Abstract
Purpose :
Ocular pharmacokinetics can be largely affected by the melanin-binding properties of the drug. The purpose of this study was to explore the retention of levofloxacin in non-pigmented and pigmented tissues in rabbit eyes and to develop a data-based ocular pharmacokinetic model for intravitreally administered levofloxacin.
Methods :
We measured ocular distribution upon intravitreal injection of levofloxacin to both New Zealand White (albino) and Dutch Belted (pigmented) rabbits. Aqueous humor, vitreous humor, plasma, retina, iris, ciliary body, and choroid samples were collected for up to 17 hours for albino and 42 days for pigmented rabbits. Drug concentrations in tissues were determined by LC-MS/MS analysis and the pharmacokinetic parameters were estimated using compartmental and non-compartmental modeling. A bottom-up simulation model was developed for ocular levofloxacin.
Results :
Pigmented ocular tissues showed substantial levofloxacin enrichment and very long retention in comparison to non-pigmented tissues in albino and pigmented rabbits. In albino rabbits, the drug exposure (AUC0 to last time of sampling) in the ocular tissues was: vitreous humor> retina >choroid>ciliary body> iris> > aqueous humor. In contrast, pigmented rabbits showed considerable enrichment and long drug retention in the pigmented tissues (choroid, ciliary body, iris). Mean residence time was in the range of 3.6-5.6 hours for choroid, ciliary body, and iris in albino rabbits, whereas it was 170-267 hours in pigmented rabbits. A simulation model was developed to quantitatively describe levofloxacin kinetics in the eye. Long retention in the pigmented tissues can potentially translate to prolonged drug action in the pigmented tissues if the free concentration in those tissues is above the threshold level for pharmacological action.
Conclusions :
This study shows the ocular pharmacokinetics of melanin-binding levofloxacin after intravitreal administration. Data was used for parameter estimation and pharmacokinetic modeling.
This abstract was presented at the 2023 ARVO Annual Meeting, held in New Orleans, LA, April 23-27, 2023.