Approaches to overcome the blood-brain barrier in the chemotherapy of primary and secondary brain tumors: Modulation of P-glycoprotein 170 and targeting of the transferrin receptor

Chemotherapy of primary and secondary brain tumors is hampered by the existence of the blood-brain-barrier (BBB) which reduce the brain concentrations of lipophilic cytostatic agents by efflux transport proteins like p-glycoprotein 170 (p-gp). Combining cytostatic agents that are p-gp substrates with inhibitors of p-gp like valspodar is a new approach to overcome the BBB. As a model of non-small cell lung cancer metastases, intracerebral implanted human NCI-H460 cells were treated with a combination of valspodar and vinblastine (50 mg/kg of valspodar p.o.; 0.8 mg/kg of vinblastine i.p., every 5 day). Brain metastases of small cell lung cancer were simulated by the intracerebral injection of DMS 114 cells and treated with a combination of valspodar and paclitaxel (50 mg/kg of valspodar p.o.; 3 mg/kg of paclitaxel i.v. on day 8, and 2 mg/kg on day 15). No therapeutic benefit was observed in the intracerebral tumor models despite restoration of chemosensitivity in vitro. The high aggressiveness of NCI-H460 cells together with the rapid cell proliferation may be the reason for the negative in vivo result. Furthermore, the increase in the vinblastine concentration in the brain, resulting from the p-gp inhibition by valspodar seemed to be extremely low. In case of intracranial DMS 114 tumors chemosensitivity of DMS 114 cells against paclitaxel seemed to be lower compared to the in vitro sensitivity. The 3rd generation p-gp modulators elacridar and tariquidar were characterized with respect to their activity to inhibit p-gp mediated efflux and their specificity against other ABC proteins present at the BBB like the ABCG2 transporter. Both compounds showed an up to 80 times higher inhibitory activity against p-gp than valspodar with IC50 values of 20.5 nM for elacridar and 26.5 nM in case of tariquidar. Elacridar and tariquidar are non-specific modulators as both compounds were able to inhibit not only the p-gp mediated efflux but also ABCG2 transport. However, they differed concerning the selectivity for the two ABC transport proteins. Also, the effect of the two modulators on the paclitaxel distribution in nude mice was examined. The paclitaxel levels were detected in liver, kidney, plasma and brain of nude mice 1.5, 3, 4.5 and 24 hours after the intravenous injection of paclitaxel by HPLC. The results were compared to the paclitaxel distribution after co-administration of the 2nd generation modulator valspodar. In contrast to valspodar, elacridar and tariquidar distribution in the nude mice was determinable. Compared to the IC50 values for p-gp inhibition determined in vitro, 500 to 2500 times higher elacridar and 100 times higher tariquidar concentrations were detected in the brain samples. Hence, sufficient modulator levels were reached for p-gp modulation at the BBB. In liver and kidney the paclitaxel concentrations were similar in the control, the elacridar and the tariquidar group, whereas in the valspodar group the paclitaxel concentrations were significantly higher. In the plasma of valspodar treated mice the paclitaxel levels were about 10 times higher than in the plasma of mice treated with tariquidar, elacridar or vehicle. The paclitaxel concentrations in the brains of the valspodar group were 5 to 8.5 times higher than in the control group, whereas in the elacridar and tariquidar groups an only 4 fold increase compared to the control group was measured. Compared with valspodar, the brain/plasma ratio of paclitaxel in mice treated with elacridar or tariquidar indicates a more specific accumulation of paclitaxel in the brain. Although the increase in paclitaxel brain concentration in case of elacridar and tariquidar co-administration was not as pronounced as in case of valspodar, a dose reduction of anticancer drugs, as in case of valspodar paclitaxel co-administration, seems unnecessary with new modulators, due to the lack of p-gp modulation in liver and kidney, and the low cytostatic drugs concentrations in plasma. Another method to increase intracerebral drug concentrations is the utilization of the transferrin receptor mediated transcytosis at the BBB for drug delivery to the brain. To be able to perform treatment studies with cytostatic drug loaded immunoliposomes targeted to the transferrin receptor both at the BBB and in tumor cells, a tumor model of primary brain tumors in the brain of nude rats was established. For this purpose, three human glioblastoma cell lines were examined, but only the U-373 MG cells turned out to be suited as an orthotopic tumor model. In summary, the administration of p-gp modulators of the 3rd generation in combination with anticancer drugs is a promising strategy for the treatment of primary and secondary brain tumors. Approaches exploiting specific transcytotic pathways at the BBB might also enhance the concentrations of cytostatic drugs in the brain, a prerequisite for a successful therapy of malignant brain tumors