paediatrics Brussels 17

Articles

Radiotherapy was withheld unless there was progressive disease defined on local imaging—age did not determine whether radiotherapy was given. For localised, non- metastatic tumours, 50 Gy in 25 daily fractions of 2 Gy per fraction, 5 days per week, was prescribed for the radiologically defined macroscopic tumour plus a margin of 2 cm. Whole neuroaxis radiotherapy was recommended for metastatic disease: for children aged 3 years and older, 35 Gy in 21 daily fractions of 1·67 Gy per fraction was prescribed to the whole neuroaxis. This was followed by a boost to the primary tumour of 20 Gy in 12 daily fractions of 1·67 Gy per fraction given to the initial tumour volume, plus a margin of 2 cm. For infants younger than 3 years, the whole neuroaxis dose was reduced to 25 Gy in 20 daily fractions of 1·25 Gy per fraction. The boost dose to the primary tumour was 20 Gy in 12 daily fractions of 1·67 Gy per fraction (ie, as for the older children). Assessment Patients were staged by full neuraxis imaging, postoperative scans (within 48 h) were recommended, but for various reasons this was achieved in most but not all cases. All patients underwent primary surgery with the aim of achieving maximal surgical resection. A complete resection was recorded when there was no visible tumour documented by the surgeon at the end of operation, a subtotal resection when visible tumour remained, and a biopsy when only sufficient tumour for diagnosis was removed. The operative notes and postoperative scans were reviewed centrally (JAGP, CM). Central radiological review of the extent of the surgical resection on postoperative scans was done according to SIOP criteria (CM, TC, WKC, RG). 17 Routine scans were requested at day 112, 224, and the end of the chemotherapy schedule. 6-monthly post- treatment scans were recommended. MRI scans of children on long-term follow-up of more than 4 years post-surgery were reviewed centrally for evidence of leucoencephalopathy as determined by white matter changes acting as a surrogate marker for methotrexate neurotoxicity. Minimum criteria for this review included T1-weighted and T2-weighted MRI scans with gadolinium enhancement. Cerebrospinal fluid samp- ling before chemotherapy was recommended. Histological slides from all patients were reviewed by DE and JI. Tumours were classified as grade II or III according to WHO criteria. 18 Ependymoblastomas, which are now classified with other primitive neuroectodermal tumours, were excluded. 18 Toxicity was assessed by the treating physician and coded in the CCLG data centre. The UKCCSG shortened listing of National Cancer Institute common toxicity criteria version 2.0 were used.

Children weighing up to 10 kg (dose by weight)

Children weighing more than 10 kg (dose by surface area)

Course 1 (day 0) Vincristine (IV bolus) Carboplatin (IV over 4 h) Course 2 (day 14) Vincristine (IV bolus)

0·05 mg/kg 20 mg/kg

1·5 mg/m² 550 mg/m²

0·05 mg/kg 250 mg/kg

1·5 mg/m²

Methotrexate

8000 mg/m²

Folinic acid

15 mg fixed dose

15 mg fixed dose

Course 3 (day 28) Vincristine (IV bolus) Cyclophosphamide

0·05 mg/kg 50 mg/kg 60 mg/kg

1·5 mg/m²

1500 mg/m² 1800 mg/m²

Mesna

Course 4 (day 42) Cisplatin (continuous infusion for 48 h)

1·3 mg/kg×2 days

80 mg/m² in two divided doses

IV=intravenous.There were seven cycles in total, each cycle was delivered over 56 days.

Table 1 : Chemotherapy schedule

Metastases at diagnosis* No Yes

Study population

80 54 11

9 4 3

Male sex

Younger than <1 year at diagnosis

Median age (range) in years

1·93 (0·05–3·16)

1·36 (0·24–2·25)

Infratentorial ependymoma Supratentorial ependymoma

69

7 2 5 4

11

Central review histology as classic (II) 54

WHO classified as anaplastic (III)

26

*Based on preoperative MRI.

Table 2 : Patient characteristics at diagnosis

Number of patients (n=89)

Haematological Grade 3

6

Grade 4 Renal Grade 3 Grade 4

83

3 0

Audiological Grade 3

3 2

Grade 4

Gastrointestinal Grade 3

23

Grade 4 Other Grade 3 Grade 4

3

27

Statistical analysis Standardised

3

received chemotherapy (SRDChemo) was calculated for each patient as the dose of

Table 3 : Cumulative toxicity

698

http://oncology.thelancet.com Vol 8 August 2007