Pharmacokinetic and histopathologic analysis of intrathecal rituximab infusion.

(A) CSF and serum profiles in cynomolgus monkeys injected with 3 to 5 mg rituximab via 1-minute intrathecal infusion. Observed CSF and serum concentrations were normalized to a 5-mg dose. Experiment numbers are shown in parentheses. Measured CSF concentrations of rituximab were maximal within 2 hours of intrathecal administration and ranged from 460 to 1550 μg/mL. CSF concentrations were approaching minimal detectable levels between 24 to 70 hours after administration in those animals with later collection time points. (B) Fit of 2-compartment intravenous infusion model to CSF concentrations from experiment no. 4 (1 monkey). Observed concentrations were normalized to a dose of 5 mg prior to fitting a 2-compartment model. (C) Fit of 2-compartment intravenous infusion model to CSF mean concentrations averaged across all experiments. Observed concentrations were normalized to a dose of 5 mg and then averaged across all experiments prior to fitting a 2-compartment model. (D) Gross and histologic analyses using hematoxylin and eosin and Luxol fast blue stains were performed between 1 and 8 months after either 1 or 2 intrathecal rituximab administrations. There was no evidence of neuronal loss, demyelination, ventriculitis, ependymitis, or vasculitis that was attributable to rituximab administration. (Di) No evidence of demyelination of white matter tracts (Luxol fast blue stain). (Dii) No evidence of neuronal loss in C1A cells of hippocampus (hematoxylin and eosin stain). Original magnifications × 200.