Phase II trial and pharmacokinetic evaluation of cytosine arabinoside for leptomeningeal metastases from breast cancer

Purpose: To determine the efficacy and pharmacokinetics of intraventricular cytosine arabinoside (Ara-C) as front-line treatment for leptomeningeal me tastases from breast cancer. Methods: Ten patients newly diagnosed with lep tomeningeal metastases (LMM) from breast cancer were treated with 100 mg in traventricular cytosine arabinoside (IVT Ara-C) via an Ommaya reservoir. Tr eatment was administered three times a week Tor 2 weeks. then once a week f or 4 weeks. and then once every 6 weeks for four cycles to responding patie nts. Nine patients were evaluable clinically, and seven patients underwent testing to determine the pharmacokinetic profile of Ara-C in the cerebrospi nal fluid (CSF). Results: Two patients had partial responses lasting 9 and 40 weeks, respectively. Two other patients had stable disease. The median s urvival duration was 30 weeks (range: 5-58 weeks). Seven patients died from LMM, Acute toxic effects associated with IVT Ara-C included meningismus, n ausea, vomiting, and myelosuppression. The median peak Ara-C level in CSF w as 16.69 +/- 6.30 mM (SD), The half life for elimination was 1.45 +/- 0.61 h (SD) There was no drugs accumulation between courses. Neuropsychological evaluations were completed in eight patients, six (75%) of whom had preexis ting cognitive deficits. Their condition generally improved over the course of treatment until the LMM progressed. No neurotoxic side effects of IVT A ra-C were observed in the two patients who had normal baseline cognitive as sessments. Conclusions: IVT Ara-C at this dose and schedule has minimal act ivity as initial treatment for LMM from breast cancer despite achievement o f high peak levels of the drug in the cerebrospinal fluid. A liposomal Ara- C formulation is currently under investigation.