An assessment of the central disposition of intranasally administered insulin lispro in the cerebrospinal fluid of healthy volunteers and beagle dogs.

Intranasally administered regular insulin and insulin aspart have shown cognitive benefit for patients with Alzheimer's disease (AD). To support development of intranasally administered insulin analogs for AD, the central disposition of intranasal insulin lispro in the cerebrospinal fluid (CSF) of healthy volunteers was investigated. Healthy volunteers (N = 8) received two sequential doses of intranasal insulin lispro (48 or 80 IU followed by 160 IU) by Aero Pump in an open-label, single-period study with serial CSF and serum sampling over 5 hours after each dose. CSF insulin lispro was also measured in beagle dogs (N = 6/dose group) that received either 24 IU/kg (equivalent local nasal (IU/cm2) dose to the human 160 IU dose) or 192 IU/kg intranasally, using the same device. Insulin lispro was measured in the CSF and serum using a validated enzyme-linked immunosorbent assay method, and pharmacokinetic parameters were calculated by standard noncompartmental methods. Intranasal administration of insulin lispro was well tolerated. Insulin lispro concentrations in the CSF of humans at all dose levels were below the limit of quantification. Serum insulin lispro concentrations were quantifiable only up to 1-2 hours in the majority of subjects. In contrast to insulin lispro in the CSF of humans, insulin lispro was detectable in the CSF at both dose levels in dogs, and serum concentrations of insulin lispro were generally higher in dogs than in healthy volunteers. The absence of insulin lispro in CSF from healthy volunteers and the lack of robust exposure-response analyses will hinder the development of intranasally administered insulin lispro for AD.