A study was undertaken to determine whether route (intraventricular vs intralumbar) of intracerebrospinal fluid (intra-CSF) drug administration influences progression-free survival in the treatment of patients with neoplastic meningitis, which occurs in 1% to 5% of patients with known cancer. Currently available treatment options result in modest responses, which is in part a reflection of obstacles to drug delivery into the leptomeningeal space.

One hundred patients with clinically and cytologically or radiographically documented neoplastic meningitis because of solid cancers received intra-CSF liposomal cytarabine or methotrexate as specified in a randomized phase 4 trial. The 2 treatment arms were well balanced for demographic and tumor-related characteristics of known prognostic importance, including age, performance status, tumor type, extent of systemic and other central nervous system (CNS) disease, prior CNS therapy, and concurrent systemic chemotherapy.

One hundred patients were randomized and treated (52 with sustained-release cytarabine, and 48 with methotrexate). Progression-free survival (the primary study endpoint) was identical between the sustained-release cytarabine and methotrexate treatment arms for all 100 patients (35 vs 37.5 days, P = .79). When progression-free survival was examined as a function of route of chemotherapy administration (lumbar vs ventricular), there was no difference for patients treated with sustained-release cytarabine (29 vs 43 days, P = .35). For patients treated with methotrexate, however, there was a statistically significant difference favoring patients receiving intraventricular therapy (19 vs 43 days, P = .048).

Site of intra-CSF chemotherapy drug administration is clinically relevant with short half-life drugs such as methotrexate.