Tumor–parenchyma interface and long-term oncologic outcomes after robotic tumor enucleation for sporadic renal cell carcinoma

ARTICLE IN PRESS

Urologic Oncology: Seminars and Original Investigations 000 (2018) 1−11

Original Article

Tumor−parenchyma interface and long-term oncologic outcomes after robotic tumor enucleation for sporadic renal cell carcinoma ,

D1X XAndrea Minervini, D2X XM.D., Ph.D.a,* 1, D3X XRiccardo Campi, D4X XM.D.a,1, D5X XFabrizio Di Maida, D6X XM.D.a, D7X XAndrea Mari, D8X XM.D.a, D9X XIlaria Montagnani, D10X XM.D.b, D1X XRiccardo Tellini, D12X XM.D.a, D13X XAgostino Tuccio, D14X XM.D., Ph.D.a, D15X XGiampaolo Siena, D16X XM.D., Ph.D.a, D17X XGianni Vittori, D18X XM.D., Ph.D.a, D19X XAlberto Lapini, D20X XM.D.a, D21X XMaria Rosaria Raspollini, D2X XM.D., Ph.D.b, D23X XMarco Carini, D24X XM.D.a a b

Department of Urology, University of Florence, Careggi Hospital, Florence, Italy Department of Pathology, University of Florence, Careggi Hospital, Florence, Italy

Received 17 April 2018; received in revised form 15 July 2018; accepted 22 August 2018

Abstract Objective: Tumor enucleation has been shown to be oncologically safe for elective treatment of renal cell carcinoma (RCC); yet, evidence on long-term oncologic outcomes after robotic tumor enucleation is lacking. In this study we provide a detailed histopathological analysis of tumor−parenchyma interface and the long-term oncologic outcomes after robotic tumor enucleation for sporadic RCC in a high-volume referral center. Materials and Methods: We selected consecutive patients undergoing robotic tumor enucleation for sporadic RCC by experienced surgeons with at least 4 years of follow-up. Pattern of pseudocapsule (PC) invasion, thickness of healthy renal margin removed with the tumor, margin status and recurrence rate were the main study endpoints. Multivariable models evaluated independent predictors of PC invasion. Results: Overall, 140 patients were eligible for the study. Of these, 127 (91%) had complete data available for analysis. Median thickness of healthy renal margin was 0.57 mm (interquartile range [IQR] 0.24−103). A distinct peritumoral PC was present in 121/127 (95%) tumors with a median thickness of 0.28 mm (IQR 0.14−0.45). In 24/121 (19.8%) cases, RCC showed complete PC invasion. At multivariable analysis, increasing tumor diameter, endophytic rate > 50% and papillary histology were significantly associated with complete PC invasion. Positive surgical margins were reported in 3/127 (2.4%) cases. At a median follow-up of 61 months (range 48−76), one patient died due to metastatic RCC. Among patients alive at follow-up, no cases of recurrence at the enucleation site were recorded, while three cases (2.4%) of renal recurrence (elsewhere in the ipsilateral kidney) and three cases (2.4%) of systemic recurrence were found. Conclusions: Distinct RCC-related features were associated with complete PC invasion. By providing a microscopic layer of healthy renal margin in almost all cases, robotic tumor enucleation achieved negative surgical margins in the vast majority of patients, even in case of complete PC invasion. At long-term follow-up, no recurrences were found at the enucleation site. Although our findings need to be confirmed by larger studies with longer follow-up, robotic tumor enucleation appears oncologically safe in experienced hands for the treatment of sporadic RCC. Ó 2018 Elsevier Inc. All rights reserved.

Keywords: Enucleation; Partial nephrectomy; Pseudocapsule; Recurrence; Renal Cell Carcinoma; Robotics

1. Introduction *Corresponding author. Tel.: +390552758011; +390557949209; +393475865716; fax: +390552758014. E-mail address: [email protected] (A. Minervini). 1 These Authors contributed equally to this work. https://doi.org/10.1016/j.urolonc.2018.08.014 1078-1439/Ó 2018 Elsevier Inc. All rights reserved.

Current guidelines recommend prioritizing partial nephrectomy (PN) for the treatment of localized renal masses when surgery is indicated [1, 2].

ARTICLE IN PRESS 2

Minervini, Campi et al. / Urologic Oncology: Seminars and Original Investigations / 00 (2018) 1−11

In this regard, adoption of robotic PN is increasing [3], especially in tertiary referral centers where it is performed for the entire spectrum of anatomic tumor complexity [4]. Tumor enucleation is a well-established technique for tumor excision during PN [5] and it is increasingly performed for T1 renal masses at high-volume centers worldwide [6]. Noteworthy, tumor enucleation has been shown to be oncologically safe for elective treatment of sporadic renal cell carcinoma (RCC) [7−11]. However, robust evidence on histological features and long-term oncologic outcomes after robotic tumor enucleation is lacking [9,10,12]. To fill this gap, in this study we provide a detailed histopathological analysis of tumor−parenchyma interface and the long-term oncologic outcomes after robotic tumor enucleation for sporadic RCC at a high-volume referral center. 2. Materials and methods 2.1. Patients and dataset After institutional Review Board approval, we retrospectively reviewed our prospectively collected RCC-database to select consecutive patients undergoing robotic tumor enucleation with no ablation of the tumor bed by experienced surgeons at our Institution as elective treatment of sporadic RCC between January 2011 and December 2013 (with at least 4 years of potential follow-up). Patients with benign renal tumors were excluded from the study. 2.2. Surgical technique Robotic partial nephrectomy was performed by four highly experienced surgeons with an enucleative intent in all patients. The rational for tumor enucleation is grounded into specific anatomic characteristics of the kidney tumor−parenchyma interface that allow definition of a constant anatomic dissection plane for tumor excision [13]. Our technique of robotic tumor enucleation has been described in detail [14] and mirrors the principles of open tumor enucleation, performed in our Institution since 1986 [7]. In brief, once a radial nephrotomy 1 to 2 mm from the lesion has been made, the natural, relatively avascular anatomic dissection plane between the peritumoral pseudocapsule and healthy renal parenchyma is developed by blunt dissection using circumferential, dynamic tractions with the two robotic arms that lift the tumor off the parenchymal bed (Fig. 1). Regardless of surgical approach, the intent of tumor enucleation is complete excision of the tumor with negative surgical margins, avoiding major injuries to both renal vasculature and urinary collecting system and maximizing preservation of healthy renal parenchyma. By developing

the anatomic cleavage plane, tumor enucleation also allows maximal vascularized nephron mass preservation. An early unclamping strategy is always adopted to minimize warm ischemia if a pure off-clamp approach is not feasible. 2.3. Histopathological analysis Handling of PN specimens and histopathological analysis were performed by dedicated uropathologists (Maria Rosaria Raspollini (MRR), Ilaria Montagnani (IM)) according to standardized Institution protocols [7,15]. Tumor histotype was assigned by the same uropathologists. Patients with benign renal tumors were excluded from the study. Tumor stage was classified according to the 2010 TNM criteria; nucleolar grading and histopathological classification of RCC followed the most recent International Society of Urological Pathology recommendations [16,17]. The entire tumor−parenchyma interface was examined by a dedicated uropathologist (MRR) as previously described [15]. Tumors were serially sectioned; then minimum, mean and maximum thickness of peritumoral PC and healthy renal margin were evaluated at 5-magnification and measured with a millimeter lens [7,15]. Then, mean pseudocapsule (PC) thickness was calculated as the mean value of 5 equally spaced independent measures, while mean healthy renal margin thickness as the mean value of 10 equally spaced independent measures on each microscopic slide [15]. PC invasion at the tumor−parenchyma interface was reported according to a standardized classification system proposed by our group in 2009 for open TE [7]. Degree of PC invasion was classified as: (a) absent (PC¡), if PC was uniformly intact without evidence of invasion by tumor cells; (b) partial (PC+), in case of tumor invading the PC layer but not the adjacent renal tissue; (c) complete (PC++), if the tumor invasion overcame the peritumoral PC into the surrounding healthy parenchyma (with or without positive surgical margins) (Fig. 2). After review of histologic slides by the uropathologists (MRR, IM), the PC¡, PC+ and PC++ invasion categories were assigned 0, 1, and 2 points, respectively, according to the recently proposed i-Cap scoring system [18]. Positive surgical margins were defined as presence of neoplastic cells directly in contact with the inked surface of the specimen. 2.4. Follow-up Follow-up after robotic tumor enucleation was performed according to a standardized Institution protocol. All patients underwent physical examination, routine laboratory tests (including evaluation of renal function) and computed tomography (CT) scan of the chest and abdomen 6 months after surgery and then yearly for the first 5 years.

ARTICLE IN PRESS Minervini, Campi et al. / Urologic Oncology: Seminars and Original Investigations 00 (2018) 1−11

3

Fig. 1. Principles of robotic tumor enucleation and of histopathological analysis of tumor−parenchyma interface. (A) Schematic view of the enucleative plane (on the left side) and intraoperative snapshots during robotic tumor enucleation (on the right side). The enucleative plane is developed by blunt dissection by using circumferentially-oriented, dynamic tractions with the two robotic arms to lift the tumor off the parenchymal bed without removal of macroscopic healthy renal tissue. (B) Schematic view and histopathological analysis of tumor−parenchyma interface after robotic tumor enucleation. The three key elements of tumor−parenchyma interface, namely the tumor (T), the peritumoral pseudocapsule and the healthy renal margin, are shown. *Tumor−parenchyma interface.

After this period, surveillance strategies were adapted considering radiation exposure, tumor’s features, and patient’s comorbidity burden. Patient’s electronic medical files, including medical reports from follow-up visits, reports, and images of CT scans and laboratory tests (all available in our Institution’s comprehensive electronic database), were reviewed in January 2018 to collect relevant oncologic data. Telephone interviews were then carried out in all patients to ensure accuracy and completeness of data collection. In selected cases an additional follow-up visit was scheduled at the out-patient clinic to collect the relevant information on recurrence status. Loco-regional recurrence was defined as recurrence in the ipsilateral retroperitoneum [2], while renal as

recurrence in the ipsilateral kidney, distant from the enucleation bed [10]. Only recurrence at the enucleation site was considered true local recurrence. Finally, systemic recurrence was defined as recurrence in distant organs. Duration of follow-up (months) was calculated from the date of surgery to either the date of the last follow-up visit or to the date of patient’s death. 2.5. Study endpoints The endpoints of the study were: (1) to evaluate the anatomic characteristics of tumor−parenchyma interface after robotic tumor enucleation for sporadic RCC, including presence and thickness of PC, thickness of healthy renal

ARTICLE IN PRESS 4

Minervini, Campi et al. / Urologic Oncology: Seminars and Original Investigations / 00 (2018) 1−11

Fig. 2. Histopathological classification of pattern of pseudocapsule (PC) invasion after robotic tumor enucleation. (A1−A2) Intraoperative snapshots during robotic tumor enucleation showing the surgeon’s view of the tumor−parenchyma interface (*) during tumor excision; (A3) Histopathological overview of tumor−parenchyma interface; (B) Histological classification (hematoxylin-eosin staining) of PC invasion in our study. From left to right: no invasion (PC¡), partial invasion into the PC layer but not beyond (PC+); complete invasion of PC layer into the surrounding healthy renal tissue (PC++). (B1−B3) Classification of PC invasion at the tumor−parenchyma interface (object of the current study); (B4−B6) Classification of PC invasion at the tumor−perirenal fat interface.

margin, and margin status; (2) to evaluate potential associations between patient- or tumor-related factors and degree of PC invasion; (3) to evaluate recurrence rate at long-term (>4 years) follow-up, assessing potential associations with PC invasion, and/or margin status.

2.6. Statistical analysis For statistical purposes, patient-related independent variables included age, gender, BMI, Charlson comorbidity index, preoperative estimated glomerular filtration rate (CKD-EPI 2009 equation, ml/min/1.73 m2) and American Society of Anesthesiologists score. Tumor-related variables included diameter at preoperative imaging, clinical stage, polar location (upper or lower pole vs. mid-renal), nearness to the renal sinus or collecting system, endophytic rate (> or < 50%), overall tumor complexity (assessed using the RENAL score and stratified as low, intermediate, and high in case of RENAL scores of 4 to 6, 7 to 9 and 10 to 12, respectively) and histopathological variables (histotype, stage, grade, pseudocapsule thickness, and pseudocapsule invasion). Descriptive statistics were obtained reporting medians (and interquartile ranges, IQR) for continuous variables,

and frequencies and proportions for categorical variables, as appropriate. Then, the Chi-square and Kruskal-Wallis tests were used to compare the distribution of independent variables among PC¡, PC+, and PC++ tumors and among tumor histotypes, as appropriate. Multinomial multivariable logistic regression analysis was used to evaluate the potential predictors of PC invasion. Statistical analyses were performed using SPSS v. 24 (IBM SPSS Statistics for Mac, Armonk, NY, IBM Corp). All tests were two-sided with a significance set at P < 0.05. 3. Results Overall, 140 patients met the inclusion criteria for this study. Of these, 127/140 (91%) had complete histopathological and follow-up data available for analysis. Median thickness of healthy renal margin removed along with the tumor was 0.57 mm (IQR 0.24−1.03). A distinct peritumoral PC was present in 121/127 (95%) tumors. Patient-, tumor-, and recurrence-related data for the overall analytic study cohort (n = 121) and stratified by pattern of PC invasion are reported in Table 1, while data for

Table 1 Descriptive patient-, tumor- and recurrence-related data for the entire cohort (n = 121) and stratified by pattern of pseudocapsule (PC) invasion. Variables

62 (51−71) 75 (62) 25.0 (23.5−27.7) 84 (70−98) 57 (47) 44 (37) 9 (7) 7 (6) 4 (3) 92 (76) 29 (24) 77 (64) 56 (46) 77 (63) 44 (37) 49 (40) 10 (8) 12 (10) 5 (4−6) 83 (69) 23 (19) 15 (12) 3.0 (2.0−3.7) 2 (1.7) 17 (12.1) 6 (5.0) 5 (4.1) 6 (5.0) 75 (62) 19 (16) 18 (15) 5 (4) 4 (3) 14 (14) 65 (63) 23 (22) 1 (1) 86 (71) 21 (17) 0(0) 14 (12) 0.19 (0.09−0.30) 0.28 (0.14−0.45)

Pattern of pseudocapsule invasion

P

No invasion (n = 48)

Partial invasion (n = 49)

Complete invasion (n = 24)

60 (47−71) 29 (60) 25.4 (23.8−28.9) 83 (69−98) 24 (50) 17 (36) 3 (6) 2 (4) 2 (4) 38 (79) 10 (21) 27 (56) 21 (44) 26 (54) 22 (46) 18 (37) 4 (8) 5 (10) 5 (5−6) 32 (67) 12 (25) 4 (8) 2.6 (1.9−3.8) 1 (2.1) 5 (10.4) 2 (4.2) 2 (4.2) 3 (6.3) 28 (58) 5 (10) 7 (16) 4 (8) 4 (8) 12 (29) 21 (51) 8 (20) 0 (0) 39 (81) 9 (19) 0(0) 0 (0) 0.24 (0.10−0.33) 0.30 (0.17−0.45)

63 (51−70) 33 (67) 25.0 (23.1−27.7) 83 (70−94) 21 (43) 20 (41) 4 (8) 3 (6) 1 (2) 39 (80) 10 (20) 34 (69) 22 (45) 33 (67) 16 (33) 20 (41) 4 (8) 7 (14) 5 (4−6) 35 (71) 5 (10) 9 (19) 3.0 (2.2−3.7) 1 (2.0) 8 (16.3) 2 (4.1) 2 (4.1) 3 (6.1) 36 (74) 6 (12) 6 (12) 1 (12) 0 (0) 1 (2) 34 (79) 8 (19) 0 (0) 39 (80) 10 (20) 0(0) 0 (0) 0.18 (0.09−0.31) 0.27 (0.14−0.46)

63 (59−76) 13 (54) 24.8 (23.2−26.6) 94 (74−105) 12 (50) 7 (30) 2 (8) 2 (8) 1 (4) 15 (63) 9 (37) 16 (67) 13 (54) 18 (75) 6 (25) 11 (46) 2 (8) 0 (0) 5 (4−7) 16 (67) 6 (25) 2 (8) 3.3 (2.5−4.8) 0(0) 4 (16.6) 2 (8.3) 1 (4.2) 2 (8.3) 11 (46) 8 (33) 5 (21) 0 (0) 0 (0) 1 (5) 10 (53) 7 (37) 1 (5) 8 (33) 2 (8) 0(0) 14 (59) 0.14 (0.03−0.22) 0.19 (0.11−0.37)

0.4 0.5 0.4 0.4 0.9

0.2 0.3 0.6 0.2 0.4 0.9 0.2 0.5 0.2

0.1 0.5 0.1 0.1 0.7 0.2 0.017

0.001

Minervini, Campi et al. / Urologic Oncology: Seminars and Original Investigations 00 (2018) 1−11

Age (years) (median, IQR) Male gender (n, %) BMI (median, IQR) Preoperative eGFR (ml/min/m2, CKD-EPI 2009) (median, IQR) Charlson comorbidity index (n, %) 0 1 2 3 4 cT stage (n, %) cT1a cT1b Right tumor side (n, %) Posterior tumor location (n, %) Polar location (n, %) Polar Midrenal >50% Endophytic rate (n, %) Contact with urinary collecting system (n, %) Contact with renal sinus (n, %) Tumor complexity (RENAL score) (median, IQR) Tumor complexity (n, %) Low (RENAL 4−6) Intermediate (RENAL 7−9) High (RENAL 10−12) Tumor diameter (cm) (median, IQR) Intraoperative complications (n, %) Overall postoperative complications (n, %) Medical complications (n, %) Minor surgical complications (n, %) (Clavien-Dindo grade I−II) Major surgical complications (n, %) (Clavien-Dindo grade III−V) Tumor histotype (n, %) ccRCC pRCC chRCC mlcRCNLMP ccpRCC Nucleolar grading (for histotypes other than chRCC) (n, %) 1 2 3 4 pT stage (n, %) pT1a pT1b pT2 pT3a PC thickness (mm) (median, IQR) Minimum Mean

Overall (n = 121)

<0.001

0.1 0.2 5 (continued on next page)

Minervini, Campi et al. / Urologic Oncology: Seminars and Original Investigations / 00 (2018) 1−11

Loco-regional Renal Local (enucleation site) Systemic Positive surgical margins (n, %) Follow-up (months) (median, IQR) Recurrence (n, %)

Tumor complexity was stratified as low, intermediate or high according to the RENAL score. BMI = body mass index; eGFR = estimated glomerular filtration rate; ccRCC = clear cell RCC; chRCC = chromophobe RCC; ccpRCC = clear cell papillary RCC; HRM = healthy renal margin; IQR = interquartile range; mlcRCNLMP = multilocular cystic renal cell neoplasm of low-malignant potential; pRCC = papillary RCC; PC = pseudocapsule; RCC = renal cell carcinoma.

0.27 (0.16−0.54) 0.25 (0.03−0.58) 0.73 (0.22−1.33) 0,98 (0.25−2.10) 3 (12.5) 61 (53−65) 0 (0) 2 (8.3) 0 (0) 1 (4)

Complete invasion (n = 24) Partial invasion (n = 49)

0.40 (0.24−0.63) 0.16 (0.07−0.36) 0.50 (0.26−0.94) 0.65 (0.39−1.55) 0 (0) 60 (53−66) 0 (0) 0 (0) 0 (0) 0 (0) 0.32 (0.23−0.57) 0.19 (0.07−0.62) 0.58 (0.25−1.21) 0.79 (0.43−1.75) 0 (0) 60 (55−61) 0 (0) 1 (2) 0 (0) 3 (6.3) 0.35 (0.21−0.59) 0.18 (0.06−0.44) 0.57 (0.24−1.03) 0.78 (0.38−1.76) 3 (2.4) 61 (56−64) 0 (0) 3 (2.4) 0 (0) 4 (3.1) Thickness of HRM (mm) (median, IQR)

Variables

Table ? 1 (Continued)

Maximum Minimum Mean Maximum

Overall (n = 121)

No invasion (n = 48)

Pattern of pseudocapsule invasion

P

0.2 0.5 0.7 0.8 0.002 0.7 0.2

6

thecohortofpatientswithtumorsnotshowinganintactPC(n = 6) inSupplementaryTable1. Overall, median tumor diameter was 3 cm (IQR 2.0 −3.7). Twenty-nine (24%) tumors were cT1b and 38 (31%) were classified of intermediate or high complexity. Median thickness of peritumoral PC was 0.28 mm (IQR 0.14−0.45). Partial (PC+/i-Cap score 2) and complete (PC++/i-Cap score 3) pseudocapsule invasion was reported in 49/121 (40.5%) and 24/121 (19.8%) cases, respectively. At multivariable analysis, increasing tumor diameter (odds ratio [OR] 2.10; 95% confidence interval [CI]: 1.18 −3.93), tumor endophytic rate > 50% (OR 10.24; 95% CI: 1.21−86.46) and papillary (compared to clear cell) RCC (OR 6.57; 95% CI: 1.11−42.75) were found to be significantly associated with complete (PC++) vs. absent (PC¡) invasion (Table 2). Likewise, the same factors were significantly associated with complete vs. partial PC invasion. Results did not change when considering a binomial pattern of PC invasion (PC++ vs. PC+ or PC¡). On the contrary, neither patient- nor tumor-related features were significantly associated with partial (PC+) vs. absent (PC¡) invasion (Table 2). In our cohort, the most prevalent tumor histotype was clear cell RCC followed by papillary and chromophobe RCC. Overall, there were 9 cases (7%) of tumors considered of low malignant potential (multilocular cystic renal cell neoplasm of low malignant potential and clear cell papillary RCC). This tumor category showed more favorable histopathological features compared to traditional RCC histotypes (Table 3). Overall, positive surgical margins were reported in 3 patients (2.5%). In these cases, PC was completely invaded (PC++). At a median follow-up of 61 months (IQR 56−64, range 48−76), three cases (2.4%) of renal recurrence (metachronous lesions elsewhere in the ipsilateral kidney) and four cases (3,1%) of systemic recurrence were reported (Supplementary Table 2). One patient died due to metastatic disease 26 months after surgery. Among patients still alive at follow-up (n = 126), no cases of local recurrence (at the enucleation site) were found. Surgical margins were negative in all patients who developed recurrence during follow-up. 4. Discussion In light of the increasing use of robotic PN [3,4,12] and the increasing adoption of tumor enucleation at high-volume centers worldwide [6], assessing the oncologic safety of robotic tumor enucleation is a key priority from both a clinical and research perspective. Indeed, while systematic reviews have confirmed the oncologic safety of tumor enucleation [9,10] and recent studies showed that it might achieve increased parenchymal mass and renal function preservation as compared to

Pattern of pseudocapsule invasion (n = 112) Covariates

Age Male vs. female gender BMI Tumor diameter (at histopathological analysis) Endophytic rate (>50% vs. <50%) Tumor complexity (RENAL score, continuous) Mid-renal vs. polar location Tumor histotype pRCC vs. ccRCC chRCC vs. ccRCC Nucleolar Grading (3−4 vs. 1−2) Mean PC thickness Mean HRM thickness

Three-tiered classification (Model 1)

Two-tiered classification (Model 2)

Complete invasion vs. no invasion

Complete invasion vs. partial invasion

Partial invasion vs. no invasion

Complete Invasion vs. partial or no invasion

OR (95%CI)

P

OR (95%CI)

P

OR (95%CI)

P

OR (95%CI)

P

1.05 (0.69−1.04) 0.55 (0.19−2.57) 0.85 (0.69−1.04) 2.10 (1.18−3.93) 10.24 (1.21−86.46) 0.52 (0.22−1.21) 0.13 (0.01−1.23) 6.57 (1.11−42.75) 2.29 (0.41−12.9) 1.63 (0.33−7.99) 0.99 (0.99−1.01) 1.00 (0.99−1.01)

0.1 0.5 0.1 0.02 0.03 0.19 0.08 0.04 0.3 0.6 0.7 0.6

1.00 (0.95−1.06) 0.48 (0.11−2.09) 0.91 (0.75−1.10) 1.95 (1.06−3.58) 8.23 (1.13−59.95) 0.57 (0.25−1.27) 0.18 (0.02−1.64) 7.85 (1.38−44.63) 2.23 (0.40−12.38) 2.40 (0.54−10.67) 0.99 (0.99−1.01) 1.00 (0.99−1.01)

0.7 0.3 0.4 0.03 0.04 0.2 0.1 0.02 0.4 0.3 0.4 0.3

1.04 (0.99−1.08) 1.13 (0.39−3.30) 0.93 (0.82−1.06) 1.07 (0.69−1.65) 1.25 (0.27−5,80) 0.92 (0.54−1.55) 0.73 (0.22−2.43) 0.84 (0.19−3.55) 1.03 (0.26−3.97) 0.68 (0.20−2.31) 1.00 (0.99−1.01) 1.00 (0.99−1.00)

0.08 0.8 0.3 0.7 0.8 0.7 0.6 0.8 0.9 0.54 0.4 0.3

1.02 (0.97−1.08) 0.53 (0.13−2.11) 0.88 (0.73−1.06) 2.00 (1.12−3.55) 8.53 (1.27−56.99) 0.56 (0.26−1.21) 0.15 (0.02−1.30) 6.91 (1.34−35.7) 2.20 (0.46−10.71) 2.056 (0.50−8.43) 0.99 (0.99−1.01) 1.00 (0.99−10.01)

0.4 0.3 0.2 0.02 0.03 0.1 0.08 0.02 0.3 0.3 0.5 0.4

The analysis was restricted to clear cell (n = 75), papillary, (n = 19) and chromophobe (n = 18) renal cell carcinoma histotypes (overall n = 112), as no cases of complete PC invasion were recorded among tumors of low-malignant potential (n = 9). To evaluate whether tumors showing complete PC invasion (PC++) significantly differed from tumor showing either partial (PC+) or absent (PC¡) PC invasion, two multivariable models were built, using either a three-tiered [Model 1: PC¡ vs. PC + vs. PC++)] or a two-tiered [Model 2: (PC++ vs. PC+ or PC¡)] classification of PC invasion. Tumor complexity was stratified as low, intermediate, or high according to the RENAL score. BMI = body mass index; ccRCC = clear cell RCC; chRCC = chromophobe RCC; CI = confidence interval; HRM = healthy renal margin; OR = odds ratio; pRCC = papillary RCC; PC = pseudocapsule; RCC = renal cell carcinoma.

Minervini, Campi et al. / Urologic Oncology: Seminars and Original Investigations 00 (2018) 1−11

Table 2 Multivariable analysis assessing predictors of PC invasion.

7

8

Tumor histotype Tumors of low malignant potential (mlcRCNLMP, ccpRCC) (n = 9) Tumor diameter (cm) (median, IQR) Tumor complexity (n, %)

cT stage (n, %) pT stage (n, %)

Nucleolar grading (for histotypes other than chRCC) (n, %)

Pattern of pseudocapsule invasion

Mean PC thickness (mm) (median, IQR) Mean thickness of HRM (mm) (median, IQR) Positive surgical margins (n, %)

Low (RENAL 4−6) Intermediate (RENAL 7−9) High (RENAL 10−12) cT1a cT1b pT1a pT1b pT3a 1 2 3 4 No invasion Partial invasion Complete invasion

2.0 (1.7−3.7) 6 (67) 3 (33) 0 (0) 7 (78) 2 (22) 8 (89) 1 (11) 0 (0) 9 (100) 0 (0) 0 (0) 0 (0) 8 (89) 1 (11) 0 (0) 0.50 (0.41−0.58) 0.41 (0.36−0.69) 0 (0)

Clear cell RCC (n = 75)

Papillary RCC (n = 19)

3.0 (2.2−3.8) 50 (67) 15 (20) 10 (13) 56 (74) 19 (26) 57 (76) 15 (20) 3 (4) 5 (7) 54 (72) 15 (20) 1 (1) 28 (37) 36 (48) 11 (15) 0.31 (0.23−0.47) 0.50 (0.23−1.02) 1 (1)

3.0 (2.0−4.5) 13 (68) 3 (16) 3 (16) 14 (74) 5 (26) 10 (53) 2 (10) 7 (37) 0 (0) 11 (58) 8 (42) 0 (0) 5 (26) 6 (32) 8 (42) 0.18 (0.11−0.28) 0.30 (0.23−0.81) 2 (11)

Chromofobe RCC (n = 18) 2.9 (2.0−3.1) 14 (78) 2 (11) 2 (11) 15 (83) 3 (17) 11 (61) 3 (17) 4 (22) /

7 (39) 6 (33) 5 (28) 0.09 (0.06−0.14) 1.18 (0.50−1.76) 0(0)

P

0.6 0.8

0.8 0.003

<0.001

0.006

<0.001 0.06 0.1

Tumor complexity was stratified as low, intermediate, or high according to the RENAL score. ccRCC = clear cell RCC; chRCC = chromophobe RCC; ccpRCC = clear cell papillary RCC; HRM = healthy renal margin; IQR = interquartile range; mlcRCNLMP = multilocular cystic renal cell neoplasm of low-malignant potential; pRCC = papillary RCC; PC = pseudocapsule; RCC = renal cell carcinoma. P value is referred to Kruskal-Wallis and Chi-square tests, as appropriate.

Minervini, Campi et al. / Urologic Oncology: Seminars and Original Investigations / 00 (2018) 1−11

Table 3 Selected descriptive tumor characteristics among different histotypes.

Minervini, Campi et al. / Urologic Oncology: Seminars and Original Investigations 00 (2018) 1−11

standard PN [5,19], there is lack of evidence on the longterm oncologic results of robotic tumor enucleation [12]. Moreover, only few studies of limited sample size evaluated the histopathological features of robotic tumor enucleation specimens [20−22]. To the best of our knowledge, this is the largest series so far reporting the histopathological and long-term oncological results of robotic tumor enucleation for sporadic RCC. A key finding from our study is that, as previously shown for our open enucleation series [7,8], robotic tumor enucleation is not a zero-margin but rather a microscopicmargin technique. Indeed, while the surgeon developed the anatomic cleavage plane without removal of a macroscopic renal margin, definitive histopathological analysis revealed a distinct microscopic silver of healthy renal margin (median thickness of 0.57 mm) in the large majority of cases (Table 1). This margin was slightly thinner than that reported in our open enucleation series (mean thickness 1.05 mm) [7]. Of note, the 3D magnification during robotic tumor enucleation may allow the surgeon to glimpse this minimal layer of healthy renal tissue during tumor excision (Fig. 1). Notably, as previously demonstrated for open tumor enucleation [23], by developing the anatomic cleavage plane in close vicinity to the tumor contours robotic tumor enucleation might allow surgeons to extend the indications for PN also in case of challenging, highly complex renal masses [1,2,12]. At histopathological analysis, a fibrous peritumoral PC was found in 95% of RCC, in accordance with previous studies [7,13,24]; yet, in our study median PC thickness was slightly lower than previously reported [7,13,24]. PC was completely invaded (PC++/i¡Cap score 3) in 19.8% of tumors, similarly to previous findings [18,24]. Of note, complete PC invasion was never reported for tumors of low malignant potential (Table 3). Of note, tumor histotype, nucleolar grading, and pathologic stage significantly differed among tumors with different pattern of PC invasion, while patient-related factors, tumor diameter, and anatomic tumor complexity did not (Table 1). At multivariable analysis, tumor diameter, endophytic rate, and tumor histotype were significantly associated with complete PC invasion (Table 2), in line with previous studies [7,13,18]. A key finding from our study is that these factors were consistent across the two multivariable models. On the contrary, there were no significant predictors of partial vs. absent PC invasion (Table 2). This suggests that a binomial reporting system of PC invasion (complete vs. partial or absent) might better capture the inherent features of RCC and might be more effective than more complex classification models [7,18]. Finally, a key finding of our study is that robotic tumor enucleation achieved negative surgical margins in 97.5% of cases and, most importantly, there was no case of local recurrence at the enucleation site after a median follow-up of 61 months. These findings highlight how oncologic

9

safety of robotic tumor enucleation is grounded into the “protective” layer of healthy renal margin beyond peritumoral PC which allows to prevent positive margins even in case of complete PC invasion (Fig. 2). In our cohort, PC invasion did not have an impact on local recurrence, as previously shown for open tumor enucleation [8]. Nonetheless, assessment of the prognostic role of PC invasion after PN requires further investigation [21,24]. Of note, patients with positive surgical margins (n = 3) did not experience local recurrence, in line with a large body of evidence supporting the relatively negligible impact of positive margins on recurrence risk [25]. In this regard, whether positive surgical margins significantly increase the risk of recurrence at the resection site after PN is still matter of debate [9,26]. Three cases of renal recurrence (distant from the enucleation site) and four cases of metastatic disease were reported during follow-up in patients with negative surgical margins (Supplementary Table 2). Even if rare, systemic recurrences after complete tumor resection have been previously reported in PN literature, independently from margin status [27,28]. This recurrence pattern highlights how different pathogenetic pathways may be involved in the occurrence of renal recurrence [29] and tumor progression after PN for localized RCC. Despite its strengths, our study is not devoid of limitations. First, the retrospective study design and the lack of key histopathological and/or follow-up data in 9% of the patient cohort might have underestimated the actual recurrence rate. Moreover, the relatively small sample size and the low number of events prevented us to evaluate the potential predictors of recurrence in our series. Second, robotic tumor enucleation was performed by four experienced surgeons at a referral high-volume center; as such, our results might not be applicable to all surgeonor center-related scenarios. Third, resection technique was not classified using a standardized reporting system. Few cases in our robotic enucleation series might have been reclassified according to the surface-intermediate-base margin score [22,30]. To this regard, however, the score was introduced in 2014 and median thickness of healthy renal margin in our series (0.57 mm) was comparable to that obtained by techniques classified as “pure or hybrid enucleation” in recent studies using the surface-intermediate-base model [15,22]. Fourth, the technique of tumor enucleation might have theoretically affected the inherent anatomic characteristics (i.e. thickness and completeness) of the peritumoral pseudocapsule layer, potentially impacting on interpretation of pseudocapsule invasion at a microscopic level. In this regard, further research is needed to evaluate the association between resection technique (enucleation vs. enucleoresection vs. resection [30]) and histopathological features of peritumoral pseudocapsule after partial nephrectomy. Finally, given the natural history of RCC, the length of follow-up in our study might have been still limited to

Minervini, Campi et al. / Urologic Oncology: Seminars and Original Investigations / 00 (2018) 1−11

10

assess potential delayed recurrences after robotic tumor enucleation. Acknowledged these limitations, our findings provide a robust foundation for significant further research in this field. In particular, larger studies with longer follow-up are needed to (a) evaluate the histopathological characteristics of peritumoral pseudocapsule and surgical margins, as well as oncologic outcomes, after robotic tumor enucleation for larger an anatomically complex renal masses; (b) confirm the longterm oncological safety of robotic tumor enucleation, comparing its results with open tumor enucleation; (c) assess the relationship between PC invasion, positive margins and true local recurrence after robotic PN and (d) evaluate the pathogenetic pathways involved in the occurrence of local and systemic recurrence after robotic partial nephrectomy. 5. Conclusions A distinct peritumoral pseudocapsule was found in most cases after robotic tumor enucleation and specific tumor characteristics were significant predictors of complete pseudocapsule invasion. By mirroring the principles of the open technique, robotic tumor enucleation achieved negative surgical margins in the vast majority of patients, even in case of complete PC invasion. At long-term follow-up, PC invasion did not have an impact on local recurrence and no recurrences were found at the enucleation site. Although they need to be confirmed by larger multicenter studies with longer follow-up, our findings provide robust evidence on the oncologic safety of robotic tumor enucleation for the treatment of sporadic RCC. Conflict of interest The authors declare that they have no conflict of interest. Funding This research did not receive any specific grant from funding agencies in the public, commercial, or not-forprofit sectors. Acknowledgments The Authors would like to thank Dr. Beatrice Orlandini (Department of Gastroenterology, University of Florence, Careggi Hospital, Florence, Italy) for the original digital drawings used in Fig. 1. Supplementary materials Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.uro lonc.2018.08.014.

References [1] Campbell S, Uzzo RG, Allaf ME, et al. Renal mass and localized renal cancer: AUA guideline. J Urol 2017;198:520–9. [2] Lyunberg B, Albiges L, Bensalah K, et al. European association of urology guidelines on renal cell carcinoma. Version 2017. https://uroweb.org/guideline/renal-cell-carcinoma/ [3] Ghani KR, Sukumar S, Sammon JD, et al. Practice patterns and outcomes of open and minimally invasive partial nephrectomy since the introduction of robotic partial nephrectomy: results from the nationwide inpatient sample. J Urol 2014 Apr;191:907–12. [4] Borghesi M, Schiavina R, Gan M, et al. Expanding utilization of robotic partial nephrectomy for clinical T1b and complex T1a renal masses. World J Urol 2013 Jun;31:499–504. [5] Marconi L, Desai MM, Ficarra V, et al. Renal preservation and partial nephrectomy: patient and surgical factors. Eur Urol Focus 2016;2:589–600. [6] Minervini A, Campi R, Mari A, et al. Mp41-11. Resection techniques for nephron sparing surgery (NSS) vary: insights from a prospectively collected multi-institutional cohort harnessing the surface−intermediate−base (s.i.b.) margin score (sib international consortium). J Urol 2016;195:e564. [7] Minervini A, di Cristofano C, Lapini A, et al. Histopathologic analysis of peritumoral pseudocapsule and surgical margin status after tumor enucleation for renal cell carcinoma. Eur Urol 2009;55:1410–8. [8] Minervini A, Raspollini MR, Tuccio A, et al. Pathological characteristics and prognostic effect of peritumoral capsule penetration in renal cell carcinoma after tumor enucleation. Urol Oncol 2014 Jan;32:50. e15−22. [9] Minervini A, Campi R, Sessa F, et al. Positive surgical margins and local recurrence after simple enucleation and standard partial nephrectomy for malignant renal tumors: systematic review of the literature and meta-analysis of prevalence. Minerva Urol Nefrol 2017 Dec;69:523–38. [10] Cao DH, Liu LR, Fang Y, et al. Simple tumor enucleation may not decrease oncologic outcomes for T1 renal cell carcinoma: a systematic review and meta-analysis. Urol Oncol 2017 Nov;35:661.e15–21. [11] Gupta GN, Boris RS, Campbell SC, Zhang Z. Tumor enucleation for sporadic localized kidney cancer: pro and con. J Urol 2015 Sep;194:623–5. [12] Carini M, Campi R, Mari A, Minervini A. Oncologic safety of robotic partial nephrectomy: setting tiles in the mosaic of evidence while designing future research projects. Eur Urol Focus 2017:[Epub ahead of print]. [13] Azhar RA, de Castro Abreu AL, Broxham E, et al. Histological analysis of the kidney tumor-parenchyma interface. J Urol 2015;193:415–22. [14] Minervini A, Tuccio A, Masieri L, et al. Endoscopic robot-assisted simple enucleation (ERASE) for clinical T1 renal masses: description of the technique and early postoperative results. Surg Endosc 2015 May;29:1241–9. [15] Minervini A, Campi R, Kutikov A, et al. Histopathological validation of the surface-intermediate-base (SIB) margin score for standardized reporting of resection technique during nephron-sparing surgery. J Urol 2015;194:916–22. [16] Delahunt B, Chevile JC, Martignoni G, et al. The International society of urological pathology (ISUP) grading system for renal cell carcinoma and other prognostic parameters. Am J Surg Pathol 2013;37:1490. [17] Srigley JR, Delahunt B, Eble JN, et al. The International Society of Urological Pathology (ISUP) Vancouver classification of renal neoplasia. Am J Surg Pathol 2013;37:1469. [18] Snarskis C, Calaway AC, Wang L, et al. Standardized reporting of microscopic renal tumor margins: introduction of the renal tumor capsule invasion scoring system. J Urol 2017;197:23–30.

Minervini, Campi et al. / Urologic Oncology: Seminars and Original Investigations 00 (2018) 1−11 [19] Dong W, Gupta GN, Blackwell RH, et al. Functional comparison of renal tumor enucleation versus standard partial nephrectomy. Eur Urol Focus 2017;3(4 5):437–43. [20] Calaway AC, Gondim DD, Flack CK, et al. Anatomic comparison of traditional and enucleation partial nephrectomy specimens. Urol Oncol 2017 May;35:221–6. [21] Wang L, Hughes I, Snarskis C, et al. Tumor enucleation specimens of small renal tumors more frequently have a positive surgical margin than partial nephrectomy specimens, but this is not associated with local tumor recurrence. Virchows Arch 2017 Jan;470:55–61. [22] Antonelli A, Furlan M, Sodano M, et al. External histopathological validation of the surface-intermediate-base margin score. Urol Oncol 2017 May;35:215–20. [23] Serni S, Vittori G, Frizzi J, et al. Simple enucleation for the treatment of highly complex renal tumors: Perioperative, functional and oncological results. Eur J Surg Oncol 2015 Jul;41:934–40. [24] Xi W, Wang J, Liu L, et al. Evaluation of tumor pseudocapsule status and its prognostic significance in renal cell carcinoma. J Urol 2018;199(4):915–20.

11

[25] Marszalek M, Carini M, Chlosta P, et al. Positive surgical margins after nephron-sparing surgery. Eur Urol 2012;61:757–63. [26] Wood EL, Adibi M, Qiao W, et al. Local tumor bed recurrence following partial nephrectomy in patients with small renal masses. J Urol 2018 Feb;199:393–400. [27] Mouracade P, Kara O, Maurice MJ, et al. Patterns and predictors of recurrence after partial nephrectomy for kidney tumors. J Urol 2017 Jun;197:1403–9. [28] Stewart-Merrill SB, Thompson RH, Boorjian SA, et al. Oncologic surveillance after surgical resection for renal cell carcinoma: a novel risk-based approach. J Clin Oncol 2015;33:4151–7. [29] Antonelli A, Furlan M, Tardanico R, et al. Features of ipsilateral renal recurrences after partial nephrectomy: a proposal of a pathogenetic classification. Clin Genitourin Cancer 2017 Oct;15:540–7. [30] Minervini A, Carini M, Uzzo RG, Campi R, Smaldone MC, Kutikov A. Standardized reporting of resection technique during nephronsparing surgery: the surface−intermediate−base margin score. Eur Urol 2014;66:803–5.