Jonas Busch, Charité University Medicine Berlin, Department of Urology, Charitéplatz 1, 10117 Berlin, Germany. e-mail: email@example.com
Study Type – Therapy (case series)
Level of Evidence 4
What's known on the subject? and What does the study add?
Over the past decade, minimally invasive laparoscopic radical prostatectomy and more recently robot-assisted laparoscopic prostatectomy have been introduced and have proven equally effective compared with open surgery in terms of mid-term cancer control and complication rates. Because long-term data is lacking, open prostatectomy is still considered the ‘gold standard’ by some authors, who argue that minimally invasive approaches have to measure up to the excellent long-term results of open surgery.
This study represents one of the largest series (1845 patients) of minimally invasive radical prostatectomy with extended follow-up (11.3 years) and detailed data on oncological outcome and postoperative incontinence. It therefore supplies previously lacking information on these details for minimally invasive prostate surgery and provides important information for patient counselling.
• To investigate biochemical recurrence (BCR) rates and data on postoperative incontinence in a large laparoscopic radical prostatectomy (LRP) cohort with extended follow-up.
MATERIALS AND METHODS
• BCR and independent predictors of BCR were identified using Kaplan–Meier and Cox regression analysis of 1845 patients who underwent LRP from 1999 to 2007.
• Urinary incontinence was evaluated by pads per day and stratified as follows: 0–1 pad: no incontinence; 2–3 pads: mild incontinence; and ≥3 pads: severe incontinence.
• Organ-confined disease, extraprostatic extension, seminal vesicle invasion and lymph node metastasis were present in 71.3%, 20.5%, 6.7% and 3.2% of patients, respectively. The positive surgical margin rate was 29.2%.
• Postoperatively, 74.9% of the patients were continent, while 9.2% had mild and 15.9% severe incontinence.
• The mean follow-up was 5 years with a maximum follow-up of 11.3 years.
• There were 51 overall deaths and six deaths from prostate cancer. The 5-year, 8-year and 10-year BCR-free survival rates were 83.9%, 78.6% and 75.6%, respectively.
• On univariate analyses preoperative D'Amico risk classification, pathological tumour stage, postoperative Gleason sum and surgical margin status were predictors of BCR (P < 0.001).
• On multivariable analysis, D'Amico classification, Gleason sum (P < 0.001), postoperative tumour stage (P < 0.001), nodal status (P < 0.001) and surgical margin status (P= 0.002) were independent predictors of BCR.
• LRP offers excellent long-term functional and oncological results with a low incidence of BCR for patients with localized disease.
• These results could be used for patient counselling before robot-assisted laparascopic prostatectomy (RALP) until long-term follow-up data for RALP is available.
Over the past decade, minimally invasive laparoscopic radical prostatectomy (LRP) and, more recently, robot-assisted laparoscopic prostatectomy (RALP) have been introduced and have proven equally effective compared with open surgery in terms of mid-term cancer control and complication rates [1–4]. In 2006, our group published complications and early outcome results of the first 1000 transperitoneal LRP cases . Owing to its relative novelty, long-term outcome data for LRP are limited. In 2009, Touijer et al.  published their oncological long-term LRP outcomes for 1564 patients with 5- and 8-year recurrence-free survival rates of 78% and 71%, respectively. Unlike other studies, they defined biochemical recurrence (BCR) as a rising PSA value of ≥0.1 ng/mL.
One of the major drawbacks of minimally invasive approaches is the lack of extended follow-up. Thus, patient counselling with respect to oncological outcome is based on long-term results from open series, which may differ from outcomes of the minimally invasive approaches. Furthermore, it has been proposed that laparoscopic procedures are associated with immune responses that are favourable with respect to cancer control ; therefore, a direct comparison of oncological outcomes from minimally invasive and open techniques is not feasible. The aim of the present study was to investigate BCR rates and data on postoperative incontinence in a large LRP cohort.
MATERIAL AND METHODS
We reviewed the records of 1860 consecutive patients who underwent LRP for localized prostate cancer (PCA) from 1999 to 2007. Patients with incomplete datasets were excluded from the analysis. Complete postoperative data was available for 1845 patients. Follow-up data were obtained by telephone interview. Data on BCR were available for 95.1% of all patients. BCR was defined as a rising PSA ≥0.2 ng/mL at least 6 weeks after surgery, confirmed by a second test. The initiation of salvage radiation or hormone therapy was also considered to represent BCR. A total of eight patients received radiation therapy within the first 3 months of surgery owing to persistent PSA levels >0.2 ng/mL which is, by definition, salvage therapy. Twenty-five patients received classical adjuvant radiation therapy for locally advanced disease with undetectable PSA. Neither of these groups was excluded from the analysis.
All data were collected according to internal review-board-approved protocol. All patients provided written informed consent to participate in the present study before undergoing surgery. LRP was performed by eight different surgeons. Two surgeons performed >400, two peformed >200 and four performed <200 LRPs in this series. The operating technique is described elsewhere [2,5]. Preservation of the neurovascular bundle was performed as an interfascial dissection.
Procession and histological evaluation were performed as previously described . Patient age and prostate size were considered continuous variables, while clinical stage, biopsy and LRP Gleason score, extraprostatic extension, seminal vesicle invasion, lymph node invasion and surgical margin status were evaluated as categorical variables. PSA, biopsy and LRP Gleason score were used as categorical variables with the following strata: PSA: ≤10.0 ng/mL, >10.0 and ≤20 ng/mL, and >20 ng/mL; biopsy and LRP Gleason score <7, 7 and >7. Clinical stage was stratified into cT1, cT2 and cT3 stages. Additionally, patients were stratified according to the D'Amico classification into low-, intermediate- and high-risk groups as described elsewhere . All patients in the high-risk group were treated with an additional pelvic lymphadenectomy. For the intermediate-risk group, this procedure was recommended and individually discussed with the patients after also consulting artificial neuronal networks or public databases. Data on lymph node counts were largely incomplete.
Urinary incontinence (UI) was evaluated by pads needed per 24 h and stratified as follows: 0–1 pad: no incontinence; 2–3 pads: mild incontinence; and ≥3 pads severe incontinence. Follow-up data on postoperative erectile function, as well as data on pre-existing erectile dysfunction and its treatment, were largely incomplete. Oncological outcomes represented by BCR were assessed using Kaplan–Meier analysis with the log-rank test and univariate and multivariable Cox regression analyses. All statistical analyses were two-tailed. A P value of <0.05 was considered to indicate statistical significance. Statistical analysis was conducted using SPSS version 18.0 (SPSS Inc., Chicago, IL, USA).
A total of 1845 patients underwent LRP between 1999 and 2007. Patient characteristics are shown in Table 1.
Table 1. Clinicopathological characteristics of patients with PCA undergoing LRP
Differences between +pLA and −pLA were significant (P < 0.001, Mann–Whitney U-test). NVB, neurovascular bundle; +pLA, with pelvic lymphadenectomy; −pLA, without pelvic lymphadenectomy.
Functional long-term outcome evaluation showed that 74.9%, 9.2%, and15.9% of patients had no, mild and severe UI, respectively. During a median (range) follow-up of 56 (0–135) months a total of 51 patients died, six of them from PCA. The mean overall survival was 130.3 months, resulting in 5-year, 8-year and 10-year overall survival rates of 97.5%, 94.5%, and 92.5%, respectively (Table 2). The overall BCR-free survival was 112.8 months (Fig. 1) with 83.9% of patients with no BCR at 5 years, 78.6% at 8 years and 75.6% at 10 years (Table 2).
Table 2. Functional and oncological long-term outcome characteristics of patients undergoing LRP
UI, n/N (%)
None (0–1 pads/24 h)
0 pads/24 h
1 pad/24 h
Mild (2 pads/24 h)
Severe (≥3 pads/24 h)
3 pads/24 h
>3 pads/24 h
Mortality, n/N (%)
Deaths from PCA
Not PCA associated
Overall survival, months
Mean (95% CI)
Mean (sem) overall survival, %,
BCR-free survival (PSA ≥0.2), months
Mean (95% CI)
Mean (sem) BCR-free survival,%
The BCR-free survival times, stratified by Gleason sum at biopsy, were 117.6 months, 105 months and 74.5 months for a Gleason sum of <7, 7 and >7, respectively (P < 0.001). After stratifying patients according to the preoperative D'Amico risk classification, BCR-free survival times (95% CI) were of 118.5 (115.9–121.1) months, 109.3 (105.2–113.3) months, and 86.0 (111.3–116.0) months for the low-, intermediate- and high-risk groups, respectively (P < 0.001; Fig. 2).
Stratification of patients according to the Gleason sum at LRP showed BCR-free survival (95% CI) of 124.8 (123.0–127.0) months, 114 (110.5–118.0) months and 79.5 (73.2–85.9) months for a Gleason sum <7, 7, and >7, respectively (P < 0.001; Fig. 3).
Stratifying patients according to pathological stage showed an overall BCR-free survival (95% CI) of 120 (117.9–122.1) months for organ-confined disease, 98.2 (92.4–104.0) months for extraprostatic extension, 71.1 (62.5–79.6) months for seminal vesicle invasion and 58.7 (39.6–78.0) months for patients with invasion of local lymph nodes (P < 0.001; Fig. 4).
On univariate analysis, D'Amico risk classification, postoperative Gleason sum, pathological stage and positive surgical margins (PSMs) were identified as predictors of BCR (Table 3). On multivariable analysis these findings were confirmed. According to the D'Amico risk stratification the intermediate- and the high-risk group had BCR with a hazard ratio (HR) of 1.48 (CI 1.13–1.95; P= 0.005) and 1.59 (CI 1.11–2.28; P= 0.011), respectively. Similarly, a postoperative Gleason sum of 7 or >7 predicted a BCR with a HR of 2.33 (CI 1.59–3.43; P < 0.001) and 4.37 (CI 2.87–6.64; P < 0.001), respectively. Extraprostatic tumour extension was associated with BCR (HR 2.03; CI 1.51–2.72; P < 0.001) as well as seminal vesicle invasion (HR 2.51; CI 1.72–3.67; P < 0.001) and lymph node involvement (HR 4.76; CI 2.65–8.54; P < 0.001). PSM status independently predicted BCR with a HR of 1.52 (CI 1.16–1.97; P= 0.002; Table 3).
Table 3. Univariate and multivariable Cox regression analysis of variables predicting BCR with PSA ≥0.2 ng/mL
Postoperative Gleason sum
Seminal vesicle invasion
Lymph node invasion
Thus far, few data on long-term follow-up after minimally invasive radical prostatectomy have been published in the world literature. This is especially true for RALP. In light of the rapidly increasing application of this technique, long-term data are essential for the counselling of patients with respect to oncological outcomes; it has been proposed that minimally invasive procedures are associated with favourable outcomes in a number of cancers owing to the induction of immune responses on several levels [7,9].
With respect to functional outcome, we showed that a notable 15.9% of patients had UI. This number is similar to the UI rates reported by other groups [10,11].
Our overall long-term BCR rates were excellent, with BCR rates of 16.1, 21.4, and 24.4% at 5-, 8-, and 10-year follow-ups, respectively. On multivariable Cox regression analyses the D'Amico risk classification, the Gleason sum at LRP, the pathological stage, and the presence of PSMs were independent predictors of BCR.
The present findings are largely consistent with other investigators' series that were published previously [6,10–12]. In 2006, Rassweiler et al.  published a large German multicentre cohort of 5824 LRP cases with a focus on complications, urinary continence and potency rates. The continence rate after 12 months was 84.9% using the ‘0-pad’ definition. Postoperative restoration of potency, defined as the ability to perform sexual intercourse, was achieved in 52.5% of all patients after a bilateral preservation of the neurovascular bundle. The PSM rate was 10.6% (3.2–18%) for pT2 and 32.7% (20–38.5%) for pT3a tumours. The mean (range) 5-year BCR-free survival was 91.4 (84.7–96) % for pT2 and 82.5 (79.4–85) % for pT3a stages . With respect to UI, the present results are similar. Although we did not achieve the same rate of continent patients using the 0-pad definition, 84.1% of patients in the present series reported mild or moderate incontinence. The differences between the present series and that of Rassweiler et al. might be partly attributable to different patient questionnaires. It is likely that a significant proportion of men in the present cohort answered the ‘pads used per day’ question with ‘1–2 pads per day’ even if these were used as safety pads in the absence of UI. The PSM rate in the present cohort was slightly higher (19.3% for pT2 and 53.5% for pT3 tumours) which did not translate into higher proportions of BCR for pT2 tumours (5-year BCR-free survival of 91.2% for pT2 tumours); however, we had significantly more patients with BCR in the cohort of patients with non-organ confined cancer (69.9% BCR-free survival for pT3a tumours). This finding could be explained by a previous policy at our institution to recommend radiation therapy for patients with locally advanced tumours and PSM in a salvage rather than in an adjuvant setting.
In 2009, Touijer et al.  reported their series of 1564 patients with 5-year and 8-year probabilities of freedom from progression of 78% and 71%. Kaplan–Meyer analysis according to a pre-treatment PCA nomogram showed probabilities of progression-free survival at 5 years of 91%, 77% and 53% for patients in the low-, intermediate- and high-risk groups, respectively. The slightly lower PSM rate of 13% could be at least partially attributable to the fact that this was a two-surgeon series with an immense amount of surgical experience. The definition of BCR at a PSA value of 0.1 ng/mL could explain the slightly lower BCR-free survival rates compared with those of the present study cohort .
In 2010, Paul et al.  published mid-term data on oncological outcome after extraperitoneal LRP. Their series included 1115 consecutive patients with a limited median follow-up of 35.6 months. Notably, the authors excluded all patients receiving neoadjuvant hormonal treatment from analysis. In their series, three surgeons produced an overall PSM rate of 26%, with a BCR-free survival rate of 83% at 5 years. PSA, Gleason score, pathological stage and PSM were identified as independent predictors of BCR. Their findings are consistent with the results obtained in the present transperitoneal surgery cohort. Unfortunately, neither the study by Touijer et al.  nor that by Paul et al.  provided any long-term data on UI; therefore, comparison with respect to this functional outcome is not feasible.
Another study from Brazil, by Mariano et al. , provided detailed analysis of their LRP experience. Within their series of 730 mostly transperitoneal LRP cases, an overall PSM rate of 19.6%, a UI of 22.1% after 1 year and a BCR rate of 10.3% after a mean follow-up of 62.5 months was reported. Unfortunately, their study did not provide detailed information on the definitions used for UI and potency, nor on the BCR rates at defined time points .
Secin et al. , in their large international multicentre study of 8544 LRP cases, primarily investigating learning curves, reported an overall PSM rate of 22%. The authors highlighted the well-known problem of a prolonged learning curve in LRP, plateauing at ∼200–250 cases. BCR-free survival rates were not investigated owing to a lack of follow-up data . Although the authors primarily looked at learning curves, the PSM rate seems fairly low; however, owing to the fact that after 200 cases a plateau was achieved and the overall number of patients included in this study was remarkable, higher PSM rates produced during the learning curve did not negatively impact the overall PSM rate of the entire cohort.
In a recently published review article by Coelho et al. , operative and postoperative data on retropubic radical prostatectomy, LRP and RALP were compared. Their meta-analysis of high-volume LRP centres showed a weighted mean continence rate at 12 months' follow-up of 84.8% and a weighted mean potency rate for patients undergoing unilateral or bilateral nerve-sparing of 31.1% and 54%, respectively. For LRP, a mean overall PSM rate of 21.3%, with 12.4% for T2 and 39.2% for T3 tumours, was described as the only oncological endpoint. Exact definitions of potency and UI were not provided .
For RALP, data on long-term oncological outcome are not available and intermediate-term data have been published by a very limited number of groups thus far. In 2010, Menon et al.  reported long-term data on 1384 consecutive RALP cases with a median follow-up of 60.2 months. The overall BCR-free survival rates at 5 and 7 years were 86.6% and 81.0%, respectively. On multivariable analyses, the preoperative D'Amico risk classification, preoperative PSA, Gleason score at biopsy, pathological stage, Gleason score at RALP, perineural invasion and the surgical margin status proved to be independent predictors of BCR . These results are similar to the findings of the present study and do represent the successful transfer of the traditional LRP to the next level of the robot-assisted technique with an at least equally effective oncological outcome.
Several limitations of the present study have to be addressed: the retrospective and single-centre nature of the present study represent a potential bias. Additionally, this a consecutive series of 1845 patients operated over a period of 8 years by eight different surgeons, reflecting the process of introducing this technique in the late 1990s and then transferring this knowledge to a second generation of LRP surgeons. Data from a substantial number of patients thereby represented the oncological outcomes of surgery during the surgeons' learning curves. Unfortunately, a large proportion of patients' datasets did not provide full information on preoperative and postoperative erectile function, therefore, a solid analysis of restoration of postoperative erectile function could not be performed.
In conclusion, LRP offers excellent long-term functional and oncological results with a low incidence of BCR for patients with localized disease. Independent predictors of BCR are preoperative D'Amico risk classification, postoperative Gleason sum, pathological tumour stage, lymph node involvement and the surgical margin status. These results could be transferred to RALP and patient counselling could be partly based on our results until long-term follow-up data for RALP are available.