Rat brain tumor models to assess the efficacy of boron neutron capture therapy: a critical evaluation.

Development of any therapeutic modality can be facilitated by the use of the appropriate animal models to assess its efficacy. This report primarily will focus on our studies using the F98 and 9L rat glioma models to evaluate the effectiveness of boron neutron capture therapy (BNCT) of brain tumors. Following intracerebral implantation the biological behavior of each tumor resembles that of human high grade gliomas in a number of ways. In both models, glioma cells were implanted intracerebrally into syngeneic Fischer rats and approximately 10-14 days later BNCT was initiated at the Brookhaven National Laboratory Medical Research Reactor. Two low molecular weight (M(r) < 210Da) 10B-containing drugs, boronophenylalanine (BPA) and/or sodium borocaptate (BSH) were used as capture agents, either alone or in combination with each other. The 9L gliosarcoma, which has been difficult to cure by means of either chemo- or radiotherapy alone, was readily curable by BNCT. The best survival data were obtained using BPA at a dose of 1200 mg/kg (64.8mg 10B), administered intraperitoneally (i.p.), with a 100% survival rate at 8 months. In contrast, the F98 glioma has been refractory to all therapeutic modalities. Tumor bearing animals, which had received 500 mg/kg (27 mg 10B) of BPA, or an equivalent amount of BSH i.v., had mean survival time (MST) of 37 and 33 days, respectively, compared to 29 days for irradiated controls. The best survival data with the F98 glioma model were obtained using BPA + BSH in combination, administered intra-arterially via the internal carotid artery (i.c.) with hyperosmotic mannitol induced blood-brain barrier disruption (BBB-D). The MST was 140 days with a cure rate of 25%, compared to a MST of 73 days with a 5% cure rate without BBB-D, and 41 days following i.v. administration of both drugs. A modest but significant increase in MST also was observed in rats that received intracarotid (i.c.) BPA in combination with Cereport (RMP-7), which produced a pharmacologically mediated opening of the BBB. Studies also have been carried out with the F98 glioma to determine whether an X-ray boost could enhance the efficacy of BNCT, and it was shown that there was a significant therapeutic gain. Finally, molecular targeting of the epidermal growth factor receptor (EGFR) has been investigated using F98 glioma cells, which had been transfected with the gene encoding EGFR and, intratumoral injection of boronated EGF as the delivery agent, followed by BNCT. These studies demonstrated that there was specific targeting of EGFR and provided proof of principle for the use of high molecular weight, receptor targeting-boron delivery agents. Finally, a xenograft model for melanoma metastatic to the brain has been developed using a human melanoma (MRA27), stereotactically implanted into the brains of nude rats, and these studies demonstrated that BNCT either cured or significantly prolonged the survival of tumor-bearing rats. It remains to be determined, which, if any, of these experimental approaches will be translated into clinical studies. Be that as it may, rat brain tumor models already have made a significant contribution to the design of clinical BNCT protocols, and should continue to do so in the future.