Histopathologic analysis of peritumoral pseudocapsule and surgical margin status after tumor enucleation for renal cell carcinoma.

Abstract

The oncologic safety of blunt tumor enucleation (TE) of renal cell carcinoma (RCC) depends on the presence of a continuous pseudocapsule (PS) around the tumor and on the possibility of obtaining negative surgical margins (SMs).
To investigate the PS and SMs after TE to define the real need to take a rim of healthy parenchyma around the tumor to avoid the risk of positive SMs. The risk of PS invasion related to other clinical and pathologic variables was also evaluated.
Between September 2006 and December 2007, data were gathered prospectively from 187 consecutive patients who had kidney surgery. Overall, 90 consecutive patients who had TE for RCC were eligible for the study. All specimens were evaluated using an image analyzer by a dedicated uropathologist.
TE was done by blunt dissection using the natural cleavage plane between the tumor and the normal parenchyma.
PS, SM, and routinely available clinical and pathologic variables were recorded.
In 60 RCC tumors (67%) the PS was intact and free from invasion (PS-) while in 30 (33%) there were signs of penetration within its layers, with or without invasion beyond it. Indeed, 26.6% had PS that had been penetrated on the parenchymal side and 6.6% had penetration on the perirenal fat tissue side. The odds of having PS penetration increased significantly with an increase in clinical tumor size. PS penetration was also significantly associated with pathologic tumor dimensions and grade. In all cases the SMs were negative after TE. The present patients, followed for >2 yr, will enable us to correlate the risk of local recurrence with PS status.
The risk of PS penetration is associated with clinical and pathologic tumor dimensions and grade. If there is PS invasion into normal parenchyma, the presence of a thin layer of tissue allows for negative SM even if a TE is performed.

Full text

Page 1

Kidney Cancer
Histopathologic Analysis of Peritumoral Pseudocapsule and
Surgical Margin Status after Tumor Enucleation for
Renal Cell Carcinoma
Andrea Minervinia,*, Claudio di Cristofanob, Alberto Lapinia, Marco Marchic,
Federico Lanzia, Gianluca Giubileia, Nicola Tosia, Agostino Tuccioa,
Massimiliano Mancinib, Carlo della Roccab, Sergio Sernia, Generoso Bevilacquad,
Marco Carinia
aDepartment of Urology, University of Florence, Careggi Hospital, Florence, Italy
bDepartment of Experimental Medicine, Sapienza University of Rome, Polo Pontino, I.C.O.T, Latina, Italy
cDepartment of Statistics, University of Florence, Italy
dDepartment of Oncology, Division of Surgical, Molecular and Ultrastructural Pathology, University of Pisa and University Hospital of Pisa, Italy
european urology 55 (2009) 1410–1418
available at www.sciencedirect.com
journal homepage: www.europeanurology.com
Article info
Article history:
Accepted July 15, 2008
Published online ahead of
print on August 3, 2008
Keywords:
Enucleation
Surgical margins
Pseudocapsule
Local recurrence
Nephron-sparing surgery
Partial nephrectomy
RCC
Abstract
Background: The oncologic safety of blunt tumor enucleation (TE) of renal cell carci-
noma (RCC) depends on the presence of a continuous pseudocapsule (PS) around
the tumor and on the possibility of obtaining negative surgical margins (SMs).
Objective: To investigate the PS and SMs after TE to define the real need to take a rim
of healthy parenchyma around the tumor to avoid the risk of positive SMs. The risk
of PS invasion related to other clinical and pathologic variables was also evaluated.
Design, setting, and participants: Between September 2006 and December 2007, data
were gathered prospectively from 187 consecutive patients who had kidney surgery.
Overall, 90 consecutive patients who had TE for RCC were eligible for the study. All
specimens were evaluated using an image analyzer by a dedicated uropathologist.
Intervention: TE was done by blunt dissection using the natural cleavage plane
between the tumor and the normal parenchyma.
Measurements: PS,SM,androutinelyavailableclinicalandpathologicvariableswere
recorded.
Results and limitations: In 60 RCC tumors (67%) the PS was intact and free from
invasion (PS?) while in 30 (33%) there were signs of penetration within its layers,
with or without invasion beyond it. Indeed, 26.6% had PS that had been penetrated
on the parenchymal side and 6.6% had penetration on the perirenal fat tissue side.
TheoddsofhavingPSpenetrationincreasedsignificantlywithanincreaseinclinical
tumor size. PS penetration was also significantly associated with pathologic tumor
dimensions and grade. In all cases the SMs were negative after TE. The present
patients, followed for >2 yr, will enable us to correlate the risk of local recurrence
with PS status.
* Corresponding author. Clinica Urologica I, Azienda Ospedaliera Careggi, Universita ` di
Firenze, Villa Monna Tessa, Viale Pieraccini 18, 50139, Firenze, Italy. Tel. +39 055 417645;
Fax: +39 055 4377755.
E-mail address: andreamine@libero.it (A. Minervini).
0302-2838/$ – see back matter # 2008 European Association of Urology. Published by Elsevier B.V. All rights reserved.doi:10.1016/j.eururo.2008.07.038

Page 2

1.Introduction
Nephron-sparing surgery (NSS) has been widely
accepted since the early 1980s [1–2] as an elective
procedure for treating single sporadic renal tumors.
Several studies have shown this procedure to offer
equally effective local control and a similar disease-
specific survival rate compared to radical nephrec-
tomy (RN) for treating renal-cell carcinoma (RCC)
tumors of <4 cm in their greatest dimension [1–4].
Moreover, recent reports have shown that NSS
achieves local tumor control equivalent to RN also
for RCC of 4–7 cm [5,6].
The concern about local recurrence due to inade-
quate tumor excision and the reported risk of
pseudocapsule (PS) invasion on the parenchymal
side [7–11] led most surgeons to remove a minimal
rim of normal-appearing parenchyma around the
tumor, and at present this is considered to be the
technique of choice in patients undergoing NSS [1,2].
Nevertheless,inrecentyearstherehavebeenseveral
reportsonthereductionofthethicknessofthesafety
margins that should be excised with the tumor to
avoid the risk of local recurrence. Others have
recently gone further, concluding that, if the tumor
is completely excised, the width of the resection
margin is irrelevant and not correlated with disease
progression,thusprovidinganintriguinginsightinto
the real need to excise an adequate rim of healthy
kidney tissue around the tumor [11–15].
The tumor enucleation (TE) technique, which
consists of excising the tumor by blunt dissection
without a visible rim of normal parenchyma, has
been reported in the treatment of benign-looking
tumors such as angiomyolipomas. But only a few
studies in the 1980s and early 1990s reported on the
use of this technique for treating small RCC tumors,
and they showed similar 5-yr survival rates to those
of partial nephrectomy [16–19]. Recently, other retro-
spective analyses confirmed that TE can be safely
usedfortreatingpT1a–pT1bRCCtumors,anditisnot
associated with any greater risk of local recurrence
than is partial nephrectomy [20–23]. Therefore, the
discrepancybetweentheoptimaloncologicresultsof
in vivo TE reported in several recent retrospective
analyses [20–23] and the pathologic concerns of
incomplete tumor excision based on data obtained
by studiesafter an ex vivo TE or tumor sectionsof RN
specimens remains an unsolved oncologic issue in
conservative kidney surgery [8,9–11].
The objective of this prospective study was to
investigate the existence, integrity, possible inva-
sion of peritumoral PS and surgical margin (SM)
status after NSS, performed as TE for the treatment
of RCC, with the aim of characterizing PS in RCC and
defining the real need to take a rim of healthy
parenchyma around the tumor to avoid the risk of a
positive SM.
2. Methods
Between September 2006 and December 2007, data were
gatheredprospectivelyfrom187consecutivepatientswhohad
kidney surgery. Overall, 104 (55.6%) had a conservative
treatment, TE, while 83 (44.4%) had RN. All but one NSS were
successful; in one case with imperative indications for a 4-cm
tumor, we decided to convert to RN for the intraoperative
detection of intrarenal vein thrombosis after TE. The patho-
logic evaluation in this case confirmed the presence of renal
vein thrombosis (pT3b). No patients had laparoscopic NSS.
ThedecisiontoproceedwithplannedNSSwasbasedonthe
patients’ preoperative imaging evaluation, medical history,
comorbidity, and age. None of the patients undergoing NSS
hadpreoperativeorintraoperativesuspicionofpositivenodes.
All patients were free from distant metastases before surgery
(M0). The preoperative evaluation in all patients included
computedtomography(CT)oftheabdomen, achestx-ray,and
ultrasonography of the kidney, ureter, and bladder.
2.1.Surgical technique: tumor enucleation
In all cases, TEwas doneaccordingto ourpreviouslydescribed
standard protocol, by blunt dissection and always using the
natural cleavage plane between the tumor and normal
parenchyma [20–22]. The kidney was directly approached
and completely separated from perirenal fat to exclude
subcapsular satellite lesions not detected by imaging. Intra-
operative ultrasonography was not used during this period.
After kidney capsule skeletonization, it was important to
visualize the limit between the healthy renal parenchyma and
the tumor, leaving the peritumoral fat in situ. The renal
pedicle was carefully isolated and usually controlled with
vascular clamps before TE. Renal hypothermia with slush was
not used during this period. The kidney capsule was then
sharply incised starting 1–2 mm away from the lesion toward
the PS, and when the PS was visually reached, the tumor was
enucleated by blunt dissection, with no visible rim of normal
parenchyma (Fig. 1). Biopsies of the surgical bed were not
taken during this period. The visible bleeding vessels and
Conclusions: The risk of PS penetration is associated with clinical and pathologic
tumor dimensions and grade. If there is PS invasion into normal parenchyma, the
presence of a thin layer of tissue allows for negative SM even if a TE is performed.
# 2008 European Association of Urology. Published by Elsevier B.V. All rights reserved.
european urology 55 (2009) 1410–1418
1411

Page 3

incidental opening of the calyces were ligated using a running
suture with 4-0 monofilament. Diathermy spray or argon-
beam coagulation of the surgical bed was not used during
this period. The parenchymal defect was closed with
horizontal interrupted sutures after sealant (Cyanoacrylate
glue, Tachosil, Floseal)and Tabotamp bolster apposition. Most
of the procedures were performed by one surgeon (M Carini,
61%).
After TE, the specimens were oriented, positioning a suture
at the deepest part of the inner pole of the tumor, and sent for
histopathologic examination.
2.2. Histopathology
After fixation in a 10% formalin solution, all specimens were
step-sectioned at 5-mm intervals, and the entire specimen
was analyzed by a dedicated uropathologist (C. di Cristofano).
The greatest diameters of the tumors were measured and
recorded. All patients were staged according to the 2002 TNM
criteria [24], and nuclear grading was assigned according to
criteria proposed by Fuhrman et al [25]. The histopathology
was reviewed according to the 2004 World Health Organiza-
tion (WHO) classification [26].
Fig. 1 – Tumor enucleations (TEs).
european urology 55 (2009) 1410–1418
1412

Page 4

PS status was carefully analyzed. PS thickness was
determined as the mean value of the four fields evaluated
for each tumor. The existence, integrity, and degree of PS
invasion as well as d the thickness of the rim of normal-
appearing parenchyma, eventually present in case of tumor
beyond PS, were evaluated, capturing the images at ?40 (3.0-
megapixel resolution), and analyzed using an image analyzer
(Motic Images Plus v.2.0).
Other variables assessed were tumor stage, histologic
subtype, Fuhrman grade, histologic necrosis, and sarcomatoid
differentiation.
2.3. Statistical analysis
The unpaired student t test and x2test were used to evaluate
the possible statistical correlation between the clinical and
pathologicvariablesandtheriskofPS+.TheriskofPSinvasion
related to other clinical and pathologic (continuous and
categorical) variables was evaluated using the logistic regres-
sion model, and odds ratios and risk ratios were calculated.
Significant-difference level was considered to be p < 0.05.
3. Results
At diagnosis, 87 of the 104 tumors (84%) were
detected incidentally in asymptomatic patients
while 17 (16%) were associated with either micro-
scopic or frank hematuria with or without flank
pain. Overall, 95 patients were treated with elective
TE (91%) while 9 patients received TE for imperative
indications (9%).
All patients with histologically confirmed RCC
were eligible for the study (90/104, 86.5%), and the 14
who had NSS for histologically confirmed benign
tumors were excluded (13.5%). At CT scan before
surgery, the mean of the greatest dimensions of the
tumors was 3.2 cm (range: 0.8–10 cm; SD: 1.5 cm;
median: 3.0 cm; interquartile range [IQR]: 2.0–
4.1 cm) and at the pathologic examination the mean
was 3.1 cm (range: 0.5–12.5 cm; SD: 1.7 cm; median:
2.9 cm; IQR: 2.1–3.8 cm), respectively.
The pathologic analysis according to the 2002
TNM classification showed that 75.6% of tumors
were pT1a, 16.7% of tumors were pT1b, 2.2% of
tumors were pT2, 4.4% of tumors were pT3a, and
1.1% of tumors were pT3b. On the basis of Fuhrman
nuclear grading, 20% of tumors were G1, 65.6% of
tumors were G2, and 14.4% of tumors were G3. The
histopathologic evaluation according to the 2004
WHO classification revealed that 75.6% of tumors
were clear-cell RCC, 17.8% were papillary; 4.4% of
tumors were chromophobe; 1.1% of tumors were
mucinous tubular and spindle-cell carcinoma, and
1.1% of tumors were unclassified RCC.
The presence of sarcomatoid differentiation was
detected in three cases (mean: 10.5%; range: 1–20%).
Histologic tumor necrosis was present in 55.5% of all
RCC tumors, and it was often found as microscopic
isolated foci. Histologic necrosis was ?20% in 47
cases (mean: 4.1%; median: 3%; range: 1–20%), while
it was >20% in only three RCC tumors.
All 90 RCC tumors were surrounded by a
continuous, nonfenestrated, fibrous PS composed
of dense connective fibrous tissue. PS thickness
presented only mild variations in every single
tumor, and in some cases there were signs of
neoplastic penetration within its layers, with or
without invasion beyond it. The mean (range)
thickness of the tumor PS was 0.39 mm (0.048–
0.798 mm).
PS status was classified as follows: PS intact and
free from invasion (PS?; Fig. 2A); PS with signs of
neoplastic infiltration within its layers, with or
without invasion beyond it (PS+). PS+ was further
divided into four categories:
? PS with signs of neoplastic infiltration on the
parenchymalkidneysidewithnoinvasionbeyond
it (PSK+; Fig. 2B)
? PS with signs of neoplastic infiltration and inva-
sion beyond it on the parenchymal kidney side
(PSK++; Fig. 2C and 2D)
? PS with signs of neoplastic infiltration on the
perirenal adipose tissue side with no invasion
beyond it (PSF+; Fig. 2E)
? PS with signs of neoplastic infiltration and inva-
sion of the adipose tissue (pT3a) (PSF++; Fig. 2F).
The PS features are reported in Table 1; in 60 RCC
tumors (66.7%) the PS was intact and free from
invasion (PS?; Fig. 2A) while in 30 tumors (33.3%)
thereweresignsof infiltrationwithinitslayers, with
or without invasion beyond it (PS+). Indeed, 26.6% of
tumors (24/90) had PS invasion on the parenchymal
side; of those, 11 tumors (12.2% overall; 11/24, 45.8%
of tumors with PS invasion) had PS penetration
(PSK+; Fig. 2B), and 13 tumors (14.4% overall; 13/24,
54.2% of tumors with PS invasion) had PS penetra-
tion and invasion beyond it (PSK++; Fig. 2C and 2D).
The remaining 6 of 90 patients (6.6%) had PS
penetration on the perirenal fat tissue side, and of
those 2 patients (2.2%) had PS penetration (PSF+;
Fig. 2E) and 4 (4.4%) had penetration and invasion
beyond it (PSF++; pT3a; Fig. 2F).
The presence of PS+ by clinical and histologic
tumor dimension, tumor stage, nuclear grade, RCC
subtype and the presence of histologic necrosis is
reported in Table 2. PS+ was significantly associated
with clinical and pathologic tumor size, the pre-
senceof histologic necrosis and nucleargrade, while
there was no statistically significant correlation
european urology 55 (2009) 1410–1418
1413

Page 5

between PS+ and both tumor stage (pT1a vs pT1b)
and the histologic subtype (clear-cell vs papillary vs
chromophobe) (Table 2).
Using logistic regression analysis, the odds of
havingaPSinvasionincreasedsignificantlywiththe
increase in clinical tumor size, as measured by CT.
TheoddsratiofortheassociationofPSinvasionwith
clinical tumor size was 1.412 (95% CI, 1.03–2.018),
indicating that each 1-cm increase in clinical tumor
size was associated with a 41% increase in the odds
of PS invasion.
The presence of PS+ was also significantly
associated with pathologic tumor dimensions. The
difference between mean (?SD) pathologic tumor
size of tumors with (3.8 ? 2.32 cm) and without
(2.8 ? 1.27 cm) PS involvement was statistically
Fig. 2 – (A) Pseudocapsule (PS) intact and free from invasion (PSS); (B) PS with signs of neoplastic infiltration on the
parenchymal kidneysidewithnoinvasionbeyond it (PSK+); (C, D) PSwith signs ofneoplastic infiltrationand invasion beyond
it on the parenchymal kidney side (PSK++); (E) PS with signs of neoplastic infiltration on the perirenal adipose tissue side with
no invasion beyond it (PSF+); (F) PS with signs of neoplastic infiltration and invasion of the adipose tissue (pT3a) (PSF++).
Abbreviations: K, kidney; T, tumor; F, perirenal fat.
european urology 55 (2009) 1410–1418
1414