Concomitant pathology in the prostate in cystoprostatectomy specimens: a prospective study and review

Abstract

OBJECTIVES

To investigate possible associated pathology in the prostate removed from patients with invasive bladder cancer and determine if there is a justification for prostate-sparing cystectomy.

PATIENTS AND METHODS

Between March 2005 and July 2007, 425 men (mean age 59 years, sd 8.23) had a cystoprostatectomy at our institute. The prostate was step sectioned at 2–3 mm intervals and any associated pathology determined; patient and tumour characteristics were correlated with prostatic pathology. The results were compared with those published previously, and the potential functional advantages of prostate sparing are reviewed and discussed.

RESULTS

Prostatic adenocarcinoma was detected in 90 of the 425 (21.2%) patients. There was no significant correlation between preoperative prostate-specific antigen level and the presence of adenocarcinoma, Gleason score or prostatic tumour stage. There was prostatic involvement as a result of direct invasion by the primary bladder tumour (contiguous) in 39 cases (9.2%). Concomitant (non-contiguous) transitional cell carcinoma of the prostatic urethra and/or ducts was detected in 27 specimens (6.4%). Additional findings were high-grade prostatic intraepithelial neoplasia in 43 patients (10.1%) and benign prostatic hyperplasia in 175 (41.2%).

CONCLUSION

We think that the potential oncological risks of prostate-sparing cystectomy outweigh any small and possible functional benefits; accordingly, the prostate should not be retained.

Abbreviation

PIN

prostatic intraepithelial neoplasia.

INTRODUCTION

Radical cystectomy is considered by many as the reference standard for managing patients with invasive bladder cancer; it provides excellent local control and long-term survival for most patients with locally advanced disease [1,2]. Some authors advocate retaining the prostate in selected patients for whom orthotopic diversion is contemplated [3–5]. They claim that most of their patients showed early recovery of continence and could maintain their potency.

To evaluate the safety and the possible oncological risks of prostate-sparing cystectomy, we conducted a prospective study to determine the associated pathological findings in the prostate for patients with invasive bladder cancer treated by radical cystoprostatectomy.

PATIENTS AND METHODS

Between March 2005 and July 2007, 425 consecutive men had a radical cystoprostatectomy for invasive bladder cancer, according to a previously described technique [6]. The clinical staging involved examination under anaesthesia, cystoscopic findings, and imaging by MRI and/or CT. Preoperative PSA values were available and recorded in most cases.

The surgical specimens were fixed in 10% formalin. Tissue sections were obtained from the primary tumour, bladder wall, ureters and lymph nodes. The TNM classification was used for the pathological staging [7], and the WHO classification for histopathological typing and grading [8]. The whole-mounted prostate was step-sectioned at 2–3 mm intervals, and these sections further divided into two to four portions according to their size. These were embedded in paraffin and a mean of 10–32 blocks were obtained from each case. The apex of the prostate and the lower third of the seminal vesicles were evaluated separately. The paraffin blocks were sectioned at 3–5 µm thickness and then stained with haematoxylin and eosin. Prostatic carcinoma, if found, was staged according to the WHO classification [9] and graded according to Gleason system [10].

The results were compared statistically using the chi-square test and Fisher’s exact test, with differences considered statistically significant at P < 0.05.

RESULTS

The clinical characteristics of the patients and the pathological features of their cystectomy specimens are given in Table 1; the associated pathological findings in the prostate are also presented in Table 1. Of 425 patients examined there was histological evidence of adenocarcinoma of the prostate in 90 (21.2%). The incidence of carcinoma of the prostate relative to some of the patient and pathological characteristics of the bladder tumour is outlined in Table 2. There was a correlation between the preoperative PSA level, stage and Gleason score of the patients with prostatic carcinoma (Table 3).

Table 1.  The clinical and pathological characteristics of 425 patients who underwent cystoprostatectomy

Characteristics	N (%)
* Nine patients had more than one significant pathology in their prostates.
Age interval, years
<50	73 (17.2)
50–60	182 (42.8)
61–70	135 (31.8)
>70	35 (8.2)
Preoperative PSA level, ng/mL (385 patients)
<4	338 (87.8)
4–<10	45 (11.7)
≥10	2 (0.5)
Histology of the primary tumour
TCC	309 (72.7)
SCC	83 (19.5)
Adenocarcinoma	17 (4.0)
Undifferentiated	16 (3.8)
pT stage of the primary tumour
pis + pT1	85 (20.0)
pT2	156 (36.7)
pT3	125 (29.4)
pT4	59 (13.8)
Grade of the primary tumour
I	54 (12.7)
II	72 (16.9)
III	299 (70.4)
Lymph node involvement
Negative	317 (74.6)
Positive	91 (21.4)
Unremoved	17 (3.8)
Tumour site
Basal/trigonal	28 (6.6)
Posterior wall	214 (50.4)
Lateral wall	151 (35.5)
Dome	32 (7.5)
Pathology
Adenocarcinoma prostate	90 (20.7)
Prostatic involvement (pT4a)	39 (8.9)
TCC of prostate (prostatic urethra/ducts)	27 (6.2)
High-grade PIN	43 (9.9)
BPH	175 (40.3)
Prostatitis/normal finding	60 (13.8)
Total	434 (100)*

Table 2.  The incidence of carcinoma of the prostate relative to some of the patient and pathological characteristics

Variable	N (%) of cases
	Involved	Uninvolved	P
Age interval, years
<50	6 (8.2)	67 (91.8)
50–60	37 (20.3)	145 (79.7)
61–70	37 (27.4)	98 (72.6)
>70	10 (28.6)	25 (71.4)	<0.01
Histology of the primary tumour
TCC	67 (21.7)	242 (78.3)
SCC	17 (20.5)	66 (79.5)
Adenocarcinoma	5 (29.4)	12 (70.6)
Undifferentiated	1 (6.3)	15 (93.7)	>0.1
pT stage of the primary tumour
pis + pT1	20 (23.5)	65 (76.5)
pT2	36 (23.4)	120 (76.9)
pT3	24 (19.2)	101 (80.8)
pT4	10 (16.4)	49 (83.1)	>0.1
Grade of the primary tumour
I	4 (7.4)	50 (92.6)
II	7 (9.7)	65 (90.3)
III	78 (26.1)	221 (73.9)	<0.01
PSA interval
<4	73 (21.6)	265 (78.4)
≥4	14 (28.8)	33 (71.2)	>0.1
Total	90 (21.2)	335 (78.8)

Table 3.  The pathology of cases with prostatic adenocarcinoma relative to preoperative PSA levels

Pathology	PSA interval			Total	P
	<4	4–10	≥10
* PSA data missing in three cases.
Gleason score
≤4	5	–	–	5
5–7	66	13	1	80
8–10	2	–	–	2	>0.1
pT stage of prostate cancer
T1	53	7	–	60
T2	20	5	1	26
T3	–	1	–	1	>0.1
Total	73	13	1	87*

There was prostatic involvement as a result of direct invasion by the primary bladder tumour (contiguous) in 39 patients (9.2%). Of these, eight were correctly staged before surgery, while the remainder were understaged. Concomitant (non-contiguous)TCC of the prostatic urethra and/or ducts was detected in 27 specimens (6.4%). The pathological findings of the primary bladder tumour among these patients are given in (Table 4). There were also additional relevant lesions, i.e. high-grade prostatic intraepithelial neoplasia (PIN) in 43 patients (10.1%) and BPH in 175 (41.2%).

Table 4.  Primary bladder pathology and concomitant TCC of the prostate

Variable	N (%)		P
	Involved	Uninvolved
* Extravesical extension not directly involving the prostate.
Histology of the primary tumour
TCC	23 (7.4)	286 (92.6)
SCC	2 (2.4)	81 (97.6)
Adenocarcinoma	1 (5.9)	16 (94.1)
Undifferentiated	1 (6.3)	15 (93.7)	>0.1
pT stage
pis + pT1	4 (5.5)	81 (94.5)
pT2	13 (8.3)	143 (91.7)
pT3	7 (5.6)	118 (94.4)
pT4	3 (5.1)*	56 (94.9)	>0.1
Grade
I	1 (1.9)	53 (98.1)
II	5 (6.9)	67 (93.1)
III	21 (7.0)	278 (93.0)	>0.1
Total	27 (6.4)	398 (93.6)

DISCUSSION

Conventional radical cystoprostatectomy involves removing the bladder, the lower parts of the ureters, the seminal vesicles, the vas and the prostate, and the pelvic lymph nodes [1]. The operation as such undermines the patient’s quality of life, as it involves a urinary diversion and results in sexual disability. Technical advances in the last two decades allowed the introduction of orthotopic substitution with or without nerve preservation. The success of the neobladder depends on preserving renal function and maintenance of continence. Good day-time continence after orthotopic substitution has been uniformly reported by most investigators. Varying degrees of nocturnal enuresis in some 20–25% of patients is still a problem [11–14]. It was suggested by several authors that preserving the prostate in whole or in part can circumvent problems related to continence, and results in the preservation of erectile function [3–5,15–21]. However, serious concerns were raised by other authorities of the possible oncological risks resulting from retaining the prostate.

A careful pathological evaluation of cystoprostatectomy specimens obtained from patients with invasive bladder cancer provides an excellent opportunity to assess the frequency of associated lesions in the prostate. The reported frequency of incidentally discovered prostate cancer is summarized in Table 5[22–49]; the frequency is variable, and can be explained by several factors. The slice thickness obtained from the prostate for pathological examination is a critical issue. Bartsch et al.[48] reported their cumulative results of 1076 patients who had a radical cystoprostatectomy between January 1986 and October 2006. From 1986 to 1994 the histopathological evaluation of the prostate was limited to four random sections, and after that a whole-mount step-sectioning technique was adopted; the incidence increased from 7.7% to 33.4%. Another important variable is the effect of the study population; Lee et al.[43], from Taiwan, reported a surprisingly low incidence (4%). A similar value was also reported by a group from Iran [47]. The age of the study population can also account for this variability; in the present study the mean patient age was 59 years, which is about a decade younger than those reported in western studies.

Table 5.  The incidence of prostate cancer in cystoprostatectomy specimens in previous reports

Ref.	Clinically significant (% of prostate cancer)	Prostate cancer (%)	Section thickness. mm	Mean age of patients, years	Study design	Sample n
R, retrospective; P, prospective; n/a, not available.
[22]	11 (50)	22 (28)	n/a	63.7	R	80
[23]	n/a	45 (27)	n/a	n/a	P	165
[24]	n/a	25 (38)	n/a	n/a	R	66
[25]	18 (60)	33 (46)	4–5	n/a	R	84
[26]	11 (20)	55 (40)	n/a	n/a	R	139
[27]	6 (33)	18 (45)	2–3	64.3	R	40
[28]	n/a	22 (46)	n/a	n/a	R	48
[29]	n/a	17 (33)	3	67.8	R	49
[30]	n/a	16 (27)	5	66.5	P	59
[31]	11 (18)	58 (44)	3	60	P	133
[32]	n/a	14 (12.6)	3	69	R	127
[33]	n/a	41 (49)	4	64	P	89
[34]	24 (48)	50 (41)	2–3.5	67.1	P	121
[35]	18 (60)	30 (23)	2–6	69	R	129
[36]	n/a	17 (28)	3–5	66.7	R	60
[37]	10 (62)	16 (23)	5–10	67	R	70
[38]	n/a	23 (18)	n/a	n/a	R	128
[39]	12 (22)	51 (51)	2.5	62	R	10
[40]	n/a	55 (42)	5	n/a	R	132
[41]	4 (20)	2 (14)	4	62	R	141
[42]	18 (31)	58 (28)	n/a	67	R	204
[43]	n/a	10 (4)	5	63	n/a	248
[44]	n/a	34 (54)	3	67	P	63
[45]	12 (57)	21 (21.6)	3	66.9	R	97
[46]	31 (53)	58 (60)	2	n/a	P	92
[47]	4 (57)	7 (14)	3–5	62.5	n/a	50
[48]	(31.6)	(33.4)	3–5	n/a	R	n/a
[49]	33 (29)	113 (48)	5	69	P	235
Present	36 (40)	90 (21.2)	2–3	59	P	425

A relevant question is which are clinically significant lesions among incidentally discovered prostatic cancers. Two main variables considered, i.e. tumour size ≥0.5 mL [26] and a Gleason score of ≥5 [50]. Based on these criteria, there were clinically significant lesions among an important subpopulation of incidentally discovered lesions (18–62%). Ruffion et al.[39] reported a mean PSA level of 2.84 ng/mL in the group with incidental prostate cancer, compared with 2.91 ng/mL in the group without prostate cancer. In the present study, the PSA levels did not indicate the presence of an incidental lesion. Furthermore, among patients with incidental pathology, the PSA level did not reflect the Gleason score or the pT stage of the prostate cancer. These findings undermine the predictive capability of preoperative PSA measurements in such cases. In addition, high-grade PIN was detected in the pathology records of 43 patients (10.1%). The natural history of such cases is enigmatic and not precisely determined. However, its presence indicates a significant risk factor for developing prostatic carcinoma [51].

The reported incidence of prostatic urothelial cancer was 12–48% (Table 6) [34,49,52–62]. Prostatic involvement might be as a result of direct invasion of the primary tumour (contiguous), or there is no continuity between the primary tumour and the prostatic urothelial tumour (non-contiguous). In our experience most prostatic involvement was a result of direct invasion. However, of these, 31 patients (80%) were understaged before surgery. The probability of prostatic involvement was also higher among TCC bladder tumours; and again, even non-TCC bladder tumours might be associated with concomitant urothelial carcinoma of the prostate.

Table 6.  The incidence of TCC in cystoprostatectomy specimens

Ref.	Study design	No. of cases	TCC N (%)		Total
			Contiguous	Non-contiguous
R, retrospective; P, prospective.
[52]	R	300	–	–	42 (14)
[53]	R	84	–	–	36 (43)
[54]	R	209	29	20	49 (23)
[55]	R	570	44	28	72 (13)
[56]	R	489	19	124	143 (29.2)
[57]	R	70	3	11	14 (20)
[58]	R	81	–	–	15 (18.5)
[59]	R	192	11	19	30 (15.5)
[60]	R	283	18	58	76 (26.8)
[34]	P	121	13	45	58 (48)
[61]	R	768	51	78	129 (17)
[62]	R	214	39	30	69 (32)
[49]	P	235	49	28	77 (33)
Present	P	425	39	27	66 (14.8)

A high incidence of benign prostatic changes are expected among men of this age group. Can BPH compromise the function of the neobladder? Patients with a neobladder empty their reservoir by increasing their intra-abdominal pressure. Would this be adequate as a driving force to empty their reservoir in the face of an enlarging prostate, or will these patients, in the long-term, need intermittent catheterization?

The functional outcomes after prostate-sparing surgery are shown in Table 7. Day-time continence rates were 59–100%, and night-time continence 31–100%. The reported rate of maintained erectile function was 70–100%. By comparison, the continence rates after radical cystoprostatectomy and orthotopic diversion compiled from large series with an adequate follow-up are given in Table 8[63,64]. The reported data show a marginal improvement that is not statistically significant, if the prostate was spared.

Table 7.  Prostate-sparing cystectomy

Method of prostatic sparing	Ref	No. of patients	Mean age, years	Functional outcome, n (%) or n
				Continence, day	Continence, night	Potency preserved
* One patient required clean intermittent catheterization (CIC); † †Five patients required CIC; ‡ 13 patients required CIC and/or TURP.
TURP	[4]	100	64	86 (97)	84 (95)	50 (82)
	[15]	27	52	27 (100)	27 (100)	27 (100)
	[16]	34	61	30 (90)	29 (85)	26 (90)
Adenectomy	[3]*	61	49	58 (95)	19 (31)	58 (95)
	[17]†	28	51	16 (59.2)	16 (59.2)	26 (92.8)
Total sparing	[5]	10	55	10	8	7
	[18]‡	20	39	20 (100)	20 (100)	40 (88)
Apical sparing	[19]	23	n/a	23 (100)	9 (40)	11 (50)
	[20]	4	26	4	4	4
	[21]	31	63	28 (90)	28 (90)	27 (87)

Table 8.  Continence after orthotopic bladder substitution

Ref.	No. of patients	Mean age, years	Follow-up, months	Continence, %
				Day	Night
[12]	290	63	57	83.7	66.3
[63]	166	62	36	98.6	79.7
[11]	344	47	38	93.3	80.0
[64]	83	63.1	12–48	92.0	82.0
[14]	169	69	33	87.0	72.0

Finally, erectile function could also be maintained after radical cystoprostatectomy with nerve-sparing; values of up to 60% for sexual functional recovery were reported if the nerve sparing was bilateral [65].

In conclusion, the incidence of prostatic pathology in cystoprostatectomy specimens was reviewed and compared with the present results. The functional outcome of orthotopic substitution of the bladder after prostate sparing and conventional radical cystectomy was also assessed and compared. The marginal reported functional advantage of prostate-sparing if weighed against the potential oncological risks, raises the question of whether it is worthwhile. We think the potential oncological risks of prostate-sparing cystectomy outweigh any small and possible functional benefits. Properly conducted radical cystoprostatectomy with orthotopic substitution should remain the standard for treating invasive bladder tumour.

CONFLICT OF INTEREST

None declared.