- Email: [email protected]
Surgical resection of pulmonary metastatic lesions in children with hepatoblastoma
Journal of Pediatric Surgery (2007) 42, 2050–2056
www.elsevier.com/locate/jpedsurg
Surgical resection of pulmonary metastatic lesions in children with hepatoblastoma Rebecka L. Meyers a,⁎, Howard M. Katzenstein b , Mark Krailo c , Eugene D. McGahren III d , Marcio H. Malogolowkin e a
Division of Pediatric Surgery, Primary Children's Medical Center, University of Utah, Salt Lake City, UT 84113, USA Division of Pediatric Hematology and Oncology, Emory Children's Center, Atlanta, GA, USA c Cure Search, Children’s Oncology Group, Arcadia, CA, USA d Division of Pediatric Surgery, University of Virginia, Charlottesville, VA, USA e Division of Pediatric Hematology and Oncology, Children's Hospital Los Angeles, Los Angeles, CA b
Received 6 August 2007; accepted 8 August 2007
Index words: Hepatoblastoma; Pulmonary metastasis; Lung metastasis; Metastasectomy; Thoracotomy
Abstract Background: Although the most common site of metastasis for hepatoblastoma is the lung, the role of thoracotomy and surgical resection of pulmonary metastasis remains uncertain. In this study, we aimed to determine the long-term outcome of patients with hepatoblastoma who have a surgical resection of pulmonary metastatic lesions. Methods: Patients with hepatoblastoma in the Children's Oncology Group INT-0098 were reviewed. Detailed reports enabling comprehensive review were available for 175 of 181 eligible patients. Prognostic factors (histology, tumor margin, surgical complications, α-fetoprotein) were also reviewed. Results: Thoracotomy for initial pulmonary metastasis: 38 patients presented with pulmonary metastasis, Children's Oncology Group stage IV. Nine of these 38 underwent thoracotomy and pulmonary metastectomy either before (2), simultaneous (5), or after (2) resection of their primary liver tumor. Eight of these 9 patients with metastectomy were long-term survivors. Thoracotomy for tumor relapse: 20 patients who had previously achieved complete tumor clearance experienced subsequent pulmonary relapse of their tumor (11 stage I or III, 9 stage IV). All 20 patients with pulmonary relapse had salvage chemotherapy; 13 also had thoracotomy and pulmonary metastectomy (8) or thoracotomy and biopsy (5). Only 4 of 13 were long-term survivors: 2 were stage I and 2 were stage IV. Conclusions: As pediatric surgeons, we varied tremendously in our timing and surgical approach to the management of pulmonary metastasis in hepatoblastoma. This large multicenter review suggests that thoracotomy should be used cautiously in the management of pulmonary relapse and perhaps more aggressively in the management of metastases present at diagnosis that persist after neoadjuvant chemotherapy. © 2007 Published by Elsevier Inc.
Presented at the 40th annual meeting of the Pacific Association of Pediatric Surgeons, Queenstown, New Zealand, April 15-19, 2007. ⁎ Corresponding author. Tel.: +1 801 588 3350; fax: +1 801 588 3364. E-mail address: [email protected] (R.L. Meyers). 0022-3468/$ – see front matter © 2007 Published by Elsevier Inc. doi:10.1016/j.jpedsurg.2007.08.030
Surgical resection of pulmonary metastatic Although the most common site of metastasis in children with hepatoblastoma is the lung, very little has been published to document the long-term outcome of pulmonary metastasectomy in these children. We do know that unilateral thoracotomy, bilateral thoracotomy, and median sternotomy are well tolerated in children with cancer [1]. We also know that children with lung metastases at diagnosis who are rendered free of all gross disease by some combination of chemotherapy, resection of the primary tumor, as well as resection of pulmonary metastases, may experience cure or long-term survival [2,3]. In International Society of Pediatric Oncology Liver Tumor Study Group (SIOPEL) multicenter studies, the results of metastectomy have been mixed. In SIOPEL-1, all 4 of the patients in whom a metastasectomy was performed survived without residual disease [2,4]. Results were not as promising in SIOPEL-2 where only 3 of the 8 patients with metastectomy were long-term survivors [5]. In these SIOPEL studies all children received neoadjuvant chemotherapy. In American studies, by contrast, not all patients have received neoadjuvant chemotherapy. The potential survival difference for patients who achieve a clinical remission of pulmonary metastases as a result of surgical resection and for those who achieve clinical remission in response to neoadjuvant chemotherapy remains unknown and needs to be defined. Moreover, no study has critically compared the potential difference in outcomes for patients who undergo metastasectomy for pulmonary lesions present at diagnosis with the outcomes for patients who undergo metastasectomy in the setting pulmonary relapse after having been in remission. We hope that review of these questions might help guide future therapy and protocol design with respect to the potential role and optimal timing of lung resection in children with metastatic hepatoblastoma.
2051 margins; Stage II, gross total resection with microscopic residual disease at the margins of disease; Stage III, biopsy only, subtotal resection, or gross total resection with tumor spill/preoperative rupture; Stage IV, metastatic disease at diagnosis with either complete or incomplete resection. Histologic diagnosis and Children's Cancer Group/Pediatric Oncology Group staging were confirmed by central review of representative tissue slides along with institutional pathology and operative reports. Surgical report forms, radiographic imaging reports, operative reports, and pathology reports were available for 175 of 181 eligible patients. Data for patients who were referred for liver transplantation were considered “off study” and were not routinely available in the research archive. Comparative analysis of the PRETEXT group was not a part of the original study, but was completed retrospectively. Data on patients' PRETEXT group enable comparison of this cohort of patients with those of the European and international studies where PRETEXT grouping is one of the primary tools for risk stratification.
1.1. Statistical analysis Comparative statistical analysis of patient survival by PRETEXT group and by COG stage was by the log-rank test. The remainder of this review is primarily descriptive and focused on the subgroup of patients whose research profile documented a thoracotomy at some point in their treatment. Because of the absence of a uniform surgical approach to the use of thoracotomy, and because of confounding variables including variation in the timing of thoracotomy, and variation in the type and amount of salvage/nonprotocol chemotherapy used, statistical analysis of thoracotomy as an independent variable was not felt to be prudent by the COG statistician.
1. Methods Patients with hepatoblastoma (N = 181) in the intergroup hepatoblastoma study INT-0098 performed by the antecedent groups of the current Children's Oncology Group (COG)—Children's Cancer Group/Pediatric Oncology Group—were retrospectively reviewed by RLM and MHM. The details regarding eligibility, treatment, evaluation, and follow-up for INT-0098 have been previously published [6]. The study protocols were approved by the institutional review boards of all participating institutions. Patients were randomized to receive one of 2 different chemotherapy regimens, cisplatin/5fluorouracil/vincristine (C5V) vs cisplatin/doxorubicin (CD). Treatment outcome was not significantly different between the 2 different chemotherapy arms of the study [6]. Stage of disease was determined by surgical and histologic criteria at the initial surgical intervention before the initiation of chemotherapy as follows: Stage I, complete gross resection with clear
2. Results 2.1. Thoracotomy for initial pulmonary metastasis Thirty-eight patients presented with pulmonary metastasis, COG stage IV. At varying times in their treatment, 9 of these 38 children underwent thoracotomy and pulmonary metastasectomy. Two of the 9 had lung resection before resection of the primary. In 5 of the 9 the thoracotomy and hepatectomy were performed simultaneously. In 2 of the 9 the lung metastasis were resected after complete resection of their primary liver tumor. Eight of these 9 patients with metastectomy were long-term survivors. Details of these 9 patients are shown in Table 1. None of these patients had additional poor prognostic indicators such as α-fetoprotein (AFP) of less than 100 at diagnosis or small cell undifferentiated histology.
2052
R.L. Meyers et al.
Table 1
Metastasectomy in hepatoblastoma lung metastases present at diagnosis (COG stage IV)
Patient PRETEXT Histology/ AFP at DX
Complication or additional surgery
Time of primary surgery
1
2
2
2
3
3
Epithelial/ mesenchymal
4
3
Embryonal/fetal
5
3
Embryonal/fetal
6
3
Embryonal/fetal
7
3
Macrotrabecular
8
4
Macrotrabecular
9
4
Epithelial/ mesenchymal
Left lobe at diagnosis Right lobe at diagnosis Right Colectomy dysplastic polyp at trisegmentectomy after neoadjuvant 10 y follow-up Left lobe after neoadjuvant chemotherapy Salivary Right pleomorphic trisegmentectomy adenoma at 9 y after neoadjuvant follow-up chemotherapy Right trisegmentectomy after neoadjuvant chemotherapy Right trisegmentectomy after neoadjuvant chemotherapy Right trisegmentectomy after neoadjuvant chemotherapy Postoperative biliary Right fistula, bile trisegmentectomy peritonitis, after neoadjuvant reoperative chemotherapy hepaticojejunostomy
Embryonal/fetal LOA, bowel obstruction Embryonal/fetal
Metastasectomy a before/simultaneous/ or after resection of primary
No. of cycles of Follow-up neoadjuvant/adjuvant/ nonprotocol chemotherapy b
Simultaneous/right wedge After/right wedge Relapse/biopsy only Before/bilateral thoracotomy
Alive 9 y
Alive 9 y
0/8/+ Lung relapse 1 y 6/2/0
Alive 12 y
4/2/+
Alive 7 y
Lung relapse 2 y 6/2/0
Alive 10 y
Simultaneous/right middle lobectomy
4/3/0
Alive 10 y
Simultaneous/right middle lobectomy
7/4/+
Alive 9 y
After/right lower and middle lobectomy
4/?/+
Alive 6 y
Simultaneous/right wedge ×4 Relapse/left wedge
5/?/+
Death tumor relapse
Simultaneous/right lung Relapse/left wedge Before/median sternotomy
Lung relapse 2 y
Alive 11 y
LOA indicates lysis of adhesions. a This table includes only stage IV patients with metastasectomy, meaning the lung resection was performed with the goal of tumor clearance, not simply as a biopsy. Patients whose only thoracotomy was for biopsy only are not included. b “Nonprotocol” chemotherapy is designated with a “+” in the table. Alterations in protocol chemotherapy were given to some of these patients for persistent lung metastasis, concern for stagnant or rising AFP levels, and/or for pulmonary relapse.
2.2. Thoracotomy for tumor relapse Twenty patients who had previously achieved complete tumor clearance experienced subsequent pulmonary relapse of their tumor (2 stage I, 9 stage III, 9 stage IV). All 20 patients with pulmonary relapse had salvage chemotherapy. Twelve of the patients with pulmonary relapse also had tumor relapse at other sites including liver, mediastinum, or bone. Thirteen of the patients with pulmonary tumor relapse underwent at least one (sometimes more) thoracotomy. In only 8 of the patients did the thoracotomy achieve complete eradication of all radiographically identifiable relapse disease; 5 of the 13 lung resection patients had persistent tumor present in either lung, liver, or bone.
Details of the select group of 8 patients whose thoracotomy achieved complete eradication of all radiographically evident relapse tumor are shown in Table 2A; these are the patients who had thoracotomy with complete metastectomy in an attempt to surgically remove all gross or radiographic evidence of relapse tumor. Only 3 of these 8 children were long-term survivors. Table 2B shows the children who had thoracotomy for pulmonary relapse with postoperative persistent tumor in either lung, liver, or bone. Only 1 of these 5 children was a long-term survivor. Overall, only 4 of 13 children with pulmonary relapse who had thoracotomy were long-term survivors; 2 were initially stage I, and 2 were initially stage IV. Surgical complications seemed unusually predominant in the group of patients with
Surgical resection of pulmonary metastatic Table 2A
2053
Thoracotomy with definitive metastasectomy of pulmonary relapse
Patient Stage/ Histology/ PRETEXT AFP at DX
Complication or Time of primary additional surgery surgery (protocol chemotherapy)
1
I/3
Embryonal/fetal N 2000 mL EBL
2
I/2
Embryonal/ macrotrabecular
3
II/2
Small cell undifferentiated
4
III/3
Mixed/ mesenchymal
5
III/3
Macrotrabecular
6
IV/3
Embryonal/fetal
7
IV/4
Epithelial/ mesenchymal
8
IV/3
Macrotrabecular
Right trisegmentectomy at diagnosis (C5V) Left lobe at diagnosis (C5V)
Survival No. of cycles of Relapse neoadjuvant/adjuvant/ complete metastasectomy nonprotocol chemotherapy ⁎⁎
Relapse/right wedge 1st relapse/left multiple wedge 2nd relapse/left lower lobectomy Preoperative Debulk, control rupture Relapse/ tumor rupture at diagnosis (CD) resected mediastinal mass and wedge LUL Left lobe after 1st relapse/right neoadjuvant wedge chemotherapy (CD) 2nd relapse/ abdominal wall 3rd relapse/right wedge 4th relapse/ mediastinal 5th relapse mediastinal 1st relapse/liver Bile fistula, Left lobe after resected + margin neoadjuvant 2nd relapse/ chemotherapy (CD) wedge lung Left lobe after Metastasectomy neoadjuvant at Dx/right chemotherapy (CD) wedge Relapse/left wedge Right trisegmentectomy Metastasectomy Postoperative at Dx/right after neoadjuvant biliary fistula, wedge ×4 biliary peritonitis chemotherapy (C5V) Relapse/left wedge + Margin, resected Right trisegmentectomy Relapse/resect at diagnosis (CD) mediastinal met lung and postoperative mediastinal sepsis, CHF mass
0/4/0
Alive 8 y
0/4/+
Alive 9 y
0/6/0
Death 16 mo
4/2/+
Death 6 y
0/6/+
Death 20 mo
4/2/+
Alive 7 years
5/?/+
Death ? years
0/1
Death 2 mo
C5V indicates cisplatin, 5fu, vincristine; CD, cisplatin doxyrubicin; EBL, estimated blood loss; LUL, left upper lobe. ⁎⁎ Nonprotocol chemotherapy is designated as a “+” in the table.
pulmonary relapse, although statistical analysis in a separate analysis of these patients [7] did not show a correlation between surgical complication and overall survival for the group as a whole. To better compare this cohort of patients with those in contemporary European (SIOPEL or German Cooperative) or Japanese (Japanese Study Group for Pediatric Liver Tumor [JPLT]) trials, overall survival and stratification by PRETEXT vs COG stage are shown in Table 3. In INT-0098,
there was a preponderance of upfront resection of the liver primary tumor for patients retrospectively classified as PRETEXT 1 and 2, and neoadjuvant chemotherapy for patients retrospectively classified as PRETEXT 3 and 4, but there were frequent exceptions to this broad theme. There were 61 stage I and II patients resected at diagnosis. Of these 61 patients, 16 (26%) were PRETEXT 1, 36 (59%) were PRETEXT 2, and 9 (15%) were PRETEXT 3. Two of 79 stage III patients had gross residual tumor after
2054
R.L. Meyers et al.
Table 2B
Thoracotomy with biopsy of pulmonary relapse
Patient Stage/ Histology/ PRETEXT AFP at DX
Complication or additional surgery
9
III/3
Bile leak Fetal/ macrotrabecular
10
III/4
Embryonal/ fetal
11
III/4
12
IV/2
Embryonal/ fetal Embryonal/ fetal
13
IV/3
IVC injury N2000 EBL HACE local relapse
Relapse biopsy
Right trisegmentectomy after only 1 cycle neoadjuvant chemotherapy (C5V) Right trisegmentectomy after neoadjuvant chemotherapy (CD) ? Operation report incomplete (C5V) Right lobe at diagnosis (CD)
Relapse/Bx 1/2/+ bone, liver, and lung
Death 2y
Relapse/Bx lung
Death 31 mo
Right + Margin, injury trisegmentectomy at LHD hepaticojejunostomy, diagnosis (CD) drain subphrenic abscess
Embryonal/ fetal
Survival No. of cycles of neoadjuvant/adjuvant/ nonprotocol chemotherapy ⁎⁎
Time of primary surgery (protocol chemotherapy)
4/2/0
Relapse/Bx 2/na/+ lung Metastasectomy 0/8/+ at Dx/right wedge Relapse/Bx lung 0/5/+ Relapse/Bx lung
Death 19 mo Alive 12 y
Death 2y
EBL, estimated blood loss; LUL, left upper lobe; IVC, inferior vena cava; HACE, hepatic artery chemoembolization; LHD, left hepatic duct; C5V, cisplatin, 5fu, vincristine; CD, cisplatin doxyrubicin. ⁎⁎ Nonprotocol chemotherapy is designated as a “+” in the table.
attempted surgical resection at diagnosis, and the remaining 77 patients had an initial biopsy followed by neoadjuvant chemotherapy.
3. Discussion Pulmonary surgery is frequently used for the treatment of metastasis in children. However, because hepatoblastoma is a relatively rare tumor, the use of metastasectomy in this tumor remains poorly characterized. In recent reviews of pulmonary metastasectomy for various tumor types in Table 3
children, Karnak et al [8] report that hepatoblastoma was the primary tumor type in only 1 of 18 patients; Abel et al [1] report hepatoblastoma as the primary tumor in 2 of 20 of their patients with metastasectomy. In neither of these reports is it clear whether or not the metastasectomy was performed for a lesion present at diagnosis and persistent after chemotherapy vs a de novo lesion that developed after therapy had been initiated (pulmonary relapse). Fifteen years ago it was noted by Feusner et al [3] that extended disease-free survival is possible for children with initial stage I hepatoblastoma who experienced a pulmonary relapse if the relapse was isolated to the lung and an aggressive surgical approach with intent to cure was used. In
Overall survival by PRETEXT group vs. COG stage
PRETEXT 1 PRETEXT 2 PRETEXT 3 PRETEXT 4 Overall survival
Stage I (pure fetal)
Stage I (unfavorable histology)
Stage II
Stage III
Stage IV
Overall survival
3 4 2 9/9 (100%)
12 26
1 6 7/7 (100%)
1 13 44 21 54/79 (68%)
2 5 24 7 16/38 (42%)
17/19 (89%) 46/54 (85%) 54/77 (70%) 9/28 (32%) ⁎P b .0001 predictive value for both PRETEXT and COG
42/45 (93%)
⁎ There were a total of 191 patients enrolled in INT0098. Of 191 patients, 181 met eligibility criteria. This review includes 178 patients in whom the radiographs and operative reports were detailed enough to determine PRETEXT group at diagnosis. The log-rank test shows both systems to be significantly related to risk of death predictive (P b .0001) of overall survival.
Surgical resection of pulmonary metastatic that analysis, 6 of 33 initial stage 1 patients experienced a tumor relapse confined to the lung, and 3 of the 6 patients were long-term survivors after a pulmonary metastasectomy. More recently, JPLT report in 8 recurrent lung tumors, surgical resection was performed completely in 6 cases with unilateral pulmonary relapse [9]. Five of the 6 JPLT patients were alive and well more than a year after recurrence, but the remaining patient experienced a second relapse in the lung and died of the disease, which proved resistant to various chemotherapeutic agents. Although in both American and German trials long-term survival after pulmonary relapse has been occasionally achieved with aggressive chemotherapy alone [10,11], in Matsunaga's review of JPLT-1 no recurrent lung tumors were cured using chemotherapy alone [9]. In this review, results of aggressive metastasectomy for pulmonary relapse in the group as a whole were not quite as promising as those seen in JPLT; only 3 of 8 patients in whom a metastasectomy was performed with curative intent were long-term survivors. However, if, like Feusner et al [3], we look only at the patients with pulmonary relapse who initially presented as stage I, 2 of 2 were long-term survivors after pulmonary metastasectomy. Moreover, long-term survivors have been reported anecdotally after repeated metastasectomy for repeated relapse over several years [12,13]. Long-term survival has even been reported anecdotally after surgical resection of isolated hepatoblastoma metastasis to the brain [14,15] showing that surgical resection must be included in the armamentarium of tumor treatment for patients whose tumor becomes resistant to repeated chemotherapy [16]. The SIOPEL 1 and 2 studies both looked at the role of thoracotomy with pulmonary metastasectomy in patients with pulmonary metastasis at diagnosis. In SIOPEL 1, 4 of 22 patients with pulmonary metastasis had surgical resection and 4 of 4 survived. In SIOPEL 2, thoracotomy was used more liberally where 8 of 25 patients with pulmonary metastasis had surgical resection and, curiously, only 3 of 8 were long-term survivors. Lung metastases respond generally sufficiently to initial chemotherapy leading in many cases to complete disappearance of the lung disease. In some cases, however, some residual disease remains visible in the lungs making surgical removal necessary. Seven stage IV patients in this study underwent unilateral thoracotomy and wedge resection or lobectomy. One had bilateral tumors removed via median sternotomy, and one had bilateral thoracotomy. The timing of these procedures relative to the surgical resection of the primary tumor varied—2 before, 5 simultaneous, and 2 after resection of the liver primary. Eight of these 9 children were long-term survivors, a very encouraging result more in keeping with the positive observation in SIOPEL-1. There is no clear limit to the number of metastases that can be resected. The anticipated result of the operation, however, has to be clear lungs. Following data from childhood sarcoma, where the number of metastases detected manually can be higher than shown by radiographic imaging, some surgeons have recommended median sternotomy with manual palpation of both lungs, even in patients with
2055 radiographically unilateral disease (Otte, personal communication). Whether or not hepatoblastoma can be assumed to be analogous to sarcoma in this respect is currently not known and needs to be investigated. The benefit of surgical resection will be greatest in tumor with chemotherapeutic resistance, like many osteosarcomas. Therefore, it stands to reason that lung nodules present at diagnosis, and persistent after several cycles of neoadjuvant chemotherapy, should probably be resected. This is particularly important if the patient's primary tumor may require liver transplantation for definitive resection—PRETEXT 3 with major venous involvement, or PRETEXT 4. The nascent COG protocol for the treatment of children with all stages of hepatoblastoma, to be opened soon, will make the following treatment recommendations regarding lung resection and will track outcome and survival from the operations closely. Thoracotomy for radiographically apparent lung lesions may be unilateral, staged bilateral, or via median sternotomy. Wedge resection is the preferred technique for the sake of preserving as much pulmonary parenchyma as possible. However, in cases where there are more than 4 lesions present in the same lobe, multifocal microscopic residual is highly possible and lobectomy may be preferable. In patients who present with metastatic disease, the new protocol will seek out new drugs with upfront window therapy and metastasectomy is probably best delayed until after all protocol chemotherapy has been received and the primary tumor has been successfully resected. The exception to this would be patients who may require liver transplantation where successful biopsy and/or complete metastasectomy of any suspicious nodules, and perhaps contralateral lung palpation, should precede transplantation. The role of metastasectomy for pulmonary relapse remains less clear. In spite of the disheartening overall results, there are many cases of long-term survival in this review and in the literature, even after multiple resections. Aggressive resection is probably indicated in patients who presented as stage I. Although there are exceptions, in stage III or IV patients pulmonary relapse portends a probable poor outcome and aggressive resection of diffuse disease may not be warranted. Alternatively, in selected cases where pulmonary relapse is solitary or focal and unilateral, the surgeon should proceed with caution.
References [1] Abel RM, Brown J, Moreland B, et al. Pulmonary metastasectomy for pediatric solid tumors. Pediatr Surg Int 2004;20:630-2. [2] Schnater JM, Aronson DC, Plaschkes J, et al. Surgical view of the treatment of patients with hapatoblastoma. Cancer 2002;94:111-1120. [3] Feusner JH, Krailo MD, Haas JE, et al. Treatment of pulmonary metastases of initial stage I hepatoblastoma in childhood. Report from the Children's Cancer Group. Cancer 1993;71:859-64. [4] Perilongo G, Brown J, Shafford E, et al. Hepatoblastoma presenting with lung metastases: treatment results of the first cooperative, prospective study of the International Society of Paediatric Oncology on childhood liver tumors. Cancer 2000;89:1845-53.
2056 [5] Czauderna P, Otte JB, Aronson DC, et al. Guidelines for surgical treatment of hepatoblastoma in the modern era—recommendations from the childhood Liver Tumour Strategy Group of the International Society of Paediatric Oncology (SIOPEL). Eur J Cancer 2005;41:1031-6. [6] Ortega JA, Douglass EC, Feusner JH, et al. Randomized comparison of cisplatin/vincristine/5-fluorouracil and cisplatin/continuous infusion doxorubicin for treatment of pediatric hepatoblastoma: a report from the Children's Cancer Group and the Pediatric Oncology Group. J Clin Oncol 2000;18:2665-75. [7] Meyers RL, Krailo M, Chen Z, et al. Pretreatment prognostic factors in children with hepatoblastoma. J Clin Oncol [in press]. [8] Karnak I, Emin Senocak M, Kutluk T, et al. Pulmonary metastases in children: an analysis of surgical spectrum. Eur J Pediatr Surg 2002;12: 151-8. [9] Matsunaga T, Sasaki F, Ohira M, et al. Analysis of treatment outcome for children with recurrent or metastatic hepatoblastoma. Pediatr Surg Int 2003;19:142-6. [10] Malagolowkin MH, Katzenstein HM, Krailo M, et al. Redefining the role of Doxorubicin for the treatment of children with hepatoblastoma. J Clin Oncol [in press].
R.L. Meyers et al. [11] Fuchs J, Bode U, von Schweinitz D, et al. Analysis of treatment efficiency of carboplatin and etoposide in combination with radical surgery in advanced and recurrent childhood hepatoblastoma: a report of the German Cooperative Pediatric Liver Tumor Study HB 89 and HB 94. Klin Padiatr 1999;211:305-9. [12] Passmore SJ, Noblett HR, Wisheart JD, et al. Prolonged survival following multiple thoracotomies for metastatic hepatoblastoma. Med Pediatr Oncol 1995;24:58-60. [13] Nishimura S, Sato T, Fujita N, et al. High dose chemotherapy in children with metastatic hepatoblastoma. Pediatr Surg Int 2002;44: 300-5. [14] Robertson PL, Muraszko KM, Axtell RA. Hepatoblastoma metastatic to brain: prolonged survival after multiple surgical resections of a solitary brain lesion. J Pediatr Hematol Oncol 1997;19:168-71. [15] Miyagi J, Kobayashi S, Kojo N, et al. Brain metastasis of hepatoblastoma: case report and review of literature. No Shinkei Geka 1984;12:753-8. [16] VonSchweinitz D, Hecker H, Harms D. Complete resection before development of drug resistance is essential for survival from advanced hepatoblastoma. J Pediatr Surg 1995;30:845-52.
www.elsevier.com/locate/jpedsurg
Surgical resection of pulmonary metastatic lesions in children with hepatoblastoma Rebecka L. Meyers a,⁎, Howard M. Katzenstein b , Mark Krailo c , Eugene D. McGahren III d , Marcio H. Malogolowkin e a
Division of Pediatric Surgery, Primary Children's Medical Center, University of Utah, Salt Lake City, UT 84113, USA Division of Pediatric Hematology and Oncology, Emory Children's Center, Atlanta, GA, USA c Cure Search, Children’s Oncology Group, Arcadia, CA, USA d Division of Pediatric Surgery, University of Virginia, Charlottesville, VA, USA e Division of Pediatric Hematology and Oncology, Children's Hospital Los Angeles, Los Angeles, CA b
Received 6 August 2007; accepted 8 August 2007
Index words: Hepatoblastoma; Pulmonary metastasis; Lung metastasis; Metastasectomy; Thoracotomy
Abstract Background: Although the most common site of metastasis for hepatoblastoma is the lung, the role of thoracotomy and surgical resection of pulmonary metastasis remains uncertain. In this study, we aimed to determine the long-term outcome of patients with hepatoblastoma who have a surgical resection of pulmonary metastatic lesions. Methods: Patients with hepatoblastoma in the Children's Oncology Group INT-0098 were reviewed. Detailed reports enabling comprehensive review were available for 175 of 181 eligible patients. Prognostic factors (histology, tumor margin, surgical complications, α-fetoprotein) were also reviewed. Results: Thoracotomy for initial pulmonary metastasis: 38 patients presented with pulmonary metastasis, Children's Oncology Group stage IV. Nine of these 38 underwent thoracotomy and pulmonary metastectomy either before (2), simultaneous (5), or after (2) resection of their primary liver tumor. Eight of these 9 patients with metastectomy were long-term survivors. Thoracotomy for tumor relapse: 20 patients who had previously achieved complete tumor clearance experienced subsequent pulmonary relapse of their tumor (11 stage I or III, 9 stage IV). All 20 patients with pulmonary relapse had salvage chemotherapy; 13 also had thoracotomy and pulmonary metastectomy (8) or thoracotomy and biopsy (5). Only 4 of 13 were long-term survivors: 2 were stage I and 2 were stage IV. Conclusions: As pediatric surgeons, we varied tremendously in our timing and surgical approach to the management of pulmonary metastasis in hepatoblastoma. This large multicenter review suggests that thoracotomy should be used cautiously in the management of pulmonary relapse and perhaps more aggressively in the management of metastases present at diagnosis that persist after neoadjuvant chemotherapy. © 2007 Published by Elsevier Inc.
Presented at the 40th annual meeting of the Pacific Association of Pediatric Surgeons, Queenstown, New Zealand, April 15-19, 2007. ⁎ Corresponding author. Tel.: +1 801 588 3350; fax: +1 801 588 3364. E-mail address: [email protected] (R.L. Meyers). 0022-3468/$ – see front matter © 2007 Published by Elsevier Inc. doi:10.1016/j.jpedsurg.2007.08.030
Surgical resection of pulmonary metastatic Although the most common site of metastasis in children with hepatoblastoma is the lung, very little has been published to document the long-term outcome of pulmonary metastasectomy in these children. We do know that unilateral thoracotomy, bilateral thoracotomy, and median sternotomy are well tolerated in children with cancer [1]. We also know that children with lung metastases at diagnosis who are rendered free of all gross disease by some combination of chemotherapy, resection of the primary tumor, as well as resection of pulmonary metastases, may experience cure or long-term survival [2,3]. In International Society of Pediatric Oncology Liver Tumor Study Group (SIOPEL) multicenter studies, the results of metastectomy have been mixed. In SIOPEL-1, all 4 of the patients in whom a metastasectomy was performed survived without residual disease [2,4]. Results were not as promising in SIOPEL-2 where only 3 of the 8 patients with metastectomy were long-term survivors [5]. In these SIOPEL studies all children received neoadjuvant chemotherapy. In American studies, by contrast, not all patients have received neoadjuvant chemotherapy. The potential survival difference for patients who achieve a clinical remission of pulmonary metastases as a result of surgical resection and for those who achieve clinical remission in response to neoadjuvant chemotherapy remains unknown and needs to be defined. Moreover, no study has critically compared the potential difference in outcomes for patients who undergo metastasectomy for pulmonary lesions present at diagnosis with the outcomes for patients who undergo metastasectomy in the setting pulmonary relapse after having been in remission. We hope that review of these questions might help guide future therapy and protocol design with respect to the potential role and optimal timing of lung resection in children with metastatic hepatoblastoma.
2051 margins; Stage II, gross total resection with microscopic residual disease at the margins of disease; Stage III, biopsy only, subtotal resection, or gross total resection with tumor spill/preoperative rupture; Stage IV, metastatic disease at diagnosis with either complete or incomplete resection. Histologic diagnosis and Children's Cancer Group/Pediatric Oncology Group staging were confirmed by central review of representative tissue slides along with institutional pathology and operative reports. Surgical report forms, radiographic imaging reports, operative reports, and pathology reports were available for 175 of 181 eligible patients. Data for patients who were referred for liver transplantation were considered “off study” and were not routinely available in the research archive. Comparative analysis of the PRETEXT group was not a part of the original study, but was completed retrospectively. Data on patients' PRETEXT group enable comparison of this cohort of patients with those of the European and international studies where PRETEXT grouping is one of the primary tools for risk stratification.
1.1. Statistical analysis Comparative statistical analysis of patient survival by PRETEXT group and by COG stage was by the log-rank test. The remainder of this review is primarily descriptive and focused on the subgroup of patients whose research profile documented a thoracotomy at some point in their treatment. Because of the absence of a uniform surgical approach to the use of thoracotomy, and because of confounding variables including variation in the timing of thoracotomy, and variation in the type and amount of salvage/nonprotocol chemotherapy used, statistical analysis of thoracotomy as an independent variable was not felt to be prudent by the COG statistician.
1. Methods Patients with hepatoblastoma (N = 181) in the intergroup hepatoblastoma study INT-0098 performed by the antecedent groups of the current Children's Oncology Group (COG)—Children's Cancer Group/Pediatric Oncology Group—were retrospectively reviewed by RLM and MHM. The details regarding eligibility, treatment, evaluation, and follow-up for INT-0098 have been previously published [6]. The study protocols were approved by the institutional review boards of all participating institutions. Patients were randomized to receive one of 2 different chemotherapy regimens, cisplatin/5fluorouracil/vincristine (C5V) vs cisplatin/doxorubicin (CD). Treatment outcome was not significantly different between the 2 different chemotherapy arms of the study [6]. Stage of disease was determined by surgical and histologic criteria at the initial surgical intervention before the initiation of chemotherapy as follows: Stage I, complete gross resection with clear
2. Results 2.1. Thoracotomy for initial pulmonary metastasis Thirty-eight patients presented with pulmonary metastasis, COG stage IV. At varying times in their treatment, 9 of these 38 children underwent thoracotomy and pulmonary metastasectomy. Two of the 9 had lung resection before resection of the primary. In 5 of the 9 the thoracotomy and hepatectomy were performed simultaneously. In 2 of the 9 the lung metastasis were resected after complete resection of their primary liver tumor. Eight of these 9 patients with metastectomy were long-term survivors. Details of these 9 patients are shown in Table 1. None of these patients had additional poor prognostic indicators such as α-fetoprotein (AFP) of less than 100 at diagnosis or small cell undifferentiated histology.
2052
R.L. Meyers et al.
Table 1
Metastasectomy in hepatoblastoma lung metastases present at diagnosis (COG stage IV)
Patient PRETEXT Histology/ AFP at DX
Complication or additional surgery
Time of primary surgery
1
2
2
2
3
3
Epithelial/ mesenchymal
4
3
Embryonal/fetal
5
3
Embryonal/fetal
6
3
Embryonal/fetal
7
3
Macrotrabecular
8
4
Macrotrabecular
9
4
Epithelial/ mesenchymal
Left lobe at diagnosis Right lobe at diagnosis Right Colectomy dysplastic polyp at trisegmentectomy after neoadjuvant 10 y follow-up Left lobe after neoadjuvant chemotherapy Salivary Right pleomorphic trisegmentectomy adenoma at 9 y after neoadjuvant follow-up chemotherapy Right trisegmentectomy after neoadjuvant chemotherapy Right trisegmentectomy after neoadjuvant chemotherapy Right trisegmentectomy after neoadjuvant chemotherapy Postoperative biliary Right fistula, bile trisegmentectomy peritonitis, after neoadjuvant reoperative chemotherapy hepaticojejunostomy
Embryonal/fetal LOA, bowel obstruction Embryonal/fetal
Metastasectomy a before/simultaneous/ or after resection of primary
No. of cycles of Follow-up neoadjuvant/adjuvant/ nonprotocol chemotherapy b
Simultaneous/right wedge After/right wedge Relapse/biopsy only Before/bilateral thoracotomy
Alive 9 y
Alive 9 y
0/8/+ Lung relapse 1 y 6/2/0
Alive 12 y
4/2/+
Alive 7 y
Lung relapse 2 y 6/2/0
Alive 10 y
Simultaneous/right middle lobectomy
4/3/0
Alive 10 y
Simultaneous/right middle lobectomy
7/4/+
Alive 9 y
After/right lower and middle lobectomy
4/?/+
Alive 6 y
Simultaneous/right wedge ×4 Relapse/left wedge
5/?/+
Death tumor relapse
Simultaneous/right lung Relapse/left wedge Before/median sternotomy
Lung relapse 2 y
Alive 11 y
LOA indicates lysis of adhesions. a This table includes only stage IV patients with metastasectomy, meaning the lung resection was performed with the goal of tumor clearance, not simply as a biopsy. Patients whose only thoracotomy was for biopsy only are not included. b “Nonprotocol” chemotherapy is designated with a “+” in the table. Alterations in protocol chemotherapy were given to some of these patients for persistent lung metastasis, concern for stagnant or rising AFP levels, and/or for pulmonary relapse.
2.2. Thoracotomy for tumor relapse Twenty patients who had previously achieved complete tumor clearance experienced subsequent pulmonary relapse of their tumor (2 stage I, 9 stage III, 9 stage IV). All 20 patients with pulmonary relapse had salvage chemotherapy. Twelve of the patients with pulmonary relapse also had tumor relapse at other sites including liver, mediastinum, or bone. Thirteen of the patients with pulmonary tumor relapse underwent at least one (sometimes more) thoracotomy. In only 8 of the patients did the thoracotomy achieve complete eradication of all radiographically identifiable relapse disease; 5 of the 13 lung resection patients had persistent tumor present in either lung, liver, or bone.
Details of the select group of 8 patients whose thoracotomy achieved complete eradication of all radiographically evident relapse tumor are shown in Table 2A; these are the patients who had thoracotomy with complete metastectomy in an attempt to surgically remove all gross or radiographic evidence of relapse tumor. Only 3 of these 8 children were long-term survivors. Table 2B shows the children who had thoracotomy for pulmonary relapse with postoperative persistent tumor in either lung, liver, or bone. Only 1 of these 5 children was a long-term survivor. Overall, only 4 of 13 children with pulmonary relapse who had thoracotomy were long-term survivors; 2 were initially stage I, and 2 were initially stage IV. Surgical complications seemed unusually predominant in the group of patients with
Surgical resection of pulmonary metastatic Table 2A
2053
Thoracotomy with definitive metastasectomy of pulmonary relapse
Patient Stage/ Histology/ PRETEXT AFP at DX
Complication or Time of primary additional surgery surgery (protocol chemotherapy)
1
I/3
Embryonal/fetal N 2000 mL EBL
2
I/2
Embryonal/ macrotrabecular
3
II/2
Small cell undifferentiated
4
III/3
Mixed/ mesenchymal
5
III/3
Macrotrabecular
6
IV/3
Embryonal/fetal
7
IV/4
Epithelial/ mesenchymal
8
IV/3
Macrotrabecular
Right trisegmentectomy at diagnosis (C5V) Left lobe at diagnosis (C5V)
Survival No. of cycles of Relapse neoadjuvant/adjuvant/ complete metastasectomy nonprotocol chemotherapy ⁎⁎
Relapse/right wedge 1st relapse/left multiple wedge 2nd relapse/left lower lobectomy Preoperative Debulk, control rupture Relapse/ tumor rupture at diagnosis (CD) resected mediastinal mass and wedge LUL Left lobe after 1st relapse/right neoadjuvant wedge chemotherapy (CD) 2nd relapse/ abdominal wall 3rd relapse/right wedge 4th relapse/ mediastinal 5th relapse mediastinal 1st relapse/liver Bile fistula, Left lobe after resected + margin neoadjuvant 2nd relapse/ chemotherapy (CD) wedge lung Left lobe after Metastasectomy neoadjuvant at Dx/right chemotherapy (CD) wedge Relapse/left wedge Right trisegmentectomy Metastasectomy Postoperative at Dx/right after neoadjuvant biliary fistula, wedge ×4 biliary peritonitis chemotherapy (C5V) Relapse/left wedge + Margin, resected Right trisegmentectomy Relapse/resect at diagnosis (CD) mediastinal met lung and postoperative mediastinal sepsis, CHF mass
0/4/0
Alive 8 y
0/4/+
Alive 9 y
0/6/0
Death 16 mo
4/2/+
Death 6 y
0/6/+
Death 20 mo
4/2/+
Alive 7 years
5/?/+
Death ? years
0/1
Death 2 mo
C5V indicates cisplatin, 5fu, vincristine; CD, cisplatin doxyrubicin; EBL, estimated blood loss; LUL, left upper lobe. ⁎⁎ Nonprotocol chemotherapy is designated as a “+” in the table.
pulmonary relapse, although statistical analysis in a separate analysis of these patients [7] did not show a correlation between surgical complication and overall survival for the group as a whole. To better compare this cohort of patients with those in contemporary European (SIOPEL or German Cooperative) or Japanese (Japanese Study Group for Pediatric Liver Tumor [JPLT]) trials, overall survival and stratification by PRETEXT vs COG stage are shown in Table 3. In INT-0098,
there was a preponderance of upfront resection of the liver primary tumor for patients retrospectively classified as PRETEXT 1 and 2, and neoadjuvant chemotherapy for patients retrospectively classified as PRETEXT 3 and 4, but there were frequent exceptions to this broad theme. There were 61 stage I and II patients resected at diagnosis. Of these 61 patients, 16 (26%) were PRETEXT 1, 36 (59%) were PRETEXT 2, and 9 (15%) were PRETEXT 3. Two of 79 stage III patients had gross residual tumor after
2054
R.L. Meyers et al.
Table 2B
Thoracotomy with biopsy of pulmonary relapse
Patient Stage/ Histology/ PRETEXT AFP at DX
Complication or additional surgery
9
III/3
Bile leak Fetal/ macrotrabecular
10
III/4
Embryonal/ fetal
11
III/4
12
IV/2
Embryonal/ fetal Embryonal/ fetal
13
IV/3
IVC injury N2000 EBL HACE local relapse
Relapse biopsy
Right trisegmentectomy after only 1 cycle neoadjuvant chemotherapy (C5V) Right trisegmentectomy after neoadjuvant chemotherapy (CD) ? Operation report incomplete (C5V) Right lobe at diagnosis (CD)
Relapse/Bx 1/2/+ bone, liver, and lung
Death 2y
Relapse/Bx lung
Death 31 mo
Right + Margin, injury trisegmentectomy at LHD hepaticojejunostomy, diagnosis (CD) drain subphrenic abscess
Embryonal/ fetal
Survival No. of cycles of neoadjuvant/adjuvant/ nonprotocol chemotherapy ⁎⁎
Time of primary surgery (protocol chemotherapy)
4/2/0
Relapse/Bx 2/na/+ lung Metastasectomy 0/8/+ at Dx/right wedge Relapse/Bx lung 0/5/+ Relapse/Bx lung
Death 19 mo Alive 12 y
Death 2y
EBL, estimated blood loss; LUL, left upper lobe; IVC, inferior vena cava; HACE, hepatic artery chemoembolization; LHD, left hepatic duct; C5V, cisplatin, 5fu, vincristine; CD, cisplatin doxyrubicin. ⁎⁎ Nonprotocol chemotherapy is designated as a “+” in the table.
attempted surgical resection at diagnosis, and the remaining 77 patients had an initial biopsy followed by neoadjuvant chemotherapy.
3. Discussion Pulmonary surgery is frequently used for the treatment of metastasis in children. However, because hepatoblastoma is a relatively rare tumor, the use of metastasectomy in this tumor remains poorly characterized. In recent reviews of pulmonary metastasectomy for various tumor types in Table 3
children, Karnak et al [8] report that hepatoblastoma was the primary tumor type in only 1 of 18 patients; Abel et al [1] report hepatoblastoma as the primary tumor in 2 of 20 of their patients with metastasectomy. In neither of these reports is it clear whether or not the metastasectomy was performed for a lesion present at diagnosis and persistent after chemotherapy vs a de novo lesion that developed after therapy had been initiated (pulmonary relapse). Fifteen years ago it was noted by Feusner et al [3] that extended disease-free survival is possible for children with initial stage I hepatoblastoma who experienced a pulmonary relapse if the relapse was isolated to the lung and an aggressive surgical approach with intent to cure was used. In
Overall survival by PRETEXT group vs. COG stage
PRETEXT 1 PRETEXT 2 PRETEXT 3 PRETEXT 4 Overall survival
Stage I (pure fetal)
Stage I (unfavorable histology)
Stage II
Stage III
Stage IV
Overall survival
3 4 2 9/9 (100%)
12 26
1 6 7/7 (100%)
1 13 44 21 54/79 (68%)
2 5 24 7 16/38 (42%)
17/19 (89%) 46/54 (85%) 54/77 (70%) 9/28 (32%) ⁎P b .0001 predictive value for both PRETEXT and COG
42/45 (93%)
⁎ There were a total of 191 patients enrolled in INT0098. Of 191 patients, 181 met eligibility criteria. This review includes 178 patients in whom the radiographs and operative reports were detailed enough to determine PRETEXT group at diagnosis. The log-rank test shows both systems to be significantly related to risk of death predictive (P b .0001) of overall survival.
Surgical resection of pulmonary metastatic that analysis, 6 of 33 initial stage 1 patients experienced a tumor relapse confined to the lung, and 3 of the 6 patients were long-term survivors after a pulmonary metastasectomy. More recently, JPLT report in 8 recurrent lung tumors, surgical resection was performed completely in 6 cases with unilateral pulmonary relapse [9]. Five of the 6 JPLT patients were alive and well more than a year after recurrence, but the remaining patient experienced a second relapse in the lung and died of the disease, which proved resistant to various chemotherapeutic agents. Although in both American and German trials long-term survival after pulmonary relapse has been occasionally achieved with aggressive chemotherapy alone [10,11], in Matsunaga's review of JPLT-1 no recurrent lung tumors were cured using chemotherapy alone [9]. In this review, results of aggressive metastasectomy for pulmonary relapse in the group as a whole were not quite as promising as those seen in JPLT; only 3 of 8 patients in whom a metastasectomy was performed with curative intent were long-term survivors. However, if, like Feusner et al [3], we look only at the patients with pulmonary relapse who initially presented as stage I, 2 of 2 were long-term survivors after pulmonary metastasectomy. Moreover, long-term survivors have been reported anecdotally after repeated metastasectomy for repeated relapse over several years [12,13]. Long-term survival has even been reported anecdotally after surgical resection of isolated hepatoblastoma metastasis to the brain [14,15] showing that surgical resection must be included in the armamentarium of tumor treatment for patients whose tumor becomes resistant to repeated chemotherapy [16]. The SIOPEL 1 and 2 studies both looked at the role of thoracotomy with pulmonary metastasectomy in patients with pulmonary metastasis at diagnosis. In SIOPEL 1, 4 of 22 patients with pulmonary metastasis had surgical resection and 4 of 4 survived. In SIOPEL 2, thoracotomy was used more liberally where 8 of 25 patients with pulmonary metastasis had surgical resection and, curiously, only 3 of 8 were long-term survivors. Lung metastases respond generally sufficiently to initial chemotherapy leading in many cases to complete disappearance of the lung disease. In some cases, however, some residual disease remains visible in the lungs making surgical removal necessary. Seven stage IV patients in this study underwent unilateral thoracotomy and wedge resection or lobectomy. One had bilateral tumors removed via median sternotomy, and one had bilateral thoracotomy. The timing of these procedures relative to the surgical resection of the primary tumor varied—2 before, 5 simultaneous, and 2 after resection of the liver primary. Eight of these 9 children were long-term survivors, a very encouraging result more in keeping with the positive observation in SIOPEL-1. There is no clear limit to the number of metastases that can be resected. The anticipated result of the operation, however, has to be clear lungs. Following data from childhood sarcoma, where the number of metastases detected manually can be higher than shown by radiographic imaging, some surgeons have recommended median sternotomy with manual palpation of both lungs, even in patients with
2055 radiographically unilateral disease (Otte, personal communication). Whether or not hepatoblastoma can be assumed to be analogous to sarcoma in this respect is currently not known and needs to be investigated. The benefit of surgical resection will be greatest in tumor with chemotherapeutic resistance, like many osteosarcomas. Therefore, it stands to reason that lung nodules present at diagnosis, and persistent after several cycles of neoadjuvant chemotherapy, should probably be resected. This is particularly important if the patient's primary tumor may require liver transplantation for definitive resection—PRETEXT 3 with major venous involvement, or PRETEXT 4. The nascent COG protocol for the treatment of children with all stages of hepatoblastoma, to be opened soon, will make the following treatment recommendations regarding lung resection and will track outcome and survival from the operations closely. Thoracotomy for radiographically apparent lung lesions may be unilateral, staged bilateral, or via median sternotomy. Wedge resection is the preferred technique for the sake of preserving as much pulmonary parenchyma as possible. However, in cases where there are more than 4 lesions present in the same lobe, multifocal microscopic residual is highly possible and lobectomy may be preferable. In patients who present with metastatic disease, the new protocol will seek out new drugs with upfront window therapy and metastasectomy is probably best delayed until after all protocol chemotherapy has been received and the primary tumor has been successfully resected. The exception to this would be patients who may require liver transplantation where successful biopsy and/or complete metastasectomy of any suspicious nodules, and perhaps contralateral lung palpation, should precede transplantation. The role of metastasectomy for pulmonary relapse remains less clear. In spite of the disheartening overall results, there are many cases of long-term survival in this review and in the literature, even after multiple resections. Aggressive resection is probably indicated in patients who presented as stage I. Although there are exceptions, in stage III or IV patients pulmonary relapse portends a probable poor outcome and aggressive resection of diffuse disease may not be warranted. Alternatively, in selected cases where pulmonary relapse is solitary or focal and unilateral, the surgeon should proceed with caution.
References [1] Abel RM, Brown J, Moreland B, et al. Pulmonary metastasectomy for pediatric solid tumors. Pediatr Surg Int 2004;20:630-2. [2] Schnater JM, Aronson DC, Plaschkes J, et al. Surgical view of the treatment of patients with hapatoblastoma. Cancer 2002;94:111-1120. [3] Feusner JH, Krailo MD, Haas JE, et al. Treatment of pulmonary metastases of initial stage I hepatoblastoma in childhood. Report from the Children's Cancer Group. Cancer 1993;71:859-64. [4] Perilongo G, Brown J, Shafford E, et al. Hepatoblastoma presenting with lung metastases: treatment results of the first cooperative, prospective study of the International Society of Paediatric Oncology on childhood liver tumors. Cancer 2000;89:1845-53.
2056 [5] Czauderna P, Otte JB, Aronson DC, et al. Guidelines for surgical treatment of hepatoblastoma in the modern era—recommendations from the childhood Liver Tumour Strategy Group of the International Society of Paediatric Oncology (SIOPEL). Eur J Cancer 2005;41:1031-6. [6] Ortega JA, Douglass EC, Feusner JH, et al. Randomized comparison of cisplatin/vincristine/5-fluorouracil and cisplatin/continuous infusion doxorubicin for treatment of pediatric hepatoblastoma: a report from the Children's Cancer Group and the Pediatric Oncology Group. J Clin Oncol 2000;18:2665-75. [7] Meyers RL, Krailo M, Chen Z, et al. Pretreatment prognostic factors in children with hepatoblastoma. J Clin Oncol [in press]. [8] Karnak I, Emin Senocak M, Kutluk T, et al. Pulmonary metastases in children: an analysis of surgical spectrum. Eur J Pediatr Surg 2002;12: 151-8. [9] Matsunaga T, Sasaki F, Ohira M, et al. Analysis of treatment outcome for children with recurrent or metastatic hepatoblastoma. Pediatr Surg Int 2003;19:142-6. [10] Malagolowkin MH, Katzenstein HM, Krailo M, et al. Redefining the role of Doxorubicin for the treatment of children with hepatoblastoma. J Clin Oncol [in press].
R.L. Meyers et al. [11] Fuchs J, Bode U, von Schweinitz D, et al. Analysis of treatment efficiency of carboplatin and etoposide in combination with radical surgery in advanced and recurrent childhood hepatoblastoma: a report of the German Cooperative Pediatric Liver Tumor Study HB 89 and HB 94. Klin Padiatr 1999;211:305-9. [12] Passmore SJ, Noblett HR, Wisheart JD, et al. Prolonged survival following multiple thoracotomies for metastatic hepatoblastoma. Med Pediatr Oncol 1995;24:58-60. [13] Nishimura S, Sato T, Fujita N, et al. High dose chemotherapy in children with metastatic hepatoblastoma. Pediatr Surg Int 2002;44: 300-5. [14] Robertson PL, Muraszko KM, Axtell RA. Hepatoblastoma metastatic to brain: prolonged survival after multiple surgical resections of a solitary brain lesion. J Pediatr Hematol Oncol 1997;19:168-71. [15] Miyagi J, Kobayashi S, Kojo N, et al. Brain metastasis of hepatoblastoma: case report and review of literature. No Shinkei Geka 1984;12:753-8. [16] VonSchweinitz D, Hecker H, Harms D. Complete resection before development of drug resistance is essential for survival from advanced hepatoblastoma. J Pediatr Surg 1995;30:845-52.