The ability of cis-DDP and several analogs to enter the CSF was investigated in rhesus monkeys that had subcutaneously implanted Ommaya reservoirs connected to catheters in each monkey's fourth ventricle. Plasma and CSF samples were analyzed for platinum content by atomic absorption spectroscopy. Plasma platinum curves were biphasic with a very slowly declining terminal phase. CSF platinum curves rose to maximum concentrations 30-40 min after an IV bolus injection and declined mono-exponentially (T 1/2 = 60 min) without displaying a detectable slow terminal phase. cis-DDP given as an IV bolus of 1.5 mg/kg or 3.0 mg/kg produced peak CSF concentrations of 0.35 and 0.78 microM platinum. The ratio of CSF platinum:plasma platinum never exceeded 0.04. When cis-DDP at 3.0 mg/kg was given as a 2- or 7-h infusion, the peak CSF concentrations were 0.28 and 0.17 microM platinum, respectively. The total CSF exposure, measured as concentration X time, was the same for bolus and for 2- and 7-h infusions. Studies with analogs showed that neither malonato 1,2-diaminocyclohexane platinum (II) nor 4-carboxyphthalato 1,2-diaminocyclohexane platinum (II) had better CSF penetrance than cis-DDP. Sulfato 1,2-diaminocyclohexane platinum (II) could not be detected in the CSF. The ratio of CSF platinum:plasma platinum was never greater tha 0.02-0.03 for any of the analogs.