Assessment of the nail penetration of antifungal agents, with different physico-chemical properties

Abstract

Onychomycosis, or fungal nail infection, is a common fungal infection largely caused by dermatophyte fungi, such as Trichophyton rubrum or Trichophyton mentagrophytes, which affects a significant number of people. Treatment is either through oral antifungal medicines, which are efficacious but have significant safety concerns, or with topical antifungal treatments that require long treatment regimens and have only limited efficacy. Thus, an efficacious topical therapy remains an unmet medical need. Among the barriers to topical delivery through the nail are the physico-chemical properties of the antifungal drugs. Here, we explore the ability of a range of antifungal compounds with different hydrophilicities to penetrate the nail. Human nail discs were clamped within static diffusion (Franz) cells and dosed with equimolar concentrations of antifungal drugs. Using LC-MS/MS we quantified the amount of drug that passed through the nail disc and that which remained associated with the nail. Our data identified increased drug flux through the nail for the more hydrophilic compounds (caffeine as a hydrophilic control and fluconazole, with LogP -0.07 and 0.5, respectively), while less hydrophilic efinaconazole, amorolfine and terbinafine (LogP 2.7, 5.6 and 5.9 respectively) had much lower flux through the nail. On the other hand, hydrophilicity alone did not account for the amount of drug associated with/bound to the nail itself. While there are other factors that are likely to combine to dictate nail penetration, this work supports earlier studies that implicate compound hydrophilicity as a critical factor for nail penetration.

Fig 1. Static diffusion (Franz) cell.

Photo (a) and schematic diagram (b) of a Franz cell, showing the sample chamber (1) and collar (2), made from stainless steel, and lower collection chamber (3) made of quartz. The nail sample was placed on a lip in the collar such that the upper surface of the nail was orientated upwards, the sample chamber was then screwed onto the collar, clamping the nail in place. Compounds were applied to the well created by the sample chamber and the top of the nail plate. The collection chamber was filled with UPW.

Fig 2. Quantification of compounds associated with nail samples.

Nail lysates were prepared from nail discs and analysed by LC-MS/MS. Caffeine could not be identified due to lack of stability in 5 M NaOH used to lyse the nail (Table C in S1 File). Data were normalised to the weight of the individual nail samples. Error bars represent standard error of the mean of data from 4–5 different nails for each compound.

Fig 3. Total drug that passed through the nail.

(a) The number of Franz cells in which drug was detected to have passed through the nail. (b) Drug flux through the nail. Error bars representing standard error of the mean are shown for compounds where at least 3 values were above the LLoQ.