR proteins in resected specimens from groups A and B. The proportion of specimens showing strong expression of Ki-67, p53 and AR proteins was significantly greater in group B than in group A (Table 2).

Discussion

Based on the findings of autopsy studies, it has been well accepted that the prevalence of silent or latent prostate cancer is much higher than that of clinically diagnosed cases [2]; however, to our knowledge, there have not been any studies focusing on the features of incidentally detected prostate cancer in Japanese men. In this study, therefore, we retrospectively analyzed clinicopathological findings of incidentally detected prostate cancer in radical cystoprostatectomy specimens from patients undergoing bladder cancer surgery compared with those of clinically detected prostate cancer in radical prostatectomy specimens.

In this series, incidental prostate cancers were found in 12.3% of cystoprostatectomy specimens, which was remarkably low compared with those in previous studies from Western countries; that is, when the outcomes of 18 studies summarized in Table 3 were combined, the proportion of incidental prostate cancer in Western countries appeared to be 28.0% [320]. Considering the well-established finding that the incidence of latent prostate cancer in Japan is similar to that in Western countries [22], these outcomes suggest that different mechanisms are involved in the development of incidental and latent cancers in Japanese and Western men.
Table 3

Published data on incidentally detected prostate cancer in radical cystoprostatectomy specimens

References

No. of patients

Mean age (year)

Section of pathological specimens (mm)

Sampling of prostate

No. of prostate cancer (%)

No. of significant cancer (%)

[3]

40

64.3

2–3

Partial

18 (45)

6 (33)

[4]

72

NA

4–5

Total

33 (46)

NA

[5]

165

NA

NA

NA

45 (27)

20 (44)

[6]

14

67.1

NA

NA

2 (14)

NA

[7]

59

66.5

5

Total

16 (27)

NA

[8]

133

NA

3

Total

58 (44)

4 (7)

[9]

127

69

3

Partial

14 (13)

NA

[10]

89

64

4

Total

41 (49)

4 (10)

[11]

129

69

NA

Partial

30 (23)

18 (60)

[12]

121

67.4

2–3.5

Total

50 (41)

24 (48)

[13]

132

61

5

Total

55 (42)

NA

[14]

100

62

2.5

Total

51 (51)

6 (12)

[15]

141

62

4

Total

20 (14)

2 (10)

[16]

128

NA

NA

NA

23 (18)

NA

[17]

248

63

5

Total

10 (4)

NA

[18]

63

67

3

Total

34 (54)

7 (20)

[19]

97

NA

2

Partial

58 (60)

31 (53)

[20]

204

67

NA

Total

58 (28)

18 (31)

Present study

251

65.3

3–5

Total

31 (12)

9 (29)

NA not available

Unsuspected prostate cancer in radical cystoprosatectomy specimens were shown to generally involve small lesions localized within the prostate [320]. The present study also confirmed this finding; that is, incidentally detected prostate cancer was regarded as having more favorable characteristics in terms of serum PSA, pathological stage, Gleason score, perineural invasion and capsular penetration than clinically detected prostate cancer. Furthermore, in this series, 29.0% of incidental prostate cancers were judged to be insignificant cancer; however, various proportions of insignificant prostate cancer in cystoprostatectomy specimens were reported in previous studies, ranging from 7 to 60%, which may reflect several methodological differences among these studies, such as the definition of insignificant disease and sectioning interval of resected specimens [320].

It is of interest to investigate the prognostic significance of incidental prostate cancer. During the observation period of this study, there was no patient with incidentally detected prostate cancer who developed biochemical recurrence. Previous studies also support this finding, which reported the lack of a prognostic impact of incidental prostate cancer, even clinically significant disease [15, 19]. In this study, to evaluate the biological aggressiveness of incidental prostate cancer, the expression patterns of potential molecular markers of prostate cancer, including Ki-67, p53 and AR, were analyzed by immunohistochemical staining, since these three molecules were shown to be significantly associated with the prognosis of patients who underwent radical prostatectomy for clinically localized prostate cancer [26, 27]. As expected, the incidences of strong expression of Ki-67, p53 and AR proteins in incidental disease are extremely low compared with those in clinically detected disease. Collectively, these findings suggest that the majority of incidentally detected prostate cancer in cystoprostatectomy specimens could be regarded as having no life-threatening impact due to the less aggressive biological phenotype.

In conclusion, the proportion of incidentally detected prostate cancers in radical cystoprostatectomy specimens from Japanese men was lower than that in Western men, although there was no evident difference in the incidence of significant disease in incidental prostate cancer between these two populations. In addition, the characteristics of incidental prostate cancer in Japanese men, including biological phenotype assessed by the expression of Ki-67, p53 and AR proteins, appeared to be more favorable than those of clinically detected prostate cancer.

Copyright information

© Springer Science+Business Media, B.V. 2009