The role of surgery in clinical stage T3 prostate cancer (cT3 PCa) is still subject to debate. We reviewed the records of 139 consecutive patients who underwent a radical prostatectomy (RP) for cT3 PCa with a mean follow-up of 8 years. All data related to surgical and perioperative complications were collected. Continence and erectile function were assessed at 12 months postoperatively and long-term oncologic outcomes were analyzed. Rectal injury and injury of the obturator nerve occurred both in 0.7% of cases. No serious in-hospital complications were noted and no reintervention was needed. Lymphatic leakage was noted in 2.2% of patients and 1.4% experienced prolonged drainage of urine. In 7.2%, wound-related problems occurred. Anastomotic stricture occurred in 2.9%. These complication rates were not different compared to surgical series of RP in localized PCa. At 12 months, complete continence was 87.8% and erectile function had fully recovered in 6% and 10% of patients who underwent a non-nerve sparing or unilateral nerve-sparing procedure, respectively. 10-year estimated biochemical PFS, clinical PFS, CSS and OS were 51.8%, 85.6%, 94.6% and 85.9%, respectively. In cT3 PCa, RP is technically feasible with morbidity comparable to RP in clinically localized PCa. Long-term oncologic control was excellent.
Locally advanced prostate cancer (PCa) is defined as cancer that has extended clinically beyond the prostatic capsule with invasion of the pericapsular tissue, the sphincter muscle, bladder neck, or seminal vesicles, but without lymph node involvement or distant metastases [
The optimal treatment of cT3 PCa has been subject to intense debate during recent years. According to the guidelines of the European Association of Urology (EAU), watchful waiting, radiation therapy (RT), Radical prostatectomy (RP), hormonal therapy (HT), and various combinations are valuable options to consider, depending on the general health status of the patient and the local extent of the tumour [
Many experts consider an RP for cT3 PCa a valid treatment option with excellent oncological outcome, but it is felt to be a burdensome procedure even for a skilled surgeon and feasibility has been questioned in the past.
In order to better define the place of surgery in cT3 PCa, we have conducted a retrospective study in 139 patients who underwent an RP for cT3 PCa. The patient files were critically reviewed and all data related to surgical and peri-operative complications were carefully collected. All data were compared to major contemporary series of RP in clinically localised disease. Additionally, functional results with respect to erectile function and continence were collected at 12 months postoperatively and long-term oncologic outcomes were assessed.
From January 1997 to December 2003 we performed an RP with bilateral pelvic lymphadenectomy in 139 patients with cT3 PCa. Ultrasound guided prostate biopsies showed a median Gleason score of 7 (range 2–10). Prostate biopsy was performed in accordance with the random systematic octant biopsy technique: lateral systematic sextant biopsies with additional bilateral transition zone biopsies [
Patient characteristics.
Number of patients | 139 |
Age (years), mean (±SD) | 61,8 (±7,0) |
cT3a | 89,9% ( |
cT3b | 10,1% ( |
Biopsy Gleason score, median (range) | 7 (2–10) |
PSA (ng/mL), mean (range) | 13,73 (3,1–97,0) |
Previous surgery | 14,4% ( |
Neo-Adjuvant Androgen Deprivation Therapy | 8,6% ( |
Non-nerve-sparing procedure | 92,8% ( |
Unilateral nerve sparing procedure | 7,2% ( |
Lymphadenectomy not performed | 7,2% ( |
Hospital stay (days), median (range) | 12 (5–27) |
pT2 | 31,1% ( |
pT3a | 51,1% ( |
pT3b | 16,3% ( |
pT4 | 1,5% ( |
PSA persistence | 10,1% ( |
Pathological Gleason score, median (range) | 7 (4–9) |
Pathological node positive | 10,1% ( |
Surgical margin positive | 13,7% ( |
Adj radiation therapy within 1 year | 7,2% ( |
Adj endocrine therapy within 1 year | 13,7% ( |
As described earlier, our surgical technique focuses on clean apical dissection, neurovascular bundle resection at least at the tumour bearing site, complete resection of the seminal vesicles, and in some cases resection of the bladder neck [
In the peri-operative period, low molecular weight heparin and compression stockings were administered as thromboembolic prophylaxis. Postoperative pain was managed for 2 days by epidural patient-controlled anaesthesia. Oral ingestion and early mobilisation was encouraged from the first postoperative day. Patients were discharged after removal of all suction drains (as soon as drainage was fewer than 15 mL per 24 h), as soon as they were on a normal diet and were fully ambulatory and pain or discomfort was manageable by oral analgesia. The urethral catheter was left in situ at discharge and was removed during a one-night hospital stay at a mean of 12 days postoperatively. Since our group has shown that pelvic floor muscle exercises shorten the duration of incontinence and improve continence rates after an RP, physiotherapy was started at catheter removal [
At 6 to 8 weeks postoperatively, patients were reassessed for the first time and serum PSA was measured. For the first postoperative year, patients were seen at 3-month intervals. For the second and third years, patients were reevaluated every 4 months and 6 months thereafter.
Patients who underwent a unilateral nerve-sparing procedure were offered treatment with 5-phosphodiesterase-inhibitors, or intracavernous prostaglandin E2 injections if the obtained effect was insufficient. Patients who underwent a non-nerve-sparing operation were offered treatment with intracavernous injections.
Further treatment strategy was based upon final histopathology and PSA evolution. In case of positive surgical margins, patients were randomised according to the EORTC 22911 protocol to receive adjuvant pelvic irradiation or not [
Patient characteristics are described in Table
Preoperatively, no ureteral or vascular injury occurred. Operative complications included one sectioning of the obturator nerve (0.7%) and one rectal laceration (0.7%). Treatment consisted of microsurgical repair of the obturator nerve and primary closure of the rectal laceration in a double layer. Long-term evolution was uneventful in both cases. No peri-operative mortality was noted.
In the peri-operative period no ureteral obstruction or urinary retention occurred. In 10 patients (7.2%) healing of the abdominal wound was delayed: 6 wound infections (4.3%) and 4 partial wound dehiscences (2.9%) occurred. Prolonged drainage in the suction drains was noted in 5 patients. Lymphatic leakage was present in 3 cases (2.2%). Two patients (1.4%) had a urinary leakage for 36 hours which resolved spontaneously with permanent suction. Prolonged drainage did not show to be prognostically relevant since all 5 patients obtained total continence at 12 months. All above mentioned patients were discharged without reintervention.
When lower urinary tract symptoms were present, a uroflowmetry was performed: within 12 months, 4 patients (2.9%) were diagnosed with an anastomotic stricture. One patient complained of a painful orgasm. Urethroscopy visualised a surgical clip at the level of the anastomosis. After removal of the clip, the dysorgasmia disappeared.
At 12 months, 98 patients were completely continent (70.5%) and 24 patients mentioned an occasional loss of a drip (17.3%). Incontinence for which protective pads were needed was only seen in 17 patients (12.2%). Of these 17 patients, one had already been treated for overactive bladder. Only 6 of these 17 patients needed more than one pad per day (4.3%). And only 2 of them complained of continuous and uncontrollable incontinence: an artificial urinary sphincter was therefore implanted (1.4%). Postoperative potency was evaluated at 12 months. 129 patients were treated by a non-nerve-sparing RP. 83.6% mentioned absence of erections; 10.4% experienced some tumescence, but not sufficient for vaginal intercourse, and 6% patients had erections, sufficient for successful vaginal intercourse. Mean age of these last patients was only 54.5 years (range 49.8 to 62.2 years). In the 10 patients who were treated with a unilateral nerve sparing procedure, erections did not recur in 40% and did recur partially though insufficiently for vaginal intercourse in 50%; 10% regained full erectile function.
Table
Complication rates after open radical retropubic prostatectomy.
Joniau | Dillioglugil et al. [ | Hisasue et al. [ | Gaylis et al. [ | Maffezzini et al. [ | Lepor et al. [ | Lerner et al. [ | |
---|---|---|---|---|---|---|---|
Number of patients | 139 | 472 | 123 | 116 | 300 | 1000 | 812 |
cT1 % (pT1 %) | 0 | 20.3 | 44.7 | 43 | (0) | 78.5 | 0 |
cT2 % (pT2 %) | 0 | 72.7 | 55.3 | 57 | (66.4) | 21.3 | 0 |
cT3 % (pT3 %) | 100 | 6.9 | 0 | 0 | (29.9) | 0.2 | 100 |
Mean age (years) | 62.0 | 63 | 66 | 66.6 | 65.5 | 60.3 | |
Mean operation time (min) | 105 | 155 | |||||
Mean blood loss (mL) | 558 | 872 | 600 | 945 | |||
Mortality % | 0 | 0 | 0 | 0.4 | |||
Rectal injury % | 0.7 | 0.6 | 4.9 | 0.9 | 0.3 | 0.5 | 1.8 |
Ureteral injury % | 0 | 0.2 | 0.8 | 0.3 | 0.1 | ||
Iliac vessel injury % | 0 | 1.1 | |||||
Obturator nerve injury % | 0.7 | 0.2 | 1.6 | 0.3 | |||
Angor/myocardial infarction % | 0.7 | 1.7 | 0.6 | 0.4 | |||
Other cardiac complications % | 0 | 10.6 | 0.8 | 0.2 | |||
Pulmonary complications % | 0 | 3.8 | 0.1 | ||||
Deep venous thrombosis/pulmonary embolism % | 0 | 2.3 | 0.8 | 3.4 | 0.3 | 0.3 | 4 |
Gastrointestinal complications % | 0 | 5.1 | 0.8 | 0.6 | |||
Neurological complications % | 1.4 | 1.5 | 0.2 | ||||
Other infectious complications % | 0 | 4.7 | 0.8 | 0 | |||
Prolonged drainage (urine, lymph, blood) % | 3.6 | 2.8 | 8.9 | 2 | 0.7 | 0.8 | |
Acute retention % | 0 | 0.6 | 2 | ||||
Reintervention % | 0 | 1.7 | 0.5 | ||||
Woundproblem % | 7.2 | 3.0 | 13.8 | 0.9 | 1 | 0.8 | 2.7 |
Anastomotic stricture at 12 months % | 2.9 | 13.8 | 0.7 | 1 | 9.2 | ||
Not dry (in need of pads) at 12 months % | 12.2 | 12.7 | 20 | 12 | 22.1 |
At final histopathology, in 19 patients, positive surgical margins were found (13.7%). Of these specimens with positive surgical margins, 2 tumours were organ confined (pT2), 12 showed extraprostatic extension (pT3a), 4 were invading the seminal vesicles (pT3b), and one had invaded the bladder neck (pT4). Table
Comparison between positive surgical margins and pathologic staging after radical retropubic prostatectomy.
Positive surgical margins | |
---|---|
All patients | 13.7% |
pT2 | 4.8% |
pT3a | 17.4% |
pT3b | 18.2% |
pT4 | 50% |
Postoperative evaluation included history, physical examination, and serum PSA measurement. PSA persistence (>0.02 ng/mL) at first follow-up was found in 14 patients (10.1%). These cases were considered surgical failures. In 10 of these 14 patients (71.4%), final histopathology revealed positive surgical margins or positive lymph nodes. Within one year, 10 patients (7.2%) underwent RT of the pelvis and 19 patients (13.7%) were started on endocrine treatment because of positive surgical margins, PSA persistence, or rising PSA (Table
The long-term oncologic outcomes were assessed by Kaplan-Meier survival estimates. The 10-year estimated biochemical progression-free survival, clinical progression-free survival, cancer specific survival, and overall survival rates were 51.8%, 85.6%, 94.6%, and 85.9%, respectively, (Figures
Kaplan-Meier plots for the oncologic outcomes of surgery for cT3a-b PCa. (a) Biochemical progression-free survival. (b) Clinical progression-free survival. (c) Cancer-specific survival. (d) Overall survival.
Treatment options for locally advanced PCa vary and the jury is still out regarding the optimal treatment [
In cT3 PCa, Thompson reported a 60 to 70% 5-year overall survival with watchful waiting [
Until the early eighties, radiotherapy was the treatment of choice for localized and locally advanced PCa. With radiotherapy as monotherapy, 10-year disease-free survival rates of 19–44% and overall survival rates of 21–54% have been reported [
In an attempt to improve disease-free survival and overall survival, the combination of RT and HT was evaluated. Laverdiere et al. had indeed shown a significant improvement in oncological outcome with adjuvant HT [
By many, the combination of external-beam RT and adjuvant HT is since considered a standard therapeutic option in patients with cT3 PCa.
Literature on the value of RP as an option for cure in cT3 PCa is limited. However, clinical evidence showing 5-year disease-specific survival rates ranging between 85% and 100% is available [
The percentage of overstaging and understaging in clinical locally advanced T3 prostate cancer.
Authors | pT2 | pT4/N+ |
---|---|---|
Van Poppel et al. [ | 13% | 8%/11% |
Van den Ouden et al. [ | 15% | 3.4%/15.6% |
Lerner et al. [ | 17% | —/33% |
Morgan et al. [ | 22% | 42% (stage D1) |
Ward et al. [ | 27% | —/27% |
Some locally advanced PCa will not be cured by surgery alone, and therefore, combinations with hormone therapy or radiotherapy have been investigated. Neoadjuvant HT did not improve biochemical or clinical progression, nor survival rates in RP [
The general impression is that complications such as rectal injury, haemorrhage, deep venous thrombosis, pulmonary embolism, urinary fistula, ureteral obstruction, stress incontinence, impotence, anastomotic stricture, and peri-operative death are more common in the cT3 patient group. Our review of literature shows that the mortality risk associated with RP is merely a theoretical risk. Other surgery-related complications such as rectal injury, ureteral obstruction, and injury to the iliac vessels or obturator nerves are encountered rarely and do not account for a significant amount of morbidity. At an incidence between 0.6% and 7.3%, all of these per-operative complications could be resolved during the same operation. Long-term consequences such as anastomotic strictures occur in 0.7% to 13.8% of patients. One single dilatation has a success rate of up to 75% [
In Table
Furthermore, in our series, functional results at 12 months show total continence (no pad necessary) in 87.8% and socially acceptable continence (max. 1 precautionary pad) in 94.2%, which is well within acceptable ranges. Finally, anastomotic stricture was encountered at a rather low rate of 2.9%. Expectedly, potency rates were poor in non-nerve-sparing RP (6% full recovery at 1 year and 10% partial recovery), while those rates were better in unilateral nerve-sparing RP (10% full recovery and 50% partial recovery). As complete recovery of erectile function can take up to 36 months, further improvement of these results may be expected [
Surgical margins after RP are of great importance in progression and oncological outcome [
Our present analysis is not devoid of limitations. First, this is a retrospective analysis of complications and functional results, using data extracted from patient files. Inherent biases are to be expected, as sometimes more discrete complications can be missed. Second, preoperative data on the functional status of the patient were not collected, limiting the interpretation of the results. Third, the complications and functional results were compared to data extracted from the literature. Indeed, a more solid approach would be to prospectively compare data on RP in cT3a-b PCa with data on RP in localized disease from the same institution. Nevertheless, we believe that our analysis has its value in outlining the incidence of complications and the functional results that can be expected after RP for locally advanced PCa. It has to be stressed that data on this subject are extremely scarce. Finally, a significant number of patients received adjuvant or salvage RT and/or HT treatment following surgery, limiting the interpretation of the results regarding the value of surgery in locally advanced PCa. Accepting this limitation, oncologic control with RP as a first step in the treatment of locally advanced PCa is excellent.
Our experience with 139 patients confirms the surgical feasibility of RP for cT3 PCa, showing complication rates comparable with RP in organ-confined PCa and showing a very low incidence of positive surgical margins and associated failure of surgery. Improvement can be expected by further defining the patient population most suitable for surgery and by further optimising adjuvant treatments such as RT and HT. Continence rates were also comparable with those achieved after RP for localized PCa. A nerve-sparing approach was only considered possible in a limited number of patients. It has to be expected, though, that modern imaging will further increase the indications for nerve-sparing surgery in locally advanced PCa.
Prospective randomized clinical trials are needed to compare oncological outcome, treatment-related complications, and quality of life in the different treatment options for cT3 PCa.
S. G. Joniau and H. P. Van Poppel were supported by the “Jozef De Wever Fonds voor prostaatkankerpreventie”.