Convection-enhanced intraparenchymal delivery (CEID) of cytosine arabinoside (AraC) for the treatment of HIV-related progressive multifocal leukoencephalopathy (PML).

AIDS-related PML continues to be a relatively common and rapidly fatal infection in patients with AIDS, and no effective therapy has been established to alleviate the effects of this disease. Through the years, isolated reports and small case studies have shown somewhat encouraging results using cytosine arabinoside (AraC) in the treatment of PML. The optimism behind the use AraC for this disease began to fade with ACTG trial 243, which suggested that AraC had no benefit in patients with HIV-related PML. In this article, we provide evidence that suggests that the failure of AraC in the ACTG trial may have been due to insufficient delivery of the drug through traditional intravenous and intrathecal routes. Furthermore, we provide evidence that convection-enhanced intraparenchymal delivery of AraC may prove to be a safe and effective means of treating this infection, and we outline a clinical trial that we have recently undertaken to test this hypothesis.

Targeting anticancer drugs to the brain: II. Physiological pharmacokinetic model of oxantrazole following intraarterial administration to rat glioma-2 (RG-2) bearing rats.

The disposition of the anticancer drug oxantrazole (OX) was characterized in rats bearing the rat glioma-2 (RG-2) brain tumor. Following intraarterial administration of 3 mg/kg of OX, serial sacrifices were completed from 5 min to 5 hr after administration. Blood and tissue samples collected at the time of sacrifice were processed and measured for OX concentrations by HPLC. The kidney had the greatest affinity for OX with the Cmax being 40.6 micrograms/ml at 15 min after administration. OX concentrations in brain tumor were higher than in normal right and left brain hemispheres, and consistent with enhanced drug blood-tumor barrier (BTB) permeability seen in experimental models for brain tumors. Observed heart, liver, lung, and spleen OX concentrations were similar, ranging from approximately 2 micrograms/ml to 20 micrograms/ml. A unique technique was used to develop a global physiological pharmacokinetic model for OX. A hybrid or forcing function method was used to estimate individual tissue compartment biochemical parameters (i.e., partition and mass transfer coefficients). A log likelihood optimization scheme was used to determine the best model structure and parameter sets for each tissue. Most tissues required a 3-subcompartment structure to adequately describe the observed data. The global model was then reconstructed with an arterial blood and rest of body compartments that provided predicted OX concentrations in agreement with the data. The model development strategy provides a systematic approach to physiological pharmacokinetic model development.

Concurrent measurements of blood flow and transcapillary transport in avian sarcoma virus-induced experimental brain tumors: implications for chemotherapy.

A blood-to-tissue transfer constant, K, and tissue blood flow, F, were measured concurrently in seven rats with a total of 19 separate brain tumors induced by intracerebral inoculation of avian sarcoma virus. Regional and local measurements of K and F were obtained using double-label quantitative autoradiography with alpha-[14C]aminoisobutyric acid and [131I]iodoantipyrine, computerized microdensitometry and image analysis. Apparent tissue extraction fractions and capillary permeability-surface area products were calculated for different tumor regions, brain adjacent to tumor and tumor-free brain. The following observations were made: five histological categories of the tumors were found; significant local and regional variations of both K and F were typical in each group, resulting in marked regional variability of permeability-surface area products but more uniform values of apparent extraction fraction; the values of F, K, permeability-surface area products and apparent extraction fraction correlated poorly with morphological features of the tumors (necrosis, cellularity, cytology, location and size); the extraction fraction of alpha-aminoisobutyric acid was usually highest in tumor centers and then decreased in a gradient from tumor periphery through adjacent brain; and regardless of classification or histological features, capillary permeability and surface area, and not tissue perfusion or blood flow, seem to determine the blood-to-tissue transport processes (delivery of bloodborne materials) in most regions of these experimental brain tumors. An operational pharmacokinetic model of drug concentration in tumor tissue is developed and the results of our analysis indicate that increases in capillary permeability such as measured in these studies would not be sufficient to deliver adequate amounts of water-soluble drugs with short plasma half-lives to tumor tissue.