Comparison of In Vitro Corneal Permeation and In Vivo Ocular Bioavailability in Rabbits of Three Marketed Latanoprost Formulations

BAK did not influence the corneal penetration of latanoprost in in vitro and in vivo studies. The formulation containing a non-ionic surfactant resulted in lower and slower ocular penetration compared with preserved or PF SF formulations. This raises questions about the relevance of BAK and some surfactants in enhancing corneal penetration of ocular formulations.

Three latanoprost formulations were tested under good laboratory practice conditions: a formulation containing benzalkonium chloride (BAK) but no surfactant (Preserved latanoprost); the same formulation except preservative-free (PF) without BAK or surfactant (SF) (PF SF latanoprost); and a different formulation without BAK but containing a non-ionic surfactant (MGHS 40 at 5%) combined with thickening agents (Carbomer 974P, Macrogol 4000) (PF latanoprost). Corneal permeation of latanoprost acid (LAT) was first determined in vitro using a reconstructed human corneal epithelium tissue. Then, in vivo pharmacokinetic studies were performed on pigmented rabbits, for which LAT concentration was measured in the aqueous humour (AH) and iris-ciliary body (ICB).

All latanoprost formulations currently available for the treatment of glaucoma or ocular hypertension contain the same concentration of latanoprost (0.005%) but differ in excipients, which may affect corneal drug permeability or stability. This study aimed at comparing corneal penetration of three marketed latanoprost solutions with different excipient formulations in in vitro and in vivo drug permeability studies.

In vitro, the cumulative transport of LAT was linear between 1 h and 4 h for preserved latanoprost and PF SF latanoprost, and LAT concentrations matched exactly at each timepoint. By contrast, the permeation of PF latanoprost was linear between 2 h and 12 h and was significantly lower than that of preserved latanoprost and PF SF latanoprost at 4 and 8 h (p < 0.001). In rabbits, the concentrations of LAT in AH and ICB were not statistically different between preserved latanoprost and PF SF latanoprost at each timepoint, except at 1 h in ICB (p = 0.005). By comparison, the LAT concentration of PF latanoprost was statistically (p < 0.05) lower than that of preserved latanoprost and PF SF latanoprost in AH and ICB from 0.5 to 3 h.