par De Witte, Olivier ;Lefranc, Florence ;Levivier, Marc ;Salmon, Isabelle ;Brotchi, Jacques ;Goldman, Serge
Référence Journal of neuro-oncology, 49, 2, page (157-163)
Publication Publié, 2000-09
Référence Journal of neuro-oncology, 49, 2, page (157-163)
Publication Publié, 2000-09
Article révisé par les pairs
Résumé : | BACKGROUND: The prognostic value of the metabolic status of cerebral gliomas determined by positron emission tomography with [18F]-fluoro-deoxy-D-glucose (FDG-PET) has been established in populations with a mixture of grades 2, 3 and 4 gliomas, but remains uncertain when only malignant gliomas are considered (grade 3 and 4). METHODS: FDG-PET performed in 30 patients with anaplastic astrocytoma (grade III) and 61 patients with glioblastoma (grade 4) were classified according to a metabolic grading. The uptake of FDG was lower in the tumor compared to white matter (WM) in grade 1 (4 glioblastoma, 4 anaplastic astrocytoma), it was intermediate between WM and cortex in grade 2 (20 glioblastoma, 22 anaplastic astrocytoma), and it was superior to cortex in grade 3 (38 glioblastoma, 4 anaplastic astrocytoma). RESULTS: Kaplan-Meier survival curves were similar in patient with grades 1 and 2, but were significantly worse (p = 0.007) in grade 3. In multivariate analysis considering age, pathological grade (anaplastic astrocytoma versus glioblastoma), and metabolic grades, the metabolic grade did not appear to be an independent prognostic factor. When anaplastic astrocytomas and glioblastomas were considered separately, metabolic grade is of predictive value only in the group of glioblastomas. CONCLUSION: In malignant gliomas, metabolic grading determined by FDG-PET was not superior to the pathological grading for survival prediction. Still, it remains of predictive value when applied to malignant gliomas histologically classified as glioblastoma. |