Oral trofosfamide and etoposide in pediatric patients with glioblastoma multiforme.

BACKGROUND

Glioblastoma multiforme in childhood is rare, and the prognosis for patients with the disease is poor. The Pediatric Oncology Society of the Germanic language group (GPOH) enrolls patients in a series of pilot trials, the first of which is reported here (HIT-GBM-A).

METHODS

Twenty-two patients with glioblastoma multiforme, World Health Organization Grade 4, between the ages of 3-15 years (45% male) were enrolled during the period 1995-1997. There were 13 supratentorial tumors, 8 brainstem tumors, and 1 cerebellar tumor. The patients underwent the following procedures: stereotactic biopsy (n = 3 patients), open biopsy (n = 1 patient), partial resection (n = 6 patients), subtotal resection (n = 4 patients), and macroscopic total resection (n = 8 patients). Adjuvant treatment consisted of oral chemotherapy with trofosfamide, 100 mg/m(2), and etoposide, 25 mg/m(2), daily or for 21-day cycles interrupted by 1-week rests. Standard fractionated radiation (54 grays) was started concurrently with the first cycle.

RESULTS

The chemotherapy was well tolerated, with no treatment-related deaths and only minor side effects. The responses in 12 evaluable patients after two cycles were as follows: 1 complete response, 1 partial response, 3 patients with stable disease, and 7 patients with progressive disease. The median overall survival was 12 months. The 1-year, 2-year, and 4-year overall survival rates were 52%, 26%, and 22%, respectively, and the event free survival rates were 26%, 22%, and 4%, respectively. None of the four surviving patients (3.2 years, 3.4 years, 4.0 years, and 4.2 years after diagnosis) is event free. Two patients are alive after tumor progression: One patient was diagnosed with a medulloblastoma, and one patient was diagnosed with an osteosarcoma as second malignancies. A control group extracted from the Surveillance, Epidemiology, and End Results data had lower survival rates: the difference between the groups was not statistically significant (P = 0.26).

CONCLUSIONS

This chemotherapy will not be used in a randomized trial of patients with glioblastoma; however, it may be evaluated for patients with tumors that have more chemoresponsive histologies.