Telephone: 442073463672; FAX: 442073463575
Liver transplantation for hepatoblastoma
Article first published online: 29 OCT 2008
Copyright © 2008 American Association for the Study of Liver Diseases
Volume 14, Issue 11, pages 1614–1619, November 2008
How to Cite
Faraj, W., Dar, F., Marangoni, G., Bartlett, A., Melendez, H. V., Hadzic, D., Dhawan, A., Mieli-Vergani, G., Rela, M. and Heaton, N. (2008), Liver transplantation for hepatoblastoma. Liver Transpl, 14: 1614–1619. doi: 10.1002/lt.21586
- Issue published online: 29 OCT 2008
- Article first published online: 29 OCT 2008
- Manuscript Accepted: 6 JUN 2008
- Manuscript Received: 5 FEB 2008
From October 1993 to February 2007, 25 liver transplantations were performed for hepatoblastoma. Of these 25, 18 children received cadaveric grafts, and 7 received left lateral segments from living donors. Fifteen patients were at level IV in the pretreatment extent of disease staging system for hepatoblastoma (PRETEXT IV; 11 received cadaveric grafts and 4 underwent living related liver transplantation [LRLT]) and 10 were level III (PRETEXT III; 7 received cadaveric grafts and 3 underwent LRLT). Preoperative chemotherapy was given according to the risk stratification system for children with hepatoblastoma protocols of the International Childhood Liver Tumour Strategy Group of the International Society of Paediatric Oncology (SIOPEL): SIOPEL I in the first 3 patients, SIOPEL II in 6, SIOPEL III in 10, and SIOPEL IV in 3 patients. Patient and graft survival after cadaveric transplantation was 91%, 77.6%, and 77.6%, at 1, 5, and 10 years, respectively, with no retransplantations. Patient and graft survival for children undergoing LRLT was 100%, 83.3%, and 83.3%, at 1, 5, and 10 years, respectively. All surviving children but 1 remain disease-free, with a median follow up of 6.8 years (range, 0.9-14.9). There were 5 deaths at a median of 13 months post-transplantation, secondary to tumor recurrence (4) and respiratory failure (1). Liver transplantation is an established treatment for unresectable hepatoblastoma confined to the liver following chemotherapy. LRLT is a therapeutic option given that the outcome is similar to that of resection and cadaveric transplantation. Liver Transpl 14:1614–1619, 2008. © 2008 AASLD.
Hepatoblastoma (HB) is the most common malignant liver tumor of early childhood, with an incidence of 0.5-1.5 per 1 million population, and accounting for approximately 60%-85% of all hepatic tumors in children.1, 2 The advent of effective chemotherapy has transformed the prognosis of this tumor over the past 25 years. Resectable tumors have an excellent prognosis, with an overall survival rate of 80% at 5 years.3–5 More than 60% of lesions that appear unresectable on initial imaging will shrink with chemotherapy and become resectable.6 However, approximately 20% of tumors remain unresectable after chemotherapy; for these, liver transplantation has been used in the absence of visible extrahepatic disease.7–9 Early reports of liver transplantation for children with unresectable HB reported a 50% 5-year survival rate, with tumor recurrence being the most common cause of death.9, 10 More recently survival has improved to approximately 80% at 5 years post-transplantation.
The majority of children have received cadaveric grafts; however, living donor liver grafts have also been used, to allow the transplant surgery to fit into the chemotherapy schedule. We have reviewed our experience with 25 liver transplantations for HB from living and cadaveric donation.
PATIENTS AND METHODS
Patients were identified by review of a prospectively acquired liver database. Recipient and donor demographics, perioperative findings, and postoperative outcome were obtained. Descriptive data are reported as medians and ranges.
From October 1993 to February 2007, 68 patients were referred for treatment of HB, and 25 received liver transplants. Thirteen of the 25 patients have been reported.11 All of the patients were treated according to the contemporary protocol guidelines from the International Childhood Liver Tumour Strategy Group of the International Society of Paediatric Oncology (SIOPEL) risk stratification system for children with hepatoblastoma for preoperative and postoperative chemotherapy, SIOPEL I-IV. The indication for liver transplantation was nonresectability of the HB and biological response to chemotherapy (fall in alpha-fetoprotein [AFP]), in the absence of lung metastases.
Twenty-five liver transplantations were performed. Of these, 18 received cadaveric grafts (12 male and 6 female, median age 2.5 years [range, 0.5-10], median weight 11 kg [range, 5.9-24.2], and median height 85 cm [range, 74-140]). Cadaveric transplants included 6 reduced grafts (left lobe in 4, right lobe in 1, and left lateral segment [LLS] in 1). Nine children received split cadaveric grafts, of which 8 were LLSs and 1 was right lobe. Three children received whole cadaveric liver grafts. Seven children received LLS grafts from living donors (5 male and 2 female, median age 2 years [range, 0.9-11], median weight 11.7 kg [range, 6.6-32.3], and median height 74.3 cm [range, 73-130]).
The AFP at the time of diagnosis was a median of 263,500 kU/L (range, 2846-3,200,000) for children who received cadaveric grafts and 75,474 kU/L (range, 538-1,300,000) for children who received living-related liver transplantation (LRLT). AFP at the time of transplantation was a median of 220 kU/L (range, 10-376,000) for cadaveric grafts and 251 kU/L (range, 26-6881) for LRLT grafts. Other blood results included serum aspartate aminotransferase median 50 IU/L (range, 19-380), bilirubin (total) 5 μmol/L (range, 2-11), gamma-glutamyl transferase 51 IU/L (range, 20-108), alkaline phosphatase 201 IU/L (range, 154-365), albumin 39 g/L (range, 34-45), creatinine 45 μmol/L (range, 33-60), hemoglobin 10.6 g/dL (range, 6.5-13.9), and international normalized prothrombin 1.03 (range, 0.89-1.06). All children underwent serial computed tomography scans to assess response to treatment and identify extrahepatic disease.
According to the pretreatment extent of disease (PRETEXT) staging system for hepatoblastoma, 15 patients were PRETEXT IV (11 for cadaveric and 4 for LRLT grafts) and 10 were PRETEXT III (7 for cadaveric and 3 for LRLT grafts). Preoperative chemotherapy was given according to SIOPEL I in the first 3 patients, SIOPEL II in 6, SIOPEL III in 10, and SIOPEL IV in 3 (Tables 1 and 2).
|Patient||Age (Years)||Sex||Serum AFP at Diagnosis||Serum AFP at Transplant||PRETEXT||SIOPEL||Pretransplantation Chemotherapy||Posttransplantation Chemotherapy||Follow-Up Time (Years)||Survival|
|1||2||Female||311,500||74||III||I||Vin, Cis, 5 FU||No||14.9||Alive|
|3||9||Male||173,000||491||IV||I||Cis, Car, Dox||Yes||10.8||Alive|
|4||2.9||Male||68,890||90||III||I||Cis, Car, Dox||No||0||Dead|
|6||2.5||Male||540,000||12,251||III||II||Cis, Car, Dox||Yes||9||Alive|
|7||2||Male||1,687,490||3,000||IV||II||Cis, Car, Dox||Yes||8.6||Alive|
|8||10||Male||668,310||16||IV||II||Cis, Car, Dox||Yes||1||Dead|
|9||9||Male||2846||10||III||II||Cis, Car, Dox||Yes||6.8||Alive|
|10||1||Male||12,050||93||IV||II||Cis, Car, Dox||Yes||5.5||Alive|
|11||2||Female||10,950||239||IV||III||Cis, Car, Dox||Yes||5||Alive|
|12||0.7||Male||213,010||155||IV||III||Cis, Car, Dox||Yes||5||Alive|
|13||2.5||Female||3,200,000||376,000||IV||III||Cis, Car, Dox||Yes||2||Dead|
|14||2.6||Male||228,000||2,270||IV||III||Cis, Car, Dox||Yes||4.3||Alive|
|15||1||Female||2,500,000||287||IV||III||Cis, Car, Dox||Yes||3.5||Alive|
|16||4||Female||10,990||1360||IV||III||Cis, Car, Dox||Yes||1.8||Dead|
|17||2.3||Female||9,979||202||IV||IV||Cis, Car, Dox||Yes||1||Alive|
|18||1.1||Male||1,100,000||50||III||IV||Cis, Car, Dox||Yes||0.9||Alive|
|Patient||Age (Years)||Sex||Serum AFP at Diagnosis||Serum AFP at Transplantation||PRETEXT||SIOPEL||Pretransplantation Chemotherapy||Posttransplantation Chemotherapy||Follow-Up Time (Years)||Survival|
|1||1.5||Male||82,500||230||IV||II||Cis, Car, Dox||Yes||9.8||Alive|
|2||2||Female||169||26||IV||II||Cis, Car, Dox||Yes||10||Alive|
|3||11||Male||5,280||251||III||III||Cis, Car, Dox||Yes||8.7||Alive|
|4||2.5||Male||1,031,309||2190||III||III||Cis, Car, Dox||Yes||8||Alive|
|5||1.5||Male||36,300||2320||IV||III||Cis, Car, Dox||Yes||1.8||Dead|
|6||0.9||Female||1,300,000||6881||II||III||Cis, Car, Dox||Yes||2||Alive|
|7||2||Male||75,474||30||IV||IV||Cis, Car, Dox||Yes||1.5||Alive|
All living donors were related and included 3 mothers, 3 fathers, and 1 uncle, with a median age of 30 years (range, 25-37) and median weight of 70 kg (range, 55.3-85). Donation occurred after a 2-stage consent protocol based on that published by Broelsch et al.12 A standard investigation protocol was performed for all donors.13 Abdominal computed tomography scan was performed to estimate the volume of the LLS and to assess the vascular supply to the potential graft. Serology was negative for human immunodeficiency virus, hepatitis C, and hepatitis B. Of the 7 donors, 5 were cytomegalovirus (CMV) immunoglobulin G (IgG)-positive and 3 were Epstein-Barr virus IgG-positive. For the transplantation, 19 children were admitted from home 4 were in the hospital, and 1 was from the pediatric intensive care unit.
Pretransplantation laparotomy was undertaken in 8 children. Of these 8, 3 children underwent attempted resection with an available liver graft, to proceed to orthotopic liver transplantation (OLT) if the tumor was unresectable. All 3 proceeded to OLT with the available graft. One child underwent extended right hepatectomy, but had positive tumor margins on histology with involvement of the portal vein and inferior vena cava and received transplantation 4 months later. One child underwent left hepatectomy followed by OLT after 9 months. Three children underwent laparotomy for resection (including 1 patient who underwent an attempted liver resection performed at another hospital). All 3 were found to be unresectable and referred for transplantation, which was performed between 10 days and 4 months later.
Overall actuarial patient and graft survival was 91%, 77.6%, and 77.6% at 1, 5, and 10 years, respectively, with no retransplantations at a median follow-up of 60 months (range, 1-179) (Fig. 1). Of the 25 cases, 5 died at a median of 13 months (range, 3 weeks to 28 months) post-transplantation, secondary to tumor recurrence in 4 and respiratory failure in 1. Patient and graft survival for LRLT was 100%, 83.3%, and 83.3% at 1, 5, and 10 years, respectively, with 1 death at 20 months from pulmonary recurrence, with a median follow-up of 96 months (range, 18-125). All surviving children, but 1, remain disease free.
The recipient operative time was a median of 6.4 hours (range, 4.5-6.6) for both groups. The median blood loss was 83.2 mL/kg (range, 27.8-540) for cadaveric and 49.5 mL/kg (range, 17-352) for the living donor group. The median cold ischemic time was 10 hours (range, 6.4-15.5) for cadaveric and 2.5 hours (range, 0.5-8) for LRLT. The median living donor operative time was 3.2 hours (range, 2.5-4.5) with median blood loss of 250 mL (range, 100-1500) and a median hospital stay of 7 days (range, 6-16). The surgical techniques of both living donor hepatectomy and transplantation were as described.14–16 The recipient surgery was started with the child first, to assess resectability and confirm the absence of extrahepatic disease in the 4 cases using a cadaveric graft. Four children had involvement of the inferior vena cava by the tumor and reconstruction was performed using cadaveric iliac venous grafts. One child had an exophytic extension of the primary tumor involving the porta hepatis and the body of the pancreas at the time of transplantation and underwent resection of the extrahepatic extension, including resection of the body and tail of the pancreas in addition to liver replacement.
There were no vascular complications in either recipient group. In the cadaveric group there were 2 bile leaks that resolved with laparotomy and drainage in 1 and percutaneous drainage in the other. One child had a laparotomy and bowel resection for small bowel obstruction on day 16 without sequelae. One child developed chylous ascites that settled spontaneously. One child developed two lung metastases 7 months after liver transplantation (previously had pulmonary metastases). One child developed acute liver and respiratory failure after attempted liver resection at another institution and despite emergency transplantation he died 3 weeks later from his lung injury with good graft function.
There were no biliary complications in the LRLT group. One child developed intestinal obstruction 13 days post-transplantation, secondary to a volvulus treated by laparotomy and small bowel resection. Two children had lung metastases pretransplantation, which resolved with chemotherapy. However, 1 recurred 1 year post-transplantation with secondary lung diseases, and despite pulmonary resection died of recurrent disease in the lungs and liver. CMV prophylaxis with ganciclovir was given to 7 CMV IgG-negative children receiving a graft from an IgG-positive donor. Three children developed (CMV) infection and responded to ganciclovir therapy.
The postoperative immunosuppression consisted of cyclosporine, steroids, and azathioprine in 6 patients, tacrolimus and steroids in 12 patients, and tacrolimus and mycophenolate mofetil (MMF) in 5. Two children were converted to MMF and steroids for renal toxicity on late follow-up. Cellular rejection was diagnosed on clinical and biochemical grounds, and confirmed by histological examination of liver biopsy. Treatment of acute cellular rejection was with a 3-day course of methylprednisolone (intravenously, 10 mg/kg/day). Of the 25 children, 20 received postoperative chemotherapy after a median interval of 5 weeks (range, 2-8) depending on the postoperative recovery of the patients.
All 7 living donors are alive with normal liver function at a median follow up of 5 years (range, 1.5-10 years). Twenty children have normal graft function without tumor recurrence at a median post-transplantation follow-up of 73 months (range, 10-178). Renal function as measured by the estimated glomerular filtration rate was a median 88 mL/minute (range, 55-115). Of 20 children, 7 showed reductions in the estimated glomerular filtration rate and were converted to MMF with steroid or low-dose tacrolimus with MMF.
HB remains unresectable by conventional surgery in up to 20% of cases despite effective chemotherapy.17, 18 Current treatment protocols have improved the outcome of HB, with several studies showing 90%-95% chemosensitivity rates, with 75%-85% of previously unresectable tumors having a good outcome.5, 19–23 Liver transplantation has been employed to resect tumors with grafts from cadaveric or living donation. Caval replacement is often required for complete clearance and can be achieved by the use of donor iliac vein interpositional grafts.24–27 The use of grafts from living donors offers some potential advantages over cadaveric donation, including shorter waiting time, better graft quality, more rapid recovery, and improved timing of transplant to comply with chemotherapy regimes.
We report 25 children with unresectable HB treated with OLT. Of these, 7 received LRLT. Similar results have been reported by others, with overall survival ranging from 62.5%-90%. (Table 3). Our results are comparable with survival rates reported after radical liver resection and cadaveric liver transplantation.25, 28–30 The most common cause of death was tumor recurrence, as is reflected in the majority of series: 20%-25% of recipients died, usually within 2 years of transplantation.9, 25 In our series, four patients (16%) developed tumor recurrence and died at a median of 13 months post-transplant. Previous lung metastases was a marker of risk for recurrence
|Authors||Year||Number of Children||Cadaveric Donors||Living Donors||Overall Survival (%)|
|Al-Qabandi et al.10||1999||8||8||0||62.5|
|Reyes et al.31||2000||12||12||0||83|
|Molmenti et al.33||2002||9||9||0||66|
|Pimpalwar et al.19||2002||12||12||0||83|
|Chardot et al.34||2002||4||0||4||75|
|Mejia et al.35||2005||10||8||2||70|
|Kasahara et al.24||2005||14||0||14||65.5|
|Otte et al.28||2005||12||2||10||90|
|Avila et al.26||2006||12||9||3||82|
|Chen et al.4||2006||9||7||2||89|
|Beaunoyer et al.36||2007||15||15||0||86.7|
The value of posttransplantation chemotherapy in this context has been questioned. Otte et al.25 reviewed the published experience of liver transplantation for HB and identified 65 children who received posttransplantation chemotherapy (82 patients did not receive any). Of the 65, 61 had pretransplantation chemotherapy with an overall survival of 77%, and a 5-year survival of 70% for those who did not. The lack of statistical difference in outcome and the added morbidity of posttransplantation chemotherapy when given with immunosuppression led to some centers avoiding such a protocol.25 We favor further chemotherapy at a time when the tumor burden is at its lowest, and the morbidity with this approach has been acceptably low. The incidence of recurrent disease compares favorably with studies without chemotherapy post-surgery, which have a 30% incidence of recurrence.32
Liver transplantation appears to be associated with better outcome when performed as a primary rather than as a salvage procedure and should be offered to all children with unresectable disease. Otte et al.25 reported 85% overall survival for 7 children undergoing primary liver transplantation and 40% for 5 children who underwent “rescue” transplantation after partial hepatectomy. In our series, 4 patients underwent “rescue” liver transplantation, of which 1 died at 20 months post-transplantation from pulmonary recurrence. The only contraindication to transplantation is the persistence of 1 or more sites of extrahepatic disease that do not respond to chemotherapy.19
LRLT is a practical therapeutic option for unresectable HB, with long-term outcome similar to that achieved for cadaveric liver transplantation and surgically-resectable tumors. Appropriate staging and early referral to centers with the expertise to optimize results provides good outcome for the majority of children with HB.
- 4Liver transplantation and chemotherapy in children with unresectable primary hepatic malignancies: development of a management algorithm. J Pediatr Gastroenterol Nutr 2006; 43: 487–493., , , , , .
- 17Hepatoblastoma in infancy and childhood: a clinical and pathological study of 32 cases. Zhonghua Min Guo Shou Yi Xue Hui Za Zhi 1991; 32: 79–87., , , , , , et al.
- 23Ifosfamide, carboplatin, and etoposide (ICE) reinduction chemotherapy in a large cohort of children and adolescents with recurrent/refractory sarcoma: the Children's Cancer Group (CCG) experience. Pediatr Blood Cancer 2005; 44: 338–347., , , , , , et al.
- 27Liver transplantation and chemotherapy in children with unresectable primary hepatic malignancies: development of a management algorithm. J Pediatr Gastroenterol Nutr 2006; 43: 487–493., , , , , .