Christopher I. McCormick

The influence of anesthetic choice, PaCO2, and other factors on osmotic blood-brain barrier disruption in rats with brain tumor xenografts.

UNLABELLED Increasing the delivery of therapeutic drugs to the brain improves outcome for patients with brain tumors. Osmotic opening of the blood-brain barrier (BBB) can markedly increase drug delivery, but achieving consistent, good quality BBB disruption (BBBD) is essential. We evaluated four experiments compared with our standard isoflurane/O2 protocol to improve the quality and consistency of BBBD and drug delivery to brain tumor and normal brain in a rat model. Success of BBBD was assessed qualitatively with the large molecular weight marker Evans blue albumin and quantitatively by measuring delivery of the low molecular weight marker [3H]-methotrexate. With isoflurane/O2 anesthesia, the effects of two BBBD drugs of different osmolalities were evaluated at two different infusion rates and infusion durations. Arabinose was superior to saline (P = 0.006) in obtaining consistent Evans blue staining in 16 of 24 animals, and it significantly increased [3H]-methotrexate delivery compared with saline in the tumor (0.388 +/- 0.03 vs 0.135 +/-0.04; P = 0.0001), brain around the tumor (0.269 +/- 0.03 vs 0.035 +/- 0.03; P = 0.0001), brain distant to the tumor (0.445 +/- 0.05 vs 0.034 +/- 0.07; P = 0.001), and opposite hemisphere (0.024 +/- 0.00 vs 0.016 +/- 0.00; P = 0.0452). Forty seconds was better than 30 s (P = 0.0372) for drug delivery to the tumor. Under isoflurane/O2 anesthesia (n = 30), maintaining hypocarbia was better than hypercarbia (P = 0.025) for attaining good BBBD. A propofol/ N2O regimen was compared with the isoflurane/O2 regimen, altering blood pressure, heart rate, and PaCO2 as covariates (n = 48). Propofol/N2O was superior to isoflurane/O2 by both qualitative and quantitative measures (P < 0.0001). Neurotoxicity and neuropathology with the propofol/N2O regimen was evaluated, and none was found. These data support the use of propofol/N2O along with maintaining hypocarbia to optimize BBBD in animals with tumors. IMPLICATIONS Propofol/N2O anesthesia may be better than isoflurane/O2 for optimizing osmotic blood-brain barrier disruption for delivery of chemotherapeutic drugs to brain tumor and normal brain.

Osmotic blood-brain barrier modification: monoclonal antibody, albumin, and methotrexate delivery to cerebrospinal fluid and brain.

In the dog, osmotic opening of the blood-brain barrier (BBB) in the posterior circulation via the vertebral artery and in the anterior circulation via the internal carotid artery was utilized to increase the delivery of three substances of varying molecular weight to the central nervous system. The cerebrospinal fluid (CSF) concentration of all three agents dramatically increased after BBB opening. In contrast to methotrexate, Evans blue-albumin and monoclonal antibody (MAb) concentrations in CSF were 6-fold greater when given after posterior rather than anterior circulation BBB opening. Conversely, MAb delivery to brain parenchyma was optimized after osmotic BBB modification via the carotid artery. This suggests that, with higher molecular weight substances, osmotic barrier opening has a differential effect on the blood-brain vs. blood-CSF barriers.

Inability of dimethyl sulfoxide and 5-fluorouracil to open the blood-brain barrier.

The inability of most chemotherapeutic agents to adequately penetrate the blood-brain barrier (BBB), in either normal brain or tumor-infiltrated brain, is a major factor limiting the use of chemotherapy in central nervous system malignancy. This barrier, however, can be opened in a reversible manner by the intra-arterial administration of hyperosmotic agents such as mannitol. It has been suggested that the intravenous administration of dimethyl sulfoxide (DMSO) or 5-fluorouracil (5-FU) can accomplish the same thing in a less invasive manner. We have not been able to confirm these findings. DMSO was administered to 25 rats intravenously at concentrations ranging from 25 to 90% or into the internal carotid artery at a concentration of 30%. The penetration of methotrexate, Evans blue-albumin, and hexosaminidase A was then evaluated at intervals ranging from 1.5 to 3.5 hours after administration. Significant barrier opening was not observed in animals receiving intravenous DMSO. Barrier modification, albeit generally modest, was obtained in animals receiving intracarotid DMSO, but this may have been the result of grand mal seizures, inasmuch as 5 of 6 of these animals had such seizures. Several of the animals receiving i.v. DMSO also had seizures, and all animals developed varying degrees of hematuria. Similarly, 5-FU was administered at a dose of 30 mg/kg i.v. and the permeability of the BBB to either Evans blue-albumin or methotrexate was evaluated. No increased permeability of the BBB to these two markers was observed. In summary, osmotic BBB opening in our hands remains the most consistent and reliable means available to open the BBB in a reversible fashion. Neither intravenous DMSO nor 5-FU seems to increase the delivery of chemotherapy or protein tracer to the central nervous system, and the use of DMSO can result in seizures and hematuria.

Effects of Gd-DTPA after osmotic BBB disruption in a rodent model : toxicity and MR findings

Objective This experiment was done to evaluate the gross neurotoxicity of intravenous Gd-DTPA administered in conjunction with osmotic blood-brain barrier (BBB) disruption and to image a human small cell lung carcinoma intracerebral tumor xenograft before and after osmotic BBB disruption. Materials and Methods Neurotoxicity studies were performed in normal Sprague-Dawley rats following osmotic BBB disruption by the injection of 25% mannitol in the right internal carotid artery and intravenous administration of Gd-DTPA (n = 10). Animals were observed for major neurologic changes such as seizure or substantial motor defects, and after death neuropathologic examination was performed. Human small cell lung carcinoma cells were implanted intracerebrally in athymic nude rats (n = 4). Gadopentetate dimeglumine was injected intravenously and serial T1-weighted images were obtained. Blood-brain barrier disruption was produced in each animal, followed by a second dose of intravenous Gd-DTPA, and imaging studies were repeated. Results No gross neurologic toxicity was observed. Tumors showed dense enhancement in a small area, and BBB disruption resulted in marked enhancement in most of the gray matter of the right cerebral hemisphere. Conclusion Gadopentetate dimeglumine appears to be safe in doses up to 21 mmol/m2 in conjunction with barrier disruption in rats. A human small cell lung carcinoma intracerebral xenograft provides a useful method to study brain tumors.

Long-term toxicity and neuropathology associated with the sequencing of cranial irradiation and enhanced chemotherapy delivery.

OBJECTIVE The goal was to evaluate, at 1 year, 75 Long-Evans rats for survival rates and toxicity associated with the sequencing of cranial irradiation and enhanced chemotherapy delivery. METHODS Seventy-five Long-Evans rats were randomized into four groups and evaluated at 1 year for survival rates and toxicity associated with the sequencing of cranial irradiation and enhanced chemotherapy delivery. Radiation (2,000 cGy) was administered as a single fraction, by using parallel opposed portals, 30 days before chemotherapy (Group 1), 24 hours before chemotherapy (Group 2), 30 days after chemotherapy (Group 3), or without chemotherapy or without radiation (control group, Group 4). Five subgroups within each treatment group included rats receiving intra-arterially administered methotrexate (1 g/m2) or intravenously administered etoposide (200 mg/m2) combined with intra-arterially administered carboplatin (200 mg/m2), administered with or without osmotic blood-rain barrier disruption, and a group receiving normal saline solution after blood-brain barrier disruption. RESULTS There was a significant increase in total toxic effects when the three experimental groups were compared with the control group (P = 0.001, 0.006, and 0.013 for Groups 1, 2, and 3, respectively). All groups receiving radiation and chemotherapy (particularly carboplatin and etoposide) had an increased incidence of hind limb paralysis, resembling experimental allergic neuritis (P = 0.053). Statistical analysis showed a trend toward increased mortality rates in Group 1 (antecedent radiation), compared with the control group (P = 0.082), and an increased incidence of intracerebral calcification (P = 0.019). No differences in mortality rates were observed for Group 2 or 3, compared with the control group. CONCLUSION Radiation before chemotherapy was a more toxic sequence and, surprisingly, carboplatin/etoposide administered in combination with radiotherapy was more detrimental than methotrexate. Additional studies are in progress to evaluate the toxicity and efficacy of sequences of cranial irradiation and enhanced chemotherapy in tumor-bearing rats.