Preclinical evaluation of 9-chloro-2-methylellipticinium acetate alone and in combination with conventional anticancer drugs for the treatment of human brain tumor xenografts.

Some ellipticine derivative salts, including 9-chloro-2-methylellipticinium (CME), have been found to have a marked selectivity against all eight brain tumor cell lines of the U.S. National Cancer Institute's disease-oriented in vitro screen. We initiated in vivo antitumor studies to explore the feasibility for further development of this class of compounds. We found that CME was extremely toxic to nude mice when given i.p. at a dose of 25 mg/kg for 3 consecutive days. Animals treated by this route experienced an increase in hepatic transaminases and histopathological changes in the liver, compatible with mitochondrial damage. In contrast, when the portal circulation was bypassed and the same dose of CME was given i.v., animals tolerated daily bolus injections for 5 consecutive days. This 5-day i.v. bolus schedule had consistent antitumor activity, with 28.1% growth delay on s.c. implanted human U251 gliomas. When the potentially high peaks of CME in the portal circulation were avoided by using a 3-day continuous infusion with osmotic minipumps implanted i.p. to release 3.4 mg kg(-1) h(-1) or 6.6 mg kg(-1) h(-1) CME, there were only modest increases in liver enzymes and leukopenia, but no meaningful antitumor activity was observed. In contrast, continuous infusion in the s.c. space was well tolerated and was accompanied by a demonstrable growth delay in s.c. U251 human gliomas of 37.8%. When CME was used in conjunction with carmustine, etoposide or cisplatin, no synergistic activities were observed, but additive effects were demonstrated. Our pharmacokinetic and disposition studies with CME argue against the notion that large and invasive tumors in the brain lack blood-brain barrier features. When CME was used in animals bearing orthotopically implanted U251 gliomas in the brain of nude mice, the survival of the treated animals was not better than vehicle controls, and the addition of CME to carmustine therapy did not improve the survival of those animals treated with carmustine alone. We conclude that, in spite of its marked cytotoxicity in vitro on a variety of human brain tumor cell lines, including U251 glioma cells, CME has a modest antitumor effect on extracranially implanted U251 glioma tumors, and no beneficial effect in animals bearing the same U251 tumor in the brain, owing to a poor penetration into the brain parenchyma.