Curiethérapie par iridium 192 des gliomes sus-tentoriels de haut grade récidivant en territoire irradié: technique du groupe hospitalier Pitié-Salpêtrière et résultats préliminaires

Brachytherapy of glioblastoma recurring in previously irradiated territory: predictive value of tumor volume

PURPOSE

To evaluate the impact of tumor volume on survival of patients reirradiated with (192)Ir for recurrent glioblastoma.

METHODS AND MATERIALS

Between 1993 and 1997, 42 patients with recurrent glioblastomas (29 males and 13 females, age 18-69 years, median age 49) were treated with (192)Ir implantation. Previous treatments included surgery, external beam radiotherapy, and chemotherapy. Maximum diameter of the recurrent tumor was 1.2-10.1 cm (median: 5.7 cm) and tumor volume was 1.6-122 cm(3) (median: 23 cm(3)). Karnofsky performance status score was 50-100 (median: 80). Brachytherapy dose was 40-60 Gy.

RESULTS

Probability of overall survival was 80% at 6 months, 48% at 1 year, and 11% at 2 years. Median survival was 50 weeks. Univariate analysis showed that both tumor volume (T < or T > or = 30 cm(3)) and Karnofsky performance status score were significant predictors of survival. Multivariate analysis showed that smaller tumor volumes were associated with a higher probability of survival (p < 0.001).

CONCLUSION

Tumor volume less than 30 cm(3) was associated with a higher probability of, and quality of, survival than larger lesions for patients reirradiated by brachytherapy for recurrent glioblastoma.

[Gross tumor volume and clinical target volume in brachytherapy of malignant tumors]

Tumours treated with radioactive implants are in most cases small and accessible to clinical examination or endoscopy. Gross tumour volume is consequently mainly determined clinically or with endoscopy. Imaging may also be useful, with classical techniques, as mammography, or with CT-scan or MRI. Clinical target volume includes gross tumour volume and a 5-10 mm margin. Distribution of dose is estimated on the basis of images performed after implantation, classically with two orthogonal films of the implant. It can then be optimised. Dose-volume histogram of gross tumour volume and critical organs can be obtained with a CT-scan or a MRI of the implant, which may be difficult to perform because of implantation material and dummy sources.

[Radiotherapy of glioblastoma]

Glioblastoma cells appear to be inherently radioresistant and to present a significant fraction of hypoxic cells. The most significant prognostic factors to compare results achieved in several series of patients are the age, performance status and quality of surgical resection. Several randomized trials have provided evidence supporting the efficacy of radiation therapy in the treatment of glioblastoma. Prescription of a 60-Gy dose delivered according to a conventional dose-fractionation scheme (single daily fractions of 1.7 to 2 Gy five times per week) in a target volume with a 2-3 cm margin of tissue surrounding the perimeter of the contrast enhancing lesion on computerized tomography and magnetic resonance imaging is derived from observations made in several retrospective and prospective studies. Evidence of improvement in survival was observed neither in patients receiving hyperfractioned and accelerated radiotherapy, nor in patients for whom radiation sensitizers such as nitroimidazole compounds or halogenated pyrimidine analogs were associated to radiation therapy. The addition of nitrosourea to radiotherapy increases the 2-year survival rate by about 10%. Combination of full-dose external beam radiotherapy and brachytherapy or radiosurgery boost in selected patients with glioblastoma leads to an increase in the median survival, while external beam radiation alone in patients with similar prognosis does not.