Abstract
Background: The therapeutic role of pelvic lymph-node dissection (PLND) for prostate cancer (PCa) is still unclear. The aim of this study was to assess the impact of the number of lymph nodes (LN) retrieved at PLND during radical prostatectomy (RP) on biochemical relapse (BCR) in node-negative patients. Patients and Methods: From October 1995 to June 2009, 1510 consecutive PCa patients underwent RP at the University of Bologna hospital. A retrospective analysis was performed on 614 pT2-4N0 patients with a minimum follow-up of 12 months. All 614 patients underwent limited or extended PLND during RP. BCR-free survival was estimated using the Kaplan-Meier method. Cox regression was applied to analyse survivals rates. Patients were divided into two groups: group 1 (n=295) had 1 to 9 retrieved LNs and group 2 (n=319) had 10 or more retrieved. The parameters analysed were age, PSA, clinical and pathological Gleason score (GS), stage, margin status and adjuvant radiotherapy (ART). BCR was defined as PSA greater than 0.2 ng/ml. Results: Mean follow-up time was 62.5±39.7 months. Group 2 showed a significantly lower BCR than group 1 (p=0.019). LN group, PSA, clinical and pathological GS, pathological stage and ART all showed an independent and significant relationship with BCR using multivariate analysis. Conclusion: In node-negative patients, a more extensive PLND affects BCR-free survival positively. A more extensive PLND may have a therapeutic role by removal of micrometastases.
- Biochemical relapse
- lymph node metastasis
- micrometastasis
- pelvic lymph node dissection
- prostate cancer
- radical prostatectomy
The role of pelvic lymph-node dissection (PLND) in patients with prostate cancer (PCa) continues to be a controversial issue. In the absence of adequate preoperative imaging techniques, PLND during radical prostatectomy (RP) is still the most accurate staging procedure (1). It has been well demonstrated that extensive PLND improves the detection of lymph-node (LN) involvement significantly compared with limited PLND (2); however, the therapeutic role of PLND and its impact on cancer outcome remains uncertain. It has been suggested that for those patients undergoing to RP and lymphadenectomy that resulted in pN0, a more extensive PLND may have some therapeutic advantage due to the removal of micrometastases (3). The aim of the present study was to assess the impact of the number of LNs retrieved at PLND during RP in node-negative patients on biochemical relapse (BCR).
Patients and Methods
The institutional PCa database of patients treated with RP at the Department of Urology at the University of Bologna was reviewed. In all, 1,510 consecutive patients had a retropubic (n=1,199) or laparoscopic (n=311) radical prostatectomy between October 1995 and June 2009. Of these patients, 1,343 who had a minimum follow-up of 12 months were evaluated. Seven patients (0.5%) had incomplete pathological data and were excluded. Four hundred and thirty-nine patients (32.7%) had Nx status and 103 patients (7.7%) had LN-positive disease and were excluded. Thus the remaining 794 patients (59.1%) had pN0 status. Eleven patients (0.8%) had pT0 disease and were excluded. A total of 81 patients (6.0%) were submitted for neoadjuvant therapy (NHT) and were excluded; among the 727 (54.1%) remaining patients, 97 (13.3%) had incomplete follow-up data and were excluded. Among the final remaining 630 patients, immediate adjuvant hormonal deprivation was administered in 16 patients (2.5%) and they were excluded. The study therefore finally included 614 patients with pT2-4N0.
Ethics. The study was performed in line with the Helsinki Declaration and national regulations. All patients provided informed consent for participation and anonymous publication of data.
Surgical technique. Until 2004, limited PLND (defined as obturator and external iliac LN) or extensive PLND (defined as obturator, external and internal iliac LN with/without common iliac LN) were offered to almost all patients undergoing RP at the Department of Urology at the University of Bologna, and the extent of the dissection varied according to the preference of the surgeon; in more recent years, the Kattan and the Briganti nomograms (4, 5) have been utilised more frequently to assess the risk of LN metastasis and low-risk patients were often spared PLND.
Histopathological details. RP specimens were routinely processed by dedicated genitourinary pathologists. Axial step sections were obtained at 3 to 4 mm intervals from the apex to the base. Tumour present at the inked margin was defined as a positive surgical margin (R1). After separation of each LN, their total number was assessed and they were analysed microscopically. LNs of 4 mm or less were completely embedded in paraffin with one hematoxylin and eosin section per paraffin block, while LNs larger than 4 mm were cut into 2-mm-thick slices. When necessary, an immunohistochemical analysis was performed. All specimens were staged using the 2002 AJCC-TNM staging system (6).
Adjuvant radiotherapy. Adjuvant radiotherapy (ART) consisted of prostatic bed irradiation and was administered in the absence of BCR within 1 year of surgery; ART was administrated to pathologically advanced or high-grade patients at the discretion of the treating physician.
Follow-up. The follow-up protocol consisted of prostate-specific antigen (PSA) measurement and digital rectal examination every 3 months during the first year. Once negative, these tests were performed every 6 months during year 2; thereafter PSA was measured annually. Imaging studies were performed when deemed necessary. All data were obtained retrospectively by clinical visits or telephone calls and were collated into a cancer database.
Patient groups. The number of LNs examined served as an indicator of the extent of PLND. Overall, the mean number of nodes obtained at RP was 10.8±6.4 (median 10, range: 1-36). The population was divided into two groups: group 1 (n=295, 48.0%) had 1 to 9 LNs, with a mean of 5.7±2.3 LNs (median: 6) and group 2 (n=319, 52.0%) had 10 or more LNs, with a mean of 15.6±5.1 LNs (median: 14).
Clinical and pathological data. Clinical and pathological data included age, PSA level at diagnosis, clinical and pathological Gleason score (GS), (divided as follows: GS≤6, GS=7, GS≥8), clinical and pathological stage, surgical margin status (R0 vs. R1) and ART (yes vs. no). Table I lists the characteristics of the two patient groups. Mean age was 65.7±5.8 years (range 44 to 78) and mean preoperative PSA was 11.5±9.2 ng/ml (range 0.2 to 81 ng/ml). When LN groups were compared, there was no significant difference regarding PSA, clinical and pathological stage, margin status or ART, while group 1 tended to have a significantly slightly higher mean age and lower clinical and pathological GS than group 2.
Primary end-point. The primary end-point was BCR, defined as a PSA level greater than 0.2 ng/ml, followed by any increase above this level or initiation of any salvage therapies due to increasing PSA.
Statistics. Means, standard deviations, medians, ranges and frequencies were used as descriptive statistics. The Fisher exact test, the linear by linear association, the Pearson chi-square and the Mann-Whitney U-test were used to compare the two groups of patients. Cancer-specific and BCR-free survival rates were estimated by the Kaplan-Meier method. Univariate and multivariate forward-stepwise Cox proportional hazards regression was applied to analyse survivals, and the hazard ratio (HR) together with the 95% confidence interval (95% CI) were evaluated. All analyses were conducted using the SPSS software (SPSS, Inc., Chicago, IL, USA; version 13.0) and two-tailed p-values of less than 0.05 were considered statistically significant.
Results
Mean follow-up was 62.5±39.7 months (range 12-159 months). Among the population of 614 node-negative patients, 5- and 10-year cancer-specific survival rates were 98.8% and 95.8% and BCR-free survival rates were 77.2% and 60.7% respectively (Figure 1). BCR was observed in 130 (21.2%) patients.
An increasing number of LNs removed predicted a slightly lower risk of BCR (HR=0.926, 95% CI 0.932-0.992 per each additional LN removed; p=0.013). Figure 2 shows the Kaplan-Meier curve for BCR-free survival with patients stratified by LN group. Group 2 showed a significantly lower risk than group 1 (HR=0.658, 95% CI: 0.464-0.934, p=0.019).
Under univariate analysis, LN group, higher PSA, higher clinical and pathological GS, higher clinical and pathological stage, presence of positive margins, and ART were significant predictors of higher risk of BCR. Under multivariate analysis, LN group, PSA, clinical and pathological GS, pathological stage and ART showed an independent and significant relationship with BCR, while age, clinical stage and margin status did not enter the analysis (Tables II and III).
Discussion
Despite recent advances in imaging techniques (7), PLND remains the gold standard procedure for LN staging in PCa. Most authors nowadays recognise the importance of PLND during RP as a staging procedure in intermediate- and high-risk patients and spare the low-risk patients from this procedure (1, 5, 8). Furthermore extensive-PLND improves significantly the detection of LN metastasis compared with limited PLND (2). However, to date, a beneficial effect of PLND on cancer outcome has not been demonstrated. Moreover, even though the major complications of PLND are rare, the minor complications are extremely frequent, especially after extensive PLND, and many urologists prefer to avoid such complications in the absence of a well-proven oncological benefit (9).
In a recent study by Murphy et al. which included 964 pN0 patients, the number of LNs removed at RP did not increase the chance of cure for patients with pT2-4N0M0 (10). Similarly, in a study by DiMarco et al. on 7,036 RPs, the extent of lymphadenectomy did not appear to affect PCa outcome in node-negative cases (11). In a study by Bhatta-Dhar et al. on 806 low-risk patients, the omission of a limited PLND did not adversely affect biochemical relapse rates at 6 years (8). The results of the present study agree with the existing literature that a limited PLND cannot have a therapeutic role in patients at low-risk of LN metastases, but these results cannot be transferred to intermediate- and high-risk populations, who can harbour LN metastases in 20-40% of cases.
By contrast, in a study of 5,038 patients, Masterson et al. found better BCR-free survival with more LNs removed in node-negative patients (p<0.01) (12). In a retrospective trial of 13,020 cases treated with RP, patients undergoing excision of at least 4 nodes (node-positive and node-negative patients) or more than 10 nodes (only node-negative patients) had lower cancer-specific death at 10 years with respect to pNx (3). Allaf et al. found better BCR-free survival for patients submitted to extensive PLND compared to limited PLND (43% vs. 10%) among men with metastatic involvement in fewer than 15% of the extracted nodes (p<0.01) (13). Moreover many studies among node-positive populations demonstrated that patients with a low volume of nodal involvement treated with RP and extensive PLND have excellent survival rates even in the absence of further adjuvant treatments, and the data from this study ultimately support a possible therapeutic effect of PLND (13-17).
In the present study, the impact of clinical and pathological characteristics as well as the impact of the extent of PLND and ART on BCR were evaluated. Under multivariate analysis, LN group, PSA, pathological stage, clinical and pathological GS and ART were independently correlated with BCR. The fact that ART was correlated with higher BCR is not surprising: in fact ART was administered to patients with locally advanced or high-grade PCa who were at high-risk of both local and systemic recurrence. Therefore, the patients submitted to ART were included in the analysis since these were the patients at higher risk of harbouring a micrometastatic disease. However, it must be emphasised that the node-negative patients receiving ART were not treated with whole pelvic irradiation, but only with a prostatic bed irradiation, having minimal effects on the pelvic LNs. By contrast, patients submitted to NHT were excluded due to the risk of hormone therapy affecting the regression of some LN metastases (18). Finally, even though patients in group 1 had lower clinical and pathological GS, they had a higher risk of BCR, further supporting a possible positive influence of a more extensive PLND on BCR.
In line with some studies (3, 12, 15) the results from this study showed a possible role of a more extensive PLND on cancer outcome after RP. The choice of separating patients into two groups using 10 retrieved LNs as a cut-off and the inclusion of patients undergoing ART may account for the differences of our results with those from other reports (8, 10, 11). In the node-negative patients, a meticulous dissection may have reduced the risk of BCR by removing small nodal micrometastases missed at histopathology. The existence of micrometastatic disease in node-negative patients was demonstrated in many studies performed with molecular and immunohistochemical techniques. In fact, when these techniques were used, 10-30% of negative LNs demonstrated positivity for the presence of micrometastatic cancer cells, especially in high-risk disease (19-22). Furthermore, about 5-15% of small (<1 mm) LN metastases may have been missed during routine analysis, but whether such amounts of metastatic disease have an adverse impact on survival remains unclear (19-24).
For many other solid tumours (such as melanoma, bladder, penile testicular, breast and colorectal cancer) the regional LN are routinely treated by some modality with an advantage for recurrence or survival. Similarly, preliminary results with radiation therapy in PCa showed a superiority of whole pelvic radiotherapy over prostate-only radiotherapy in high-risk patients (25). Therefore, it is not surprising that PLND during RP may have a therapeutic role. However these conclusions cannot be absolute: in fact a relationship between the extent of PLND and cancer outcome can be also influenced by the well known ‘Will Rogers phenomenon’(26), in other words the stage shift: after a limited dissection, the risk of ignored micro- and macro-metastases is higher, and these patients may have been understaged because they were classified as pN0 but they will have a behaviour similar to pN+. It must be emphasised however, that the ‘Will Rogers phenomenon’ might also influence the results of each well-designed, prospective randomised clinical trial, comparing extended vs. limited or no PLND. It can be concluded that the positive effect of the extent of PLND and the BCR may be related both to the removal of some micrometastatic disease and to the stage shift.
The principal limitation of the present study was its retrospective nature; furthermore, no histopathological revision of the LN specimens was performed to assess the presence of micrometastases. In addition, the number of retrieved LNs was assumed to be an indicator of the extent of PLND. Finally, the BCR was used as an indicator of cancer outcome, but both local relapse and skeletal metastasis may have accounted for some cases of BCR and the relationship between cancer-specific survival and BCR was not determined precisely.
Conclusion
In these node-negative patients, a more extensive PLND positively affected BCR-free survival. Even though the extent of PLND may have influenced BCR indirectly due to the stage shift (the ‘Will Rogers’ phenomenon), it is reasonable to conclude that a more extensive PLND during RP may have a therapeutic effect due to the removal of micrometastases. The assessment of the influence of the extent of PLND on survival should be examined more fully in trials with longer follow-up and prospective trials comparing extensive vs. limited or no PLND are awaited.
- Received January 28, 2010.
- Revision received March 28, 2010.
- Accepted March 28, 2010.
- Copyright© 2010 International Institute of Anticancer Research (Dr. John G. Delinassios), All rights reserved