• brainstem tumors;
  • radiotherapy;
  • chemotherapy;
  • children


  1. Top of page
  2. Abstract


The current Phase II study was conducted to evaluate the survival and toxicity observed in children with newly diagnosed brainstem gliomas who were treated with the daily radiotherapy with topotecan used as a radiosensitizer.


Eligible patients were those ages 3–18 years with previously untreated tumors arising in the pons diagnosed within the previous 6 months. Histologic confirmation was not mandatory provided that the clinical and magnetic resonance imaging findings were typical for a diffusely infiltrating brainstem lesion. Treatment was comprised of a 6-week course of topotecan administered intravenously at a dose of 0.4 mg/m2/day over 30 minutes within 1 hour before irradiation. Radiotherapy was comprised of a once-daily treatment of 1.8 grays (Gy) per fraction to a total dose of 54 Gy.


Thirty-two patients were included in the current study between August 2000 and October 2002. All patients completed the combined treatment in accordance with the treatment design. Only partial responses were observed, occurring in 40% of the patients. The 9-month and 12-month survival rates were 34.4% ± 8% and 25.5% ± 8%, respectively. The median duration of survival for these 32 patients was 8.3 months. An intratumoral cystic/necrotic change was observed in five patients, with clinical impairment noted in two patients. One intratumoral hemorrhage occurred during radiotherapy, and was associated with transitory neurologic impairment.


The findings of the current study regarding newly diagnosed brainstem glioma patients treated with topotecan given as a radiosensitizing agent did not reproduce the encouraging results obtained in preclinical studies. Therefore, the concomitant combination of topotecan and radiotherapy at this schedule and these doses cannot be recommended for the treatment of patients with brainstem gliomas. Cancer 2005. © 2005 American Cancer Society.

Malignant diffuse brainstem gliomas, which constitute at least 10% of all childhood brain tumors, still retain a dismal prognosis regardless of the treatment used. Surgery is not feasible without a high risk of life-threatening neurologic complications and is not recommended in patients with intrinsic tumors.1 Conventional radiotherapy, delivering 55–60 grays (Gy), results in a transitory improvement of neurologic signs and symptoms and is associated with a median survival of 9 months.2, 3 Several trials using hyperfractionated radiotherapy have attempted to improve the prognosis, without success.4

The use of chemotherapy also has been assessed using conventional sequential regimens,5–7 a continuous regimen before and during radiotherapy,8 chemotherapy concomitant with hyperfractionated radiotherapy,9 and high-dose chemotherapy followed by hematopoietic stem cell rescue.10, 11 To our knowledge, the role of chemotherapy delivered daily, concomitant with each ionizing radiation fraction as radiosensitizer, has not yet been assessed for the treatment of brain stem gliomas in a Phase II trial.

We previously tested the efficacy of this type of approach in preclinical models and demonstrated that topotecan, a topoisomerase II inhibitor, could lead to a synergistic effect when given concomitantly with radiotherapy.12, 13 Topotecan is a water-soluble derivative of camptothecin, stabilizing the DNA topoisomerase I complex and thereby interfering with DNA replication and DNA repair.14 Taking into account the radiosensitizing properties of topotecan that have been reported both in vitro and in vivo by other authors,15, 16 we designed a Phase II trial for children diagnosed with a diffuse brainstem glioma. Radiotherapy was delivered daily at a conventional dose, after the administration of a low dose of topotecan, which was used as a radiosensitizer. The daily dose of topotecan was chosen according to the maximum tolerated dose observed in the Children's Cancer Group-0952 Phase I study, in which topotecan and radiotherapy were administered over 6 weeks.17 In the current study, we report the results of the French Society of Paediatric Oncology (SFOP) prospective, multicentric study, which to our knowledge is the first to assess the role of concomitant chemotherapy and radiotherapy in a Phase II study for the treatment of malignant brain stem gliomas occurring in children.


  1. Top of page
  2. Abstract

Inclusion Criteria

Newly diagnosed patients ages 3–18 years with malignant brainstem tumors were eligible for the current study. The diagnosis of diffuse brainstem glioma was based on magnetic resonance imaging (MRI) criteria. A biopsy was not mandatory except in the case of an atypical presentation or an uncertain diagnosis. If hydrocephalus was found to be present on imaging and was associated with signs of increased cranial pressure, cerebrospinal fluid diversion then was recommended before initiating the protocol.

No performance criteria were required for entry onto the study. No treatment other than corticosteroids was allowed; it was recommended that corticosteroids be tapered as clinically tolerated. Information was provided according to French law and informed written consent was required before the initiation of the therapy. The protocol was approved by the Brain Tumor committee of the SFOP, the ethical committee, and by each local institutional review board.


The concomitant chemoradiotherapy was to be initiated as soon as possible after diagnosis.


Strict guidelines for the administration of radiotherapy were given. Patients were to receive a single daily fraction of 1.8 Gy given 5 days per week for 30 fractions; the treatment volume was to include the extent of the tumor based on MRI assessment, with a minimum margin of 2 cm beyond the tumor in all dimensions. The planned total dose was 54 Gy delivered over the course of 6 weeks.


Topotecan (GlaxoSmithKline, Marly-le-Roi, France) was administered intravenously over 30 minutes within 1 hour before radiotherapy, at a dose of 0.4 mg/m2/day.

Assessment of Response and Toxicity

A detailed neurologic and general clinical examination was performed at baseline and then every week during the treatment, followed by monthly or more according to clinical signs and symptoms. The dose level of steroid needed was registered. A complete blood count was required once or twice per week according to hematologic toxicity. In cases of neutropenia with an absolute neutrophil count of < 0.5 × 109/mm3, no treatment with granulocyte–colony-stimulating factor was recommended. Platelet transfusions were recommended in instances of thrombocytopenia < 50 × 109/mm3.

Response was assessed based on clinical signs and on MRI scans performed 1 month after the completion of the treatment, and every 3 months thereafter. The response rate was definitively established by central imaging review.

Statistical Analysis

The statistical analysis of this Phase II study was based on a triangular test.18 The α risk (error of the first kind) was 10% and the β risk was 5%. A 9-month survival rate of < 50% or > 70% was considered to be the upper limit for ineffective treatment and the lower limit for effective treatment, respectively. The 9-month survival rate was analyzed for every 5 patients followed for at least 9 months. When performing an interim analysis, cumulative results were reported on the triangular figure. The trial was to be stopped if either the upper boundary was crossed (rejection of H0, benefit from the proposed strategy [i.e., a 9-month survival rate > 50%]), or if the lower boundary was crossed (nonrejection of H0, no evidence of benefit). Provided no boundary was crossed, additional patients were to be included and/or followed.


  1. Top of page
  2. Abstract

Between August 2000 and October 2002, 32 patients from 10 centers were entered onto the study. No patient was excluded from the analysis. There were 14 males and 18 females, ranging in age from 3–16.8 years (median, 7 yrs). All the patients had at least 2 of the three typical neurologic signs, namely, cranial nerve deficits, long tract signs, and ataxia of less than 6 months' duration (median, 2 mos). The baseline median Lansky score was 60 (range, 20–100). All 32 tumors involved at least two-thirds of the entire brainstem. Seven patients underwent biopsy. The histologic findings were high-grade glioma in all patients. Two patients underwent a cerebrospinal fluid diversion procedure at the time of diagnosis.

All the patients completed the combined treatment in accordance with the study design.


A clinical improvement was observed in 16 patients during the 2 months of treatment. Fourteen patients did not receive steroid therapy during this period. In the 18 patients treated with steroids, the steroids were withdrawn or decreased in 30% of the patients but could not be decreased in 43% and had to be initiated in 20% of patients.

Neuroimaging from all the patients with the exception of four who were not reported as responders was centrally reviewed. Only partial responses were achieved and were observed in 40% of the patients. It is interesting to note that an intratumoral cystic/necrotic change was observed in five patients, with clinical impairment reported in two patients. One intratumoral hemorrhage was reported to occur during radiotherapy and was associated with transitory neurologic impairment.


The 9-month survival rate was 34.4% ± 8% and the 12-month survival rate was 25.5% ± 8%. The median survival duration of the 32 patients in the current study was 8.3 months (Fig. 1). Although none of the limits as defined by the initial statistical design had been reached, we decided to close the study because of the lack of improvement noted in patient survival (Fig. 2). At the time the study was stopped, 23 patients had a follow-up of 9 months or longer; among these patients, 13 were dead at 9 months of follow-up. None of the 32 study patients survived longer than 29 months.

thumbnail image

Figure 1. Overall survival of all patients enrolled in the current study who were treated with topotecan and radiotherapy.

Download figure to PowerPoint

thumbnail image

Figure 2. Triangular test corresponding to the hypotheses, H0 (P ≤ 0.50) and H1 (P > 0.50), with p1 = 0.70 (α5 = 0.10 and β = 0.05). The upper and lower boundaries are given by Z = S - Np0 and V = Np0 (1 - p0), with S representing the total number of patients free of the event considered (death only or death or disease progression), and N representing the total number of patients participating in the analysis. If the path crosses the upper boundary, it is then concluded that there is a benefit from the proposed strategy. If the path crosses the lower boundary, it is then concluded that there is no evidence of a benefit from the proposed strategy.

Download figure to PowerPoint


Episodes of Grade 4 neutropenia, anemia, and thrombocytopenia were reported to occur in 11%, 34%, and 0% of the patients, respectively. Neutropenia was associated with a nonlife-threatening infection in five patients (one patient with bacteremia and four patients with urinary infections).

One death was the result of intratumoral hemorrhage that was not associated with an increase in the tumor volume or thrombocytopenia.


  1. Top of page
  2. Abstract

Diffuse, infiltrating gliomas of the brainstem currently have a grim prognosis regardless of the treatment used and new strategies are therefore needed. Death typically occurs after local disease progression within the field of irradiation. The use of chemotherapy reportedly resulted in only very poor response rates (cumulative response rate of < 5%)19 and did not consistently increase the median survival, which remains between 8–12 months.20 Various therapeutic approaches have been tested using sequential, continuous, or high-dose chemotherapy regimens.20 The use of radiosensitizers or chemoradiotherapy has to our knowledge been reported only rarely.9, 21 Several in vitro and in vivo studies have demonstrated the radiosensitizing properties of topotecan, a topoisomerase I inhibitor with activity against malignant gliomas16 and the ability to penetrate the blood-brain barrier (approximately 30%).22 This activity appears to be schedule dependent, S-phase specific, and not related directly to drug cytotoxicity.23 Finally, the combination of topotecan and radiotherapy was found to provide a radiopotentiation ratio of ≥ 2 in a preclinical model.16 To evaluate the efficacy and toxicity of the concomitant daily administration of topotecan and radiotherapy, we performed preclinical studies on high-grade glioma xenografts.12, 13 In a first study, a radiopotentiation ratio of 1.8 was obtained on the U87 glioblastoma xenografted onto nude mice using radiotherapy at a dose of 20 Gy given over 5 consecutive days (4 Gy/day × 5) combined with topotecan administered concomitantly at a daily dose of 1.5 mg/kg. Moreover, the tumor growth delay observed with this regimen was equivalent to that obtained with radiotherapy alone at a total dose of 41 Gy. Subsequently, we performed preclinical studies on 2 high-grade glioma xenografts to determine whether the detrimental effect of the prolongation of overall treatment time (40 Gy delivered over 1 week, 2 weeks, 4 weeks, or 6 weeks, with the goal of being closer to actual clinical practice) on the efficacy of radiotherapy could be compensated by the concomitant administration of topotecan.12 We demonstrated that the combined treatment not only allowed for compensation of the detrimental time effect, but induced a synergistic effect (radiopotentiation ratio of 3.7) on the most radioresistant glioma model (U87) as well. We then designed a Phase II trial based on the encouraging results observed in these preclinical studies. The dose of daily topotecan was chosen according to the maximum tolerated dose observed in the Children's Cancer Group-0952 Phase I trial, using the same regimen (but with a total dose of radiation of 59.4 Gy) in which the limiting factor was hematologic.16 Based on this toxicity and the potential significant radiopotentiation, we decided not to increase the daily dose of topotecan out of concern for brainstem necrosis and/or hemorrhages.

The results of the current study were disappointing, with a 9-month survival rate of 34% ± 8% and a median survival of 8.3 months. In addition, these results did not compare favorably with those of our previous study using different strategies.12, 13 In the study in which sequential courses of carboplatin were administered (2 courses prior to radiotherapy followed by a chemoradiotherapy phase including 5 courses of carboplatin), the 9-month survival rate was 65.8% ± 15% and the median survival for the 38 study patients was 11 months.5 Using a strategy of continuous administration of procarbazine (at a dose of 150 mg/m2/day for 28 days followed by conventional radiotherapy and 3 additional courses of procarbazine in case of response) assessed in 22 patients, the survival rate at 9 months was 54% ± 20%.8

Using a high-dose chemotherapy strategy (busulfan at a dose of 600 mg/m2 over 4 days plus thiotepa at a dose of 900 mg/m2) followed by autologous bone marrow transplantation in 21 eligible patients, the median survival time was 10 months (range, 3–26 mos).10 The same results were recently reported from studies in which carboplatin was administered twice weekly along with radiotherapy (with a median survival of 12 mos reported),24 tamoxifen was administered during and after radiotherapy (with a median survival of 10 mos reported),25 trophosphamide plus etoposide was given (with a median survival of 8 mos reported),26 and etanidazole was administered concomitantly with radiotherapy (with a median survival of 8.5 mos reported).26 A summary of these results are presented in Table 1.

Table 1. Overall Survival Reported in Recent Series Using Cytotoxic Agents in the Treatment of Patients with Malignant Brainstem Gliomas
StudiesCytotoxic agents used9-mo OSMedian OS in mos
  1. OS: overall survival; SFOP: French Society of Paediatric Oncology; VP-16: etoposide; CBDCA: carboplatin; VCR: vincristine; CDDP: cisplatin; CPM: cyclophosphamide.

Current studyTopotecan41%8.5
SFOP strategies   
Sequential regimen5Carboplatin67%11
Continuous regimen8Procarbazine54%10
High-dose chemotherapy10Busulfan + thiotepa58%10 
Other institutional strategies   
Broniscer et al.25Tamoxifen 10
Allen et al., 199924Carboplatin 12
Wolff et al.,200226Trophosphamide + VP-16 8
Jennings et al.33CBDCA/VP-16/VCR 8.5

In the Children's Cancer Group-0952 study using the combination of topotecan and radiotherapy in 16 patients with brainstem gliomas, a median survival of 15 months was reported (95% confidence interval, 9.6–19 mos).28 The estimated 1-year probability of survival was 53%. This Phase I trial tested the concomitant association between topotecan given as a 30-minute intravenous infusion 30–60 minutes before each radiation treatment daily for 33 days (rather than 30 days, as in the current study) with a total irradiation dose of 59.4 Gy (vs. 55 Gy in the current study). However, these relatively good results could be due to a sample effect rather than the three supplementary doses of the topotecan and radiotherapy regimen.

The disappointing results of the current study could be explained by the relatively low dose of daily topotecan administered. However, the radiosensitivity properties of topotecan have also been tested at similar dosages in two Phase I studies and one Phase II study of adult patients treated for high-grade glioma. In the series by Grabenbauer et al., the recommended dose for topotecan was 0.6 mg/m2/day.29 The dose-limiting toxicity was hematologic. Of 17 patients who were evaluable for response and toxicity, 3 achieved a partial response and 12 achieved stable disease. In the series by Lesimple et al., topotecan was administered as a continuous infusion from Day 3 to Day 5 on Weeks 1, 3, and 5, and during radiotherapy (60 Gy given in 30 fractions over 40 days).30 The dose-limiting toxicity was hematologic; the recommended dose was 0.9 mg/m2/day. The survival data assessed regarding 20 patients reported a median time to disease progression of 17 weeks for those patients who underwent a partial resection. In the Phase II study conducted by the Radiation Therapy Oncology Group (RTOG 9513 study), 87 adult patients received topotecan at a dose of 1.5 mg/m2/day over the course of 5 days every 3 weeks for 3 cycles both during and after radiotherapy (60 Gy given as 30 fractions over 6 weeks).31 The toxicity was primarily hematologic but two Grade 4 neurotoxicities were reported. The median survival was 9.3 months. No statistically significant difference was noted between the survival of patients in the current study and that of the matched patients from the RTOG database. No child in the current study died of radiation-induced brain necrosis, although the cause of death in one child was found to be related to intratumoral hemorrhage without thrombocytopenia. The cystic/necrotic changes observed in the current study were not unexpected because, in previous studies, a syndrome of intralesional cystic/necrotic changes noted on MRI scan within the first 6 weeks after the completion of radiotherapy, associated with clinical deterioration, also was reported after radiotherapy alone, particularly when hyperfractionated, high-dose radiotherapy was used.32 It is not possible to conclude that the increase in toxicity resulting from radiotherapy noted in the current study is different from our previous Phase II study in the same patients.


The results of the current study, with a 9-month survival rate of 34.4%, do not compare favorably with those obtained in our 3 previous SFOP studies, in which different regimens of chemotherapy were combined with radiotherapy. Therefore, the concomitant combination of topotecan and radiotherapy at the schedule and doses used in the current study cannot be recommended for the treatment of patients with brainstem gliomas.


  1. Top of page
  2. Abstract
  • 1
    Albright AL, Packer RJ, Zimmermann R, et al. Magnetic resonance scans should replace biopsies for the diagnosis of diffuse brainstem gliomas: a report of the Children's Cancer Group. Neurosurgery. 1993; 33: 10261029.
  • 2
    Kaplan AM, Albright AL, Zimmerman RA, et al. Brainstem gliomas in children. A Children's Cancer Group review of 119 cases. Pediatr Neurosurg. 1999; 24: 185192.
  • 3
    Guillamo JS, Doz F, Delattre JY. Brain stem gliomas. Curr Opin Neurol. 2001; 14: 711715.
  • 4
    Mandell L, Kadota R, Freeman C, et al. There is no role for hyperfractionated radiotherapy in the management of children with newly diagnosed diffuse intrinsic brainstem tumors: results of a Pediatric Oncology Group phase III trial comparing conventional vs. hyperfractionated radiotherapy. Int J Radiat Oncol Biol Phys. 1999; 43: 959964.
  • 5
    Doz F, Neuenschwander S, Bouffet E, et al. Carboplatin before and during radiation therapy for the treatment of malignant brain stem tumours: a study by the Societe Francaise d'Oncologie Pediatrique. Eur J Cancer. 2002; 38: 815819.
  • 6
    Lashford LS, Thiesse P, Jouvet A, et al. Temozolomide in malignant gliomas of childhood: a United Kingdom Children's Cancer Study Group and French Society for Pediatric Oncology Intergroup Study. J Clin Oncol. 2002; 20: 46844691.
  • 7
    Broniscer A, Iacono L, Chintagumpala M, et al. Role of temozolomide after radiotherapy for newly diagnosed diffuse brainstem glioma in children: results of a multiinstitutional study (SJHG-98). Cancer. 2005; 103: 133139.
  • 8
    Gentet JC, Frappaz D, Pautard B, et al. Procarbazine in children with malignant brainstem glioma: a SFOP phase II study [abstract P-37]. Med Pediatr Oncol. 1999; 33: 205.
  • 9
    Freeman CR, Kepner J, Kun LE, et al. A detrimental effect of a combined chemotherapy-radiotherapy approach in children with diffuse intrinsic brain stem gliomas? Int J Radiat Oncol Biol Phys. 2000; 47: 561564.
  • 10
    Bouffet E, Raquin M, Doz F, et al. Radiotherapy followed by high dose busulfan and thiotepa: a prospective assessment of high dose chemotherapy in children with diffuse pontine gliomas. Cancer. 2000; 88: 685692.
  • 11
    Dunkel IJ, Garvin JH Jr., Goldman S, et al. High dose chemotherapy with autologous bone marrow rescue for children with diffuse pontine brain stem tumors. Children's Cancer Group. J Neurooncol. 1998; 37: 6773.
  • 12
    Chastagner P, Kozin SV, Taghian A. Topotecan selectively enhances the radioresponse of human small-cell lung carcinoma and glioblastoma multiforme xenografts in nude mice. Int J Radiat Oncol Biol Phys. 2001; 50: 777782.
  • 13
    Chastagner P, Pinel S, Merlin JL, et al. Influence of overall treatment time of topotecan and radiotherapy administered alone or in combination in two xenografted human glioblastoma multiforme. Proc Am Soc Clin Oncol. 2001; 20: 60a.
  • 14
    Kingsbury WD, Boehm JC, Jakas DR, et al. Synthesis of water-soluble (aminoalkyl)camptothecin analogues: inhibition of topoisomerase I and antitumor activity. J Med Chem. 1991; 34: 98107.
  • 15
    Marchesini R, Colombo A, Caserini C, et al. Interaction of ionizing radiation with topotecan in two human tumor cell lines. Int J Cancer. 1996; 66: 342346.
  • 16
    Lamond JP, Mehta MP, Boothman DA. The potential of topoisomerase l inhibitors in the treatment of CNS malignancies: report of a synergistic effect between topotecan and radiation. J Neurooncol. 1996; 30: 16.
  • 17
    Needle MN, Mehlta M, Krailo M, et al. Phase I study of topotecan as a radiosensitizer prior to daily involved field irradiation in children with intrinsic pontin glioma [abstract]. Proc Am Soc Clin Oncol. 1999: 18: 558a.
  • 18
    Bellissant E, Benichou J, Chastang C. Application of the triangular test to phase II cancer clinical trials. Stat Med. 1990; 9: 907917.
  • 19
    Chastagner P, Bouffet E, Grill J, Kalifa C. What have we learnt from previous phase II trials to help in the management of childhood tumours? Eur J Cancer. 2001; 37: 19811993.
  • 20
    Broniscer A, Gajjar A. Supratentorial high-grade astrocytoma and diffuse pontine brainstem glioma: two challenges for the pediatric oncologist. Oncologist 2004; 9: 197206.
  • 21
    Benesch M, Lackner H, Moser A, et al. Outcome and long-term side effects after synchronous radiochemotherapy for childhood brain stem gliomas. Pediatr Neurosurg. 2001; 35: 173180.
  • 22
    Blaney SM, Cole DE, Balis FM, Godwin K, Poplack DG. Plasma and cerebrospinal fluid pharmacokinetic study of topotecan in nonhuman primates. Cancer Res. 1993; 53: 725727.
  • 23
    Chen AY, Choy H, Rothenbery ML. DNA topoisomerase I-targeting drugs as radiation sensitizers. Oncology (Huntingt). 1990; 13 (Suppl): 3946.
  • 24
    Allen J, Siffert J, Donahue B, et al. A phase I/II study of carboplatin combined with hyperfractionated radiotherapy for brainstem gliomas. Cancer. 1999; 86: 10641069.
  • 25
    Broniscer A, Leite CC, Lanchote VL, et al. Radiation therapy and high-dose tamoxifen in the treatment of patients with diffuse brainstem gliomas: results of a Brazilian cooperative study. Brainstem Glioma Cooperative Group. J Clin Oncol. 2000; 18: 12461253.
  • 26
    Wolff JE, Westphal S, Molenkamp G, et al. Treatment of paediatric pontine glioma with oral trophosphamide and etoposide. Br J Cancer. 2002; 21: 945949.
  • 27
    Marcus KJ, Dutton SC, Barnes P, et al. A phase I trial of etanidazole and hyperfractionated radiotherapy in children with diffuse brainstem glioma. Int J Radiat Oncol Biol Phys. 2003; 55: 11821185.
  • 28
    Sanghavi SN, Needle MN, Boyett JM, et al. A phase I study of topotecan as radiosensitizer for brainstem gliomas of childhood: first report of the Children's Cancer Group-0952. Neuro-oncol. 2003; 5: 813.
  • 29
    Grabenbauer GG, Anders K, Fietkau RJ, et al. Prolonged infusional topotecan and accelerated hyperfractionated 3d-conformal radiation in patients with newly diagnosed glioblastoma-a phase I study. J Neurooncol. 2002; 60: 269275.
  • 30
    Lesimple T, Hassel MB, Gedouin D, et al. Phase I study of topotecan in combination with concurrent radiotherapy in adults with glioblastoma. J Neurooncol. 2003; 65: 141148.
  • 31
    Fisher B, Won M, MacDonald D, et al. Phase II study of topotecan plus cranial radiation for glioblastoma multiforme: results of Radiation Therapy Oncology Group 9513. Int J Radiat Oncol Biol Phys. 2002; 53: 980986.
  • 32
    Packer RJ, Prados M, Phillips P, et al. Treatment of children with newly diagnosed brain stem gliomas with intravenous recombinant beta-interferon and hyperfractionated radiation therapy: a childrens cancer group phase I/II study. Cancer. 1996; 77: 21502156.
  • 33
    Jennings MT, Sposto R, Boyett JN, et al. Preradiation chemotherapy in primary high-risk brainstem tumors: phase II study CCG-9941 of the Children's Cancer Group. J Clin Oncol. 2002; 20: 36313637.