SEARCH

SEARCH BY CITATION

Keywords:

  • prostatic neoplasm;
  • biochemical recurrence;
  • minimally invasive;
  • prostatectomy;
  • positive surgical margins

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Conflict of Interest
  9. References

Objective

  • To assess the prognostic significance of positive surgical margins (PSMs) after minimally invasive radical prostatectomy (MIRP) in interaction with other established prognosis factors.

Patients and Methods

  • We retrospectively analysed data prospectively collected between 1998 and 2010 for 4628 consecutive patients who underwent MIRP for clinically localized prostate cancer.
  • The impact of PSM on biochemical recurrence (BCR), defined as prostate-specific antigen (PSA) >0.2 ng/mL, was evaluated using multivariable Cox proportional hazards regression.
  • Estimates of BCR-free survival were generated using the Kaplan–Meier method and compared among groups using the log-rank test.

Results

  • The median follow-up was 55 months.
  • On multivariable analysis, PSM was an independent prognostic factor for BCR (adjusted hazard ratio: 2.14 for PSMs vs negative surgical margins (NSMs); 95% confidence interval [CI]: 1.86–2.45; P < 0.001). Other independent predictors for BCR were preoperative PSA, date of surgery, pT stage, Gleason score and lymph node involvement (all P < 0.001).
  • The 5-year BCR-free probability was 80.6% (95% CI: 79–82.2) for NSMs vs 51% (95% CI: 47–55) for PSMs (log-rank P < 0.001).
  • Patients with pT2 and pT3a PSMs had a similar prognosis to those with pT3a and pT3b NSMs, respectively (log-rank P ≥ 0.05).

Conclusion

  • A PSM after MIRP is associated with 2.14-fold increased risk of BCR. In patients with pT2 and pT3a disease, a PSM could be considered a pathological upstaging.

Abbreviations
PSM

positive surgical margin

NSM

negative surgical margin

BCR

biochemical recurrence

RP

radical prostatectomy

LRP

laparoscopic RP

RALP

robot-assisted laparoscopic prostatectomy

MIRP

minimally invasive radical prostatectomy

ORP

open radical prostatectomy

IQR

interquartile range

Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Conflict of Interest
  9. References

The ultimate goal of radical prostatectomy (RP) is to safely achieve complete resection of the tumour, with full recovery of continence and potency. Since the first report of standardized and reproducible laparoscopic RP (LRP) [1] and the introduction of robot-assisted laparoscopic prostatectomy (RALP) [2], minimally invasive RP (MIRP) has supplanted open RP (ORP) as the most common surgical approach to RP [3]. A positive surgical margin (PSM) after RP represents incomplete resection of the tumour, resulting in a high probability of salvage treatment as PSMs are an important predictor of biochemical recurrence (BCR) [4], with an estimated 5-year BCR-free survival ranging from 37 to 86% [5]. Recent advances in MIRP in the last 20 years have led to an improvement in morbidity with acceptable cancer control, and PSM rates and pelvic lymph node dissection quality for MIRP are similar to those of ORP [6]. An analysis of a national sample of 5% of US Medicare beneficiaries suggested that men treated with MIRP had a 3.6-fold higher risk of salvage therapies within 6 months of surgery than those having ORP [7], but the risk for these unfavourable outcomes decreases in high-volume surgical centres. Pathological data were not available but the results of that analysis suggest inadequate training and a rapid adoption of MIRP in small-volume centres with a higher PSM rate. Recently, Masterson et al. [8] reported a notably lower PSM rate in pT3 disease after RALP than after ORP and a significantly shorter linear length of PSMs, regardless of the stage; however, Ficarra et al. [9] did not find any difference in terms of PSM rate between MIRP and ORP in a cumulative analysis of published comparative studies.

Long-term oncological outcomes after ORP and, in particular, the impact of PSMs in interaction with intrinsic characteristics of the disease were recently reported by several groups in the literature, whereas such data after MIRP with long-term follow-up are lacking.

The aim of the present study was to assess the relationship between PSM and other established prognostic factors in terms of recurrence-free survival in a large contemporary series of patients undergoing MIRP.

Patients and Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Conflict of Interest
  9. References

Study Design and Population

After institutional review board approval, we retrospectively analysed data prospectively collected between 1998 and 2010 in 4889 consecutive patients with clinically localized prostate cancer who underwent RP at L'Institut Mutualiste Montsouris (IMM) in Paris, France. Patients who received neoadjuvant hormonal therapy (n = 87) or with missing clinical (n = 44) or follow-up information (n = 130) were excluded, leaving 4628 patients for analysis. Clinical and pathological variables, including age at surgery, preoperative PSA level, body mass index, prostate volume, pathological T stage, Gleason score, surgical margin status and lymph node involvement, were prospectively recorded in a digitized database.

Surgical Technique

All patients underwent LRP or RALP according to the standardized technique described by Guillonneau et al. [1]. The indication for and use of a unilateral or bilateral nerve-sparing technique varied according to the surgeon's preference and depended on individual patient characteristics. Standard lymphadenectomy, limited to the external iliac nodal group, was performed in patients with Gleason score ≥4 + 3 or PSA >20 ng/mL or clinical stage ≥ T2c.

Pathological Evaluation

Specimens were evaluated by dedicated genitourinary pathologists according to the Stanford technique [10]. Briefly, all RP specimens were totally inked and fixed in 10% formalin. Prostates were transversally step-sectioned perpendicular to the posterior rectal surface at 3- mm intervals. Pathological reporting included: pathological T stage, Gleason score, seminal vesicle invasion, extraprostatic extension and surgical margin status. A PSM was defined as any cancer in contact with the inked margin of resection in the RP specimen.

Postoperative Assessment

Patients were followed up with DRE and serum PSA measurement at 6 weeks, then semi-annually for the first 3 years, and annually thereafter. BCR was defined as serum PSA >0.2 ng/mL, with a subsequent confirmatory increase or a single detectable PSA value followed by salvage therapy (radiation or androgen deprivation therapy).

Statistical Analysis

Statistical analyses were conducted using SPSS version 17.0 statistical software (IBM Corp., Somers, NY, USA). Patients were censored when the BCR criterion was fulfilled, e.g. at the date of confirmatory PSA increase or initiation of salvage therapy, or at last follow-up or death if the event of interest had not been attained. The impact of PSM on BCR was evaluated using Cox proportional hazards regression after controlling for the effects of preoperative PSA, date of surgery, pathological T stage, Gleason sum and lymph node involvement. After stratifying patients into three stage groups (pT2, pT3a and pT3b), estimates of BCR-free survival were generated according to the surgical margin status using the Kaplan–Meier method and compared among groups using the log-rank test (pairwise over strata). The same analysis was performed after stratifying patients into three groups according to the Gleason score (<7, = 7 and >7). We used the chi-squared test to compare the surgical margin status among pT stage groups. All tests were two-sided, with a P value of <0.05 considered to indicate statistical significance.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Conflict of Interest
  9. References

A total of 4628 patients were included in the study with a median (interquartile range [IQR]) follow-up of 55 (22–84) months. Clinical and pathological characteristics of the cohort are summarized in Table 1. According to the D'Amico risk group classification, 48% (n = 2241) of patients were at low risk, 43% (n = 1967) at intermediate risk and 9% (n = 420) at high risk. All procedures were performed using a laparoscopic approach including 1068 (23%) cases with robotic assistance. Nerve-sparing surgery was performed in a total of 3850 patients: bilateral in 69% (n = 2640) and unilateral in 31% (n = 1210).

Table 1. Demographics and pathological characteristics of the cohort
VariableAll patients, N = 4628
  1. BMI, body mass index.

Median (IQR) age at surgery, years62 (57–66)
Median (IQR) BMI, kg/m225 (23–27)
Median (IQR) PSA, ng/mL7.3 (5.5–10.2)
Median (IQR) prostate volume, mL50 (41–64)
Pathological T stage, n (%) 
pT23258 (70.4)
pT3a1019 (22)
pT3b351 (7.6)
Lymph node involvement, n (%) 
pN0883 (19.1)
pN135 (0.8)
pNx3710 (80.1)
Gleason sum, n (%) 
<71580 (34.1)
72764 (59.7)
>7284 (6.1)
Positive surgical margins, n (%)818 (17.7)

Of the 818 patients (17.7%) with PSMs, pathological stage was T2 in 13%, T3a in 25% and T3b in 34% (P < 0.001). PSMs were 1 mm in size in 28% (n = 230), 2 mm in 23% (n = 185) and ≥3 mm in 49% (n = 403) of cases.

Overall, 107 patients died during follow-up including 21 from prostate cancer, and 952 had BCR, including 334 of the 818 (41%) patients with PSMs. In univariate analyses, preoperative PSA (P < 0.001), date of surgery (P = 0.002), pathological T stage (P < 0.001), lymph node involvement (P < 0.001), Gleason sum (P < 0.001) and surgical margin status (P < 0.001) were all associated with BCR (Table 2). In multivariable analysis, PSM was an independent prognostic factor for recurrence (adjusted hazard ratio [HR]: 2.14 for PSMs vs negative surgical margins [NSMs]; 95% CI: 1.86–2.45; P < 0.001). Other independent predictors for BCR were preoperative PSA, date of surgery, pathological T stage, Gleason sum and lymph node involvement (all P < 0.05). The 5-year BCR-free probability was 80.6% (95% CI: 79–82.2) for NSMs vs 51% (95% CI: 47–55) for PSMs (log-rank P < 0.001). Figure 1 shows the Kaplan–Meier BCR-free survival curves after stratifying by pathological T stage and surgical margin status. Interestingly, patients with pT2 and pT3a PSMs had similar survival rates to those with pT3a and pT3b NSMs, respectively (log-rank P = 0.3). The cohort was further subdivided into Gleason score <7, 7 or >7, according to surgical margin status, and Kaplan–Meier recurrence-free survival curves were generated (Fig. 2). These showed that increasing Gleason score was positively correlated with a higher risk of cancer recurrence. Moreover, patients with Gleason score <7 and PSMs had a worse prognosis than those with Gleason score 7 and NSMs.

figure

Figure 1. Kaplan–Meier curve comparing pathological subgroups as stratified according to tumour stage and surgical margin status (PSM vs NSM). Significant (P < 0.05): pT2-NSM vs pT2-PSM, pT3a-NSM vs pT3a-PSM and pT3b-NSM vs pT3b-PSM. Nonsignificant (P ≥ 0.05): pT2-PSM vs PT3a-NSM and pT3a-PSM vs PT3b-NSM.

Download figure to PowerPoint

figure

Figure 2. Kaplan–Meier curve comparing pathological Gleason score (GS) subgroups (<7, = 7 and >7), stratified according to tumour surgical margin status (PSM vs NSM). All P values <0.05 (pairwise over strata).

Download figure to PowerPoint

Table 2. Univariate and multivariate Cox regression analysis predicting biochemical recurrence after MIRP
CovariableUnivariate analysisMultivariable analysis
HR (95% CI)PHR (95% CI)P
Preoperative PSA (continuously coded), ng/mL1.03 (1.01–1.04)<0.0011.023 (1.013–1.034)<0.001
RP date, by year0.9 (0.88–0.95)0.0020.92 (0.9–0.95)0.01
Pathological T stage    
pT3a vs pT23.06 (2.64–3.5)<0.0012.25 (1.94–2.62)<0.001
pT3b vs pT28.2 (7–9.6) 4.65 (3.88–5.56) 
Lymph node involvement    
pN1 vs pN05.54 (3.77–8.14)<0.0013.8 (2.57–5.63)<0.001
pNx vs pN00.46 (0.4–0.5) 0.93 (0.79–1.09)0.4
Surgical margin status, positive vs negative3 (2.6–3.4)<0.0012.14 (1.86–2.45)<0.001
Gleason sum    
7 vs <72.66 (2.24–3.15)<0.0011.81 (1.52–2.16)<0.001
>7 vs <78.42 (6.8–10.42) 3.66 (2.9–4.6) 

We re-ran the analysis after excluding patients with NSMs (Table 3). The 5-year BCR-free survival probability was 68.8% for pT2 (95% CI: 63.2–74), 41.3% for pT3a (95% CI: 33.9–48.7) and 17.4% for pT3b disease (95% CI: 11.4–23.4; log-rank P < 0.001). In multivariable analyses, PSM size ≥3 mm was an independent prognostic factor for BCR (adjusted HR: 1.56; 95% CI: 1.18–2.04; P = 0.001). Pathological T stage, lymph node involvement and Gleason sum remained independently associated with BCR (all P < 0.05).

Table 3. Multivariate Cox regression analysis predicting BCR after MIRP among patients with PSMs (N = 818)
CovariableHR (95% CI);P
  1. BMI, body mass index.

Preoperative PSA, ng/mL (continuously coded)1.01 (0.98–1.015)0.1
Pathological T stage  
pT3a vs pT22.3 (1.77–2.99)<0.001
pT3b vs pT24.5 (3.3–6.1) 
Lymph node involvement <0.001
pN1 vs pN02.68 (1.48–4.8)0.001
pNx vs pN00.82 (0.63–1.07)0.1
Gleason score <0.001
7 vs <71.2 (0.89–1.6)0.2
>7 vs <72.07 (1.4–3.04)<0.001
Surgical margin size 0.006
2 vs 1 mm1.36 (0.98–1.88)0.06
≥3 vs 1 mm1.56 (1.18–2.04)0.001

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Conflict of Interest
  9. References

After RP for localized prostate cancer, the pathological examination of the surgical specimen provides crucial information regarding prognostic factors that include tumour stage, Gleason score and surgical margin status. Surgical margin status is a surrogate for surgical removal quality [11], but it is also closely related to intrinsic tumour characteristics, particularly tumour stage.

To our knowledge, the present study is one of the largest modern series reporting long-term outcomes after MIRP. The single-centre nature of our study is an advantage as it limits the potential bias resulting from discrepancies in pathological assessment and the inter-observer variability that can be associated with multicentre studies, especially where there is no centralized review of histological slides. The interesting finding of the present study is that PSMs are associated with an increased risk of BCR that is similar to that of a pathological upstaging, regardless of the location and the size of the PSM.

Our PSM rate of 18% is consistent with the PSM rate after MIRP reported in the literature, which ranges from 9.6 to 30% [12-16]. In a recent study involving 8095 patients treated with RALP with an overall PSM rate of 15.7% (n = 1272), pathological stage (4.5-fold increased risk for pT2 vs pT3) and preoperative PSA (2.9-fold increased risk for PSA of ≤4 vs >10 ng/mL) were the most important predictive factors for PSM in multivariable analysis [13]. Also, the impact of PSM on oncological outcomes is influenced by tumour characteristics because it increases the risk for BCR to a greater extent in patients with intermediate- and high-risk disease than in those with low-risk disease [17]. We confirmed the negative impact of a PSM in the present study, with a 2.14-fold increased risk of BCR regardless of stage. Similarly to the report by Preston et al. [18], which involved a series of >8000 RP cases, we found that preoperative PSA, Gleason score, pT stage and lymph node involvement were also independent predictors of BCR. In addition, when we stratified patients according to pathological stage and surgical margin status, we noted a significant impact of PSM in patients with pT2 and pT3a disease that increased the risk of BCR and upstaged these two categories to pT3a and pT3b NSMs, respectively. Budaus et al. [12] tested the interaction between PSM, pT stage and Gleason score in a series of >4000 patients with a median follow-up of 47 months. In multivariate models, only the pT stage and surgical margin status interaction achieved independent predictor status in their report. The estimated 5-year BCR-free survival rates decreased with more advanced stage and/or PSM status and all intergroup differences achieved statistical significance in pairwise comparisons. The series by Budaus et al. involved patients with an inclusion period starting in 1992 and, given the stage migration of prostate cancer attributable to the widespread use of PSA, their different findings may be explained by the larger tumours with similar pathological stages found in cases included at the beginning of their study period, e.g. at the beginning of the PSA era [19].

We also conducted an analysis to evaluate and quantify the effect of prognostic factors on BCR in patients with PSMs. Only 41% of patients with PSMs experienced BCR during follow-up. This is consistent with the report by Ahyai et al. [20], with only 20% of patients with PSMs experiencing BCR during a relative short follow-up of 35 months. We also confirmed that a PSM ≥3 mm was independently associated with a 1.5-fold increased risk of BCR.

Our study highlights the continuing importance of complete surgical removal of both organ-confined and non-organ confined prostate cancers in the era of minimally invasive surgery. Efforts should be made to minimize the risk of PSMs during MIRP, if necessary with the help of preoperative modern imaging that can guide the surgeon to resect further into the tumour area in organ-confined disease and to resect widely enough around the prostate in the case of non-organ confined disease.

Potential limitations of our study include its retrospective design and its single-centre nature. Additionally, we did not analyse the impact of the number (solitary vs multiple) and the location of PSMs on BCR; however, Stephenson et al. [21], using a multivariable model, followed >7000 patients in a large multi-institutional study and found that adding the number and location of PSMs did not improve the predictive accuracy of their nomogram compared with one in which surgical margin status was modelled as positive vs negative, and concluded that there appears to be no justification for the use of this variable in the management of PSM.

In conclusion, a PSM after MIRP is associated with a 2.14-fold increased risk of BCR. In patients with pT2 and pT3a disease, a PSM could be considered a pathological upstaging. Our finding highlights the continuing importance of complete surgical removal of both organ-confined and non-organ-confined prostate cancers in this era of minimally invasive surgery.

Acknowledgments

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Conflict of Interest
  9. References

The authors would like to thank Dr Pierre Colin (Department of Urology, CHU, Lille, France) for his help with the statistical analysis.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Patients and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Conflict of Interest
  9. References
  • 1
    Guillonneau B, Cathelineau X, Barret E, Rozet F, Vallancien G. Laparoscopic radical prostatectomy. Preliminary evaluation after 28 interventions. Presse Med 1998; 27: 15701574
  • 2
    Abbou CC, Hoznek A, Salomon L et al. Laparoscopic radical prostatectomy with a remote controlled robot. J Urol 2001; 165 (6 Pt 1): 19641966
  • 3
    Trinh QD, Sammon J, Sun M et al. Perioperative outcomes of robot-assisted radical prostatectomy compared with open radical prostatectomy: results from the nationwide inpatient sample. Eur Urol 2012; 61: 679685
  • 4
    Chuang AY, Nielsen ME, Hernandez DJ, Walsh PC, Epstein JI. The significance of positive surgical margin in areas of capsular incision in otherwise organ confined disease at radical prostatectomy. J Urol 2007; 178 (4 Pt 1): 13061310
  • 5
    Swanson GP, Basler JW. Prognostic factors for failure after prostatectomy. J Cancer 2012; 2: 119
  • 6
    Touijer K, Eastham JA, Secin FP et al. Comprehensive prospective comparative analysis of outcomes between open and laparoscopic radical prostatectomy conducted in 2003 to 2005. J Urol 2008; 179: 18111817; discussion 1817
  • 7
    Tewari AK, Jhaveri JK, Surasi K, Patel N, Tan GY. Benefit of robotic assistance in comparing outcomes of minimally invasive versus open radical prostatectomy. J Clin Oncol 2008; 26: 49995000; author reply 5001–2
  • 8
    Masterson TA, Cheng L, Boris RS, Koch MO. Open vs. robotic-assisted radical prostatectomy: a single surgeon and pathologist comparison of pathologic and oncologic outcomes. Urol Oncol 2012; [Epub ahead of print]
  • 9
    Ficarra V, Novara G, Artibani W et al. Retropubic, laparoscopic, and robot-assisted radical prostatectomy: a systematic review and cumulative analysis of comparative studies. Eur Urol 2009; 55: 10371063
  • 10
    Stamey TA, Yemoto CM, McNeal JE, Sigal BM, Johnstone IM. Prostate cancer is highly predictable: a prognostic equation based on all morphological variables in radical prostatectomy specimens. J Urol 2000; 163: 11551160
  • 11
    Eastham JA, Kattan MW, Riedel E et al. Variations among individual surgeons in the rate of positive surgical margins in radical prostatectomy specimens. J Urol 2003; 170 (6 Pt 1): 22922295
  • 12
    Budaus L, Isbarn H, Eichelberg C et al. Biochemical recurrence after radical prostatectomy: multiplicative interaction between surgical margin status and pathological stage. J Urol 2010; 184: 13411346
  • 13
    Patel VR, Coelho RF, Rocco B et al. Positive surgical margins after robotic assisted radical prostatectomy: a multi-institutional study. J Urol 2011; 186: 511516
  • 14
    Mottrie A, Van Migem P, De Naeyer G, Schatteman P, Carpentier P, Fonteyne E. Robot-assisted laparoscopic radical prostatectomy: oncologic and functional results of 184 cases. Eur Urol 2007; 52: 746750
  • 15
    Patel VR, Thaly R, Shah K. Robotic radical prostatectomy: outcomes of 500 cases. BJU Int 2007; 99: 11091112
  • 16
    Curto F, Benijts J, Pansadoro A et al. Nerve sparing laparoscopic radical prostatectomy: our technique. Eur Urol 2006; 49: 344352
  • 17
    Alkhateeb S, Alibhai S, Fleshner N et al. Impact of positive surgical margins after radical prostatectomy differs by disease risk group. J Urol 2010; 183: 145150
  • 18
    Preston MA, Carriere M, Raju G et al. The prognostic significance of capsular incision into tumor during radical prostatectomy. Eur Urol 2011; 59: 613618
  • 19
    Mouraviev V, Villers A, Bostwick DG, Wheeler TM, Montironi R, Polascik TJ. Understanding the pathological features of focality, grade and tumour volume of early-stage prostate cancer as a foundation for parenchyma-sparing prostate cancer therapies: active surveillance and focal targeted therapy. BJU Int 2011; 108: 10741085
  • 20
    Ahyai SA, Zacharias M, Isbarn H et al. Prognostic significance of a positive surgical margin in pathologically organ-confined prostate cancer. BJU Int 2010; 106: 478483
  • 21
    Stephenson AJ, Wood DP, Kattan MW et al. Location, extent and number of positive surgical margins do not improve accuracy of predicting prostate cancer recurrence after radical prostatectomy. J Urol 2009; 182: 13571363