CLINICAL STUDY

A Retrospective Study of the Preoperative

Treatment of Advanced Wilms Tumor in Children

with Chemotherapy versus Transcatheter Arterial

Chemoembolization Alone or Combined with

Short-term Systemic Chemotherapy

Min-Ju Li, Professor, MD, Yin-Bao Zhou, Professor,Yong Huang, Associate Professor, Da-Xing Tang, MD, Shan Xu, Professor,

De-Hua Wu, Associate Professor, Yi-Ying Zhang, Professor,Hong-Feng Tang, Professor

ABSTRACT

Purpose: To evaluate the therapeutic effect of preoperative transcatheter arterial chemoembolization (TACE) combined withshort-term systematic chemotherapy in the treatment of advanced Wilms tumor.

Materials and Methods: This was a retrospective study on 66 patients with unilateral advanced Wilms tumor, age 5 months to11 years (median, 2.9 years; 30 boys and 36 girls), treated at our institution between 1995 and 2007. Characteristics of the patientpopulation were maximal tumor diameter � 10 cm, or involvement of periaortic lymph nodes, or inferior vena cava invasion, or distalmetastasis, or tumor with anaplastic histology. Patients were divided into three groups. Twenty patients were treated with conventionalpreoperative chemotherapy (PC group) using vindesine, actinomycin D, and pirarubicin for 4 weeks; 21 patients were treated in theTACE group with preoperative renal arterial chemoembolization using Lipiodol-pirarubicin-vindesine emulsion; and 25 patients weretreated with preoperative chemoembolization combined with short-term systematic chemotherapy (T�S) for 2 weeks.

Results: No drug-induced cardiotoxicity, nephrotoxicity, or hepatic dysfunction was observed. Complete surgical removal of thetumor was achieved in 12 (65.0%), 17 (80.9%), and 22 (88.0%) patients in the PC, TACE, and T�S groups, respectively (T�S groupvs PC group, P � .030). The 2-year relapse-free survival rates were 65.0%, 80.9%, and 100.0% in the PC, TACE, and T�S groups,respectively (T�S vs PC, P � .001).

Conclusions: From our experience, preoperative chemoembolization combined with short-term systematic chemotherapy is able toachieve higher rates of complete tumor resection and relapse-free survival in the treatment of advanced Wilms tumor.

ABBREVIATIONS

ADR � Adriamycin, FH � favorable histology, NWTS � National Wilms’ Tumor Study, PC � conventional preoperativechemotherapy, RFS � relapse-free survival rate, TACE � preoperative transcatheter arterial chemoembolization, THP � 4=-tetrahydropyranyldoxorubicin, T�S � preoperative chemoembolization combined with short-term systematic chemotherapy,

UH � unfavorable histology

WciuaWpsaun

From the Division of Pediatric Surgical Oncology, Department of PediatricSurgery (M-J.L, Y.H., D-X.T., S.X., D-H.W.), Department of Radiology (Y-B.Z.,Y-Y.Z.), and Department of Pathology (H-F.T.), Children’s Hospital, ZhejiangUniversity School of Medicine, Hangzhou 310003, China. Received May 15,2009; final revision received October 15, 2010; accepted November 16, 2010.Address correspondence to M-J.L.; E-mail: minjuli@hzcnc.com

This study was supported by a grant from the Foundation for Medical andHealth Care of Zhejiang Province, China (No.2003ZC027). None of the authorshave identified a conflict of interest.

© SIR, 2011

J Vasc Interv Radiol 2011; 22:279–286

bDOI: 10.1016/j.jvir.2010.11.025

ilms tumor is the most common malignant renal tumor inhildren. The outcome for patients with Wilms tumor hasmproved remarkably during the past decades owing to these of adjuvant chemotherapy and neoadjuvant chemother-py (1). However, the treatment of children with advancedilms tumor remains a challenge (2). Several untoward

rognostic factors have been identified, such as large tumorize, difficult operation, incomplete excision, peritonealdhesions, tumor extending to vena cava, metastases, andnfavorable histology (2–4). Novel treatment strategies areeeded to maximize survival and minimize long-term mor-

idity for the patients with advanced Wilms tumor.

ok

tw(paaps2nupttdpes

sa

v(ipC2

280 � Preoperative TACE in Treatment of Advanced Wilms Tumor Li et al � JVIR

The value of transcatheter arterial chemoembolization(TACE) has been accepted in the treatment of liver tumors(5,6). However, no randomized, validated studies appear in theliterature for the application of preoperative chemotherapyversus chemoembolization alone or combined with chemo-therapy in the treatment of renal tumors in children (7–12).

In an attempt to improve the outcome of patients withadvanced Wilms tumor, we have performed preoperativechemoembolization since 1995 (13–16). The purpose ofthis retrospective study is to review the therapeutic efficacyof preoperative renal chemoembolization combined withshort-term systematic chemotherapy.

MATERIALS AND METHODS

From January 1995 to December 2007, a total of 66 patientswith unilateral advanced Wilms tumor were treated at ourhospital. All patients underwent computed tomography(CT)/magnetic resonance imaging (MRI) and ultrasoundscan. A core-needle biopsy for histologic diagnosis wasperformed before the treatment. Histology results wereclassified as unfavorable if diffuse anaplastic features werepresent (UH) and favorable (FH) if absent. This study wasapproved by our institutional review board; the need forinformed consent for publication of data was waived.

Characteristics of the patient population were as fol-lows: maximal tumor diameter greater than 10 cm, and/orinvolvement of periaortic lymph nodes, or inferior venacava invasion, or distal metastasis based on the imagingstudies, or tumor with diffuse anaplastic histology (UH)according to the biopsy report.

Patients with bilateral renal tumors, congenital mesoblas-tic nephroma, malignant rhabdoid tumor, clear cell sarcoma,

Table 1. Demographics of Patient Groups

Characteristic

Median age, y (range)

Sex, M:F ratio

Tumor side, right/left

Tumor characteristics on admission (%)

Greatest dimension � 10 cm

Involvement of periaortic lymph nodes

Inferior vena cava invasion

Distal metastasis

Diffuse anaplastic histology

Duration of preoperative treatment, day (�x � s) (range)

Postoperative stage (%)

I

II

III

IV

Histology (%)

FH

UH

and renal cell carcinoma were excluded from this study. c

Eligible patients were between 5 months and 11 yearsf age (median, 2.9 years; 30 boys and 36 girls). The rightidney was treated in 36 and the left in 30 patients.

According to the modality of preoperative treatment,he patients were divided into three groups. Twenty patientsere treated with conventional preoperative chemotherapy

PC group) from January 1995 through December 2007; 21atients were treated with preoperative renal transcatheterrterial chemoembolization (TACE group) between Janu-ry 1995 and December 2002; and 25 patients underwentreoperative chemoembolization combined with short-termystematic chemotherapy (T�S group) between January003 and December 2007. The selection of the patients wasot exactly randomized, but each surgeon decided individ-ally which therapeutic modality should be applied in theatient he managed. In 1995 we introduced the TACEherapy after approval by the institutional review board, sohis therapy was available for application according to theecision of the surgeon. This method was applied in 21atients until 2002. To further improve the therapeuticffect, we have complemented the TACE therapy with ahort-term systematic chemotherapy since 2003.

Patients in the three groups were similar in terms ofex, age, tumor characteristics at diagnosis, and presence ofnaplasia (Table 1).

Patients in the PC group received conventional intra-enous preoperative chemotherapy consisting of vindesineMinsheng Pharmaceuticals Inc, Hangzhou, China) admin-stered four times as 3 mg/m2 injections once weekly,irarubicin (Main Luck Pharmaceuticals Inc, Shenzhen,hina) administered two times in single doses 20 mg/m2 onconsecutive days, and actinomycin D (Haizheng Pharma-

(n � 20) TACE (n � 21) T�S (n � 25)

1 (0.8–8) 3.3 (0.75–11) 2.9 (0.5–8)

9:11 7:14 14:11

6/14 13/7 12/13

2 (60.0) 13 (61.9) 16 (64.0)

6 (30) 7 (33.3) 7 (28.0)

1 (5.0) 0 (0) 1 (4.0)

1 (5.0) 1 (4.8) 3 (12.0)

1 (5.0) 2 (9.5) 2 (8.0)

4.3 (20–38) 12.0 � 2.52 (7–16) 15.3 � 1.8 (14–19)

2 (10.0) 1 (4.8) 0 (0.0)

6 (30.0) 9 (42.9) 11 (44.0)

1 (55.0) 10 (47.6) 11 (44.0)

1 (5.0) 1 (4.8) 3 (12.0)

9 (95.0) 19 (90.5) 23 (92.0)

1 (5.0) 2 (9.5) 2 (8.0)

PC

3.

1

29.2 �

1

1

euticals Inc, Taizhou, China) administered 10 �g/kg on 5

tflwpha

fdarbc

wt

p1rt

psns

Volume 22 � Number 3 � March � 2011 281

consecutive days. The timing of surgery in this group was4 weeks after starting chemotherapy.

In the TACE group, patients underwent transcatheter ar-terial chemoembolization under basal and caudal epidural an-esthesia. The catheter for selective renal arterial catheterizationwas a 4-F or 5-F Cobra catheter (Cook Vascular Incorporated,Pennsylvania). The chemoembolization emulsion was pre-pared immediately before injection with vigorous mixing forseveral minutes. Initial abdominal aortography and renal arte-riography were performed to define the tumor blood supply forthe purpose of planning the chemoembolization. The chemo-embolization emulsion consisted of iodized oil (Lipiodol;Guerbet, Aulna-Sons-B, France), 0.5 mL per tumor maximaldiameter (in centimeters); pirarubicin, 40 mg/m2; and vin-desine 3 mg/m2 that was infused into the renal artery flushinghe kidney tumor blood vessels, which were visible underuoroscopy (Fig 1). Intravenous hydration and alkalizationere administered during the procedure. After treatment, theatients received continued hydration and antibiotics for 48ours. Postprocedure nausea and vomiting were treated with

ntiemetics, but parenteral analgesics were not needed. o

In the T�S group, chemoembolization was used first,ollowed by 2 weeks of systemic chemotherapy with vin-esine 3 mg/m2 once and actinomycin D 10 �g/kg daily in5-day course before surgery. Surgical resection was car-

ied out 2 weeks after TACE. So the total dose of piraru-icin, vindesine and actinomycin D was similar to that ofonventional preoperative systemic chemotherapy.

There were five, three, and three patients with bodyeight � 12 kg in the PC, TACE, and T�S groups, respec-

ively, and the dosage of drugs was reduced to two thirds.The mean duration and standard deviation (�x � s) of

reoperative treatment were 29.2 � 4.3, 12.0 � 2.52, and5.3 � 1.8 days in the PC, TACE, and T�S groups,espectively; range, 20–38, 7–16, and 14–19 days, respec-ively (Table 1).

To minimize the side effects of anthracycline, we usedirarubicin (4=-O-tetrahydropyranyldoxorubicin, [THP]) in-tead of doxorubicin hydrochloride in this study. THP is aew anthracycline antibiotic with an antitumor efficacyimilar to that of doxorubicin but less cardiotoxity because

Figure 1. TACE performed on a 3-year-old girl with right renal Wilms tumor. (a)

Renal artery angiography found a largehypervascular lesion in the right kidneywith abundant neovascularity. (b) Che-moembolization emulsion consisting ofiodized oil was infused into the renalartery. (c) Lipiodol deposits in the embo-lized tumor after TACE by CT scan.

f its different pharmacodynamic properties (17).

0trltrTG

swS

awt

mpxa

aespSIfrb

(tlGoali

(atdrltcllid

aaibiimpm

RTbaggtt

282 � Preoperative TACE in Treatment of Advanced Wilms Tumor Li et al � JVIR

Tumor volumes were measured on ultrasound scansusing the ellipsoid formula (length � thickness � depth �.523). The measurement was performed before preopera-ive therapy and repeated before surgery. Tumor volumeeduction in comparison with the initial volume was calcu-ated. Two groups of responders were defined according tohe reduction of tumor volume: (i) poor response (� 40%eduction) and (ii) good response (� 40% reduction) (18).oxicity was scored according to the Children’s Cancerroup toxicity grading system (19).

After tumor resection, the patients were assigned tourgical stage. Histopathologic classification and stagingere done according to the Third National Wilms’ Tumortudy (NWTS) Group (20).

Postoperative treatment was based on tumor histologynd its postsurgical stage. The treatment protocol wasorked out according to the NWTS protocol modified by

he Beijing Children’s Hospital (21).Postoperative follow-up was performed at the first

onth and then every 3 months after surgery, includinghysical, imaging (abdominal sonography and CT, chest-ray, electrocardiogram) and laboratory screening (bloodnd urine analysis, and liver and renal function tests).

Standard methods were used for the analysis of censorednd noncensored data, including the Kaplan-Meier method forstimating survival curves, the log-rank test for comparison ofurvival curves, and the Fisher exact test for comparing inde-endent proportions. All analyses were carried out using thePSS 13.0 software statistical system (SPSS Inc., Chicago,llinois). The survival time was measured in months, startingrom the date of diagnosis and ending with either the date ofecurrence or metastasis or on a censoring date, or the patientseing still alive on 31 December 2008.

RESULTS

Toxicity and Complications of TreatmentNo preoperative death, preoperative tumor rupture, drug-induced cardiotoxicity, renal insufficiency, or hepaticdysfunction were found in any of the three groups. Oral

Table 2. Complications of Preoperative Treatment in Each Gr

Complication PC (n � 20) TAC

Preoperative death 0

Renal insufficiency 0

Hepatic dysfunction 0

Cardiotoxicity 0

Moderate fever 1 (5.0%) 12

Mucositis 5 (25.0%) 0

Grade I–II marrow suppression 6 (30.0%) 1

Note.—NS � not significant.

mucositis developed in five (25.0%), zero (0%), and one C

4.0%) patients in groups PC, TACE, and T�S, respec-ively (Table 2). The latter two groups had significantlyower instances than the PC group (P � .034, P � .021).rade I-II marrow suppression developed in six (30.0%),ne (4.8%), and two (8.0%) patients in groups PC, TACE,nd T�S, respectively (Table 2). The latter two groups hadower instances than the PC group but did not show signif-cant differences.

In the TACE group and T�S group, 12 (57.1%) and 1456.0%) patients became moderately febrile (about 38°C)fter chemoembolization; this was controlled with symp-omatic treatment (Table 2). Two weeks after TACE, io-ized oil deposits were still visible within the tumor in theepeated CT scan. Iodized oil deposition in the paraaorticymph nodes was observed in three cases, although paraaor-ic lymph nodes were not detectable in the CT scans beforehemoembolization of these patients. Postoperative histo-ogic examination of these marked lymph nodes confirmedymph node metastases with necrosis (Fig 2). This find-ng implies that the chemoembolization agent flowedirectly into the paraaortic lymph node metastases.

One patient required repeated embolization 2 weeksfter the first chemoembolization due to the presence of anccessory renal artery. This patient showed absence ofodized oil accumulation in the inferior limit of the tumory CT scan after chemoembolization. Upon review of thenitial treatment, a missed accessory renal artery wasdentified. That vessel was treated during a second che-oembolization. The tumor further shrank and was com-

letely resected. That patient remained alive at 72onths of follow-up (after surgery).

esponse to Treatmentumor volumes were significantly reduced after chemoem-olization combined with short-term systematic chemother-py. There was significant difference between the T�Sroup and PC group in regard to their proportion of poor orood responders (P � .013) (Table 3). The rate of completeumor resection in the T�S group was significantly higherhan that in PC group (P � .030), as shown in Table 3.

21) T�S (n � 25)

P Value

T�S vs PC TACE vs PC

0 NS NS

0 NS NS

0 NS NS

0 NS NS

) 14 (56.0%) .001 .001

1 (4.0%) .034 .021

2 (8.0%) NS NS

oup

E (n �

0

0

0

0

(57.1%

(0%)

(4.8%)

omparison of the TACE group with the PC group showed

sg4v7(hsFf

sgTceg

D

Ahtt

Volume 22 � Number 3 � March � 2011 283

no significant difference; however, it showed a trend ofhigher rate of complete tumor resection in the TACE group.

PathologyIn the patients who underwent chemoembolization, patho-logic examination of the specimen found massive necrosisin the tumor and increased thickness of the fibrous envelopearound the tumor. Yellowy residues of iodized oil emulsionand foaming histiocyte infiltration were observed in thetumor (Fig 3). Necrosis was seen not only in the maintumor but also in the periaortic lymph node metastases withLipiodol deposition (Fig 2c) (22). There was a significantdifference between the T�S group and PC group in regardto the rate of tumor total necrosis (P � .044) (Table 3).

Postoperative staging and histologic typing within thethree groups are shown in Table 1.

OutcomesNone of the patients had secondary malignancy occurrenceat the last follow-up on December 31, 2008. In the PCgroup, one patient died during surgery. One patient in theTACE group was lost to follow-up 2 years after surgery. Allpatients in the T�S group have been on follow-up so far.The patients in the PC, TACE, and T�S groups werefollowed up for 3–156 months (mean � SD, 89.1 � 63.6),3–163 months (86.9 � 52.8), and 13–67 (40.7 � 18.3)

months, respectively (Table 4). a

The patients in the PC group had a 2-year relapse-freeurvival rate (RFS) of 65.0% versus 80.9% in the TACEroup and 100.0% in the T�S group (Table 4) (Fig 4). The-year RFS estimates were 59.6% (95% confidence inter-al, 48.5%–70.7%) and 80.0% (95% confidence interval,2.4%–89.5%) in the PC and TACE groups, respectivelyTable 4) (Fig 4). In the T�S group, all of the 25 patientsad disease-free survival, including FH stage II in 10, UHtage II in one, FH stage III in 10, UH stage III in one, andH stage IV in three patients. Nine patients had beenollowed up for more than 4 years.

The log-rank test comparing the survival curveshowed significant difference in survival rate of the T�Sroup compared with that of PC group (P � .001) andACE group (P � .024), respectively. The TACE groupompared with the PC group showed no significant differ-nce (P � .153); however, a trend of increase in the TACEroup’s survival rate was noted (Table 4).

ISCUSSION

lthough the survival rate of the patients with Wilms tumoras greatly improved, the treatment of advanced Wilmsumor remains a challenge. Some of the advanced Wilmsumors have been regarded as “inoperable” or “unresect-

igure 2. Iodized oil deposition in the periaorticmph nodes metastases, which was undetected byT scan on admission. (a) Advanced Wilms tumor in2-year-old boy. CT scan on admission showed the

umor in the right kidney but no periaortic lymphode metastasis. (b) CT scan after chemoemboliza-

ion. Lipiodol deposition was seen not only in theain tumor but also in the periaortic lymph nodeetastasis (arrow). (c) Microscopic examination

ound massive necrosis in the periaortic lymph nodend a small quantity of yellow Lipiodol depositionarrow) (hematoxylin and eosin stain, � 40). (Avail-ble in color online at www.jvir.org.)

F

lyCatntmmfa(a

ble” (2–4). Radiation therapy was one of the early adju-

Uyct1cb

in and

� 4.4

284 � Preoperative TACE in Treatment of Advanced Wilms Tumor Li et al � JVIR

vant treatments when operative removal of the tumor wastechnically difficult (23). The International Society of Pe-diatric Oncology studies largely focus on the issue of pre-operative chemotherapy to facilitate surgery of a shrunkentumor and to treat metastasis as early as possible. Theduration of conventional preoperative chemotherapy is 4 or8 weeks (24–26). Preoperative chemotherapy is also used

Table 3. Tumor Response, Operative Blood Loss, Operating TResection, and Total Necrosis in Each Group

Variable PC (n � 20)

Tumor volume, mL

On admission (�x � s) 378.7 � 118.3

Before operation (�x � s) 291.5 � 112.8

Response

Good response (� 40% reduction) 7 (35.0%)

Poor response (� 40% reduction) 13 (65.0%)

Operative blood loss, mL (�x � s) 212.5 � 55.9

Operating time, min (�x � s) 159.5 � 24.8

Rapture during operation 4/20 (20.0%)

Microscopic residual 4/20 (20.0%)

Complete tumor resection 12/20 (60.0%)

Tumor total necrosis 2/20 (10.0%)

Note.—NS � not significant.

Figure 3. Pathologic examination of the surgical specimen. (a)

the tumor (arrow) and thickening of the tumor fibrous capsule.Lipiodol emulsion residues were found in the tumor (hematoxyl

Table 4. Follow-up and Cumulative Survival of Patient Group

Variable PC (n � 20) TACE

Death during surgery 1

Median follow-up, mo (range) 114 (3–156) 89 (

2-year RFS, % (� SE) 65.0 � 10.7 80.9

4-year RFS, % (� SE) 59.6 � 9.6 80.0

for treatment of “inoperable” Wilms tumor by NWTS and t

nited Kingdom Children’s Cancer Study group in recentears (27,28). However, some patients did not respond toonventional preoperative chemotherapy and died beforehe excision of primary tumor. Ritchey et al (3) reviewed31 children in NWTS-3 who had received preoperativehemotherapy for unresectable tumors or judged inoperabley imaging. Thirteen of them did not respond to chemo-

apture During Operation, Microscopic Residual, Complete

n � 21) T�S (n � 25)

P Value

T�S vs PC TACE vs PC

153.2 381.7 � 138.8 NS NS

127.1 230.0 � 92.8 NS NS

52.4%) 18 (72.0%) .013 NS

65.0%) 7 (28.0%) .013 NS

64.5 170.8 � 60.6 .022 NS

24.8 138.4 � 27.5 .011 NS

9.5%) 1/25 (4.0%) NS NS

9.5%) 2/25 (8.0%) NS NS

80.9%) 22/25 (88.0%) .030 NS

23.8%) 9/25 (36.0%) .044 NS

scopic examination of the specimen found massive necrosis incroscopically, extensive and homogenous necrosis and yelloweosin stain, � 200). (Available in color online at www.jvir.org.)

1) T�S (n � 25)

P Value

T�S vs PC T�S vs TACE

0

42 (13–67)

100.0 � 0.0 .001 .024

ime, R

TACE (

391.9 �

276.1 �

11 (

10 (

173.8 �

144.8 �

2/21 (

2/21 (

17/21 (

5/21 (

Macro(b) Mi

s

(n � 2

0

3–163)

� 8.6

herapy, but the disease progressed. Eight children died

br

lclprnamtmth

fl

lapslie

A

TasDf

R

Volume 22 � Number 3 � March � 2011 285

before the removal of primary tumor (3). Ora et al (29)reported tumor progression during preoperative chemother-apy in 57 of 1,090 patients (5%) with localized Wilmstumors. Patients whose tumors do increase in size havepoorer event-free and overall survival rates (29) indepen-dently of stage distribution and histopathologic risk group.

The merits of renal chemoembolization are basedon the concept that the blood supply of tumor only comesfrom the renal artery. The anticancer drug and embolizingmaterial are injected into the tumor feeding artery, increas-ing the effect of the chemotherapy agents in the ischemictissue and reducing the risk of bleeding during surgery (30).Animal experiments revealed that renal arterial chemoem-bolization can maintain high local concentrations of theanticancer drug while maintaining low blood levels of theanticancer drug (31–33). In our study, we found the depo-sition of Lipiodol-containing chemotherapy drugs withinthe tumor and in the metastasis of periaortic lymph nodesand thus extend the contact time of chemotherapeutic agentwith tumor cells and induce ischemia and necrosis, whileavoiding concomitant systemic toxicity. Preoperative che-moembolization may reduce intraoperative blood loss andtumor rupture. Nevertheless, the survival rate of chemoem-bolization alone is hardly ideal. Three patients in the TACEgroup died from pulmonary metastasis, and one died fromlocal tumor recurrence after surgery. The first was a 6-year-old girl with anaplastic histology stage-III tumor. She un-derwent surgery 1 week after chemoembolization. The tu-mor was ruptured and incompletely resected duringsurgery, and lung metastasis emerged 6 months after oper-ation. This is possibly a result of performing surgery tooearly after chemoembolization. As a result, we delayedresection to 2 weeks after chemoembolization. The secondand third patients were a 2-year-old girl and a 3-year-oldgirl with favorable histology stage-III tumors. They suf-fered from pulmonary metastasis 3 and 12 months after

Figure 4. Kaplan-Meier survival estimates for patients. (Avail-able in color online at www.jvir.org.)

surgery, respectively. The fourth patient was a 3-year-old

oy with stage-II tumor who had an intraabdominal tumorelapse 20 months after surgery.

In this study, we found that preoperative chemoembo-ization combined with short-term systemic chemotherapyan induce more massive necrosis of tumor and periaorticymph nodes metastasis, further improve the tumor com-lete resection rate, and achieve a high disease-free survivalate. This approach is the preoperative multimodal combi-ation therapy consisting of localized arterial chemother-py, arterial embolization, and intravenous systemic che-otherapy. This combination is effective in that it causes

umor necrosis while eliminating the possible existence oficro-metastasis and gives excellent surgical clearance of

umor mass with minimal risk of recurrence, thereby en-ancing survival.

There were limitations in this study. As a result ofewer cases in this group and a shorter observation time,ong-term effects warrant further investigation.

These findings suggest that preoperative chemoembo-ization combined with short-term systematic chemother-py is a safe and effective modality for the treatment ofatients with advanced Wilms tumor. This therapy offerseveral advantages, including optimal tumor and periaorticymph node metastasis, shorter preoperative treatment time,ncreased complete tumor resection rate of the tumor, andxcellent prognosis.

CKNOWLEDGEMENTS

he authors thank all the children and their parents forllowing us to publish the data collected during this re-earch project. The authors also thank the staff of theepartment of Pediatric Surgical Oncology and Radiology

or their patient care and organization.

EFERENCES

1. D’Angio GJ. The National Wilms Tumor Study: a 40 year perspective.Lifetime Data Anal 2007; 13:463–470.

2. Grundy RG, Hutton C, Middleton H, et al. Outcome of patients withstage III or inoperable WT treated on the second United Kingdom WTprotocol (UKWT2); a United Kingdom Children’s Cancer Study Group(UKCCSG) study. Pediatr Blood Cancer 2004; 42:311–319.

3. Ritchey ML, Shamberger RC, Haase G, Horwitz J, Bergemann T, BreslowNE. Surgical complications after primary nephrectomy for Wilms’ tu-mor: report from the National Wilms’ Tumor Study Group. J Am Coll Surg2001; 192:63–68.

4. Ritchey ML, Pringle KC, Breslow NE, et al. Management and outcomeof inoperable Wilms’ tumor. A report of National Wilms Tumor Study-3.Ann Surg 1994; 220:683–690.

5. Oue T, Fukuzawa M, Kusafuka T, Kohmoto Y, Okada A, Imura K.Transcatheter arterial chemoembolization in the treatment of hepatoblas-toma. J Pediatr Surg 1998; 33:1771–1775.

6. Takaki Y, Kaminou T, Shabana M, Ihaya T, Otsubo K, Ogawa T, et al.Suitable blending method of lipiodol-cisplatin in transcatheter arterialembolization for hepatocellular carcinoma: evaluation of sustained re-lease and accumulation nature. Hepatogastroenterology 2008; 55:202–206.

7. Harrison MR, de Lorimier AA, Boswell WO. Preoperative angiographic

embolization for large hemorrhagic Wilms’ tumor. J Pediatr Surg 1978;13:757–758.

1

1

1

1

1

1

1

1

1

2

2

2

2

2

2

2

2

2

2

3

3

3

3

286 � Preoperative TACE in Treatment of Advanced Wilms Tumor Li et al � JVIR

8. Danis RK, Wolverson MK, Graviss ER, O’Connor DM, Joyce PF, CradockTV. Preoperative embolization of Wilms’ tumors. Am J Dis Child 1979;133:503–506.

9. Gapchenko AS, Kononenko NG, Iugrinov OG, Galakhin KA, Siniuta BF,Shevchenko NV. Chemoembolization of blood vessels in the complextreatment of children with nephroblastoma. Klin Khir 1992; 5:18–21.

10. Zupancic B, Bradic I, Batinica S, et al. Our 10-year experience withembolized Wilms’ tumor. Eur J Pediatr Surg 1995; 5:88–91.

1. Chitnis M, Chowdhary SK, Lazarus C. Preoperative angioembolisationfor life-threatening haemorrhage from Wilms’ tumour: a case report.Pediatr Surg Int 2004; 20:290–291.

2. Smith NP, Jesudason EC, McDowell HP, Rowlands P, Ashworth M,Losty PD. Emergent embolisation to control severe haematuria inWilms’ tumour. Pediatr Surg Int 2005; 21:313–315.

3. Li MJ, Tang DX, Zhou YB, et al. Preoperative interventional therapy forchildren with advanced Wilms’ tumor. Chin J Pediatr Surg 2001; 22:10–13 (in Chinese).

4. Li MJ, Zhou YB, Shen LG. Prospective study of preoperative transcath-eter arterial chemo-embolization for Wilms’ tumor. Zhejiang Da Xue XueBao Yi Xue Ban 2003; 32:69–71 (in Chinese).

5. Li MJ, Huang Y, Tang DX, et al. Treatment of advanced Wilms’ tumor.Zhonghua Zhong Liu Za Zhi 2006; 28:791–795 (in Chinese).

6. Liu WG, Gu WZ, Zhou YB, Tang HF, Li MJ, Ma WX. The prognosticrelevance of preoperative transcatheter arterial chemoembolization(TACE) and PCNA/VEGF expression in patients with Wilms’ tumour. EurJ Clin Invest 2008; 38:931–938.

7. Munck JN, Riggi M, Rougier P, et al. Pharmacokinetic and pharmaco-dynamic advantages of pirarubicin over adriamycin after intraarterial he-patic administration in the rabbit VX2 tumor model. Cancer Res 1993;53:1550–1554.

8. Weirich A, Ludwig R, Graf N, et al. Survival in nephroblastoma treatedaccording to the trial and study SIOP-9/GPOH with respect to relapse andmorbidity. Ann Oncol 2004; 15:808–820.

9. Trotti A, Colevas A, Setser A, et al. CTCAE v3.0: development of acomprehensive grading system for the adverse effects of cancer treat-ment. Semin Radiat Oncol 2003; 13:176–181.

0. D’Angio GJ, Breslow N, Beckwith JB, et al. Treatment of Wilms’tumor. Results of the Third National Wilms’ Tumor Study. Cancer 1989;64:349–360.

1. Bai JW. Wilms’ tumor, In: Zhang JZ (ed). Modern Pediatric Oncologic

Surgery. Beijing: Beijing Science Publishing House, 2003; 245–254.

2. Lin L, Tang HF, Li MJ, et al. Pathological alterations of Wilm=s tumorafter interventional chemotherapy. J Clin Exp Pathol 2007; 23:313–316.

3. Wagget J, Koop CE. Wilms’ tumor preoperative radiotherapy and che-motherapy in the management of massive tumors. Cancer 1970; 26:338–340.

4. Graf N, Tournade MF, de Kraker J. The role of preoperative chemo-therapy in the management of Wilms’ tumor. The SIOP studies. Inter-national Society of Pediatric Oncology. Urol Clin North Am 2000; 27:443–454.

5. Reinhard H, Semler O, Burger D, et al. Results of the SIOP 93-01/GPOH trial and study for the treatment of patients with unilateral non-metastatic Wilms’ Tumor. Klin Pediatr 2004; 216:132–140.

6. Tournade MF, Com-Nougué C, de Kraker J, et al. Optimal duration ofpreoperative therapy in unilateral and nonmetastatic Wilms’ tumor inchildren older than 6 months: results of the Ninth International Society ofPediatric Oncology Wilms’ Tumor Trial and Study. J Clin Oncol 2001;19:488–500.

7. Ritchey ML. The role of preoperative chemotherapy for Wilms’ tumor:the NWTSG perspective. National Wilms’ Tumor Study Group. SeminUrol Oncol 1999; 17:21–27.

8. Mitchell C, Pritchard-Jones K, Shannon R, et al. Immediate nephrec-tomy versus preoperative chemotherapy in the management of non-metastatic Wilms’ tumour: results of a randomised trial (UKW3) by theUK Children’s Cancer Study Group. Eur J Cancer 2006; 42:2554–2562.

9. Ora I, van Tinteren H, Bergeron C, de Kraker J. Progression of localisedWilms’ tumour during preoperative chemotherapy is an independentprognostic factor: a report from the SIOP 93-01 nephroblastoma trial andstudy. Eur J Cancer 2007; 43:131–136.

0. Bakal CW, Cynamon J, Lakritz PS, Sprayregen S. Value of preoperativerenal artery embolization in reducing blood transfusion requirementsduring nephrectomy for renal cell carcinoma. J Vasc Intervent Radiol1993; 4:727–731.

1. Fujiwara K, Hayakawa K, Nagata Y, et al. Experimental embolization ofrabbit renal arteries to compare the effects of poly L-lactic acid micro-spheres with and without epirubicin release against intraarterial injectionof epirubicin. Cardiovasc Intervent Radiol 2000; 23:218–223.

2. Kurzidem M, Seidensticker P, Rassweiler J. Renal chemoembolizationwith mitomycin c/Ethibloc: pharmacokinetics and efficacy in an animalmodel. J Endourol 2002; 16:515–518.

3. Kónya A, Choi BG, Van Pelt CS, Wright KC. Transcatheter arterial

embolization of renal VX-2 carcinoma: ethiodol-ethanol: capillary emboli-zation combined with carboplatin. Korean J Radiol 2007; 8:136–147.