Marges chirurgicales après prostatectomie radicale laparoscopique et robot assistée pour tumeur localisée à la glande : l’indice de masse corporelle fait il la différence ?

25 juin 2018

Auteurs : S. Albisinni, J. Grosman, F. Aoun, T. Quackels, A. Peltier, R. Van Velthoven, T. Roumeguère
Référence : Prog Urol, 2018, 8, 28, 434-441




 




Introduction


Prostate cancer (PCa) is a major health concern worldwide, being the second most common neoplasm and sixth cause of cancer-related death in the world [1]. Robotic-assisted prostatectomy (RALP) is becoming the new standard of treatment in the local control of disease, with comparable oncologic outcomes to the open approach [2]. The ability to fully resect the prostate and avoid positive surgical margins (PSMs) is one of the most important objectives of prostatectomy, although there is still debate concerning the impact of a PSMs on long-term oncologic outcomes [3, 4]. Indeed, patients with PSMs are at increased risk of biochemical recurrence [3, 5] and receive more frequently adjuvant radiotherapy. These can impact on patients' quality of life [6] and be a source of stress and anxiety [7].


Multiple risk factors for PSMs have been identified, including prostate volume, tumor volume, elevated Gleason score and, most of all, pT3 disease [3, 4, 8]. Obesity has also been associated to PSMs in several studies performed on open prostatectomy series [9, 10, 11]. The minimally invasive approach (RALP and laparoscopic radical prostatectomy, LRP) may ease surgical access and visibility in obese men, simplifying the procedure in this population. This could result in a reduction of PSMs after minimally invasive RP in the obese population. Recently, Suardi et al reported a reduction of PSMs after RALP compared to open radical prostatectomy [8]: whether these findings can be extended to obese men remains unknown as there is a paucity of data concerning the impact of body habitus and obesity on the risk of PSMs after RALP.


Aim of the current study is to explore PSMs is a cohort of patients undergoing minimally invasive RP (RALP and LRC), and determine whether obesity may be a risk factor for PSMs and their localization.


Patients and methods


We have retrospectively identified 1005 patients treated by minimally invasive radical prostatectomy between January 2005 and December 2015 in two academic hospitals. The study received institutional review board approval and patients signed written consent to manage their data. All patients operated until 2010 underwent a standard laparoscopic procedure (i.e. LRP), while beginning in January 2010 men were operated by a robotic-assisted approach (i.e. RALP). Institutional review board approval was obtained in both centres. Before surgery, patients' weight and height were measured and BMI calculated as kg/m2. Patients were then categorized in BMI<25kg/m2, BMI 25-29.9kg/m2 and BMI >30kg/m2.


We excluded from our analysis patients with pT3 disease on final specimen analysis, in order to reduce disease specific risk factors [8] for PSMs and focus on surgical technique and body habitus. This included 398 men with pT3 in our study. Three patients with pN+ were also excluded as well as 65 due to missing data, leaving a cohort of 539 men for final analysis.


Surgeries were performed by four experimented surgeons (TR, TQ, AP, RVV). All procedures were completed via a transperitoneal approach following the same technique. Pathologic examination was performed in two laboratories and total embedding specimens analyzed by dedicated uro-pathologists. A positive surgical margin was defined as the presence of at least one malignant gland in contact with the inked surface of the specimen, and was measured in millimeters.PSMs were further dichotomized in <4mm and ≥4mm, as suggested by Epstein et al. [12]. The outcome measured was rate of PSM according to the BMI and surgical approach.BCR was defined by a PSA levels>0.2ng/ml and subsequently rise [1]. The date of the first PSA ≥0.2ng/ml was used to define biochemical recurrence.


Statistical analysis


Variables were compared across BMI groups using Kruskall-Wallis or &khgr; 2 test, as appropriate. Uni- and multivariate logistic regression models were constructed to assess the impact of BMI and surgical technique on PSM risk. Variables tested included PSA (continuous,logarithmically transformed due to non-parametric distribution), BMI (continuous), prostate volume (continuous, logarithmically transformed due to non-parametric distribution), surgical technique (categoric, LRP vs RALP), Gleason score (categoric, 6 vs 7 vs ≥8) and pT2 substage (categoric, 2a vs 2b vs 2c).


To explore risk factors for BCR, univariate and multivariate Cox proportional hazard models were performed. The same variables as above were tested as risk factors for BCR, including surgical margin status. Kaplan-Meier curves were calculated to estimate BCR-free survival rates for the entire cohort and stratified by BMI group and margin status.


All analyses and graphics were performed using the Stata software version 12. A two-sided p <0.05 defined statistical significance.


Results


The clinical and pathological characteristics of the 539 patients are listed in Table 1. Mean follow-up was 42±32months. Pelvic lymph node dissection was performed in 149 (28%) patients. All these patients had no evidence of lymph node invasion.


Overall, 127 (24%) of men had PSM detected on final specimen evaluation. Mean PSM length was 3.9±3.4mm, and 30 (6%) men presented margins ≥4mm (Table 2). The vast majority of PSM were localized either at the apex (58%) or on the posterior (24%) surface of the prostate, while basal (9%) and lateral (9%) PSMs were less frequent. Moreover, 19/127 (15%) patients had a second PSM on their specimen, maintaining the same anatomic distribution (apical>posterior>basal>lateral).


When exploring patients according to BMI status, no significant difference was found across group with regards to PSA, prostate volume, pT and Gleason score (all P >0.12) (Table 2). Of note, obese patients (BMI>30kg/m2) were more likely to be upstaged at final pathologic examination (Table 2), although the difference with lean patients did not reach statistical significance. Analysing the rate of PSMs within BMI categories, no significant association between increased BMI and PSMs was detected (Table 3). Nonetheless, when comparing obese men to their lean counterpart, a non-significant trend was observed, as obese men had an absolute higher rate of PSM (29% vs 22%), longer margins (4.1mm vs 3.4mm) and higher rate of margins ≥4mm (8% vs 4%) (Table 3).


PSM were also explored as a function of minimally invasive technique (i.e. laparoscopy vs robotic-assisted surgery). Globally, PSMs were more frequent in LRP (27%) compared to RALP (22%), though this difference was non-significant (P =0.15). Moreover, in LRP PSMs were longer (4.3mm vs 3.6mm) and we detected a higher rate of margins ≥4mm (8% vs 4%), though these differences across techniques did not reach statistical significance (P ≥0.15).


On uni- and multivariate logistic regression, PSA (P =0.039) and prostate volume (P <0.001) were significantly associated to PSMs. BMI was not a statistically significant risk factor for PSM (P =0.14), neither was the minimally invasive technique (LRC vs RALP) used (P =0.54).


During the follow-up period, 53/539 (9.8%) of patients experienced BCR. Biochemical recurrence-free survival (bRFS) rates were 89% (95%CI: 84-93%) at 5 years. bRFS significantly varied according to specimen Gleason score (P <0,001) (Figure 1a) and margin status (P =0.01) (Figure 1b), while BMI (P =0.42) had no impact on bRFS (Figure 1c). PSA (P <0.001), margin status (P =0.007) and Gleason score (P <0.001) were significantly associated to BCR on logistic regression (Table 4).


Figure 1
Figure 1. 

Biochemical recurrence free survival according to Gleason score (1a), margin status (1b) and BMI (1c).





Discussion


RP is a common and frequently performed procedure in urology and avoiding a PSM is one of the surgeon's most cardinal thoughts for localized disease. There are multiple recognized risk factors for PSMs: the most important are disease specific, including pT stage, Gleason score, tumor volume [3, 4, 8], while others are non-disease related. In particular, BMI has emerged as a significant risk factor in several studies exploring PSMs during open RP [9, 10, 11, 13]. Minimally-invasive surgery has surged to a standard in the field of PCa, with open RP rates dropping significantly in US and Europe in the last ten years [14, 15]. The magnified vision, increased access to the pelvis and high dexterity provided by the surgical robot are all reasons for the rise of robotic-assisted surgery in the recent years [14]. Moreover, investigators have found that RALP could be superior to open RP in terms of PSM rate [8, 16]. Albeit RALP in obese men is definitely feasible [17], data are lacking concerning the rate of PSM in such population when prostatectomy is performed via a minimally invasive approach. We herein report an in depth analysis of PSMs after RALP and LRP as a function of body habitus and BMI, exploring PSM length and anatomical localization. In the current study we did not detect a statistically significant difference in terms of PSMs after minimally invasive RP across various BMI categories. Moreover, PSM anatomical distribution was also similar no matter body habitus, as well as length of the margin. Nonetheless, a non-significant tendency toward higher rates of PSM in obese men (BMI>30kg/m2) was seen, with a slightly higher incidence of PSM, longer margins and higher rate of significant margins (≥4mm), though all these differences did not reach statistical significance.


Obesity did not appear to be a significant risk factor for PSMs in the current study. This is in line with published data exploring perioperative outcomes of obese men during minimally invasive RP [18, 19, 20]. Xu et al. reviewed current studies on RALP in the obese and, analyzing 1821 obese vs 4801 non-obese men, did not detect a higher incidence of PSMs in the former group. Sundi et al analyzed 987 men undergoing RALP of which 193 presented a BMI>30kg/m2, finding no significant difference (P =0.14) in PSM rate across BMI categories. Finally, Zilberman et al. reported a PSM rate of 18.5% in 146 obese men harboring pT2 disease, with no significant difference compared to the lean counterpart (P =0.35) [21]. Most investigators account the enhanced vision, the increased dexterity and the improved access to the pelvis of the robotic-assistance to explain the absence of difference in PSMs seen in the minimally invasive approach compared to open surgery [11]. Moreover, we did not record significantly longer PSM in the overweight and obese population of our cohort.


Concerning localization, in the present study we detected mainly apical (58%) and posterior (24%) PSMs. This anatomical localization was quite homogenous across BMI categories, with no significant difference across lean, overweight or obese men (P =0.75). Similarly, in the 19 patients in whom a second PSM was detected, the same anatomical distribution was found, with apical PSMs being the most frequent, followed by posterior margins. Interestingly, apical PSMs were numerically more frequent in lean patients than in the obese cohort, although the apical dissection is considered one of the most challenging in open RP in the obese [11]. Zilberman et al. also reported similar anatomical distribution of PSM, with obese men presenting a lower rate of apical margins compared to lean patients [21].


Sooriakumaran et al. recently published a large study comparing PSM rate across open, laparoscopic and robotic-assisted RP [16]. Indeed, both RALP (14%) and LRP (16%) presented lower rates of PSMs compared to the open approach (23%). Similarly Evans et al., analyzing 2175 consecutive RP specimens, reported lower rates of PSMs after RALP (20%) and LRP (25%) compared to open RP (33%) [22]. Moreover, Suardi et al. analyzed risk factors for PSM in a large retrospective cohort from the San Raffaele hospital in Milan [8], and found that RALP was a significant protective factor (OR0.89, 95%CI 0.82-0.97, P =0.01). In our study we lack of an open RP arm for comparison, however, we evaluated whether the type of minimally approach (i.e. laparoscopy vs robotic-assisted laparoscopy) could have been associated to an increase in PSM. Notably, the absence of tactile feedback has been advanced by investigators as a possible detriment of the robotic platform, and potentially a risk factor of inadvertent tumor effraction. On the other hand, robotic-assisted surgery gives the operator increased dexterity and manoeuvrability compared to conventional laparoscopy, easing dissection. Indeed, we did not observe any statistically significant association between type of minimally invasive approach and PSM risk. This is supported by other studies in which LRP and RALP yielded similar results in terms of PSM rate: Busch et al reported 17.4% PSM in LRP vs 22.9% in RALP in men with pT2 PCa and a BMI>30kg/m2 [19]. Furthermore, in a review and cumulative analysis of comparative studies, Ficarra et al. found that PSM rates were similar between RALP and LRP (RR: 0.97; 95% CI of RR: 0.65-1.46; P =0.9) [23].


Our study is not devoid of limitations. First, the lack of an open RP arm does not allow us to draw comparisons between our minimally invasive outcomes and those of open approach. Second, the study is retrospective in nature, thus its results must be managed with care. Finally, the short follow-up available for a PCa study is an obstacle to a thorough analysis of long-term outcomes of PSMs.


Conclusions


Obese men were not at significant risk of having PSMs at the time of minimally invasive RP compared to lean and overweight men in this retrospective study. Localization of PSMs was also similar no matter the body habitus. Moreover, we did not observe a significant difference in PSM rate between a laparoscopic versus a robotic-assisted approach. Larger prospective studies are necessary to validate our findings.


Disclosure of interest


The authors declare that they have no competing interest.




Table 1 - General characteristics of the cohort (n =539).
Age (years)    
Median (IQR)  63 (58-67) 
Mean±SD  62±
BMI (kg/m 2 )    
Median (IQR)  26.1 (24.3-28.41) 
Mean±SD  26.5±3.6 
PSA (ng/ml)    
Median (IQR)  6.5 (4.6-9.2) 
Mean±SD  7.8±5.5 
Prostate volume (ml)    
Median (IQR)  52 (44-65) 
Mean±SD  57±22 
ASA score    
54 (10%) 
325 (60%) 
31 (6%) 
Unknown  129 (24%) 
Surgical technique    
Laparoscopy  192 (36%) 
Robotic-assisted  347 (64%) 
Biopsy Gleason score    
368 (68%) 
3+ 114 (21%) 
4+ 31 (6%) 
>8  26 (5%) 
pT    
2a  51 (9%) 
2b  42 (8%) 
2c  446 (83%) 
Pathologic Gleason score    
333 (62%) 
3+ 148 (27%) 
4+ 37 (7%) 
>8  21 (4%) 
Lymphadenectomy    
Nx  390 (72%) 
N0  149 (28%) 
Surgical margins    
R0  412 (76%) 
R1  127 (24%) 





Table 2 - Characteristics of the cohort according to BMI.
  BMI<25kg/m2  BMI 25-29.9kg/m2  BMI≥30kg/m2  P  
Number of patients (n)   192  266  84   
Age (years)          
Median (IQR)  62 (57-67)  63 (58-68)  63 (58-67)  0.52 
Mean±SD  62± 63± 62±  
PSA(ng/ml)         
Median (IQR)  6.1 (4.4-8.9)  6.5 (4.7-9.4)  7.4 (5.4-9.2)  0.12 
Mean±SD  7.5±5.2  8.0±6.2  8.2±4.9   
Prost vol (ml)          
Median (IQR)  51 (42-63)  53 (44-65)  52 (45-75)  0.39 
Mean±SD  57±22  57±21  60±25   
ASA Score         0.5 
21 (11%)  29 (11%)  4 (5%)   
115 (60%)  156 (59%)  54 (64%)   
11 (6%)  14 (5%)  6 (7%)   
unknown  45 (23%)  67 (25%)  20 (24%)   
Surgical technique        0.64 
Laparoscopy  64 (33%)  95 (36%)  33 (39%)   
Robotic-assisted  127 (67%)  169 (64%)  51 (61%)   
Biopsy Gleason score         
132 (69%)  176 (66%)  60 (71%)   
3+ 42 (22%)  58 (22%)  15 (18%)   
4+ 12 (7%)  12 (5%)  6 (7%)   
≥8  5 (3%)  18 (7%)  3 (4%)   
pT        0.89 
2a  16 (8%)  26 (10%)  9 (11%)   
2b  17 (9%)  18 (7%)  7 (8%)   
2c  158 (83%)  220 (83%)  68 (81%)   
Pathologic Gleason score         
125 (65%)  158 (59%)  50 (59%)   
3+ 49 (26%)  74 (28%)  25 (30%)   
4+ 13 (7%)  17 (6%)  7 (9%)   
≥8  4 (2%)  15 (7%)  2 (2%)   
Gleason upstaging  27 (14%)  45 (17%)  18 (21%)  0.31 
Lymphadenectomy        0.85 
Nx  141 (74%)  189 (72%)  60 (71%)   
N0  50 (26%)  75 (28%)  24 (29%)   





Table 3 - Margin status of entire cohort and according to BMI.
  Entire cohort  BMI<25kg/m2  BMI 25-29.9kg/m2  BMI ≥30kg/m2  P  
Positive margin            
Overall margins           
R0  412 (76%)  149 (78%)  203 (77%)  60 (71%)  0.48 
R1  127 (24%)  42 (22%)  61 (23%)  24 (29%)   
Margin lengths (mm)  3.9±3.4  3.4±3.1  4.1±3.5  4.1±3.5  0.58 
PSM ≥4mm  30 (6%)  7 (4%)  16 (6%)  7 (8%)  0.65 
Margin localisation          0.75 
Apical/Anterior  73 (58%)  26 (63%)  35 (58%)  12 (50%)   
Posterior  31 (24%)  6 (13%)  17 (28%)  8 (34%)   
Basal  11 (9%)  5 (12%)  4 (6%)  2 (8%)   
Lateral  12 (9%)  5 (12%)  5 (8%)  2 (8%)   
Second positive margin            
2nd positive margin  19/127 (15%)  8 (19%)  9 (15%)  2 (8%)  0.5 
2nd margin length  3.1±2.5  1.8±0.4  4.3±3.3  2.5±0.7  0.38 
2nd margin localisation          0.24 
Apical/Anterior  11 (58%)   
Posterior  5 (26%)   
Basal  2 (11%)   
Lateral  1 (5%)   





Table 4 - Uni- and multivariate logistic regression exploring risk factors for positive margins. All patients harbored pT2 disease.
  Univariate 
Multivariate 
  OR  95%CI  P   OR  95%CI  P  
PSA   1.28  0.90-1.84  0.17  1.54  1.02-2.33  0.039 
BMI   1.04  0.98-1.09  0.21  1.04  0.99-1.10  0.14 
Prostate volume   0.97  0.95-0.98  <0.001  0.96  0.95-0.98  <0.001 
Surgical tech a  0.74  0.49-1.11  0.15  0.86  0.53-1.39  0.54 
Gleason              
Ref  Ref  Ref  Ref  Ref  Ref 
1.37  0.9-2.07  0.14  0.98  0.63-1.5  0.95 
> 1.48  0.55-3.9  0.44  1.19  0.41-3.4  0.75 
pT              
2a  Ref  Ref  Ref  Ref  Ref  Ref 
2b  1.47  0.51-4.2  0.48  1.21  0.40-3.66  0.73 
2c  1.76  0.80-3.86  0.16  1.51  0.66-3.45  0.33 



[a] 
LRC vs RALP.


References



Center M.M., Jemal A., Lortet-Tieulent J., Ward E., Ferlay J., Brawley O., et al. International variation in prostate cancer incidence and mortality rates Eur Urol 2012 ;  61 (6) : 1079-1092 [cross-ref]
Heidenreich A., Bastian P.J., Bellmunt J., Bolla M., Joniau S., van der Kwast T., et al. EAU guidelines on prostate cancer. Part 1: screening, diagnosis, and local treatment with curative intent-update 2013 Eur Urol 2014 ;  65 (1) : 124-137 [cross-ref]
Boorjian S.A., Karnes R.J., Crispen P.L., Carlson R.E., Rangel L.J., Bergstralh E.J., et al. The impact of positive surgical margins on mortality following radical prostatectomy during the prostate specific antigen era J Urol 2010 ;  183 (3) : 1003-1009 [cross-ref]
Mauermann J., Fradet V., Lacombe L., Dujardin T., Tiguert R., Tetu B., et al. The impact of solitary and multiple positive surgical margins on hard clinical end points in 1712 adjuvant treatment-naive pT2-4 N0 radical prostatectomy patients Eur Urol 2013 ;  64 (1) : 19-25 [cross-ref]
Karakiewicz P.I., Eastham J.A., Graefen M., Cagiannos I., Stricker P.D., Klein E., et al. Prognostic impact of positive surgical margins in surgically treated prostate cancer: multi-institutional assessment of 5831 patients Urology 2005 ;  66 (6) : 1245-1250 [inter-ref]
Suardi N., Gallina A., Lista G., Gandaglia G., Abdollah F., Capitanio U., et al. Impact of adjuvant radiation therapy on urinary continence recovery after radical prostatectomy Eur Urol 2014 ;  65 (3) : 546-551 [cross-ref]
Tavlarides A.M., Ames S.C., Diehl N.N., Joseph R.W., Castle E.P., Thiel D.D., et al. Evaluation of the association of prostate cancer-specific anxiety with sexual function, depression and cancer aggressiveness in men 1 year following surgical treatment for localized prostate cancer Psychooncology 2013 ;  22 (6) : 1328-1335 [cross-ref]
Suardi N., DellOglio P., Gallina A., Gandaglia G., Buffi N., Moschini M., et al. Evaluation of positive surgical margins in patients undergoing robot-assisted and open radical prostatectomy according to preoperative risk groups Urol Oncol Semin Orig Investig 2016 ;  34 (2) : [57e1-57e7].
Freedland S.J., Grubb K.A., Yiu S.K., Nielsen M.E., Mangold L.A., Isaacs W.B., et al. Obesity and capsular incision at the time of open retropubic radical prostatectomy J Urol 2005 ;  174 (5) : 1798-1801[discussion 1801].
 [cross-ref]
Siddiqui S.A., Inman B.A., Sengupta S., Slezak J.M., Bergstralh E.J., Leibovich B.C., et al. Obesity and survival after radical prostatectomy: a 10-year prospective cohort study Cancer 2006 ;  107 (3) : 521-529 [cross-ref]
Database Study group SEARCH, Jayachandran J., Aronson W.J., Terris M.K., Presti J.C., Amling C.L., et al. Obesity and positive surgical margins by anatomic location after radical prostatectomy: results from the shared equal access regional cancer hospital database BJU Int 2008 ;  102 (8) : 964-968
Brimo F., Partin A.W., Epstein J.I. Tumor grade at margins of resection in radical prostatectomy specimens is an independent predictor of prognosis Urology 2010 ;  76 (5) : 1206-1209 [inter-ref]
Lee S.E., Lee W.K., Jeong M.S., Abdullajanov M., Kim D.S., Park H.Z., et al. Is body mass index associated with pathological outcomes after radical prostatectomy in Korean men? BJU Int 2011 ;  107 (8) : 1250-1255
Stolzenburg J.U., Kyriazis I., Fahlenbrach C., Gilfrich C., Günster C., Jeschke E., et al. National trends and differences in morbidity among surgical approaches for radical prostatectomy in Germany World J Urol 2016 ;
Hu J.H., O'Malley P., Chughtai B., Isaacs A., Mao J., Wright J.D., et al. Comparative effectiveness of cancer control and survival after robot-assisted versus open radical prostatectomy J Urol 2016 ;
Sooriakumaran P., Srivastava A., Shariat S.F., Stricker P.D., Ahlering T., Eden C.G., et al. A multinational, multi-institutional study comparing positive surgical margin rates among 22393 open, laparoscopic, and robot-assisted radical prostatectomy patients Eur Urol 2014 ;  66 (3) : 450-456 [cross-ref]
Ellimoottil C., Roghmann F., Blackwell R., Kadlec A., Greco K., Quek M.L., et al. Open versus robotic radical prostatectomy in obese men Curr Urol 2015 ;  8 (3) : 156-161
Xu T., Wang X., Xia L., Zhang X., Qin L., Zhong S., et al. Robot-assisted prostatectomy in obese patients: how influential is obesity on operative outcomes? J Endourol Endourol Soc 2015 ;  29 (2) : 198-208 [cross-ref]
Busch J., Gonzalgo M.L., Leva N., Ferrari M., Cash H., Kempkensteffen C., et al. Matched comparison of robot-assisted, laparoscopic and open radical prostatectomy regarding pathologic and oncologic outcomes in obese patients World J Urol 2015 ;  33 (3) : 397-402 [cross-ref]
Sundi D., Reese A.C., Mettee L.Z., Trock B.J., Pavlovich C.P. Laparoscopic and robotic radical prostatectomy outcomes in obese and extremely obese men Urology 2013 ;  82 (3) : 600-605 [inter-ref]
Zilberman D.E., Tsivian M., Yong D., Ferrandino M.N., Albala D.M. Does body mass index have an impact on the rate and location of positive surgical margins following robot assisted radical prostatectomy? Urol Oncol 2012 ;  30 (6) : 790-793 [cross-ref]
Evans S.M., Millar J.L., Frydenberg M., Murphy D.G., Davis I.D., Spelman T., et al. Positive surgical margins: rate, contributing factors and impact on further treatment: findings from the Prostate Cancer Registry BJU Int 2014 ;  114 (5) : 680-690 [cross-ref]
Ficarra V., Novara G., Artibani W., Cestari A., Galfano A., Graefen M., et al. Retropubic, laparoscopic, and robot-assisted radical prostatectomy: a systematic review and cumulative analysis of comparative studies Eur Urol 2009 ;  55 (5) : 1037-1063 [cross-ref]






© 2018 
Elsevier Masson SAS. Tous droits réservés.