Journal of Neuro-Oncology 67: 101–106, 2004.© 2004 Kluwer Academic Publishers. Printed in the Netherlands.

Clinical Study

High-dose chemotherapy with autologous stem cell rescue asfirst line of treatment in young children with medulloblastoma andsupratentorial primitive neuroectodermal tumors

Antonio Perez-Martınez, Victor Quintero, Marta Gonzalez Vicent, Julian Sevilla,Miguel Angel Dıaz and Luis MaderoDepartment of Pediatric Hematology and Oncology, Hospital Nino Jesus, UniversidadAutonoma de Madrid, Madrid, Spain

Key words: autologous peripheral blood progenitor cell transplantation, high-dose chemotherapy,medulloblastoma, supratentorial primitive neuroectodermal tumors, young children

Summary

In order to improve the dismal prognosis of patients younger than 4 years old with medulloblastoma and supraten-torial primitive neuroectodermal tumors (stPNET) seven young children were treated with high-dose chemotherapy(HDCT) and autologous stem cell rescue in our center.

All patients underwent surgical debulking and standard chemotherapy. None of them received irradiation. TheHDCT included busulfan 16 mg/kg, orally over 4 days (from days −5 to −2) in 6 hourly divided doses, andmelphalan at a dose of 140 mg/m2 given by intravenous infusion over 5 min on day −1. Three patients additionallyreceived thiotepa 250 mg/m2 given by intravenous infusion daily over 2 days (from day −2 to −1) and two patientsadditionally received topotecan 2 mg/m2 given by intravenous infusion daily over 30 min for 5 days (from day −11to −7).

Patients’ stem cells were mobilized with granulocyte colony-stimulating factor at a dose of 12 µg/kg twice dailysubcutaneously for four consecutive days. Cryopreserved peripheral blood progenitor cells were reinfused 48 h aftercompletion of chemotherapy.

With a median follow-up of 21 months (range 5–64) five complete responses were observed; one patient hadpartial response and one had stable disease. There was no treatment-related mortality. The 2 year event-free survivalwas 71.43 ± 17%.

Therefore we conclude that HDCT as consolidation regimen may improve the cure rates in very young childrenwith medulloblastoma/stPNET avoiding long-term sequelae of radiotherapy.

Introduction

Treatment for children with medulloblastoma orsupratentorial primitive neuroectodermal tumors(stPNET) includes surgical resection, radiotherapy andchemotherapy. The survival of young children withmedulloblastoma or stPNET is significantly worsethan for older children and adults [1]. The likely expla-nations for the poorer survival of these young childrenare the different biologic characteristics of their tumorsand the fact that they received a lower therapeuticradiotherapy dosage [2].

Recently, high-dose chemotherapy (HDCT) withautologous stem cell rescue has been used as an adju-vant therapy or as salvage therapy to treat pediatricpatients with brain tumors, and to avoid deleteri-ous side-effects of radiotherapy for infants and veryyoung children [3–6]. However, few studies havedemonstrated the utility of HDCT/ASCT in the frontline treatment of young unirradiated children withmedulloblastoma and stPNET [7,8].

We report our experience with HDCTand ASCT in seven children younger than4 years old.

102

Table 1. Patient characteristics

Sex Age Diagnosis Staging Surgery Radiotherapy Standard Status at(months) chemotherapy transplantation

Patient 1 Male 47 MB T2 Total NO SFOP CRPatient 2 Male 34 stPNET T2 Partial NO SFOP SDPatient 3 Male 29 MB T1 Subtotal NO SFOP CRPatient 4 Male 14 MB T2 Total NO SFOP CRPatient 5 Female 31 MB T1 Total NO SFOP CRPatient 6 Male 47 stPNET T2 Partial NO SFOP PRPatient 7 Male 36 MB T1 Total NO SFOP CR

Diagnosis: MB: medulloblastoma; stPNET: supratentorial primitive neuroectodermal tumors.Induction chemotherapy: SFOP: French Society of Pediatric Oncology.Status at transplantation: CR: complete remission; SD: stable disease; PR: partial remission.

Patients and methods

Patients

Between 1997 and 2002, seven children younger than4 years old with pathologic confirmation of medullo-blastoma or stPNET underwent a treatment sched-ule that consisted of surgical debulking, standardchemotherapy and HDCT and autologous stem cellrescue as first line treatment.

Maximal surgical resection of the tumor wasattempted in all patients. None received craniospinalirradiation. After surgery, all children received stan-dard chemotherapy according to the French Society ofPediatric Oncology (SFOP) that comprised cycles ofalternating courses given every 21 days [9]. Course 1consisted of carboplatin (15 mg/kg on day 1) given in1 h by intravenous infusion and procarbazine (4 mg/kgon days 1–7) given orally. Course 2 consisted of etopo-side (5 mg/kg in days 1 and 2) given in 1 h intravenousinfusion and cisplatin (1 mg/kg on days 1 and 2) givenin 4 h intravenous infusion. Course 3 consisted ofvincristine (0.05 mg/kg on day 1) given by rapid intra-venous infusion, and cyclophosphamide (50 mg/kg onday 1) given in 1 h by intravenous infusion withMesna.

Tumor staging at the time of initial diagnosis wasdone according to the Chang system [10].

Before HDCT normal cardiac function (ejectionfraction >45% or shortening fraction >30% byechocardiogram), normal renal function (serum creati-nine<1 mg/dl or creatinine clearance>70 ml/min/m2),normal liver function (AST < 1.5 times normal andbilirubin <1.5 mg/dl), normal bone marrow func-tion (hemoglobin >10 g/dl, WBC count >3000 µl,

absolute neutrophil count (ANC) >1500 µl andplatelets >100,000 mm3) and a Lansky score >70 wererequired [11].

The main characteristics of the patients are shown inTable 1.

Transplant procedure

All patients had PBSCs mobilized with granulo-cyte colony-stimulating factor (G-CSF, Neupogen®,Amgen, Thousand Oaks, CA, USA) at a dose of12 µg/kg twice daily subcutaneously for four con-secutive days before starting apheresis, as previouslyreported [12].

Peripheral blood progenitor cells collection by large-volume leukapheresis were performed on day +5 aftermobilization by a Cobe Spectra cell separator (CobeBCT, Lakewood, CO, USA).

Each apheresis product was analyzed for CD34+cell content by flow cytometry using an Epics Eliteflow cytometer (Coulter Corporation, Hialeah, Florida,USA). The final product containing 10% dimethyl sul-foxide was frozen using a computer-controlled freezerand stored in liquid nitrogen at −196◦C.

Conditioning regimen consisted of busulfan4 mg/kg/day, orally over 4 days (from days −5 to−2) in 6 hourly divided doses and melphalan ata dose of 140 mg/m2/day by intravenous infusionover 5 min on day −1. Three patients addition-ally received thiotepa 250 mg/m2/day intravenouslyover 2 days (from days −3 to −2) and two patientsadditionally received topotecan 2 mg/m2/day over5 days by intravenous infusion over 30 min (fromdays −11 to −7).

103

On day 0, collected peripheral blood progenitor cellswere re-infused after rapid thawing at 37◦C.

All patients received G-CSF intravenously(10 µg/kg/day) starting on day +1 until an ANC>1 × 109/l was maintained for two consecutive days.Version 2.0 of National Cancer Institute CommonToxicity Criteria was used to grade toxicity [13].

Supportive care

After placement of a central venous line, patientswere admitted to the hospital for HDCT and werenursed under strict protective isolation in barrier nurs-ing units with HEPA-filtered air. Infection prophylaxiswas provided with cotrimoxazole.

Intravenous clonazepam was prescribed from thefirst day of busulfan to the day of infusion in order toavoid busulfan-related seizures.

Platelet transfusions were administered to main-tain a platelet count higher than 20,000 mm3, andtransfusions of irradiated packed red blood cells wereadministered to maintain a hematocrit concentrationhigher than 25%. All blood derived transfusions wereirradiated prior to use. Empiric broad-spectrum anti-biotic treatment was initiated as soon as fever of >38◦Coccurred. Amphotericin (1 mg/kg/day) was added iffever and neutropenia continued for 5 days after anti-biotics were started. In addition, any patient with a>10% weight loss from the time of starting therapyreceived nutritional support using parenteral nutritionor nasogastric feeding.

Definitions

The extent of surgical resection was defined as follows:a partial resection if reduction was greater than 10% butless than 90% of the tumor mass; a subtotal resection ifreduction was greater than 90% of the tumor mass, butvisible tumor remained; and a gross total resection if notumor was visible on postoperative magnetic resonanceimaging or computed tomographic scan [7].

Neutrophil recovery was defined as the numberof days taken to achieve an ANC >0.5 × 109/l forthree consecutive days. Platelet recovery was definedas the time taken to achieve >20 × 109/l for threeconsecutive days without requiring transfusion.

Standard radiographic criteria were used to deter-mine response. Complete response (CR) was definedas complete resolution of all lesions and no clini-cal progression. Partial response (PR) was defined as

a >50% reduction in the product of the maximumperpendicular diameters of all lesions and no clinicalprogression. Stable disease (SD) was defined as <50%reduction and <25% increase in size with no clinicalprogression, and progressive disease was defined as>25% increase in size [7].

Informed consent

Potential treatment risks and benefits, supportive careand therapeutic options were explained and discussed.Signed informed consent was obtained for each childfrom parents or legal guardians.

Statistical analysis

Data are expressed as median and range. Event-freesurvival (EFS) was assessed from the date ofASCT to the date of disease progression or death.Distributions of EFS were estimated using the methodof Kaplan–Meier [14].

Results

Standard treatment

Between August 1997 and December 2002, seven chil-dren (six boys and one girl) younger than 4 years withnewly diagnosed medulloblastoma (five patients) orstPNET (two patients), were enrolled. Median age atdiagnosis was 34 months (14–48).

Complete surgical resection was achieved in fourpatients; subtotal resection was achieved in one patientand two patients had partial resections. All childrenreceived standard chemotherapy. None was irradiated.At entry into consolidation with HDCT, five patientshad CR, one had PR and one SD.

Toxicity of consolidation chemotherapy withHDCT and engraftment kinetics

No patient developed treatment-related mortality.A median of 7.1 (2.4–4.8) × 106 kg peripheral

blood progenitor cells were re-infused on day 0.Hematological recovery was achieved in all patients.The median time to neutrophil count recovery was9 days (7–11) and the median time to platelet recoverywas 12 days (8–60).

104

All patients experienced myelosuppression grade IV,requiring a median of three packed red blood cellsand two platelet transfusions and G-CSF treatment. Allpatients developed febrile neutropenic episodes andreceived treatment with broad-spectrum antibiotics andone patient required the addition of amphotericin B forpersistent fever. However, no patient had documentedfungal infections. Only two febrile episodes duringthe neutropenia period were confirmed by clinicalobservation or positive culture: one patient developeda cutaneous vesicular eruption that resembled chicken-pox and another one had catheter-related infection bycoagulase negative staphylococci. Both children weresuccessfully treated.

Nausea, vomiting and mucositis were universal. Allpatients required parenteral narcotics for pain con-trol, and intravenous alimentation was required in fivepatients.

One patient developed engraftment syndromerequiring mechanical ventilation and high dose steroidsto resolve it [15]. One patient developed veno-occlusivedisease that was resolved with supportive treatment3 weeks later.

The main characteristics of the toxicity andengraftment kinetics are shown in Table 2.

Response to consolidation chemotherapy withHDCT and survival

All patients kept the disease status at transplanta-tion: five had CR, one had PR and one SD. As ofMarch 2003, four patients are alive and tumor free(medulloblastomas), one patient has PR (stPNET)and two patients (one medulloblastoma and onestPNET) died from disease progression at 12 and5 months after HDCT, respectively. Among the sur-viving patients, the median follow-up post-HDCT was

21 months. The 2-year Kaplan–Meier estimates of EFSrates was 71.43 ± 17%.

Discussion

Treatment of very young children with medulloblas-toma and stPNET is particularly challenging becauseof the aggressive nature of these tumors and thepotential long-term sequelae related to treatment [16].Disease-free survival for these brain tumors in childrenyounger than 4 years old is associated with a survivalrate of less than 20% [3]. Furthermore, the adverseconsequences of craniospinal irradiation, in particu-lar, developmental delay, central endocrinopathies andintellectual dysfunction, are markedly increased in thisgroup of patients and they should be unacceptable [7].

All our patients were defined as high risk medul-loblastoma and stPNET according to age criteria atdiagnosis (less than 4 years old) [1].

Generally, HDCT regimen has been used in somerefractory and relapsed pediatric brain tumors [7,17].Published experience with HDCT as first line oftreatment for children younger than 4 years, diag-nosed with medulloblastoma/stPNET and unirradiatedis scanty [7,8]. Megatherapy with hematopoietic rescuefor patients younger than 4 years with medulloblastomaas first line treatment has not been systematically stud-ied, as it has been for patients with other solid tumors[18,19].

This paper describes the outcome of a homogeneousgroup of pediatric patients that were treated withHDCT. In our experience, seven children receivedHDCT with peripheral stem cell rescue, and five ofthem are alive without sequelae. Disease-free sur-vival was 71.43 ± 17% with a median of follow-upof 21 months (5–11). Although our study population

Table 2. Conditioning regimens, engraftment kinetics and course

Conditioning CD34+/kg NR PTR Mucositis Follow-up Outcomeregimen (days) (days) (months)

Patient 1 BU-TI-ME 5.2 × 106 8 60 IV 64 CRPatient 2 BU-ME 7.1 × 106 11 12 II 12 DODPatient 3 BU-TI-ME 16.6 × 106 7 10 II 31 CRPatient 4 BU-TI-ME 24.8 × 106 9 10 IV 21 CRPatient 5 BU-TO-ME 3.27 × 106 8 13 IV 21 CRPatient 6 BU-TO-ME 2.4 × 106 10 8 III 21 PRPatient 7 BU-ME 13.5 × 106 11 14 IV 5 DOD

NR: neutrophil recovery; PTR: platelet recovery; BU-TI-ME: busulfan-thiotepa-melphalan; BU-ME: busulfan-melphalan;BU-TO-ME: busulfan-topotecan-melphalan; CR: complete remission; PR: partial remission; DOD: died of disease.

105

is small our results and those of other groups indi-cate that HDCT could play an important role as firstline treatment in children less than 4 years old withmedulloblastoma/stPNET [7,20].

Transplant-related toxicities and mortality have beenreported to be higher in children with brain tumorsthan other solid tumors. However, in our series, therewere no toxicity-related deaths and the procedure waswell-tolerated.

Patient with good general condition and stageof disease at transplantation may have reduced ourtoxicity-related complications.

To date, the best conditioning regimen for thesepatients has not yet been established. A busulfan/melphalan-based conditioning regimen may be effec-tive in medulloblastoma and stPNET treatment, as ithas been reported in other solid tumors of childhood[21]. The addition of a third drug such as an thiotepaor topotecan may increase tumor response withoutincreasing transplant-related toxicity [22].

Interestingly, neither patient with stPNET achieveda CR to induction chemotherapy and one of the tworelapsed early. The foregoing finding could be dueto the tumor localization. The outcome is known tobe worse for the supratentorial than the infratentorialtumor localization.

Furthermore, our patients who were in CR at trans-plantation remained in this status after transplanta-tion which indicates that HDCT may be useful asconsolidation therapy.

Our preliminary results suggest that childrenyounger than 4 years with medulloblastoma/stPNETmay benefit from HDCT as consolidation therapyallowing a delay or elimination of irradiation and withwell-tolerated toxicities.

Acknowledgements

The authors thank Dr. Manuel Ramırez for his help inwriting the manuscript.

References

1. Allon JC, Epstein F: Medulloblastoma and other primarymalignant neuroectodermal tumor of the CNS: the effectof age and the patients extent of disease on prognosis.J Neurosurg 57: 446–451, 1982

2. Zeltzer P, Boyett J, Finlay J, Albright A, Rorke L, Milstein J,Allen J, Stevens K, Stanley P, Li H, Wisoff J, Geyer R,

McGuire-Cullen P, Stehbens J, Shurin S, Packer R: Metas-tasis stage, adjuvant treatment, and residual tumor areprognostic factors for medulloblastoma in children: conclu-sions from the Children’s Cancer Group 921 randomizedphase III study. J Clin Oncol 17: 832–845, 1999

3. Dunkel IJ, Finlay JL: High-dose chemotherapy with autol-ogous stem cell rescue for brain tumors. Crit Rev OncolHematol 41: 197–204, 2002

4. Graham ML, Herndon JE II, Casey JR, Chaffee S,Ciocci GH, Krischer JP, Kurtzberg J, Laughlin MJ,Longee DC, Olson JF, Paleologus N, Pennington CN,Friedman HS: High-dose chemotherapy with autologousstem-cell rescue in patients with recurrent and high riskpediatric brain tumors. J Clin Oncol 15: 1814–1823, 1997

5. Kalifa C, Valteau D, Pizer B, Vassal G, Grill J, Hartmann O:High-dose chemotherapy in childhood brain tumours.Child’s Nerv Syst 15: 498–505, 1999

6. Jenkin D, Danjoux C, Greenberg M: Subsequent quality oflife for children irradiated for a brain tumor before age fouryears. Med Pediatr Oncol 31: 506–511, 1998

7. Mason WP, Grovas A, Halpern S, Dunkel IJ, Garvin J,Heller G, Rosenblum M, Gardner S, Lyden D, Sands S,Puccetti D, Lindsley K, Merchant TE, O’Malley B, Bayer L,Petriccione MM, Allen J, Finlay JL: Intensive chemotherapyand bone marrow rescue for young children with newly diag-nosed malignant brain tumors. J Clin Oncol 16: 210–221,1998

8. Finlay JL, Grovas Am, Garvin J, Dunkel I, Bayer L,Pucetti D, Halpern S, Drachtman R, Goldman S, Tugal O,Jayabose S, Weinblatt M, Sadhev I, Allen JC: The‘head start’ regimen for children less than 6 years of agenewly-diagnosed with malignant brain tumors. Med PediatrOncol 25: 250 (# O-69), 1995

9. Marec-Berare P, Jouvet A, Thiesse P, Kalifa C, Koz F,Frappaz D: Supratentorial embryonal tumors in childrenunder 5 years of age: an SFOP study of treatment withpostoperative chemotherapy alone. Med Pediatr Oncol 38:83–90, 2002

10. Laurent J, Chang CH, Cohen M: A classification system forprimitive neuroectodermal tumors (medulloblastoma) of theposterior fossa. Cancer 56: 1807–1809, 1985

11. Lansky SB, List MA, Lansky LL, Ritter-Sterr C, Miller D:The measurement of performance in childhood cancerpatients. Cancer 60: 1651–1656, 1987

12. Sevilla J, Gonzalez-Vicent M, Madero L, Garcıa-Sanchez F,Diaz MA: Granulocyte colony-stimulating factor alone at12 µg/kg twice a day for 4 days for peripheral blood pro-genitor cell priming in pediatric patients. Bone MarrowTransplant 30: 417–420, 2002

13. Trotti A, Byhardt R, Stetz J, Gwede C, Corn B, Fu K,Gunderson L, McCormick B, Morrisintegral M, Rich T,Shipley W, Curran W: Common toxicity criteria: version 2.0.An improved reference for grading the acute effects ofcancer treatment: impact of radiotherapy. Int J Radiat OncolBiol Phys 47: 13–47, 2000

14. Kaplan EL, Meier P: Non-parametric estimation fromincomplete observations. J Am Stat Assoc 3: 457–481, 1958

15. Madero L, Vicent MG, Sevilla J, Prudencio M, Rodrıguez F,Dıaz MA: Engraftment syndrome in children undergoing

106

autologous peripheral blood progenitor cell transplantation.Bone Marrow Transplant 30: 355–358, 2002

16. Mulhern R, Kepner J, Thomas P, Armstrong F, Friedman H,Kun L: Neuropsychologic functioning of survivors of child-hood medulloblastoma randomized to received conven-tional or reduced-dose craniospinal irradiation: A PediatricOncologic Group study. J Clin Oncol 16: 1723–1728, 1998

17. Guruangan S, Dunkel IJ, Goldman S, Garvin JH,Rosenblum M, Boyett JM, Gardner S, Merchant TE,Gollamudi S, Finlay JL: Myeloablative chemotherapywith autologous bone marrow rescue in young childrenwith recurrent malignant brain tumors. J Clin Oncol 16:2486–2493, 1998

18. Pinkerton Cr: ENSG 1-randomised study of high-dosemelphalan in neuroblastoma. Bone Marrow Transplant7(Suppl 3): 112–113, 1991

19. Meyers PA, Gorlick R, Heller G, Casper E, Lane J,Huvos AG, Healey JH: Intensification of preoperativechemotherapy for osteogenic sarcoma: results of theMemorial Sloan-Kettering (T12) protocol. J Clin Oncol 16:2452–2458, 1998

20. Schmandt S, Kuhl J: Chemotherapy as prophylaxis andtreatment of meningosis in children less than 3 years ofage with medulloblastoma. J Neuro-Oncol 38: 187–192,1998

21. Dıaz MA, Vicent MG, Madero L: High-dose busulfan/melphalan as conditioning for autologous PBPC transplan-tation in pediatric patients with solid tumors. Bone MarrowTransplant 11: 1157–1159, 1999

22. Kalifa C, Hartmann O, Demeocq F, Vassal G, Couanet D,Terrier-Lacombe MJ, Valteau D, Brugieres L, Lemerle J:High-dose busulfan and thiotepa with autologous bone mar-row transplantation in childhood malignant brain tumors:a phase II study. Bone Marrow Transplant 9: 227–233,1992

Address for offprints: L. Madero, Department of PediatricHematology and Oncology, Hospital Nino Jesus, MenendezPelayo 65, 28009 Madrid, Spain; Tel.: 34 91-5035900/478; Fax:34 91-5744669; E-mail: lmadero.hnjs@salud.madrid.org