- Email: [email protected]
Residual Benign Prostatic Glands at the Urethrovesical Anastomosis after Radical Retropubic Prostatectomy: Prediction and Impact on Disease Outcome
European Urology
European Urology 46 (2004) 321–326
Residual Benign Prostatic Glands at the Urethrovesical Anastomosis after Radical Retropubic Prostatectomy: Prediction and Impact on Disease Outcome Roger Paul*, Mathias Hoppmann, Heiner van Randenborgh, Hubert Ku¨bler, Michael Alschibaja, Markus Gu¨nther, Rudolf Hartung Department of Urology, Technische Universita¨t Munich, Klinikum rechts der Isar, Ismaninger Str. 22, 81675 Munich, Germany Accepted 27 April 2004 Available online 18 May 2004
Abstract Objective: Biochemical failure after radical prostatectomy (RP) for localized prostate cancer (PC) is the first evidence of disease recurrence. If residual benign prostatic glands are left behind on RP a theoretical PSA production from benign glands or residual neoplastic tissue could explain PSA failure. This study investigates the prediction and impact on disease outcome of residual benign glands at the urethrovesical anastomosis. Material and methods: 802 patients who underwent RP were retrospectively evaluated with special focus on residual benign glands (Bþ) at the urethrovesical anastomosis. B-status was defined from a biopsy of the urethral stump at 9, 12 and 3 o’clock position. Results: From 802 patients 73.6% were classified as Bþ, 26.4% B0. 92.0% of Bþ patients demonstrated only isolated glands (B1), 8.0% showed abundant glands (B2). There was no difference in disease outcome for B0 and Bþ patients. Patients with early PC who are candidates for nerve sparing procedures are more likely for Bþ status. Conclusions: Benign prostatic glands at the apical margin of the RP specimen are a common finding, but neither isolated nor abundant glands have an impact on disease outcome. We think that a precise apical dissection to improve continence rates is possible, although these patients are at risk for residual benign tissue at the apex. # 2004 Elsevier B.V. All rights reserved. Keywords: Prostate cancer; Radical prostatectomy; Prognosis; Surgical margins; Benign prostatic glands
1. Introduction PSA recurrence after radical prostatectomy for localized prostate cancer is a surrogate parameter for tumor recurrence. Unfortunately a significant subset of patients will experience disease progression after radical prostatectomy [1–3]. Theoretically residual benign prostatic tissue could explain elevated PSA levels in serum in a significant subset of patients as suggested by Djavan et al. [4]. On the other hand residual benign prostatic tissue could also mean that radical surgery is compromised, and *
Corresponding author. Tel. þ49-89-4140-2507; Fax: þ49-89-4140-2585. E-mail address: [email protected] (R. Paul).
therefore tumor recurrence could be more often found in patients with residual benign prostatic glands. In this meaning Ravery [5] concludes that an elevated PSA level characterizes biochemical failure after radical prostatectomy rather than residual benign prostatic glands. We evaluated in a retrospective analysis the impact of this residual benign tissue on disease outcome and tried to identify patients at risk for residual benign glands at the urethrovesical anastomosis.
2. Materials and methods We evaluated in a retrospective analysis n ¼ 1247 patients who underwent a radical retropubic prostatectomy from 1986 to 2003 at our institution for localized prostate cancer after the anatomical
0302-2838/$ – see front matter # 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.eururo.2004.04.023
322
R. Paul et al. / European Urology 46 (2004) 321–326
method described by Walsh et al. [6]. The careful apical dissection was carried out preparing a long urethral stump. After incision of the urethra we obtained three small specimens from the urethral stump at the 9, 12 and 3 o’clock position. These specimens were sent for frozen section to evaluate neoplastic surgical margins at the apex. Additionally these specimens were analyzed regarding benign glands (B) on frozen sections and definite pathology. The amount of benign glands found in these biopsy specimens was categorized in three categories: no benign glands (B0), isolated glands (B1) or abundant glands (B2). From these 1247 patients altogether 802 patients were evaluable in our retrospective analysis. From all 802 patients a recent followup was obtained, with special focus on postoperative PSA values, local recurrence and survival. We analyzed the influence of benign glands at the urethrovesical anastomosis on PSA-free survival— defined as elevated serum PSA levels above the cut-off value of 0.5 ng/ml). We also looked for local recurrence-free survival, defined as either histological proven local recurrence or in case of suspicion of local recurrence and local radiotherapy, if decreasing PSA values after radiation were noted. For metastases-free
survival an event was defined, if bone or visceral metastases were detected by bone scan or computed tomography. Disease-specific and overall survival was also calculated. Statistical analysis was carried out to test for differences in the subgroups using the w2-test for categorical parameters and the Mann–Whitney U and Kruskal–Wallis test for continuous variables. Kaplan–Meier analysis and Log-rank test were performed to evaluate disease outcome. Statistical significance was accepted at 5% ( p ¼ 0:05).
3. Results 3.1. Identification of patients with benign prostatic glands at the urethrovesical anastomosis Mean follow up of the 802 patients is 48.3 months with a range of 0.6–187.5 months. Mean age at the time of operation was 64.5 years. There was no statistical
Table 1 Patient characteristics stratified after status of benign glands (B) at the urethrovesical anastomosis Parameter
B0
Bþ
p
Bþ
p
isolated glands
abundant glands
n n (%) mean-age (yrs) median PSA (ng/ml) median PSA Density mean prostate size (ml)
212 26.4 64.8 9.4 0.158 62.5
590 73.6 64.5 7.8 0.121 65.9
nd nd p ¼ 0.57 p ¼ 0.002 p < 0.001 p ¼ 0.07
543 92.0 64.5 7.8 0.121 65.8
47 8.0 64.3 8.0 0.110 66.6
nd nd p ¼ 0.59 p ¼ 0.94 p ¼ 0.48 p ¼ 0.68
DRE Not suspicious Suspicious
31.4 68.6
34.5 65.5
p ¼ 0.42
34.1 65.9
38.3 61.7
p ¼ 0.56
TRUS Not suspicious Suspicious
23.3 76.7
27.7 72.3
p ¼ 0.22
27.9 72.1
25.0 75.0
p ¼ 0.68
pT-stage pT2b pT3a pT3b
61.2 19.6 19.2
72.4 17.1 10.4
p ¼ 0.0014
74.0 15.5 10.5
54.9 35.3 9.8
p ¼ 0.0014
pN-stage pN0 pNþ
94.6 5.4
96.0 4.0
p ¼ 0.41
96.1 3.9
95.1 4.9
p ¼ 0.76
Grading G1–2 G3–4
71.7 28.3
76.0 24.0
p ¼ 0.21
76.9 23.1
66.7 33.3
p ¼ 0.10
Gleason score 6 7 8
61.0 23.3 15.7
66.2 20.1 13.7
p ¼ 0.47
66.3 20.0 13.6
64.3 21.4 14.3
p ¼ 0.07
R-Status R0 R1
70.1 29.9
84.2 15.8
p < 0.0001
85.0 15.0
74.5 25.5
p ¼ 0.0735
Nerve sparing RP Without ns With ns
63.3 36.7
53.4 46.6
p ¼ 0.0097
53.0 47.0
57.7 42.3
p ¼ 0.52
Statistical significant differences were investigated by Mann–Whitney U and Kruskal–Wallis test for continuous and w2-test for categorical variables.
R. Paul et al. / European Urology 46 (2004) 321–326
difference between B0 and Bþ patients. The characteristics of B0 and Bþ, as well as B1 and B2 patients are summarized in Table 1. Median preoperative PSA of overall patients is 8.8 ng/ml, there were statistically significant differences between B0 (9.4 ng/ml) and Bþ patients (7.8 ng/ml). A similar result was obtained for PSA density. The only differences between B0 and Bþ patients, except PSA and PSA density, were pT-stage ( p ¼ 0:0014), R-status ( p < 0:0001) and the surgical technique regarding nerve-sparing procedure ( p ¼ 0:0097). There was no difference between B0 and Bþ patients looking at the results of DRE and TRUS preoperatively, pN-stage, WHO grading and Gleason score. There was a trend but no a statistical significance regarding prostatic size. For B0 patients the mean prostate volume was 62.5 ml compared to 65.9 ml for Bþ patients. Patients with benign prostatic glands at the urethrovesical anastomosis are characterized in our investigation by tumors with low PSA values preoperatively, organ confined tumors (pT2b), negative surgical margins (R0) with nerve sparing procedures. In summary patients with very early stage tumors and patients who are candidates for a limited surgical procedure will have more often residual benign tissue at the apical margin. 3.2. Influence of benign prostatic glands on disease progression Benign prostatic glands at the urethrovesical anastomosis were not a statistical significant factor in PSAfree survival ( p ¼ 0:62) in our analysis (Fig. 1A). Looking at the subgroup of patients with organ confined prostate cancer (pT2b, p ¼ 0:42) and negative surgical margins (R0, p ¼ 0:81) only (Fig. 1B and C) there was also no statistical difference in PSA-free survival for B0 and Bþ patients. No differences were found for the subgroups of locally advanced tumors (pT3a, pT3b), for patients with highly or poorly differentiated tumors (Gleason score 6, 7 and 8, WHO grading G1–2, G3–4) and lymph node positive and negative patients (data not shown). If we look for differences between B0, B1 and B2 patients in PSA-free survival, the results are identical in that we do not find differences in the biochemical failure rate (Fig. 2A). If we look at local recurrence rate there is a statistical significant difference ( p ¼ 0:047) for B0 and Bþ patients (see Fig. 2B), but not for B0, B1 and B2 patients (Fig. 2c). The calculated 5 and 10-year local recurrence free survival rate was identical with 98.9% for Bþ patients and also for B0 with 96.5% (Fig. 2b). This difference, although statistical significant, is clini-
323
(A)
1 ,8
B0
,6
B+
,4 ,2
Chi-Square
DF
,25 1
0 0
20
P-Value
1 40
60
0 ,62 80
1 00
1 20
1 40
(B)
1
B0
,8 ,6 ,4 ,2
B+ Chi-Square
DF P-Value
,65 1
0 0
20
1 40
0 ,42 60
80
1 00
1 20
1 40
(C)
1 ,8 ,6
B0 B+
,4 ,2
Chi-Square
DF P-Value
,06 0
0 0
20
1 40
0 ,81 60
80
1 00
1 20
1 40
Fig. 1. Kaplan–Meier analysis and Log-rank test for patients with (Bþ) and without (B0) benign prostatic glands at the urethrovesical anastomosis demonstrating no differences in PSA-free survival (A) in all patients, (B) in patients with organ-confined prostate cancer (pT2b) only and (C) in patients with negative surgical margins (R0) only.
cally not relevant and also not intuitive, because we would have expected more local recurrences in Bþ patients. In this respect B2 patients with abundant prostatic glands at the urethrovesical anastomosis show
324
R. Paul et al. / European Urology 46 (2004) 321–326
(A)
1
B2 B0
,8
In summary our results suggest that neither isolated nor abundant benign glands at the apical margin have an impact on PSA-free, local-recurrence-free, metastases-free, disease-specific and overall survival.
,6
B1
,4 ,2 Chi-Square
DF
P-Value
2
0 ,88
,25 3
0 0
20
40
60
80
1 00
1 20
1 40
(B)
1
B0
B+
,8 ,6 ,4 ,2 Chi-Square
DF
3 ,9 38
0 0
20
P-Value
1 40
0 ,04 7 60
80
1 00
(C)
1 20
1 40
B2 1
B0
B1
,8 ,6 ,4 ,2 0 0
20
40
60
80
1 00
1 20
1 40
Fig. 2. Kaplan–Meier analysis and Log-rank test for patients with isolated (B1) or abundant benign prostatic glands (B2) compared to patients without glands (B0) at the urethrovesical anastomosis demonstrating (A) no difference in PSA-free survival and, (B þ C) local recurrence free survival.
no local recurrence at all (Fig. 2c). We also analyzed metastases-free ( p ¼ 0:71), disease-specific ( p ¼ 0:50), and overall survival p ¼ 0:12) of B0 and Bþ patients, but as expected these analyses did not show any statistical significant differences.
4. Discussion Benign prostatic glands left behind after radical prostatectomy can theoretically influence PSA-free survival and may therefore have also an impact on metastases-free, disease-specific and overall survival in prostate cancer patients, as shown by Pound et al. [7]. Theoretically residual benign prostatic tissue could produce enough PSA to elevate PSA levels in serum, especially if a low cut-off of the PSA value is used. On the other hand residual benign prostatic tissue could also mean that radical surgery is compromised, and therefore tumor recurrence could be more often found in patients with residual benign prostatic glands. With our investigation we tried to predict which patients are at risk for residual benign tissue and tried to establish the impact of benign prostatic glands found at the urethrovesical anastomosis after radical retropubic prostatectomy on disease outcome. Our results demonstrate that patients with benign prostatic glands had a statistical significant lower PSAvalue, a lower pT-stage and were operated on more often in a nerve sparing procedure. There was trend but no statistical difference for the prostatic size, which was a little bit higher in Bþ patients compared to B0 patients (65.9 ml vs. 62.5 ml). These results suggest that in early prostate cancer with special focus on potency preservation the risk of leaving benign tissue at the apical margin seems to increase, whereas in patients with advanced disease and non-nerve-sparing procedure the surgeon tends to perform a more ‘‘radical’’ procedure, therefore a reduced rate of benign glands at the apical margins is found. We think this finding can be explained by the length of the urethral stump, which correlates with postoperative continence. Patients with low tumor burden who are candidates for nerve sparing procedures are operated on more carefully trying to keep the urethral stump as long as possible without compromising radicality of the procedure, which was controlled in our institution by the biopsies from the urethral stump. In patients with large tumor burden, or apical tumors the surgeon tends to be a little more ‘‘radical’’ at the apical dissection, therefore the percentage of benign glands is reduced in these patients. The goal of the radical retropubic prostatectomy in localized prostatic neoplasm is cure of cancer;
R. Paul et al. / European Urology 46 (2004) 321–326
therefore, if residual benign tissue is more frequently found in patients with early stage curable tumors the impact of benign prostatic glands on disease outcome has to be established very carefully. It is known that a significant subset of patients will experience disease progression after radical prostatectomy [1–3]. PSA progression after reaching the nadir, which may be defined as low as 0.1 ng/ml after prostatectomy is generally considered biochemical failure [8], which may precede metastases and death of prostate cancer [7]. Several factors like i.e. tumor stage, Gleason score, preoperative PSA value, seminal vesicle invasion and positive surgical margins are well known to influence PSA-free survival [9,10]. However it is unclear if benign prostatic glands will also influence disease outcome. In this respect Djavan et al. [4] reported of benign prostatic glands at the surgical margins in about 27% of all radical prostatectomy specimens (n ¼ 351). They conclude that 79% of all pT2 cancers with benign glands at the surgical margins demonstrated elevated PSA levels postoperatively, which was not related to disease progression. They suggest that these patients may falsely be categorized as biochemical failure. Shah et al. [11] investigated the frequency of benign glands at the surgical margins from 119 patients and described this finding in only 13% of patients. Biochemical failure, which was observed in only two patients, was not related to benign prostatic glands. They conclude that benign glands at the surgical margins would be an unusual cause of PSA progression. The discrepancies regarding the frequency compared to our study may be explained by the analysis of the tissue from the remaining urethral stump. We think this is prove of residual prostatic tissue at the apical margin, in contrary benign glands at the surgical margins may be due to surgical technique, tissue handling and processing and does not necessarily predict residual benign tissue left behind as suggested by Djavan et al. [4] and Shah et al. [11]. We think that residual benign tissue after radical prostatectomy cannot be assessed from the surgical specimen. Wood et al. [12] reported about benign glands at the bladder neck. They also evaluated residual benign tissue from biopsies taken
325
after the prostate has been removed. They report of 7% of only benign glands at the bladder margin if performing a bladder neck sparing surgery. They could demonstrate that benign glands are found more often in bladder neck sparing surgery. Comparing our data we think it is obvious that the bladder neck dissection, even if performed as a bladder neck sparing technique, is more radical than the apical dissection, which is probably due to the poorly defined apical region of the prostate and the close relation to the pelvic floor and external sphincter region. In this respect we found benign glands ten times more often at the distal part of the prostate compared to the results of Wood et al. [12] at the proximal margin of the prostate. Neither isolated nor abundant prostatic glands at the apical margin were predictive of PSA failure in our analysis. Therefore we cannot agree with the finding of Djavan et al. [4] that PSA recurrence can be attributed in a significant subset of patients due to benign prostatic glands. These results are confirmed by Shah et al., who could also not detect a correlation of benign glands at the surgical margins with PSA failure. Ravery et al. [5] investigated the PSA doubling time after radical prostatectomy and they conclude from these data that PSA recurrence must be attributed to neoplastic tissue and only in very few cases to benign tissue.
5. Conclusion Patients with early stage localized prostate cancer are at higher risk for residual benign prostatic tissue at the urethrovesical anastomosis. The reason is a subtler than radical surgery associated with nerve sparing technique at the prostatic apex and/or bladder neck sparing procedures at the bladder margin. Neither isolated nor abundant glands are associated with an unfavorable prognosis in prostate cancer. Our analysis demonstrates that residual benign prostate tissue has no impact on PSA-free survival, local-recurrence-free survival, overall- and disease-free survival, therefore a very subtle surgical technique of nerve sparing technique is feasible without compromising disease outcome in prostate cancer.
References [1] Augustin H, Hammerer PG. Disease recurrence after radical prostatectomy. Contemporary diagnostic and therapeutical strategies. Minerva Urol Nefrol 2003;55:251–61. [2] Nelson JB, Lepor H. Prostate cancer: radical prostatectomy. Urol Clin North Am 2003;30:703–23.
[3] Zhang Y, Glass A, Bennett N, Oyama KA, Gehan E, Gelmann EP. Longterm outcomes after radical prostatectomy performed in a communitybased health maintenance organization. Cancer 2004;100:300–7. [4] Djavan B, Sesterhann I, Hruby St, Susani M, Haitel A, Etemad, Mostofi FK, McLeod DG, Marberger M. Benign prostatic glands in
326
[5]
[6]
[7]
[8]
R. Paul et al. / European Urology 46 (2004) 321–326 the surgical margin of radical retropubic Prostatectomies: redefining PSA nadir. J Urol 2000;163A:624. Ravery V. The significance of recurrent PSA after radical prostatectomy: benign versus malignant sources. Semin Urol Oncol 1999;17:127–9. Walsh PC, Lepor H, Eggleston JC. Radical prostatectomy with preservation of sexual function: anatomical and pathological considerations. Prostate 1983;4:473–85. Pound CR, Partin AW, Eisenberger MA, Chan DW, Pearson JD, Walsh PC. Natural history of progression after PSA elevation following radical prostatectomy. JAMA 1999;281:1591–7. Ravery V, Hermieu JF, Hoffmann P, Delmas V, Boccon-Gibod L. Post-treatment PSA, indicator of radical treatment effectiveness of localized cancer of the prostate. Prog Urol 1996;6: 981–6.
[9] Blute ML, Bergstralh EJ, Iocca A, Scherer B, Zincke H. Use of Gleason score, prostate specific antigen, seminal vesicle and margin status to predict biochemical failure after radical prostatectomy. J Urol 2001;165:119–215. [10] Salomon L, Levrel O, Anastasiadis AG, Irani J, De La Taille A, Saint F, et al. Prognostic significance of tumor volume after radical prostatectomy: a multivariate analysis of pathological prognostic factors. Eur Urol 2003;43:39–44. [11] Shah R, Bassily N, Wei J, Mucci NR, Montie JE, Sanda MG, et al. Benign prostatic glands at surgical margins of radical prostatectomy specimens: frequency and associated risk factors. Urology 2000;56:721–5. [12] Wood DP, Peretsman SJ, Seay TM. Incidence of benign and malignant prostate tissue in biopsies of the bladder neck after a radical prostatectomy. J Urol 1995;154:1443–6.
European Urology 46 (2004) 321–326
Residual Benign Prostatic Glands at the Urethrovesical Anastomosis after Radical Retropubic Prostatectomy: Prediction and Impact on Disease Outcome Roger Paul*, Mathias Hoppmann, Heiner van Randenborgh, Hubert Ku¨bler, Michael Alschibaja, Markus Gu¨nther, Rudolf Hartung Department of Urology, Technische Universita¨t Munich, Klinikum rechts der Isar, Ismaninger Str. 22, 81675 Munich, Germany Accepted 27 April 2004 Available online 18 May 2004
Abstract Objective: Biochemical failure after radical prostatectomy (RP) for localized prostate cancer (PC) is the first evidence of disease recurrence. If residual benign prostatic glands are left behind on RP a theoretical PSA production from benign glands or residual neoplastic tissue could explain PSA failure. This study investigates the prediction and impact on disease outcome of residual benign glands at the urethrovesical anastomosis. Material and methods: 802 patients who underwent RP were retrospectively evaluated with special focus on residual benign glands (Bþ) at the urethrovesical anastomosis. B-status was defined from a biopsy of the urethral stump at 9, 12 and 3 o’clock position. Results: From 802 patients 73.6% were classified as Bþ, 26.4% B0. 92.0% of Bþ patients demonstrated only isolated glands (B1), 8.0% showed abundant glands (B2). There was no difference in disease outcome for B0 and Bþ patients. Patients with early PC who are candidates for nerve sparing procedures are more likely for Bþ status. Conclusions: Benign prostatic glands at the apical margin of the RP specimen are a common finding, but neither isolated nor abundant glands have an impact on disease outcome. We think that a precise apical dissection to improve continence rates is possible, although these patients are at risk for residual benign tissue at the apex. # 2004 Elsevier B.V. All rights reserved. Keywords: Prostate cancer; Radical prostatectomy; Prognosis; Surgical margins; Benign prostatic glands
1. Introduction PSA recurrence after radical prostatectomy for localized prostate cancer is a surrogate parameter for tumor recurrence. Unfortunately a significant subset of patients will experience disease progression after radical prostatectomy [1–3]. Theoretically residual benign prostatic tissue could explain elevated PSA levels in serum in a significant subset of patients as suggested by Djavan et al. [4]. On the other hand residual benign prostatic tissue could also mean that radical surgery is compromised, and *
Corresponding author. Tel. þ49-89-4140-2507; Fax: þ49-89-4140-2585. E-mail address: [email protected] (R. Paul).
therefore tumor recurrence could be more often found in patients with residual benign prostatic glands. In this meaning Ravery [5] concludes that an elevated PSA level characterizes biochemical failure after radical prostatectomy rather than residual benign prostatic glands. We evaluated in a retrospective analysis the impact of this residual benign tissue on disease outcome and tried to identify patients at risk for residual benign glands at the urethrovesical anastomosis.
2. Materials and methods We evaluated in a retrospective analysis n ¼ 1247 patients who underwent a radical retropubic prostatectomy from 1986 to 2003 at our institution for localized prostate cancer after the anatomical
0302-2838/$ – see front matter # 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.eururo.2004.04.023
322
R. Paul et al. / European Urology 46 (2004) 321–326
method described by Walsh et al. [6]. The careful apical dissection was carried out preparing a long urethral stump. After incision of the urethra we obtained three small specimens from the urethral stump at the 9, 12 and 3 o’clock position. These specimens were sent for frozen section to evaluate neoplastic surgical margins at the apex. Additionally these specimens were analyzed regarding benign glands (B) on frozen sections and definite pathology. The amount of benign glands found in these biopsy specimens was categorized in three categories: no benign glands (B0), isolated glands (B1) or abundant glands (B2). From these 1247 patients altogether 802 patients were evaluable in our retrospective analysis. From all 802 patients a recent followup was obtained, with special focus on postoperative PSA values, local recurrence and survival. We analyzed the influence of benign glands at the urethrovesical anastomosis on PSA-free survival— defined as elevated serum PSA levels above the cut-off value of 0.5 ng/ml). We also looked for local recurrence-free survival, defined as either histological proven local recurrence or in case of suspicion of local recurrence and local radiotherapy, if decreasing PSA values after radiation were noted. For metastases-free
survival an event was defined, if bone or visceral metastases were detected by bone scan or computed tomography. Disease-specific and overall survival was also calculated. Statistical analysis was carried out to test for differences in the subgroups using the w2-test for categorical parameters and the Mann–Whitney U and Kruskal–Wallis test for continuous variables. Kaplan–Meier analysis and Log-rank test were performed to evaluate disease outcome. Statistical significance was accepted at 5% ( p ¼ 0:05).
3. Results 3.1. Identification of patients with benign prostatic glands at the urethrovesical anastomosis Mean follow up of the 802 patients is 48.3 months with a range of 0.6–187.5 months. Mean age at the time of operation was 64.5 years. There was no statistical
Table 1 Patient characteristics stratified after status of benign glands (B) at the urethrovesical anastomosis Parameter
B0
Bþ
p
Bþ
p
isolated glands
abundant glands
n n (%) mean-age (yrs) median PSA (ng/ml) median PSA Density mean prostate size (ml)
212 26.4 64.8 9.4 0.158 62.5
590 73.6 64.5 7.8 0.121 65.9
nd nd p ¼ 0.57 p ¼ 0.002 p < 0.001 p ¼ 0.07
543 92.0 64.5 7.8 0.121 65.8
47 8.0 64.3 8.0 0.110 66.6
nd nd p ¼ 0.59 p ¼ 0.94 p ¼ 0.48 p ¼ 0.68
DRE Not suspicious Suspicious
31.4 68.6
34.5 65.5
p ¼ 0.42
34.1 65.9
38.3 61.7
p ¼ 0.56
TRUS Not suspicious Suspicious
23.3 76.7
27.7 72.3
p ¼ 0.22
27.9 72.1
25.0 75.0
p ¼ 0.68
pT-stage pT2b pT3a pT3b
61.2 19.6 19.2
72.4 17.1 10.4
p ¼ 0.0014
74.0 15.5 10.5
54.9 35.3 9.8
p ¼ 0.0014
pN-stage pN0 pNþ
94.6 5.4
96.0 4.0
p ¼ 0.41
96.1 3.9
95.1 4.9
p ¼ 0.76
Grading G1–2 G3–4
71.7 28.3
76.0 24.0
p ¼ 0.21
76.9 23.1
66.7 33.3
p ¼ 0.10
Gleason score 6 7 8
61.0 23.3 15.7
66.2 20.1 13.7
p ¼ 0.47
66.3 20.0 13.6
64.3 21.4 14.3
p ¼ 0.07
R-Status R0 R1
70.1 29.9
84.2 15.8
p < 0.0001
85.0 15.0
74.5 25.5
p ¼ 0.0735
Nerve sparing RP Without ns With ns
63.3 36.7
53.4 46.6
p ¼ 0.0097
53.0 47.0
57.7 42.3
p ¼ 0.52
Statistical significant differences were investigated by Mann–Whitney U and Kruskal–Wallis test for continuous and w2-test for categorical variables.
R. Paul et al. / European Urology 46 (2004) 321–326
difference between B0 and Bþ patients. The characteristics of B0 and Bþ, as well as B1 and B2 patients are summarized in Table 1. Median preoperative PSA of overall patients is 8.8 ng/ml, there were statistically significant differences between B0 (9.4 ng/ml) and Bþ patients (7.8 ng/ml). A similar result was obtained for PSA density. The only differences between B0 and Bþ patients, except PSA and PSA density, were pT-stage ( p ¼ 0:0014), R-status ( p < 0:0001) and the surgical technique regarding nerve-sparing procedure ( p ¼ 0:0097). There was no difference between B0 and Bþ patients looking at the results of DRE and TRUS preoperatively, pN-stage, WHO grading and Gleason score. There was a trend but no a statistical significance regarding prostatic size. For B0 patients the mean prostate volume was 62.5 ml compared to 65.9 ml for Bþ patients. Patients with benign prostatic glands at the urethrovesical anastomosis are characterized in our investigation by tumors with low PSA values preoperatively, organ confined tumors (pT2b), negative surgical margins (R0) with nerve sparing procedures. In summary patients with very early stage tumors and patients who are candidates for a limited surgical procedure will have more often residual benign tissue at the apical margin. 3.2. Influence of benign prostatic glands on disease progression Benign prostatic glands at the urethrovesical anastomosis were not a statistical significant factor in PSAfree survival ( p ¼ 0:62) in our analysis (Fig. 1A). Looking at the subgroup of patients with organ confined prostate cancer (pT2b, p ¼ 0:42) and negative surgical margins (R0, p ¼ 0:81) only (Fig. 1B and C) there was also no statistical difference in PSA-free survival for B0 and Bþ patients. No differences were found for the subgroups of locally advanced tumors (pT3a, pT3b), for patients with highly or poorly differentiated tumors (Gleason score 6, 7 and 8, WHO grading G1–2, G3–4) and lymph node positive and negative patients (data not shown). If we look for differences between B0, B1 and B2 patients in PSA-free survival, the results are identical in that we do not find differences in the biochemical failure rate (Fig. 2A). If we look at local recurrence rate there is a statistical significant difference ( p ¼ 0:047) for B0 and Bþ patients (see Fig. 2B), but not for B0, B1 and B2 patients (Fig. 2c). The calculated 5 and 10-year local recurrence free survival rate was identical with 98.9% for Bþ patients and also for B0 with 96.5% (Fig. 2b). This difference, although statistical significant, is clini-
323
(A)
1 ,8
B0
,6
B+
,4 ,2
Chi-Square
DF
,25 1
0 0
20
P-Value
1 40
60
0 ,62 80
1 00
1 20
1 40
(B)
1
B0
,8 ,6 ,4 ,2
B+ Chi-Square
DF P-Value
,65 1
0 0
20
1 40
0 ,42 60
80
1 00
1 20
1 40
(C)
1 ,8 ,6
B0 B+
,4 ,2
Chi-Square
DF P-Value
,06 0
0 0
20
1 40
0 ,81 60
80
1 00
1 20
1 40
Fig. 1. Kaplan–Meier analysis and Log-rank test for patients with (Bþ) and without (B0) benign prostatic glands at the urethrovesical anastomosis demonstrating no differences in PSA-free survival (A) in all patients, (B) in patients with organ-confined prostate cancer (pT2b) only and (C) in patients with negative surgical margins (R0) only.
cally not relevant and also not intuitive, because we would have expected more local recurrences in Bþ patients. In this respect B2 patients with abundant prostatic glands at the urethrovesical anastomosis show
324
R. Paul et al. / European Urology 46 (2004) 321–326
(A)
1
B2 B0
,8
In summary our results suggest that neither isolated nor abundant benign glands at the apical margin have an impact on PSA-free, local-recurrence-free, metastases-free, disease-specific and overall survival.
,6
B1
,4 ,2 Chi-Square
DF
P-Value
2
0 ,88
,25 3
0 0
20
40
60
80
1 00
1 20
1 40
(B)
1
B0
B+
,8 ,6 ,4 ,2 Chi-Square
DF
3 ,9 38
0 0
20
P-Value
1 40
0 ,04 7 60
80
1 00
(C)
1 20
1 40
B2 1
B0
B1
,8 ,6 ,4 ,2 0 0
20
40
60
80
1 00
1 20
1 40
Fig. 2. Kaplan–Meier analysis and Log-rank test for patients with isolated (B1) or abundant benign prostatic glands (B2) compared to patients without glands (B0) at the urethrovesical anastomosis demonstrating (A) no difference in PSA-free survival and, (B þ C) local recurrence free survival.
no local recurrence at all (Fig. 2c). We also analyzed metastases-free ( p ¼ 0:71), disease-specific ( p ¼ 0:50), and overall survival p ¼ 0:12) of B0 and Bþ patients, but as expected these analyses did not show any statistical significant differences.
4. Discussion Benign prostatic glands left behind after radical prostatectomy can theoretically influence PSA-free survival and may therefore have also an impact on metastases-free, disease-specific and overall survival in prostate cancer patients, as shown by Pound et al. [7]. Theoretically residual benign prostatic tissue could produce enough PSA to elevate PSA levels in serum, especially if a low cut-off of the PSA value is used. On the other hand residual benign prostatic tissue could also mean that radical surgery is compromised, and therefore tumor recurrence could be more often found in patients with residual benign prostatic glands. With our investigation we tried to predict which patients are at risk for residual benign tissue and tried to establish the impact of benign prostatic glands found at the urethrovesical anastomosis after radical retropubic prostatectomy on disease outcome. Our results demonstrate that patients with benign prostatic glands had a statistical significant lower PSAvalue, a lower pT-stage and were operated on more often in a nerve sparing procedure. There was trend but no statistical difference for the prostatic size, which was a little bit higher in Bþ patients compared to B0 patients (65.9 ml vs. 62.5 ml). These results suggest that in early prostate cancer with special focus on potency preservation the risk of leaving benign tissue at the apical margin seems to increase, whereas in patients with advanced disease and non-nerve-sparing procedure the surgeon tends to perform a more ‘‘radical’’ procedure, therefore a reduced rate of benign glands at the apical margins is found. We think this finding can be explained by the length of the urethral stump, which correlates with postoperative continence. Patients with low tumor burden who are candidates for nerve sparing procedures are operated on more carefully trying to keep the urethral stump as long as possible without compromising radicality of the procedure, which was controlled in our institution by the biopsies from the urethral stump. In patients with large tumor burden, or apical tumors the surgeon tends to be a little more ‘‘radical’’ at the apical dissection, therefore the percentage of benign glands is reduced in these patients. The goal of the radical retropubic prostatectomy in localized prostatic neoplasm is cure of cancer;
R. Paul et al. / European Urology 46 (2004) 321–326
therefore, if residual benign tissue is more frequently found in patients with early stage curable tumors the impact of benign prostatic glands on disease outcome has to be established very carefully. It is known that a significant subset of patients will experience disease progression after radical prostatectomy [1–3]. PSA progression after reaching the nadir, which may be defined as low as 0.1 ng/ml after prostatectomy is generally considered biochemical failure [8], which may precede metastases and death of prostate cancer [7]. Several factors like i.e. tumor stage, Gleason score, preoperative PSA value, seminal vesicle invasion and positive surgical margins are well known to influence PSA-free survival [9,10]. However it is unclear if benign prostatic glands will also influence disease outcome. In this respect Djavan et al. [4] reported of benign prostatic glands at the surgical margins in about 27% of all radical prostatectomy specimens (n ¼ 351). They conclude that 79% of all pT2 cancers with benign glands at the surgical margins demonstrated elevated PSA levels postoperatively, which was not related to disease progression. They suggest that these patients may falsely be categorized as biochemical failure. Shah et al. [11] investigated the frequency of benign glands at the surgical margins from 119 patients and described this finding in only 13% of patients. Biochemical failure, which was observed in only two patients, was not related to benign prostatic glands. They conclude that benign glands at the surgical margins would be an unusual cause of PSA progression. The discrepancies regarding the frequency compared to our study may be explained by the analysis of the tissue from the remaining urethral stump. We think this is prove of residual prostatic tissue at the apical margin, in contrary benign glands at the surgical margins may be due to surgical technique, tissue handling and processing and does not necessarily predict residual benign tissue left behind as suggested by Djavan et al. [4] and Shah et al. [11]. We think that residual benign tissue after radical prostatectomy cannot be assessed from the surgical specimen. Wood et al. [12] reported about benign glands at the bladder neck. They also evaluated residual benign tissue from biopsies taken
325
after the prostate has been removed. They report of 7% of only benign glands at the bladder margin if performing a bladder neck sparing surgery. They could demonstrate that benign glands are found more often in bladder neck sparing surgery. Comparing our data we think it is obvious that the bladder neck dissection, even if performed as a bladder neck sparing technique, is more radical than the apical dissection, which is probably due to the poorly defined apical region of the prostate and the close relation to the pelvic floor and external sphincter region. In this respect we found benign glands ten times more often at the distal part of the prostate compared to the results of Wood et al. [12] at the proximal margin of the prostate. Neither isolated nor abundant prostatic glands at the apical margin were predictive of PSA failure in our analysis. Therefore we cannot agree with the finding of Djavan et al. [4] that PSA recurrence can be attributed in a significant subset of patients due to benign prostatic glands. These results are confirmed by Shah et al., who could also not detect a correlation of benign glands at the surgical margins with PSA failure. Ravery et al. [5] investigated the PSA doubling time after radical prostatectomy and they conclude from these data that PSA recurrence must be attributed to neoplastic tissue and only in very few cases to benign tissue.
5. Conclusion Patients with early stage localized prostate cancer are at higher risk for residual benign prostatic tissue at the urethrovesical anastomosis. The reason is a subtler than radical surgery associated with nerve sparing technique at the prostatic apex and/or bladder neck sparing procedures at the bladder margin. Neither isolated nor abundant glands are associated with an unfavorable prognosis in prostate cancer. Our analysis demonstrates that residual benign prostate tissue has no impact on PSA-free survival, local-recurrence-free survival, overall- and disease-free survival, therefore a very subtle surgical technique of nerve sparing technique is feasible without compromising disease outcome in prostate cancer.
References [1] Augustin H, Hammerer PG. Disease recurrence after radical prostatectomy. Contemporary diagnostic and therapeutical strategies. Minerva Urol Nefrol 2003;55:251–61. [2] Nelson JB, Lepor H. Prostate cancer: radical prostatectomy. Urol Clin North Am 2003;30:703–23.
[3] Zhang Y, Glass A, Bennett N, Oyama KA, Gehan E, Gelmann EP. Longterm outcomes after radical prostatectomy performed in a communitybased health maintenance organization. Cancer 2004;100:300–7. [4] Djavan B, Sesterhann I, Hruby St, Susani M, Haitel A, Etemad, Mostofi FK, McLeod DG, Marberger M. Benign prostatic glands in
326
[5]
[6]
[7]
[8]
R. Paul et al. / European Urology 46 (2004) 321–326 the surgical margin of radical retropubic Prostatectomies: redefining PSA nadir. J Urol 2000;163A:624. Ravery V. The significance of recurrent PSA after radical prostatectomy: benign versus malignant sources. Semin Urol Oncol 1999;17:127–9. Walsh PC, Lepor H, Eggleston JC. Radical prostatectomy with preservation of sexual function: anatomical and pathological considerations. Prostate 1983;4:473–85. Pound CR, Partin AW, Eisenberger MA, Chan DW, Pearson JD, Walsh PC. Natural history of progression after PSA elevation following radical prostatectomy. JAMA 1999;281:1591–7. Ravery V, Hermieu JF, Hoffmann P, Delmas V, Boccon-Gibod L. Post-treatment PSA, indicator of radical treatment effectiveness of localized cancer of the prostate. Prog Urol 1996;6: 981–6.
[9] Blute ML, Bergstralh EJ, Iocca A, Scherer B, Zincke H. Use of Gleason score, prostate specific antigen, seminal vesicle and margin status to predict biochemical failure after radical prostatectomy. J Urol 2001;165:119–215. [10] Salomon L, Levrel O, Anastasiadis AG, Irani J, De La Taille A, Saint F, et al. Prognostic significance of tumor volume after radical prostatectomy: a multivariate analysis of pathological prognostic factors. Eur Urol 2003;43:39–44. [11] Shah R, Bassily N, Wei J, Mucci NR, Montie JE, Sanda MG, et al. Benign prostatic glands at surgical margins of radical prostatectomy specimens: frequency and associated risk factors. Urology 2000;56:721–5. [12] Wood DP, Peretsman SJ, Seay TM. Incidence of benign and malignant prostate tissue in biopsies of the bladder neck after a radical prostatectomy. J Urol 1995;154:1443–6.