Search

Objective: To evaluate the pharmacokinetics of methotrexate in ventricular cerebrospinal fluid and serum after osmotic blood-brain barrier disruption and intra-arterial administration compared with intravenous or simple intra-arterial infusion in patients with primary central nervous system lymphoma., Methods: Serum and ventricular cerebrospinal fluid were sampled after methotrexate administration in 12 patients. Blood-brain barrier disruption was induced on 2 sequential days by mannitol (25%) infusion delivered to the vertebral or internal carotid artery territories followed by intra-arterial methotrexate (dose, 1.4 g/m2; 47 treatments). Sixteen treatments were given without barrier disruption by intravenous (3.5 g/m2; nine treatments) or intra-arterial (2.8 g/m2; seven treatments) infusion., Results: Ventricular cerebrospinal fluid-methotrexate peak levels after blood-brain barrier disruption of the vertebral and the internal carotid arteries territories were 19.3 +/- 2.9 and 8.5 +/- 0.7 micromol/L (P < .001), and the area under the curve from time 0 to infinity was 178.0 +/- 21.3 and 110.0 +/- 12.4 [micromol/L x h, respectively (P < .01). No significant differences were observed in serum levels. After intra-arterial infusion was performed without disruption, the serum peak level was higher than that achieved by intravenous treatment (518.2 +/- 67.7 versus 180.6 +/- 31.8 micromol/L; P < .001). No differences were observed in cerebrospinal fluid concentrations, which dropped below 1 micromol/L at 6 hours. The cerebrospinal fluid/serum ratio [AUC(%)] of methotrexate after blood-brain barrier disruption was three to four times greater than that by systemic administration., Conclusion: Enhanced methotrexate delivery to the central nervous system can be attained by intra-arterial administration combined with osmotic disruption of the blood-brain barrier compared with simple intra-arterial or intravenous administration.

Background: The authors have developed a new method of drug delivery into the brain using implantable biodegradable microspheres. In this study, this method was used to provide localized and sustained delivery of 5-fluorouracil (5-FU) after the surgical resection of glioblastoma. This antimetabolite and radiosensitizing drug was selected in an attempt to decrease the rate of local recurrence of the tumor., Methods: Eight patients with newly diagnosed glioblastoma were included in the study and 2 increasing amounts of 5-FU were studied (70 mg and 132 mg). After surgical resection of the tumor, poly(D-L lactide-co-glycolide) 5-FU-loaded microspheres with an average dimension of 45 microm were implanted in the wall of the surgical bed. External beam radiation (59.4 grays) was initiated before the seventh postsurgical day. Patients were followed by clinical examination, magnetic resonance imaging, and 5-FU assays in the blood and cerebrospinal fluid (CSF)., Results: 5-FU assays confirmed sustained concentrations in the CSF for at least 1 month. Concentrations of 5-FU in the blood were lower and transitory. Systemic tolerance to the treatment was good; one case of recurrent brain swelling was observed at the higher dose studied. At the time of last follow-up the overall median survival time was 98 weeks from the time of implantation and 2 patients had achieved disease remission at 139 and 153 weeks, respectively., Conclusions: This study demonstrates that biodegradable microspheres are efficient systems for drug delivery into the brain and may have future application in the treatment of brain tumors. Further studies are needed to confirm the potential of 5-FU-loaded microspheres for the radiosensitization of glioblastoma. [Please see editorial on pages 197-9, this issue].