paediatrics Brussels 17

280

Ependymomas in babies and infants

Maintenance Chemotherapy until RT

VCR 1.5mg/m2 MTX 5g/m2 CF-rescue

VCR 1.5mg/m2 MTX 5g/m2 CF-rescue

MTX 5g/m2 CF-rescue VCR 1.5mg/m2

Procarbazine 100 mg/m2/d

HIT - SKK 87

VCR 1.5mg/m2

VCR 1.5mg/m2

Day 1 (= 57) 8

14 15

29

43

Induction Chemotherapy before Maintenance

Ifosfamide 3g/m2/d x 3 Mesna 3g/m2/d x 5 Etoposide 150mg/m2/d x 3

HIT - SKK 87

Cisplatin 40mg/m2/d x 3 Cytarabine 400mg/m2/ d x 3

MTX 5g/m2 CF-rescue

MTX 5g/m2 CF-rescue

Procarbazine 100mg/d x 10

Weeks 1

3

5

7

9

2

12

14

16

18

Postoperative Chemotherapy MTX intraventricular 2mg/d1-4

MTX Intraventricular 2mg/d1-4 Carboplatin i.v. 200 mg/m2/d1-3 Etoposid i.v. 150 mg/m2/d1-3

MTX intraventricular 2mg/d1-2 MTX i.v. /24h 5 g/m2/d 1 VCR i.v. 1.5 mg/m2/d1

MTX intraventricular 2mg/d1-2 MTX i.v. /24h 5 g/m2/d 1 VCR i.v. 1.5 mg/m2/d1

HIT - SKK 92

Cyclophosphamide i.v. 800 mg/m2/d1-3 VCR i.v. 1.5 mg/m2/d1

Weeks 1 (= 10 = 19) 3

5

7

Fig. 2. Chemotherapy schedules.

evaluation of prognostic factors and survival. Patients were treated in 23 centers. Documentation of disease was performed by the treating centers. The clinical data was monitored at the Children’s Hospital, University of Wu¨rzburg, Germany. Additional data on radiotherapy was collected and monitored by the Department of Radio-Oncology, University of Tu¨bingen, Germany. The follow-up period was defined as extending from the date of surgery to the latest patient contact or event. The length of survival was calculated from the date of surgery. Terminal events were defined as date of death from any cause (overall survival) or the date of first progression or relapse after surgery (progression-free survival (PFS)). For all patients alive without events, the length of survival was censored for the statistical analysis as the last date of documented contact with the patient. Data for patients who died without evidence of progression was censored. The Kaplan–Meier method was used to estimate overall survival, and the log-rank test was applied for statistical comparison of survival estimates. Data is presented with nominal two-tailed p-values and 95% confidence intervals. All analysis was carried out with the SAS Institute system for Windows, version 8 software (SAS Institute, Cary, NC).

experimental chemotherapy was recommended to follow, and radiotherapy was recommended for children older than 18 months. Radiotherapy Infratentorial and metastatic tumors were to be treated by irradiation of the neuraxis, followed by a boost to the posterior fossa. For supratentorial ependymoma, treatment volume should encompass the primary tumor site only, unless the tumor was involving the ventricular system. The prescribed total dose for the neuraxis covering the whole subarachnoidal space was 35.2 Gy (1.6 Gy per fraction, five times weekly). The posterior fossa was to receive a boost dose of 20.0 Gy (2.0 Gy per fraction, five times per week). For local radiotherapy, the prescribed total dose was 54.0 Gy (2.0 Gy per fraction). The PTV should encompass the primary tumour volume plus additionally a 2 cm safety margin. For children without any residual disease or dissemination, total dose to the neuraxis was allowed to be reduced to 24.0 Gy. The choice was at the discretion of the local radiotherapist. At the time of onset of the HIT-SKK trial no further detailed guidelines for radiotherapy were included. In 1991, the guidelines were specified and a quality assurance program was integrated. Methods and results have been published already elsewhere [25,26] Statistical considerations The data of children with anaplastic ependymoma, as confirmed by the institutional pathologists, included in the HIT-SKK87 and 92 trials served as basis for statistical

Results Patient population

Thirty-four children with ependymoma were eligible (median age, 20.5 months). Histopathologic findings were