- Email: [email protected]
GPOH study
ARTICLE IN PRESS Surgical Oncology (2008) 17, 33–40
Available at www.sciencedirect.com
journal homepage: www.elsevier.com/locate/suronc
REVIEW
Surgical implications for liver metastases in nephroblastoma—Data from the SIOP/GPOH study$ Joerg Fuchsa,,1, Philipp Szavaya,1, Tobias Luithlea, Rhoikos Furtwa ¨nglerb, Norbert Grafb a
Department of Pediatric Surgery, Children’s Hospital, University of Tuebingen, Hoppe-Seyler-Strasse 3, 72076 Tuebingen, Germany b Department of Pediatric Hematology and Oncology, University of Homburg/Saar, 66421 Homburg/Saar, Germany Accepted 28 August 2007
KEYWORDS Nephroblastoma; Wilms’tumor; Surgery; Treatment; Metastases; Hepatic; Liver; Recurrence; Survival
Summary Background: In children with Wilms’ tumor, the 5-year overall survival rate is over 90% in the SIOP/GPOH study group. However, a small group of patients have tumor lesions in the liver at the time of initial diagnosis or as a recurrence. This group seems to have a worse prognosis in terms of survival. The treatment and outcome of patients with a hepatic recurrence were analyzed compared to previously published data of patients with primary hepatic metastases. Patients and methods: We reviewed the records of 45 out of 1365 patients enrolled in the SIOP 93-01/GPOH study and the SIOP 2001/GPOH study between April 1, 1994 and September 30, 2004. Median age at diagnosis was 6.49 years (1.37–34.16 years) in 29 patients who were initially presented with hepatic metastases (group I) with 9 males and 20 females. In 16 children who had a recurrence of a nephroblastoma in the liver (group II), median age at diagnosis was 4.62 years (1.84–31.08 years) with 9 males and 7 females.
Abbreviations: ACT-D, dactinomycin; ADR, adriamycin; AVD, ACT-D, VCR, DOX; CARBO, carboplatin; CCE, CARBO, CYCLO, VP 16, G-CSF; CCED, CARBO, CYCLO, VP 16, DOX; CCEV, CARBO, CYCLO, VP 16, VCR; CR, complete remission; CRF, case report form; CT, computer tomography; CYCLO, cyclophosphamide; DOX, doxorubicin; G-CSF, granulocyte stimulating factor; GPOH, German Association of Pediatric Oncology and Hematology; Gy, Gray; ICE, CARBO, VP 16, IFO, G-CSF; IFO, ifosfamide; MRI, magnetic resonance imaging; NWTS, National Wilms’ Tumor Study; NWTSG, National Wilms’ Tumor Study Group; SIOP, International Society of Paediatric Oncology; VCR, vincristine; VOD, veno-occlusive disease; VP 16, etoposide $ Presented in part at the 37th annual conference of The International Society of Paediatric Oncology SIOP 2005, Vancouver, Canada, September 21–24, 2005. Corresponding author. Tel.: +49 7071 298 6621; fax: +49 7071 29 4046. E-mail address: [email protected] (J. Fuchs). 1 Contributed equally in this study. 0960-7404/$ - see front matter & 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.suronc.2007.08.011
ARTICLE IN PRESS 34
J. Fuchs et al. Results: In group I out of 29 patients, 11 died at a median of 13.07 months. Overall survival in group I was 62.58%. In group II, 9 patients died at a median 52 months. Overall survival in group II was 54.7%. Conclusion: This report suggests that when complicated by metastases of the liver, Wilms’ tumor has a less favorable outcome. Chemotherapy and radiotherapy play a definitive role in the treatment of these children. The importance of complete resection of hepatic lesions in both groups should be emphasized. & 2007 Elsevier Ltd. All rights reserved.
Contents Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 Methods and patients . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 General remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 Classification and staging system . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35 Recommendation of the treatment protocols in cases with liver metastases . . . . . . . . . . . . . . . . . . . . . . . . . 35 Principles of chemotherapy and radiation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35 Statistical methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35 Patient data and grouping . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35 Diagnostic evaluation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36 Group I . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36 Metastases pattern . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36 Surgery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36 Chemotherapy and radiation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36 Histological classification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 Outcome . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 Group II. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 Metastases pattern . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 Surgery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 Chemotherapy and radiation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 Histological classification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 Outcome . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 Overall results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 Conflict of Interest Statement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
Introduction Striking improvement in treatment for children with a Wilms’ tumor has been achieved, leading to 5-year overall survival rates above 90% in the SIOP/GPOH study group [1,2]. Despite this fact, several issues remain subject to concern and ongoing discussion: relapsing nephroblastoma, metastatic disease and risk stratification. In this study, we focus on a small group of patients who had tumor lesions in the liver at the time of initial diagnosis or developed a hepatic recurrence of the disease. The purpose of this study is to evaluate the data of the SIOP/ GPOH 93-01 and 2001 study group in order to identify factors in surgical treatment that have an effect on the prognosis of these patients. Data of children with a hepatic recurrence of Wilms’ tumor were compared to previously published data of patients with primary hepatic metastases [3].
Methods and patients In order to define survival and success of treatment of this group of patients, we reviewed the records of 1365 patients
enrolled in the SIOP 93-01/- and the SIOP 2001/Wilms’ tumor study of the German Society of Pediatric Oncology and Hematology between April 1, 1994 and September 30, 2004. These studies conform to the Declaration of Helsinki and were approved by the ethic commissions of all participating centers (leading IRB: Ethikkommission der Landesaeztekammer des Saarlandes, Germany, IRB00004360).
General remarks All children from Germany, Austria and Switzerland suffering from renal tumors were prospectively registered in the Wilms’ tumor studies SIOP 93-01 and 2001 of the German Society of Pediatric Hematology and Oncology (GPOH). For each patient a case report form (CRF) for registration, preoperative chemotherapy, surgical treatment of the primary tumor and/or metastases, postoperative chemotherapy, and radiation was acquired. Pre-treatment CT scans or MRI images were reviewed by reference radiologists of the study (J. Tro ¨ger, Heidelberg, Germany and Unsinn,
ARTICLE IN PRESS Surgical implications for liver metastases in nephroblastoma Innsbruck, Austria). The tumor specimens were analyzed by the local pathologist and then sent for review to the SIOP/ GPOH Pediatric Tumor Registry (D. Harms and I. Leuschner, Kiel, Germany). For regular follow-up, patients received physical examination, imaging diagnostics and laboratory screening up to 5 years after initial diagnosis.
Classification and staging system Histological classification for nephroblastoma was carried out according to the SIOP criteria, dividing tumors into low, intermediate and high-risk groups. Between SIOP 93-01 and SIOP 2001, blastemal predominant tumors were shifted into the high-risk group according to their poor prognosis, whereas focal anaplasia was classified as intermediate risk in SIOP 2001 [5–7]. Staging was based on the post-surgical staging system according to the SIOP protocol [4]. In stage I, the tumor is limited to the kidney and completely removed. The renal capsule has an intact outer surface. In stage II, the tumor extends beyond the kidney, but is completely removed. There may be an extrarenal tumor extension, e.g. penetration of the tumor capsule or a tumor thrombus in the renal vein. The involvement of local lymph nodes is possible (stage II N+ [SIOP 93-01/GPOH], stage III [SIOP 2001/GPOH]). Stage III is a tumor rupture or macroscopic tumor remnant after surgery. Stage IV is metastatic disease (lung, liver, bone, brain) and stage V refers to the phenomenon of bilateral disease.
Recommendation of the treatment protocols in cases with liver metastases All children with primary liver metastases should receive a preoperative chemotherapy consisting of dactinomycin (ACT-D), vincristine (VCR) and doxorubicin (DOX). In cases, where metastases do not disappear, a complete resection of liver metastases is recommended during tumor nephrectomy or 14 days later. Radiation is indicated in cases where liver metastases did not disappear after chemotherapy and could not be resected, or in children with incomplete resection of metastases. Dosage for liver radiation varies from 15 to 30 Gy (single dose 1.5 Gy). All children with tumor recurrence in the liver should initially be treated with chemotherapy followed by a riskadapted approach depending on tumor response and risk factors regarding initial treatment (local stage, histology, time to relapse). Surgery for liver metastasis should be performed as soon as possible after chemotherapy. Only general recommendations for tumor resection exist, without a specific timeinterval during treatment.
Principles of chemotherapy and radiation All children from 6 months to 16 years of age with unilateral nephroblastoma and liver metastases at the time of initial diagnosis (stage IV) receive the same 3 drug preoperative chemotherapy in both studies. Chemotherapy consists of dactinomycin (ACT-D; 15 mg/kg body weight; maximal bolus
35 injection 0.5 mg; day 1, 2, 3, week 1, 3, 5 [SIOP 93-01/ GPOH] or 45 mg/kg body weight as a single injection on 1 day, week 1, 3, 5 [SIOP 2001/GPOH], vincristine (VCR; 1.5 mg/ m2; maximal bolus injection 2.0 mg; day 1 from week 1–6) and adriamycin (ADR; 50 mg/m2 as 4-h-infusion, day 1 from week 1, 5). For small children (infancy or body weight o12 kg) the drug dosage is generally reduced to two-thirds. Postoperative chemotherapy depends on risk stratification after tumor resection. In cases with complete tumor resection and low or intermediate risk, the children receive ACT-D, VCR and DOX (AVD) without radiation. All children with incomplete tumor resection or children in the high-risk group receive etoposide (VP 16), CARBO, DOX, cyclophosphamide (CYCLO) and additional radiotherapy (maximum 30 Gy). In cases of tumor recurrence, the study differentiates between a 3-drug and a 4-drug regimen. AVD, the 3-drug regimen (ACT, VCR, DOX), is administered for 27 weeks to patients with stage I after tumor nephrectomy. The 4-drug regimen, CCEV (CYCLO, CARBO, VP 16, VCR) or CCED (CYCLO, CARBO, VP 16, DOX), depending on whether DOX was administered initially, is administered for 34 weeks to patients with all other stages after initial surgery. The highrisk group (including initial stages III and IV, high malignancy such as diffuse anaplasia, and early recurrence or recurrence in areas of radiation) receives chemotherapy with two courses of ICE (ifosfamide (IFO), carboplatin (CARBO), etoposide (VP 16)) or CCE (CYCLO, CARBO, VP 16), depending on the primary chemotherapy. After this multiagent therapy, high-dose chemotherapy with stem-cell rescue is recommended in case of CR [4].
Statistical methods Standard methods were used for the analysis of the censored and non-censored data, including Kaplan–Meier curves, and Fisher’s exact test for comparing independent proportions. Overall or relapse free survival was assessed with the proportional hazards model. Post-relapse survival was defined from the time of initial relapse until death. All analyses were carried out using SPSS for Windows. Statistical significance was assumed for po0.05.
Patient data and grouping Altogether, 45 out of 1365 patients had liver lesions deriving from a nephroblastoma, either at the time of initial diagnosis or when relapsing.
Diagnostic evaluation Liver lesions were diagnosed with abdominal ultrasonography and CT or MRI. All patients received more than one of the above-mentioned investigations. Forty patients had a unilateral tumor, 4 patients had a bilateral Wilms’ tumor, and one patient had an extrarenal Wilms’ tumor. Twenty-three of the unilateral tumors were left-sided and 17 were right-sided. In addition to hepatic metastases, 28 patients had metastatic lesions in the lungs. Other sites of metastases included bones (n ¼ 4), mediastinum (n ¼ 3), and diaphragm (n ¼ 2). Three patients developed inferior cava vein thrombosis.
ARTICLE IN PRESS 36
J. Fuchs et al.
According to the SIOP-classification, out of 45 patients, 5 had a stage I tumor, 11 with stage II, 13 with stage III, 12 with stage IV and 4 patients with stage V (see Table 1). The patients with primary liver metastases (group I) and those with secondary liver metastases (group II) were analyzed separately. Data of group I have been published previously [3]: Group I: 29 out of 1365 patients (2.12%) had liver metastases at the time of initial diagnosis. The median age at diagnosis for these 29 patients was 6.49 years (range 1.37–34.16 years) with 9 males and 20 females in the group. Group II: 16 out of 1365 patients had a recurrence in the liver at a median of 12.5 months (2–35 months) after initial diagnosis. Median age at initial diagnosis was 4.62 years (1.84–31.08 years) with 9 males and 7 females in the group.
Results Group I Metastases pattern All primary liver metastases were distant, not-contiguous metastases.
Table 1 Tumor-stage after preoperative chemotherapy in patients with primary (group I) and secondary (group II) hepatic metastases of nephroblastoma, n ¼ 45. Stage according to SIOP Stage Stage Stage Stage Stage
I II III IV V
Table 2
Group 1 (n ¼ 29)
Group 2 (n ¼ 16)
Patients
%
Patients
%
0 9 8 10 2
0 31 28 34 7
5 2 5 2 2
31.25 12.5 31.25 12.5 12.5
Surgery In group I, all patients but 2 underwent tumor-nephrectomy. Liver metastases were operated in 11 children at time of nephrectomy. Primary complete resection was achieved in 6 patients, while in 5 children surgery of liver lesions was incomplete. Complete resection of hepatic lesions was achieved in 3 patients in a second attempt. In one patient, a secondary resection still was incomplete. In 14 patients, surgery for liver lesions was not undertaken (see Table 2).
Chemotherapy and radiation Nineteen out of 29 patients received preoperative chemotherapy; 14 according to the study protocols for stage IV disease of the SIOP 93-01/GPOH and SIOP 2001/GPOH. Due to small age or weight, 5 children received standard risk chemotherapy without doxorubicin. Seven out of 29 patients did not receive chemotherapy preoperatively. In these patients, treatment was not according to the guidelines of the treatment protocol. This group comprised the 6 patients who underwent primary nephrectomy and 1 patient with a Smith-Lemli-Opitz-Syndrome, in whom no treatment was initiated. For 3 out of 29 children, insufficient data were available. Thirteen out of 29 did not undergo surgical treatment for their liver metastases and were only treated with chemotherapy or/and radiotherapy. There were 7 survivors in this group (3 with chemotherapy alone and 4 with a combination of chemotherapy and radiation). In 3 cases, no further treatment could be initiated due to tumor progression; all 3 patients died. In 26 patients chemotherapy was administered after tumor nephrectomy. Within this group, 16 patients received chemotherapy according to the high-risk protocol (CARBO, ADR, IFO, VP 16) and 10 patients according to the intermediate-risk protocol (ACT-D, VCR, DOX). In 4 children, treatment was followed by autologous stem-cell rescue. Out of these 26 children, 16 patients survived, with 10 patients in the high-risk group and 6 patients in the intermediate risk group. Two survivors in the group had undergone autologous stem-cell rescue. A total of 16 patients received additional radiotherapy with a median dose of 21.75 Gy ranging from 10.5 to 30.6 Gy.
Impact of surgery on hepatic lesions in patients with hepatic metastases of nephroblastoma, n ¼ 45.
Surgery
Group 1
Group 2
Patients (n ¼ 29)
%
Patients survived
Patients (n ¼ 16)
%
Patients survived
Primary Complete Incomplete
11 6 5
37.9 (20.7) (17.2)
7 6 1
6 2 4
37.5 (18.8) (25)
2 2 0
Secondary Complete Incomplete
4 3 1
13.8 (10.3) (3.4)
3 3 0
(2a)
(18.8)
(0)
(2a)
(18.8)
(0)
48.3
7
10
62.5
5
None
14
The bold numbers underline the importance of complete resection, where all patients survived. The italic numbers underline that those patients are included in primary resection and therefore do not count for the total number. a The two patients with secondary incomplete resection had both also primary incomplete resection and are included in the group with primary incomplete resection.
ARTICLE IN PRESS Surgical implications for liver metastases in nephroblastoma
Table 3
Histological subtype and survival in patients with hepatic metastases of nephroblastoma, n ¼ 45.
Histological subtype
Low risk Intermediate risk High risk Unclassified a b
37
Group 1
Group 2
Patients
%
Patients survived
Patients
%
Patients survived
4 18 4a 3
13.8 62.1 13.8 10.3
2 14 1 1
0 11 5b 0
– 68.7 31.3 –
– 6 1 –
Including two patients with diffuse anaplasia. Including three patients with diffuse anaplasia, one patient with clear cell sarcoma.
Radiotherapy was administered locally in four cases, to the abdominal site (i.e. liver) in seven cases and to the lungs in two cases. No documentation of the radiation field was available in nine patients. Histological classification Histology was classified according to SIOP as low risk in 4, intermediate risk in 18 and high risk in 4, with two children having diffuse anaplastic tumors. In 3 children, the tumor could not be classified histologically due to declining treatment or to death prior to initiation of treatment. In 10 cases, nephroblastomatosis was found additionally (see Table 3). Outcome Median follow-up was 64 months (range 3–157 months; Figure 1). At the time of the last follow-up, 17 patients were alive and in complete remission. One patient was lost to follow-up, but was reported to be alive. Eleven patients died in the course, at a median of 13.07 months (0.25–42 months) after initial diagnosis. These included 6 patients who never had surgery for their liver lesions, and 5 patients who had incomplete resections of their metastatic liver lesions. All patients who underwent complete resection of hepatic metastases (n ¼ 9 out of 29) survived. One child survived, despite incomplete surgery, with postoperative chemotherapy and radiation. A total of 7/14 children survived with a non-surgical treatment of the liver metastasis. All seven of these patients showed a complete remission of metastatic disease after preoperative chemotherapy. Nine children died due to progressive disease, whereas one child died due to cardiorespiratory failure and another child due to thrombocytopenia and consecutive cerebral bleeding. The 5-year survival rate was 62.6%. Survival rate after 7 years was 59.4%.
Group II Metastases pattern In 7 patients, metastases were local, contiguous as well as distant, hematogeneous to the liver, while in nine patients, metastases were only distant, hematogeneous to the liver. Surgery In group II, all patients underwent tumor-nephrectomy. Liver metastases were operated in 6/16 children. Complete resection was achieved in 2 patients. In 4 children, surgery of liver lesions was incomplete. Nine patients
did not undergo surgery for their liver lesions. In one patient, records were missing regarding surgical therapy (see Table 2). Chemotherapy and radiation Eleven patients received preoperative chemotherapy according to the protocols of the SIOP 93-01/GPOH and the SIOP 2001/GPOH Wilms’ tumor study. In 15 patients, postoperative chemotherapy was administered. Within this group, 10 patients received chemotherapy according to the high-risk protocol (CARBO, ADR, IFO, VP 16), with 6 survivors. Three patients were treated according to the intermediate-risk protocol (ACT-D, VCR, DOX). In 2 children, treatment was according to protocols of different, nonnephroblastoma tumor studies. Two patients underwent additional autologous stem transplantation. A total of 8 patients received additional radiotherapy with a median dose of 30 Gy (18–40 Gy). Histological classification Histology was classified as intermediate risk in 11 and high risk in 5, with 3 children having diffuse anaplastic tumors and 1 patient with a clear cell sarcoma. In one case, nephroblastomatosis was found additionally (see Table 3). Outcome Median follow-up was 71 months (2–111 months). At the time of the last follow-up 9 of the 16 patients had died due to progressive disease at a median of 52 months (12–108 months) after initial diagnosis. These included 5 out of 10 patients who did not undergo surgical resection of their liver lesions, and 4 out of 6 patients who did. Both patients who had a complete resection of hepatic metastases survived, whereas the 4 children with incomplete surgical resection died. 5 children survived with non-surgical treatment of their liver recurrence. All 9 children died due to progressive disease. The 5-year survival rate was 54.7% for the group with liver recurrence (Figure 1). Nevertheless, 2 children died late in the further course, one without resection of the liver lesions, the other with incomplete resection despite five surgical attempts. Survival in this group decreases to 18.2% after 7 years. Overall results Of all 45 patients with liver metastases, 25 children are alive and in remission, and 20 patients died. The 5-year survival
ARTICLE IN PRESS J. Fuchs et al. 1.00
1.00
0.90
0.90
0.80
0.80
0.70
0.70
0.60
0.60
Survival
Survival
38
0.50 0.40
0.50 0.40
0.30
0.30
0.20
0.20
0.10
0.10
0.00
0
12 24 36 48 60 72 84 96 108 120 132 144 156 168
0.00
12
0
24
Months after Diagnosis
36
48
60
72
84
Months after Diagnosis
n = 29, deaths =11, p = 0.63
n = 16, deaths = 9, p = 0.55
Kaplan-Meier survival curve for patients with primary hepatic metastases in nephroblastoma
Kaplan-Meier survival curve for patients with hepatic metastases as relapse of nephroblastoma
Figure 1 Kaplan–Meier survival curves for patients with hepatic metastases of nephroblastoma, 5-year survival.
0.90
1.00
0.80
0.90
0.70
0.80 0.70
0.60 Survival
Survival
1.00
0.50 0.40 0.30
0.60 0.50 0.40 0.30
0.20
0.20
0.10
0.10
0.00
0
12
24
36
48
60
72
84
96 108 120 132 144 156 168
Months after Diagnosis
0.00
0
12
24
36
48
60
72
84
96 108 120
Months after Diagnosis
n = 45, deaths= 20, p = 0.5931
n = 1365, deaths = 113, p = 0.9179
Kaplan-Meier survival curve for all 45 patients with hepatic metastases in nephroblastoma
Kaplan-Meier survival curve for all 1365 patients in the SIOP 93-01/GPOH and SIOP 2001/GPOH Wilms’ Tumor Study
Figure 2 Kaplan–Meier survival curves for all patients with hepatic metastases of nephroblastoma and for all patients enrolled in the SIOP 93-01/GPOH and SIOP 2001/GPOH Wilms’ Tumor study, 5-year survival.
rate of all 45 patients with liver metastases was 59.3% (Figure 2). However, as mentioned above, survival in group II decreases to 18.2% after 7 years. Although there seems to be a trend that survival in group II is worse than in group I,
differences are statistically not significant (p ¼ 0.140, Log Rank (Mantel–Cox)). Therefore, comparison between patient groups can only be descriptive.
ARTICLE IN PRESS Surgical implications for liver metastases in nephroblastoma
Discussion In NWTS treatment protocols as well as in SIOP treatment protocols, there are only general recommendations for treatment of liver metastases. For metastases located in the liver, radical surgical excision has been successful for tumor control in the SIOP studies 6 and 9 [8]. Children with liver metastases of nephroblastoma still have a worse prognosis than with metastases to other sites [9,10]. Among children with abdominal recurrence of nephroblastoma, the risk of death is 40 greater than in children with a non-abdominal relapse [11]. This represents an important survival factor prognostically and was observed in different international trials [11–13]. Independent of the site of metastases, overall survival in the SIOP9/GPOH-study was 87.5% for children with completely excised metastases and 47.4% for incompletely resected or inoperable metastases. Similar results were achieved in the NWTS 3 study [14,27]. In the group investigated, the overall 5-year survival was 59.3%, which seems to reflect the more serious prognosis in children with hepatic metastases. In contrast to the NWTSG consensus, preoperative chemotherapy is recommended for children with a Wilms’ tumor within the various trials of the SIOP study group [3,15,16]. The main argument against preoperative chemotherapy is its compromising effect on tumor staging and histology. Preoperative chemotherapy may not only cause significant tumor shrinkage, but may facilitate the surgical approach and it offers the advantage of treating metastases up. In patients with primary hepatic metastases, preoperative chemotherapy proved to be effective. In group I, liver metastases disappeared completely in 5 patients out of the 19 treated with preoperative chemotherapy. For relapsed nephroblastoma, there are encouraging results with high dose chemotherapy and hematopoietic stem-cell rescue [17,18]. Also for patients with hepatic metastases, results have been published suggesting effectiveness of treatment [19]. In the group investigated, 2 patients underwent hematopoietic stem-cell rescue, one of which survived. Complete resection of liver lesions had not been achieved in the child who died. Radiotherapy has a role, both in adjuvant therapy of metastatic disease and incomplete resection as well as in palliation, where it has been proven to be effective for management of pain [11]. The NWTSG suggests that radiotherapy has a major impact in the cure of complicated or relapsed Wilms’ tumor. In group II, radiotherapy did not appear to improve the outcome. Out of 8 children who had additional radiotherapy, 6 died. Furthermore, there are still serious concerns regarding potential severe side and late effects as a consequence of radiotherapy [12]. The possibility of long-term effects, including liver fibrosis, cholestasis, veno-occlusive disease (VOD) and lung fibrosis has to be given serious consideration, particularly following radiotherapy to the liver and lung [20–23]. While a combination of abdominal radiotherapy and multi-agent regimens may achieve a high tumor-control rate [13,24], surgical radicalism could lead to a reduction of the postoperative need for radiotherapy. The ‘‘key factor in the management of Wilms’ tumor, however, is the resection of the primary tumor’’ [25,27]. This is clearly also true for
39 metastatic tumor sites. In group I, all patients survived, when a complete resection of liver lesions had been achieved, either primarily or when operated on secondarily, whereas out of 6 children with incomplete surgical resection of hepatic metastases, 5 patients died [1]. In children with a recurrence in the liver, those with a primary complete resection of the liver lesion survived, whereas the patients with an incomplete resection at either first or second attempt all died. For children with a hepatic recurrence of Wilms’ tumor, prognosis after incomplete resection seems to be even worse when compared to those with primary hepatic metastases and decreases to below 20% after 7 years. Early relapse did not have an impact on survival in the patients we investigated. Median time for relapse in the whole group was 15.5 months. Two children with hepatic relapse after 2 and 4 months, respectively, both survived. Due to the small numbers of patients even in this large study, this can only be considered as a trend, and is obviously not of statistical significance. In adult patients with liver metastases from renal malignancies, those who underwent a curative liver resection survived longer than patients with a palliative liver resection [26]. Our data suggest that successful complete surgical resection of the primary tumor and of liver metastases in children with Wilms’ tumor improves survival. For obvious reasons, the question as to why complete resection of liver lesions was not always possible is hard to answer. The aim should therefore be to refer those few children with primary liver metastases or relapsing nephroblastoma to the liver to centers of excellence with adequate surgical experience of pediatric solid tumors. Common protocols of both the NWTSG and the SIOP study group should be anticipated in order to improve treatment options for these children.
Conflict of Interest Statement All authors hereby disclose any financial and personal relationships with other people or organizations that could inappropriately influence (bias) this publication. This excludes employment, consultancies, stock ownership, honoraria, paid expert testimony, patent applications/ registrations, and grants or other funding.
References [1] Weirich A, Ludwig R, Graf N, Abel U, Leuschner I, Vujanic GM, et al. Survival in nephroblastoma treated according to the trial and study SIOP9/GPOH with respect to relapse and morbidity. Annals of Oncology 2004;15:808–20. [2] Graf N, Tournade MF, de Kraker J. The role of preoperative chemotherapy in the management of Wilms’ tumor—the SIOP studies. The Urology Clinics of North America 2000; 27:443–54. [3] Szavay P, Luithle T, Graf N, Furtwangler R, Fuchs J. Primary hepatic metastases in nephroblastoma—a report of the SIOP/ GPOH study. Journal of Pediatric Surgery 2006;41:168–72. [4] SIOP/GPOH (Gesellschaft fu ¨r Pa ¨diatrische Onkologie und Ha ¨matologie), SIOP 2001/GPOH. Therapieoptimierungsstudie
ARTICLE IN PRESS 40
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
J. Fuchs et al. zur Behandlung von Kindern und Jugendlichen mit einem Nephroblastom; 2003. Delemarre JFM, Sandstedt B, Harms D, Boccon-Gibod L, Vujanic GM. The new SIOP (Stockholm) working classification of renal tumours of childhood. Medical and Pediatric Oncology 1996; 26:145–6. Vujanic GM, Harms D, Sandstedt B, Weirich A, de Kraker J, Delemarre JFM. New definitions of focal and diffuse anaplasia in Wilms’ tumor: the International Society of Paediatric Oncology (SIOP) experience. Medical and Pediatric Oncology 1999;32:317–23. Vujanic GM, Sandstedt B, Harms D, Kelsey A, Leuschner I, de Kraker J. Revised International Society of Paediatric Oncology (SIOP) working classification of renal tumors of childhood. Medical and Pediatric Oncology 2002;38: 79–82. Godzinski J, Tournade MF, Weirich A, de Kraker J, Lemerle J, Voute PA, et al. For the SIOP Nephroblastoma Committee: Stage IV nephroblastoma with extrapulmonary metastatic involvement in the SIOP 6 and 9 study. Medical and Pediatric Oncology 1991;19:371. Firoozi F, Kogan BA. Follow-up and management of recurrent Wilms’ tumor. The Urology Clinics of North America 2003;30: 869–79. Varan A, Bu ¨yu ¨kpamukcu N, Caglar M, Yavuz K, Bilgehan Y, Canan A, et al. Prognostic significance of metastatic site at diagnosis in Wilms’ tumor. Journal of Pediatric Hematology Oncology 2005;27:188–91. Paulino AC. Relapsed Wilms’ tumor—is there a role for radiation therapy? American Journal of Clinical Oncology (CCT) 2001;24:408–13. D’Angio GJ. Pre- or post-operative treatment for Wilms’ tumor? Who, what, when, where, how, why—and which. Medical and Pediatric Oncology 2003;41:545–9. Grundy RG, Hutton C, Middleton H, Imeson J, Pritchard J, Kelsey A, et al. United Kingdom Children’s Cancer Study Group: outcome of patients with stage III or inoperable WT treated on the Second United Kingdom WT Protocol (UKWT2)—A United Kingdom Children’s Cancer Study Group (UKCCSG) Study. Pediatric Blood and Cancer 2004;42: 311–9. Green DM, Breslow NE, Evans I, Moksness J, D’Angio GJ. Treatment of children with stage IV favorable histology Wilms’ tumor: a report from the National Wilms’ Tumor Study Group. Medical and Pediatric Oncology 1996;26:147–52. Burger D, Moorman-Voestermans CG, Mildenberger H, Lemerle J, Voute PA, Tournade MF, et al. The advantages of preoperative therapy in Wilms’ tumor. Zeitschrift fur Kinderchirurgie 1985;40:170–5.
[16] Godzinski J, Tournade MF, de Kraker J, Lemerle J, Voute PA, Weirich A, et al. Rarity of surgical complications after postchemotherapy nephrectomy for nephroblastoma. Experience of the International Society of Paediatric Oncology—Trial and Study ‘‘SIOP 9’’. International Society of Paediatric Oncology Nephroblastoma Trial and Study Committee. European Journal of Pediatric Surgery 1998;8:83–6. [17] Pein F, Michon J, Valteau-Couanet D, Quintana E, Frappaz D, Vannier JP, et al. High-dose melphalan, etoposide and carboplatin followed by autologous stem-cell rescue in pediatric high-risk recurrent Wilms’ tumor: a French Society of Pediatric Oncology study. Journal of Clinical Oncology 1998;16:3295–301. [18] Garaventa A, Hartmann O, Bernard JL, Zucker JM, Pardo N, Castel V, et al. Autologous bone marrow transplantation for pediatric Wilms’ tumor: the experience of the European Bone Marrow Transplantation Solid Tumor Registry. Medical and Pediatric Oncology 1994;22:11–4. [19] Campbell AD, Cohn SL, Reynolds M, Seshadri R, Morgan E, Geissler G, et al. Treatment of relapsed Wilms’ tumor with high dose therapy and autologous hematopoietic stem-cell rescue: the experience at Children’s Memorial Hospital. Journal of Clinical Oncology 2004;22:2885–90. [20] Bisogno G, de Kraker J, Weirich A, Masiero L, Ludwig R, Tournade MF, et al. Veno-occlusive disease of the liver in children treated for Wilms’ tumor. Medical and Pediatric Oncology 1997;29:245–51. [21] Davies-Johns T, Chidel M, Macklis RM. The role of radiation therapy in the management of Wilms’ tumor. Seminars in Urologic Oncology 1999;17:46–54. [22] Paulino AC, Wen BC, Brown CK, Tannous R, Mayr NA, Zhen WK, et al. Late effects in children treated with radiation therapy for Wilms’ tumor. International Journal of Radiation Oncology, Biology, Physics 2000;46:1239–46. [23] Trobs RB, Hansel M, Friedrich T. A 23-year experience with malignant renal tumors in infancy and childhood. European Journal of Pediatric Surgery 2001;11:92–8. [24] Green DM. The treatment of stages I–IV favorable histology Wilms’ tumor. Journal of Clinical Oncology 2004;22:1366–72. [25] Grosfeld JL. Risk-based management: current concept of treating malignant solid tumors of childhood. Journal of the American College of Surgeons 1999;189:407–25. [26] Alves A, Adam R, Majno P, Delvart V, Azoulay D, Castaing D, et al. Hepatic resection for metastatic renal tumors: is it worthwhile? Annals of Surgical Oncology 2003;10:705–10. [27] Thomas PRM, Shochat SJ, Norkool P, Beckwith JB, Breslow NE, D’Angio GJ. Prognostic implications of hepatic adhesion, invasion and metastases at diagnosis of Wilms’ tumor. Cancer 1991;68:2486–8.
Available at www.sciencedirect.com
journal homepage: www.elsevier.com/locate/suronc
REVIEW
Surgical implications for liver metastases in nephroblastoma—Data from the SIOP/GPOH study$ Joerg Fuchsa,,1, Philipp Szavaya,1, Tobias Luithlea, Rhoikos Furtwa ¨nglerb, Norbert Grafb a
Department of Pediatric Surgery, Children’s Hospital, University of Tuebingen, Hoppe-Seyler-Strasse 3, 72076 Tuebingen, Germany b Department of Pediatric Hematology and Oncology, University of Homburg/Saar, 66421 Homburg/Saar, Germany Accepted 28 August 2007
KEYWORDS Nephroblastoma; Wilms’tumor; Surgery; Treatment; Metastases; Hepatic; Liver; Recurrence; Survival
Summary Background: In children with Wilms’ tumor, the 5-year overall survival rate is over 90% in the SIOP/GPOH study group. However, a small group of patients have tumor lesions in the liver at the time of initial diagnosis or as a recurrence. This group seems to have a worse prognosis in terms of survival. The treatment and outcome of patients with a hepatic recurrence were analyzed compared to previously published data of patients with primary hepatic metastases. Patients and methods: We reviewed the records of 45 out of 1365 patients enrolled in the SIOP 93-01/GPOH study and the SIOP 2001/GPOH study between April 1, 1994 and September 30, 2004. Median age at diagnosis was 6.49 years (1.37–34.16 years) in 29 patients who were initially presented with hepatic metastases (group I) with 9 males and 20 females. In 16 children who had a recurrence of a nephroblastoma in the liver (group II), median age at diagnosis was 4.62 years (1.84–31.08 years) with 9 males and 7 females.
Abbreviations: ACT-D, dactinomycin; ADR, adriamycin; AVD, ACT-D, VCR, DOX; CARBO, carboplatin; CCE, CARBO, CYCLO, VP 16, G-CSF; CCED, CARBO, CYCLO, VP 16, DOX; CCEV, CARBO, CYCLO, VP 16, VCR; CR, complete remission; CRF, case report form; CT, computer tomography; CYCLO, cyclophosphamide; DOX, doxorubicin; G-CSF, granulocyte stimulating factor; GPOH, German Association of Pediatric Oncology and Hematology; Gy, Gray; ICE, CARBO, VP 16, IFO, G-CSF; IFO, ifosfamide; MRI, magnetic resonance imaging; NWTS, National Wilms’ Tumor Study; NWTSG, National Wilms’ Tumor Study Group; SIOP, International Society of Paediatric Oncology; VCR, vincristine; VOD, veno-occlusive disease; VP 16, etoposide $ Presented in part at the 37th annual conference of The International Society of Paediatric Oncology SIOP 2005, Vancouver, Canada, September 21–24, 2005. Corresponding author. Tel.: +49 7071 298 6621; fax: +49 7071 29 4046. E-mail address: [email protected] (J. Fuchs). 1 Contributed equally in this study. 0960-7404/$ - see front matter & 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.suronc.2007.08.011
ARTICLE IN PRESS 34
J. Fuchs et al. Results: In group I out of 29 patients, 11 died at a median of 13.07 months. Overall survival in group I was 62.58%. In group II, 9 patients died at a median 52 months. Overall survival in group II was 54.7%. Conclusion: This report suggests that when complicated by metastases of the liver, Wilms’ tumor has a less favorable outcome. Chemotherapy and radiotherapy play a definitive role in the treatment of these children. The importance of complete resection of hepatic lesions in both groups should be emphasized. & 2007 Elsevier Ltd. All rights reserved.
Contents Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 Methods and patients . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 General remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 Classification and staging system . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35 Recommendation of the treatment protocols in cases with liver metastases . . . . . . . . . . . . . . . . . . . . . . . . . 35 Principles of chemotherapy and radiation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35 Statistical methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35 Patient data and grouping . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35 Diagnostic evaluation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36 Group I . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36 Metastases pattern . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36 Surgery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36 Chemotherapy and radiation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36 Histological classification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 Outcome . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 Group II. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 Metastases pattern . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 Surgery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 Chemotherapy and radiation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 Histological classification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 Outcome . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 Overall results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 Conflict of Interest Statement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
Introduction Striking improvement in treatment for children with a Wilms’ tumor has been achieved, leading to 5-year overall survival rates above 90% in the SIOP/GPOH study group [1,2]. Despite this fact, several issues remain subject to concern and ongoing discussion: relapsing nephroblastoma, metastatic disease and risk stratification. In this study, we focus on a small group of patients who had tumor lesions in the liver at the time of initial diagnosis or developed a hepatic recurrence of the disease. The purpose of this study is to evaluate the data of the SIOP/ GPOH 93-01 and 2001 study group in order to identify factors in surgical treatment that have an effect on the prognosis of these patients. Data of children with a hepatic recurrence of Wilms’ tumor were compared to previously published data of patients with primary hepatic metastases [3].
Methods and patients In order to define survival and success of treatment of this group of patients, we reviewed the records of 1365 patients
enrolled in the SIOP 93-01/- and the SIOP 2001/Wilms’ tumor study of the German Society of Pediatric Oncology and Hematology between April 1, 1994 and September 30, 2004. These studies conform to the Declaration of Helsinki and were approved by the ethic commissions of all participating centers (leading IRB: Ethikkommission der Landesaeztekammer des Saarlandes, Germany, IRB00004360).
General remarks All children from Germany, Austria and Switzerland suffering from renal tumors were prospectively registered in the Wilms’ tumor studies SIOP 93-01 and 2001 of the German Society of Pediatric Hematology and Oncology (GPOH). For each patient a case report form (CRF) for registration, preoperative chemotherapy, surgical treatment of the primary tumor and/or metastases, postoperative chemotherapy, and radiation was acquired. Pre-treatment CT scans or MRI images were reviewed by reference radiologists of the study (J. Tro ¨ger, Heidelberg, Germany and Unsinn,
ARTICLE IN PRESS Surgical implications for liver metastases in nephroblastoma Innsbruck, Austria). The tumor specimens were analyzed by the local pathologist and then sent for review to the SIOP/ GPOH Pediatric Tumor Registry (D. Harms and I. Leuschner, Kiel, Germany). For regular follow-up, patients received physical examination, imaging diagnostics and laboratory screening up to 5 years after initial diagnosis.
Classification and staging system Histological classification for nephroblastoma was carried out according to the SIOP criteria, dividing tumors into low, intermediate and high-risk groups. Between SIOP 93-01 and SIOP 2001, blastemal predominant tumors were shifted into the high-risk group according to their poor prognosis, whereas focal anaplasia was classified as intermediate risk in SIOP 2001 [5–7]. Staging was based on the post-surgical staging system according to the SIOP protocol [4]. In stage I, the tumor is limited to the kidney and completely removed. The renal capsule has an intact outer surface. In stage II, the tumor extends beyond the kidney, but is completely removed. There may be an extrarenal tumor extension, e.g. penetration of the tumor capsule or a tumor thrombus in the renal vein. The involvement of local lymph nodes is possible (stage II N+ [SIOP 93-01/GPOH], stage III [SIOP 2001/GPOH]). Stage III is a tumor rupture or macroscopic tumor remnant after surgery. Stage IV is metastatic disease (lung, liver, bone, brain) and stage V refers to the phenomenon of bilateral disease.
Recommendation of the treatment protocols in cases with liver metastases All children with primary liver metastases should receive a preoperative chemotherapy consisting of dactinomycin (ACT-D), vincristine (VCR) and doxorubicin (DOX). In cases, where metastases do not disappear, a complete resection of liver metastases is recommended during tumor nephrectomy or 14 days later. Radiation is indicated in cases where liver metastases did not disappear after chemotherapy and could not be resected, or in children with incomplete resection of metastases. Dosage for liver radiation varies from 15 to 30 Gy (single dose 1.5 Gy). All children with tumor recurrence in the liver should initially be treated with chemotherapy followed by a riskadapted approach depending on tumor response and risk factors regarding initial treatment (local stage, histology, time to relapse). Surgery for liver metastasis should be performed as soon as possible after chemotherapy. Only general recommendations for tumor resection exist, without a specific timeinterval during treatment.
Principles of chemotherapy and radiation All children from 6 months to 16 years of age with unilateral nephroblastoma and liver metastases at the time of initial diagnosis (stage IV) receive the same 3 drug preoperative chemotherapy in both studies. Chemotherapy consists of dactinomycin (ACT-D; 15 mg/kg body weight; maximal bolus
35 injection 0.5 mg; day 1, 2, 3, week 1, 3, 5 [SIOP 93-01/ GPOH] or 45 mg/kg body weight as a single injection on 1 day, week 1, 3, 5 [SIOP 2001/GPOH], vincristine (VCR; 1.5 mg/ m2; maximal bolus injection 2.0 mg; day 1 from week 1–6) and adriamycin (ADR; 50 mg/m2 as 4-h-infusion, day 1 from week 1, 5). For small children (infancy or body weight o12 kg) the drug dosage is generally reduced to two-thirds. Postoperative chemotherapy depends on risk stratification after tumor resection. In cases with complete tumor resection and low or intermediate risk, the children receive ACT-D, VCR and DOX (AVD) without radiation. All children with incomplete tumor resection or children in the high-risk group receive etoposide (VP 16), CARBO, DOX, cyclophosphamide (CYCLO) and additional radiotherapy (maximum 30 Gy). In cases of tumor recurrence, the study differentiates between a 3-drug and a 4-drug regimen. AVD, the 3-drug regimen (ACT, VCR, DOX), is administered for 27 weeks to patients with stage I after tumor nephrectomy. The 4-drug regimen, CCEV (CYCLO, CARBO, VP 16, VCR) or CCED (CYCLO, CARBO, VP 16, DOX), depending on whether DOX was administered initially, is administered for 34 weeks to patients with all other stages after initial surgery. The highrisk group (including initial stages III and IV, high malignancy such as diffuse anaplasia, and early recurrence or recurrence in areas of radiation) receives chemotherapy with two courses of ICE (ifosfamide (IFO), carboplatin (CARBO), etoposide (VP 16)) or CCE (CYCLO, CARBO, VP 16), depending on the primary chemotherapy. After this multiagent therapy, high-dose chemotherapy with stem-cell rescue is recommended in case of CR [4].
Statistical methods Standard methods were used for the analysis of the censored and non-censored data, including Kaplan–Meier curves, and Fisher’s exact test for comparing independent proportions. Overall or relapse free survival was assessed with the proportional hazards model. Post-relapse survival was defined from the time of initial relapse until death. All analyses were carried out using SPSS for Windows. Statistical significance was assumed for po0.05.
Patient data and grouping Altogether, 45 out of 1365 patients had liver lesions deriving from a nephroblastoma, either at the time of initial diagnosis or when relapsing.
Diagnostic evaluation Liver lesions were diagnosed with abdominal ultrasonography and CT or MRI. All patients received more than one of the above-mentioned investigations. Forty patients had a unilateral tumor, 4 patients had a bilateral Wilms’ tumor, and one patient had an extrarenal Wilms’ tumor. Twenty-three of the unilateral tumors were left-sided and 17 were right-sided. In addition to hepatic metastases, 28 patients had metastatic lesions in the lungs. Other sites of metastases included bones (n ¼ 4), mediastinum (n ¼ 3), and diaphragm (n ¼ 2). Three patients developed inferior cava vein thrombosis.
ARTICLE IN PRESS 36
J. Fuchs et al.
According to the SIOP-classification, out of 45 patients, 5 had a stage I tumor, 11 with stage II, 13 with stage III, 12 with stage IV and 4 patients with stage V (see Table 1). The patients with primary liver metastases (group I) and those with secondary liver metastases (group II) were analyzed separately. Data of group I have been published previously [3]: Group I: 29 out of 1365 patients (2.12%) had liver metastases at the time of initial diagnosis. The median age at diagnosis for these 29 patients was 6.49 years (range 1.37–34.16 years) with 9 males and 20 females in the group. Group II: 16 out of 1365 patients had a recurrence in the liver at a median of 12.5 months (2–35 months) after initial diagnosis. Median age at initial diagnosis was 4.62 years (1.84–31.08 years) with 9 males and 7 females in the group.
Results Group I Metastases pattern All primary liver metastases were distant, not-contiguous metastases.
Table 1 Tumor-stage after preoperative chemotherapy in patients with primary (group I) and secondary (group II) hepatic metastases of nephroblastoma, n ¼ 45. Stage according to SIOP Stage Stage Stage Stage Stage
I II III IV V
Table 2
Group 1 (n ¼ 29)
Group 2 (n ¼ 16)
Patients
%
Patients
%
0 9 8 10 2
0 31 28 34 7
5 2 5 2 2
31.25 12.5 31.25 12.5 12.5
Surgery In group I, all patients but 2 underwent tumor-nephrectomy. Liver metastases were operated in 11 children at time of nephrectomy. Primary complete resection was achieved in 6 patients, while in 5 children surgery of liver lesions was incomplete. Complete resection of hepatic lesions was achieved in 3 patients in a second attempt. In one patient, a secondary resection still was incomplete. In 14 patients, surgery for liver lesions was not undertaken (see Table 2).
Chemotherapy and radiation Nineteen out of 29 patients received preoperative chemotherapy; 14 according to the study protocols for stage IV disease of the SIOP 93-01/GPOH and SIOP 2001/GPOH. Due to small age or weight, 5 children received standard risk chemotherapy without doxorubicin. Seven out of 29 patients did not receive chemotherapy preoperatively. In these patients, treatment was not according to the guidelines of the treatment protocol. This group comprised the 6 patients who underwent primary nephrectomy and 1 patient with a Smith-Lemli-Opitz-Syndrome, in whom no treatment was initiated. For 3 out of 29 children, insufficient data were available. Thirteen out of 29 did not undergo surgical treatment for their liver metastases and were only treated with chemotherapy or/and radiotherapy. There were 7 survivors in this group (3 with chemotherapy alone and 4 with a combination of chemotherapy and radiation). In 3 cases, no further treatment could be initiated due to tumor progression; all 3 patients died. In 26 patients chemotherapy was administered after tumor nephrectomy. Within this group, 16 patients received chemotherapy according to the high-risk protocol (CARBO, ADR, IFO, VP 16) and 10 patients according to the intermediate-risk protocol (ACT-D, VCR, DOX). In 4 children, treatment was followed by autologous stem-cell rescue. Out of these 26 children, 16 patients survived, with 10 patients in the high-risk group and 6 patients in the intermediate risk group. Two survivors in the group had undergone autologous stem-cell rescue. A total of 16 patients received additional radiotherapy with a median dose of 21.75 Gy ranging from 10.5 to 30.6 Gy.
Impact of surgery on hepatic lesions in patients with hepatic metastases of nephroblastoma, n ¼ 45.
Surgery
Group 1
Group 2
Patients (n ¼ 29)
%
Patients survived
Patients (n ¼ 16)
%
Patients survived
Primary Complete Incomplete
11 6 5
37.9 (20.7) (17.2)
7 6 1
6 2 4
37.5 (18.8) (25)
2 2 0
Secondary Complete Incomplete
4 3 1
13.8 (10.3) (3.4)
3 3 0
(2a)
(18.8)
(0)
(2a)
(18.8)
(0)
48.3
7
10
62.5
5
None
14
The bold numbers underline the importance of complete resection, where all patients survived. The italic numbers underline that those patients are included in primary resection and therefore do not count for the total number. a The two patients with secondary incomplete resection had both also primary incomplete resection and are included in the group with primary incomplete resection.
ARTICLE IN PRESS Surgical implications for liver metastases in nephroblastoma
Table 3
Histological subtype and survival in patients with hepatic metastases of nephroblastoma, n ¼ 45.
Histological subtype
Low risk Intermediate risk High risk Unclassified a b
37
Group 1
Group 2
Patients
%
Patients survived
Patients
%
Patients survived
4 18 4a 3
13.8 62.1 13.8 10.3
2 14 1 1
0 11 5b 0
– 68.7 31.3 –
– 6 1 –
Including two patients with diffuse anaplasia. Including three patients with diffuse anaplasia, one patient with clear cell sarcoma.
Radiotherapy was administered locally in four cases, to the abdominal site (i.e. liver) in seven cases and to the lungs in two cases. No documentation of the radiation field was available in nine patients. Histological classification Histology was classified according to SIOP as low risk in 4, intermediate risk in 18 and high risk in 4, with two children having diffuse anaplastic tumors. In 3 children, the tumor could not be classified histologically due to declining treatment or to death prior to initiation of treatment. In 10 cases, nephroblastomatosis was found additionally (see Table 3). Outcome Median follow-up was 64 months (range 3–157 months; Figure 1). At the time of the last follow-up, 17 patients were alive and in complete remission. One patient was lost to follow-up, but was reported to be alive. Eleven patients died in the course, at a median of 13.07 months (0.25–42 months) after initial diagnosis. These included 6 patients who never had surgery for their liver lesions, and 5 patients who had incomplete resections of their metastatic liver lesions. All patients who underwent complete resection of hepatic metastases (n ¼ 9 out of 29) survived. One child survived, despite incomplete surgery, with postoperative chemotherapy and radiation. A total of 7/14 children survived with a non-surgical treatment of the liver metastasis. All seven of these patients showed a complete remission of metastatic disease after preoperative chemotherapy. Nine children died due to progressive disease, whereas one child died due to cardiorespiratory failure and another child due to thrombocytopenia and consecutive cerebral bleeding. The 5-year survival rate was 62.6%. Survival rate after 7 years was 59.4%.
Group II Metastases pattern In 7 patients, metastases were local, contiguous as well as distant, hematogeneous to the liver, while in nine patients, metastases were only distant, hematogeneous to the liver. Surgery In group II, all patients underwent tumor-nephrectomy. Liver metastases were operated in 6/16 children. Complete resection was achieved in 2 patients. In 4 children, surgery of liver lesions was incomplete. Nine patients
did not undergo surgery for their liver lesions. In one patient, records were missing regarding surgical therapy (see Table 2). Chemotherapy and radiation Eleven patients received preoperative chemotherapy according to the protocols of the SIOP 93-01/GPOH and the SIOP 2001/GPOH Wilms’ tumor study. In 15 patients, postoperative chemotherapy was administered. Within this group, 10 patients received chemotherapy according to the high-risk protocol (CARBO, ADR, IFO, VP 16), with 6 survivors. Three patients were treated according to the intermediate-risk protocol (ACT-D, VCR, DOX). In 2 children, treatment was according to protocols of different, nonnephroblastoma tumor studies. Two patients underwent additional autologous stem transplantation. A total of 8 patients received additional radiotherapy with a median dose of 30 Gy (18–40 Gy). Histological classification Histology was classified as intermediate risk in 11 and high risk in 5, with 3 children having diffuse anaplastic tumors and 1 patient with a clear cell sarcoma. In one case, nephroblastomatosis was found additionally (see Table 3). Outcome Median follow-up was 71 months (2–111 months). At the time of the last follow-up 9 of the 16 patients had died due to progressive disease at a median of 52 months (12–108 months) after initial diagnosis. These included 5 out of 10 patients who did not undergo surgical resection of their liver lesions, and 4 out of 6 patients who did. Both patients who had a complete resection of hepatic metastases survived, whereas the 4 children with incomplete surgical resection died. 5 children survived with non-surgical treatment of their liver recurrence. All 9 children died due to progressive disease. The 5-year survival rate was 54.7% for the group with liver recurrence (Figure 1). Nevertheless, 2 children died late in the further course, one without resection of the liver lesions, the other with incomplete resection despite five surgical attempts. Survival in this group decreases to 18.2% after 7 years. Overall results Of all 45 patients with liver metastases, 25 children are alive and in remission, and 20 patients died. The 5-year survival
ARTICLE IN PRESS J. Fuchs et al. 1.00
1.00
0.90
0.90
0.80
0.80
0.70
0.70
0.60
0.60
Survival
Survival
38
0.50 0.40
0.50 0.40
0.30
0.30
0.20
0.20
0.10
0.10
0.00
0
12 24 36 48 60 72 84 96 108 120 132 144 156 168
0.00
12
0
24
Months after Diagnosis
36
48
60
72
84
Months after Diagnosis
n = 29, deaths =11, p = 0.63
n = 16, deaths = 9, p = 0.55
Kaplan-Meier survival curve for patients with primary hepatic metastases in nephroblastoma
Kaplan-Meier survival curve for patients with hepatic metastases as relapse of nephroblastoma
Figure 1 Kaplan–Meier survival curves for patients with hepatic metastases of nephroblastoma, 5-year survival.
0.90
1.00
0.80
0.90
0.70
0.80 0.70
0.60 Survival
Survival
1.00
0.50 0.40 0.30
0.60 0.50 0.40 0.30
0.20
0.20
0.10
0.10
0.00
0
12
24
36
48
60
72
84
96 108 120 132 144 156 168
Months after Diagnosis
0.00
0
12
24
36
48
60
72
84
96 108 120
Months after Diagnosis
n = 45, deaths= 20, p = 0.5931
n = 1365, deaths = 113, p = 0.9179
Kaplan-Meier survival curve for all 45 patients with hepatic metastases in nephroblastoma
Kaplan-Meier survival curve for all 1365 patients in the SIOP 93-01/GPOH and SIOP 2001/GPOH Wilms’ Tumor Study
Figure 2 Kaplan–Meier survival curves for all patients with hepatic metastases of nephroblastoma and for all patients enrolled in the SIOP 93-01/GPOH and SIOP 2001/GPOH Wilms’ Tumor study, 5-year survival.
rate of all 45 patients with liver metastases was 59.3% (Figure 2). However, as mentioned above, survival in group II decreases to 18.2% after 7 years. Although there seems to be a trend that survival in group II is worse than in group I,
differences are statistically not significant (p ¼ 0.140, Log Rank (Mantel–Cox)). Therefore, comparison between patient groups can only be descriptive.
ARTICLE IN PRESS Surgical implications for liver metastases in nephroblastoma
Discussion In NWTS treatment protocols as well as in SIOP treatment protocols, there are only general recommendations for treatment of liver metastases. For metastases located in the liver, radical surgical excision has been successful for tumor control in the SIOP studies 6 and 9 [8]. Children with liver metastases of nephroblastoma still have a worse prognosis than with metastases to other sites [9,10]. Among children with abdominal recurrence of nephroblastoma, the risk of death is 40 greater than in children with a non-abdominal relapse [11]. This represents an important survival factor prognostically and was observed in different international trials [11–13]. Independent of the site of metastases, overall survival in the SIOP9/GPOH-study was 87.5% for children with completely excised metastases and 47.4% for incompletely resected or inoperable metastases. Similar results were achieved in the NWTS 3 study [14,27]. In the group investigated, the overall 5-year survival was 59.3%, which seems to reflect the more serious prognosis in children with hepatic metastases. In contrast to the NWTSG consensus, preoperative chemotherapy is recommended for children with a Wilms’ tumor within the various trials of the SIOP study group [3,15,16]. The main argument against preoperative chemotherapy is its compromising effect on tumor staging and histology. Preoperative chemotherapy may not only cause significant tumor shrinkage, but may facilitate the surgical approach and it offers the advantage of treating metastases up. In patients with primary hepatic metastases, preoperative chemotherapy proved to be effective. In group I, liver metastases disappeared completely in 5 patients out of the 19 treated with preoperative chemotherapy. For relapsed nephroblastoma, there are encouraging results with high dose chemotherapy and hematopoietic stem-cell rescue [17,18]. Also for patients with hepatic metastases, results have been published suggesting effectiveness of treatment [19]. In the group investigated, 2 patients underwent hematopoietic stem-cell rescue, one of which survived. Complete resection of liver lesions had not been achieved in the child who died. Radiotherapy has a role, both in adjuvant therapy of metastatic disease and incomplete resection as well as in palliation, where it has been proven to be effective for management of pain [11]. The NWTSG suggests that radiotherapy has a major impact in the cure of complicated or relapsed Wilms’ tumor. In group II, radiotherapy did not appear to improve the outcome. Out of 8 children who had additional radiotherapy, 6 died. Furthermore, there are still serious concerns regarding potential severe side and late effects as a consequence of radiotherapy [12]. The possibility of long-term effects, including liver fibrosis, cholestasis, veno-occlusive disease (VOD) and lung fibrosis has to be given serious consideration, particularly following radiotherapy to the liver and lung [20–23]. While a combination of abdominal radiotherapy and multi-agent regimens may achieve a high tumor-control rate [13,24], surgical radicalism could lead to a reduction of the postoperative need for radiotherapy. The ‘‘key factor in the management of Wilms’ tumor, however, is the resection of the primary tumor’’ [25,27]. This is clearly also true for
39 metastatic tumor sites. In group I, all patients survived, when a complete resection of liver lesions had been achieved, either primarily or when operated on secondarily, whereas out of 6 children with incomplete surgical resection of hepatic metastases, 5 patients died [1]. In children with a recurrence in the liver, those with a primary complete resection of the liver lesion survived, whereas the patients with an incomplete resection at either first or second attempt all died. For children with a hepatic recurrence of Wilms’ tumor, prognosis after incomplete resection seems to be even worse when compared to those with primary hepatic metastases and decreases to below 20% after 7 years. Early relapse did not have an impact on survival in the patients we investigated. Median time for relapse in the whole group was 15.5 months. Two children with hepatic relapse after 2 and 4 months, respectively, both survived. Due to the small numbers of patients even in this large study, this can only be considered as a trend, and is obviously not of statistical significance. In adult patients with liver metastases from renal malignancies, those who underwent a curative liver resection survived longer than patients with a palliative liver resection [26]. Our data suggest that successful complete surgical resection of the primary tumor and of liver metastases in children with Wilms’ tumor improves survival. For obvious reasons, the question as to why complete resection of liver lesions was not always possible is hard to answer. The aim should therefore be to refer those few children with primary liver metastases or relapsing nephroblastoma to the liver to centers of excellence with adequate surgical experience of pediatric solid tumors. Common protocols of both the NWTSG and the SIOP study group should be anticipated in order to improve treatment options for these children.
Conflict of Interest Statement All authors hereby disclose any financial and personal relationships with other people or organizations that could inappropriately influence (bias) this publication. This excludes employment, consultancies, stock ownership, honoraria, paid expert testimony, patent applications/ registrations, and grants or other funding.
References [1] Weirich A, Ludwig R, Graf N, Abel U, Leuschner I, Vujanic GM, et al. Survival in nephroblastoma treated according to the trial and study SIOP9/GPOH with respect to relapse and morbidity. Annals of Oncology 2004;15:808–20. [2] Graf N, Tournade MF, de Kraker J. The role of preoperative chemotherapy in the management of Wilms’ tumor—the SIOP studies. The Urology Clinics of North America 2000; 27:443–54. [3] Szavay P, Luithle T, Graf N, Furtwangler R, Fuchs J. Primary hepatic metastases in nephroblastoma—a report of the SIOP/ GPOH study. Journal of Pediatric Surgery 2006;41:168–72. [4] SIOP/GPOH (Gesellschaft fu ¨r Pa ¨diatrische Onkologie und Ha ¨matologie), SIOP 2001/GPOH. Therapieoptimierungsstudie
ARTICLE IN PRESS 40
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
J. Fuchs et al. zur Behandlung von Kindern und Jugendlichen mit einem Nephroblastom; 2003. Delemarre JFM, Sandstedt B, Harms D, Boccon-Gibod L, Vujanic GM. The new SIOP (Stockholm) working classification of renal tumours of childhood. Medical and Pediatric Oncology 1996; 26:145–6. Vujanic GM, Harms D, Sandstedt B, Weirich A, de Kraker J, Delemarre JFM. New definitions of focal and diffuse anaplasia in Wilms’ tumor: the International Society of Paediatric Oncology (SIOP) experience. Medical and Pediatric Oncology 1999;32:317–23. Vujanic GM, Sandstedt B, Harms D, Kelsey A, Leuschner I, de Kraker J. Revised International Society of Paediatric Oncology (SIOP) working classification of renal tumors of childhood. Medical and Pediatric Oncology 2002;38: 79–82. Godzinski J, Tournade MF, Weirich A, de Kraker J, Lemerle J, Voute PA, et al. For the SIOP Nephroblastoma Committee: Stage IV nephroblastoma with extrapulmonary metastatic involvement in the SIOP 6 and 9 study. Medical and Pediatric Oncology 1991;19:371. Firoozi F, Kogan BA. Follow-up and management of recurrent Wilms’ tumor. The Urology Clinics of North America 2003;30: 869–79. Varan A, Bu ¨yu ¨kpamukcu N, Caglar M, Yavuz K, Bilgehan Y, Canan A, et al. Prognostic significance of metastatic site at diagnosis in Wilms’ tumor. Journal of Pediatric Hematology Oncology 2005;27:188–91. Paulino AC. Relapsed Wilms’ tumor—is there a role for radiation therapy? American Journal of Clinical Oncology (CCT) 2001;24:408–13. D’Angio GJ. Pre- or post-operative treatment for Wilms’ tumor? Who, what, when, where, how, why—and which. Medical and Pediatric Oncology 2003;41:545–9. Grundy RG, Hutton C, Middleton H, Imeson J, Pritchard J, Kelsey A, et al. United Kingdom Children’s Cancer Study Group: outcome of patients with stage III or inoperable WT treated on the Second United Kingdom WT Protocol (UKWT2)—A United Kingdom Children’s Cancer Study Group (UKCCSG) Study. Pediatric Blood and Cancer 2004;42: 311–9. Green DM, Breslow NE, Evans I, Moksness J, D’Angio GJ. Treatment of children with stage IV favorable histology Wilms’ tumor: a report from the National Wilms’ Tumor Study Group. Medical and Pediatric Oncology 1996;26:147–52. Burger D, Moorman-Voestermans CG, Mildenberger H, Lemerle J, Voute PA, Tournade MF, et al. The advantages of preoperative therapy in Wilms’ tumor. Zeitschrift fur Kinderchirurgie 1985;40:170–5.
[16] Godzinski J, Tournade MF, de Kraker J, Lemerle J, Voute PA, Weirich A, et al. Rarity of surgical complications after postchemotherapy nephrectomy for nephroblastoma. Experience of the International Society of Paediatric Oncology—Trial and Study ‘‘SIOP 9’’. International Society of Paediatric Oncology Nephroblastoma Trial and Study Committee. European Journal of Pediatric Surgery 1998;8:83–6. [17] Pein F, Michon J, Valteau-Couanet D, Quintana E, Frappaz D, Vannier JP, et al. High-dose melphalan, etoposide and carboplatin followed by autologous stem-cell rescue in pediatric high-risk recurrent Wilms’ tumor: a French Society of Pediatric Oncology study. Journal of Clinical Oncology 1998;16:3295–301. [18] Garaventa A, Hartmann O, Bernard JL, Zucker JM, Pardo N, Castel V, et al. Autologous bone marrow transplantation for pediatric Wilms’ tumor: the experience of the European Bone Marrow Transplantation Solid Tumor Registry. Medical and Pediatric Oncology 1994;22:11–4. [19] Campbell AD, Cohn SL, Reynolds M, Seshadri R, Morgan E, Geissler G, et al. Treatment of relapsed Wilms’ tumor with high dose therapy and autologous hematopoietic stem-cell rescue: the experience at Children’s Memorial Hospital. Journal of Clinical Oncology 2004;22:2885–90. [20] Bisogno G, de Kraker J, Weirich A, Masiero L, Ludwig R, Tournade MF, et al. Veno-occlusive disease of the liver in children treated for Wilms’ tumor. Medical and Pediatric Oncology 1997;29:245–51. [21] Davies-Johns T, Chidel M, Macklis RM. The role of radiation therapy in the management of Wilms’ tumor. Seminars in Urologic Oncology 1999;17:46–54. [22] Paulino AC, Wen BC, Brown CK, Tannous R, Mayr NA, Zhen WK, et al. Late effects in children treated with radiation therapy for Wilms’ tumor. International Journal of Radiation Oncology, Biology, Physics 2000;46:1239–46. [23] Trobs RB, Hansel M, Friedrich T. A 23-year experience with malignant renal tumors in infancy and childhood. European Journal of Pediatric Surgery 2001;11:92–8. [24] Green DM. The treatment of stages I–IV favorable histology Wilms’ tumor. Journal of Clinical Oncology 2004;22:1366–72. [25] Grosfeld JL. Risk-based management: current concept of treating malignant solid tumors of childhood. Journal of the American College of Surgeons 1999;189:407–25. [26] Alves A, Adam R, Majno P, Delvart V, Azoulay D, Castaing D, et al. Hepatic resection for metastatic renal tumors: is it worthwhile? Annals of Surgical Oncology 2003;10:705–10. [27] Thomas PRM, Shochat SJ, Norkool P, Beckwith JB, Breslow NE, D’Angio GJ. Prognostic implications of hepatic adhesion, invasion and metastases at diagnosis of Wilms’ tumor. Cancer 1991;68:2486–8.