Background In a previous study a new hydrosoluble nail lacquer (P-3051) containing 8% ciclopirox (CPX) showed higher nail penetration compared to a water-insoluble 5% amorolfine (MRF) lacquer. To our knowledge, in vivo human data on a similar topic are not available.
Objectives To compare fingernail penetration of P-3051 with that of MRF reference in humans and to evaluate their predicted efficacy against Trichophyton rubrum and Candida parapsilosis.
Methods Single centre, randomized, multiple dose, open label, within subjects study. Test and reference were self-applied to all fingernails of either hand for 28 days. At baseline and after 15 and 25 days, the nail free edge was collected for analysis. Efficiency coefficients were calculated for T. rubrum and C. parapsilosis as ratios of nail concentration/minimum inhibitory concentration. The coefficients were classified as very high, high or poor.
Results Nail concentrations after 15 days were 2.82 ± 0.58 μg/mg for CPX and 0.64 ± 0.11 μg/mg for MRF. At day 25 there was a non-significant decline (1.85 ± 0.31 μg/mg, P = 0.077) for CPX and a highly significant (0.13 ± 0.03 μg/mg, P = 0.0002) 80% decline for MRF. Efficiency coefficients were very high/high in all subjects treated with P-3051 against both T. rubrum and C. parapsilosis; they were significantly lower for MRF reference against both pathogens at both observation points.
Conclusions P-3051 exhibited better penetration and higher predicted efficacy after in vivo multiple application to human fingernails when compared to MRF reference. These in vivo data are in good agreement with our previous in vitro study.
Onychomycosis, a common fungal infection of the nail unit affecting 2–10% of adult population, is currently managed with broad-spectrum antifungals delivered via systemic or topical routes.1–3 Topical agents penetrate and accumulate in the nail plate, with decreasing concentrations from the surface to the inner layer. In fact, the keratinized structure of the nail presents a high resistance to drug permeation, so that topical monotherapy is currently recommended only for mild to moderate disease, without matrix involvement.3–5 If delivery systems allowing more extensive permeation were available, topical monotherapy would probably result useful also in more severe stages of the disease, now treated with systemic compounds that can be associated with the risk of serious adverse reactions and drug interactions.6–8 Nail lacquers represent the most suitable topical formulations instead of creams and gels, as the permanence at the site of action is a critical factor for therapeutic success.5 Since the beginning of the 1990s, first generation medicated nail lacquers became commercially available in Europe. They are based on a water insoluble vehicle with 8% ciclopirox (CPX) or 5% amorolfine (MRF) as active ingredient. The water-insoluble polymers, similar to those of coloured nail polishes, create a thick, glossy and hard film on nail surface that entraps the active ingredient and obstructs diffusion from subsequent applications, thus requiring frequent removal by means of organic solvents and nail filing, possibly resulting in skin irritation and nail structure damages. We recently reported a study on a water-soluble nail lacquer (P-3051; 8% ciclopirox), based on hydroxypropyl chitosan (HPCH) that forms an invisible, non-irritating film, easily removed by water washing and applicable to perungual skin. This new formulation appeared more efficient in promoting in vitro CPX penetration through bovine hoof slices (as a model of human nail) compared to first generation CPX lacquer.9 Accordingly, the new formulation showed superior clinical efficacy in the treatment of onychomycosis.10 In a further in vitro study on bovine hoof slices we showed that transungual permeation and the antimycotic activities of P-3051 in terms of efficiency coefficients were definitely superior to those of MRF nail lacquer.11 On the contrary, no information is available about the kinetic behaviour of these two products in nails after in vivo administration to human subjects. Aim of the present study was to evaluate in vivo whether the different nail penetration seen in vitro11 after a single application of the two different nail lacquers is predictive of a similar behaviour following multiple applications to humans under the approved dosage schedule.
Materials and methods
The nail concentration study was performed in the frame of a phase I study directed to the investigation of clinical and instrumental endpoints. Fingernails were chosen rather than toenails for practical reasons and in order to obtain more homogeneous results. The end-points were: concentrations of the two active principles at the site of activity and efficacy against nail pathogens; growth rate of thumbnails in comparison with baseline; safety profile and suitability of the two treatments. Only data relative to nail penetration, predictive microbiological efficacy and safety are reported here.
Study design and treatment procedures
The whole study was performed at Cross Research S.A., Phase I Unit (Arzo, Switzerland), including random allocation sequence, enrolment and assignment of participants to interventions. The investigators screened 26 subjects and enrolled 24 subjects. First screening visit was performed on 14 April 2007 and follow-up ended on 31 May 2007.
The design was single centre, randomized, multiple dose, open label, within subjects. The study was fully GCP and ICH topic E6 compliant and the protocol was approved by the institutional ethics committee according to the local regulations. P-3051 (test) and MRF (reference) medicated nail lacquers were self-applied to all fingernails of either hand for 28 days. All subjects received both treatments, one treatment per hand. The subjects were randomized to receive test treatment on all fingernails of one hand and reference treatment on all fingernails of the other hand for the whole study duration. The random allocation sequence was computer-generated using the PLAN procedure of the validated SAS/STAT® software (SAS Institute Inc., Cary, NC, USA). P-3051 was applied once daily by a brush (dose regimen according to approved labelling); the reference was applied twice weekly by using a spatula (maximum dose as per the approved labelling). Both test and reference products were applied in the evening. The subjects were recommended not to wash their hands for at least 6 h after the application. Before each test application, the subjects were instructed to wash their hands with water and neutral soap and to dry them accurately; before each reference application, the subjects had to remove the previous product layer with an isopropyl alcohol swab. At baseline and after 15 and 25 days of application, after removing the residual product by the above mentioned procedures, the free edge of the nails was cut and collected for analysis (pool of five nails of each hand). According to the application schedule (once daily for P-3051 and twice weekly for the reference), on day 15 nail trimming occurred after 12–20 h from the application (both hands), while on day 25 it still occurred after 12–20 h from the application of P-3051, but after 84–92 h from the reference application. From the available literature data, steady state ciclopirox nail concentrations are reached in 2 weeks,12 while amorolfine is reported to form a 1-week depot just after a single application onto the nail.13.
The study was performed on 24 Caucasian healthy males, aged 21–40 years (mean 31.8), with a mean body weight of 77.3 kg and no current conditions or medical history likely to affect the outcome of the study or to represent a potential risk to the subject. Subjects of the same gender were used to minimize the individual variability of nail thickness. As literature reference was not available, the sample size was twice the minimum number of subjects foreseen by the CHMP guideline for bioequivalence studies.14
Drugs and chemicals
The composition of the clinical study drugs as declared by the manufacturers was as follows: (i) P-3051 (Polichem S.A., Lugano, Switzerland): 8% CPX, water, ethanol, hydroxypropyl chitosan, cetostearyl alcohol, ethyl acetate; (ii) Loceryl® (Galderma International S.A., France), MRF reference: 5% MRF, Eudragit RL100, triacetine, butyl acetate, ethyl acetate, ethyl alcohol.
The laboratory standards were as follows: CPX (Erregierre, S. Paolo D`Argon, Italy); ciclopirox-D5 (internal standard, Polichem S.A., Lugano, Switzerland); MRF hydrochloride (Ranbaxy, New Delhi, India); triazolam (1M methanolic solution in amber vial; internal standard, Cerilliant, Round Rock, TX, USA).
The analytical work was performed at IPAS (Ligornetto, Switzerland). CPX and MRF were assayed in human fingernail samples by liquid chromatography with tandem in space mass spectrometry detection (LC-MS-MS). The assay procedures were previously developed and validated according to international guidelines in the range 0.1–50 ng/mg nail for both CPX and MRF. Data were subjected to linear least square regression analysis: the mean correlation coefficients (r) were 0.9978 and 0.9973 for CPX and MRF, respectively. The limit of quantitation (LOQ) was 0.1 ng/mg nail for both substances. The mean absolute recovery from human nail over the investigated concentration range was 73.7 and 71.3% for ciclopirox and internal standard (I.S.), and 88.8% and 91.4% for MRF and I.S., respectively. Subsequently and only for CPX quantitation, 1000, 2000, 5000, 20 000, 50 000 fold dilutions have been validated in order to allow the ordinary analysis of out of range nail samples.
After an incubation period of 16 h at 37°C in a water bath with 1M NaOH solution, nail specimens plus IS were extracted with a mixture of triethylamine and n-hexane (CPX treated samples) or with diethylether. The organic phases were then separated and evaporated to dryness under stream of nitrogen at 50°C. The residue was dissolved in the appropriate mobile phase. Blank control hydrolysed nail solution was prepared following the same procedure.
The CPX or MRF amount recovered into the nails at baseline and after 15 and 25 days was determined as μg drug/mg nail. The efficiency coefficients were calculated as ratios between the drug recovered in the nail after application of P-3051 or reference at each time point and the minimum inhibitory concentration (MIC) for Trichophyton rubrum and Candida parapsilosis, as previously reported by ourselves.11 The obtained values were classified as very high when ≥5000 x MIC, high when ≥500 < 5000 x MIC, and poor when < 500 x MIC. Statistical analysis was performed by GraphPad Prism software (GraphPad Software Inc., San Diego, CA, USA).
The evaluation included descriptive analysis and between treatment comparisons by means of the Student’s two-tailed paired t-test (drug nail concentrations: day 25 vs. day 15, Fig. 1; efficiency coefficients: P-3051 vs. MRF reference, Figs 2 and 3) and Fisher’s exact test (incidence of subjects showing very high/high vs. poor efficiency coefficients, Table 3). Differences were considered statistically significant at P <0.05. Safety was evaluated by a local tolerability assessment on days 15 and 28. Irritation at the application site (left and right hand, respectively) was assessed using the 0–7 score scale of the US Guidance for irritation testing according to Table 1.15 Moreover, subjective symptoms (i.e. itching, burning and pain) were recorded.
Table 1. Irritation score scale
No evidence of irritation
Minimal erythema, barely perceptible
Definite erythema, readily visible; minimal oedema or minimal papular response
Erythema and papules
Erythema, oedema and papules
Strong reaction spreading beyond treatment area
All 24 subjects completed the treatment and performed all protocol procedures. Therefore, 24 individual sets of results were evaluated at the completion of the study. After opening the randomisation code, it was verified that the two formulations were applied on an equal number (60 each) of right and left hand fingernails.
CPX and MRF concentrations in the fingernail clippings were below the LOQ (0.1 ng/mL for both active principles) in all samples collected at baseline. During treatment, CPX was measurable at all time points. In the MRF group the active principle was measurable in all samples on day 15, while on day 25 appreciable levels could be measured only in 22/24 subjects. As shown in Fig. 1, nail concentrations for CPX were 2.82 ± 0.58 μg/mg after 15 days and declined by 34% to 1.85 ± 0.31 μg/mg on day 25: this difference was within the range of variability and not statistically significant (P = 0.077). Nail concentrations for MRF were 0.64 ± 0.11 μg/mg on day 15 and declined by 80% to 0.13 ± 0.03 μg/mg on day 25. This difference was highly significant (P = 0.0002).
The MIC values of CPX and MRF against T. rubrum and C. parapsilosis are reported in Table 2. These values are usually expressed as μg/mL. Since 1 mL of the culture media can be assumed to weigh about 1 g, the MIC is expressed as μg/g as reported by other authors,16 to be homogeneous with the nail concentrations.
Table 2. Minimum inhibitory concentration defined as 100% growth inhibition
Figures 2 and 3 show the efficiency coefficients of the drugs against the two pathogens. In all cases the efficiency coefficients of P-3051 were significantly higher compared to MRF reference. As shown in Fig. 2, against T. rubrum the P values were = 0.019 (day 15) and <0.0001 (day 25). As shown in Fig. 3, against C. parapsilosis the P values were = 0.0002 (day 15) and <0.0001 (day 25). Classification of the efficiency coefficients in very high, high or poor is reported in Table 3. They were very high or high in all subjects treated with P-3051 for both pathogens at day 15 and 25. Lower efficiency coefficients were found for the reference in comparison to P-3051: in particular, they were very high/high against T. rubrum in 23/24 (P = 1.00) and in 15/24 (P = 0.0008) subjects at day 15 and 25, against C. parapsilosis in 15/24 (P = 0.0008) and 1/24 (P < 0.0001) subjects at day 15 and 25, respectively. Both treatments were well tolerated, with only five adverse events occurring during the study in five different individuals. Two of them were unrelated: one allergic rhinitis and one headache. One occurred in the P-3051 applied hand: one mild and transient episode of yellow nail discoloration (judged as possibly related to the test treatment). Two subjects showed nail breakage (severity not reported) in the reference product applied thumb. An increase in nail fragility is a listed adverse event in the labelling of the reference. As regards the irritation test, the irritation score was 0 in all 24 subjects receiving either P-3051 or reference at both time points. Moreover, no subjective symptom was reported by any of the 24 study subjects.
Table 3. Classification based on the efficiency coefficients of either drug after 15 and 25 day multiple application of ciclopirox P-3051 or amorolfine reference to 24 healthy male volunteers
Very high ≥5000 x MIC
High ≥500 < 5000 x MIC
Poor <500 x MIC
Comparison between treatments by means of the Fisher’s exact test (subjects with very high/high vs. poor efficiency coefficients) showed no difference against T. rubrum after 15 days (P = 1.00) and significant superiority of P-3051 against T. rubrum after 25 days (P = 0.0008) and against C. parapsilosis after 15 days (P = 0.0008) and after 25 days (P < 0.0001).
MIC, minimum inhibitory concentration.
Number of subjects
P-3051 (MIC 0.25 μg/mL)
Amorolfine (MIC 0.1 μg/mL)
Number of subjects
P-3051 (MIC 0.5 μg/mL)
Amorolfine (MIC 0.8 μg/mL)
Our results met the objective to confirm in an in vivo study in humans the better nail penetration of ciclopirox after P-3051 multiple application compared to MRF reference. Two recent papers reported a comparative evaluation of the pharmacokinetic properties of the two topical antifungal agents. Neubert et al.17 evaluated their physicochemical characteristics and their release from commercial lacquer formulations using the Fourier transform infrared attenuated total reflection-technique. This study suggested that MRF may be more suitable for drug delivery to human nails, because it penetrates into the fingernails via the hydrophilic pathway. Since the nail is mainly constituted of hydrophilic materials, drug hydrophilic properties are a very important prerequisite for drug penetration. On the other side, using an in vitro hoof membrane model we showed11 that ciclopirox P-3051 permeates more easily than MRF and smaller amounts of subungual fluids collected after application of P-3051 lacquer than of MRF were required for a complete inhibition of fungal growth. The indications arising from these two different approaches are contrasting, as it often happens due to the well known pitfalls associated with extrapolations from both theoretical and in vitro data. The results of the present study, performed in vivo in humans, confirmed the far better penetration of ciclopirox into human nails respect to amorolfine. Consequently, the bovine hooves appear an adequate model of human nails, while the physicochemical properties described by Neubert17 do not appear sufficient to predict the kinetic behaviour in human nails.
A question arose about the appropriateness of the study design assuming that the steady state is reached within 2-week period for both drugs. So far, no in vivo study addressed the penetration of MRF in human nails measuring nail concentrations by means of a chemical analytical method. Nevertheless, evidence of MRF nail penetration has been reported using in vitro models or microbiological methods.17–20 Another paper claims MRF to provide long lasting depot, being one single application sufficient to cover 7 days.13 As regards CPX, steady state nail concentrations are known to be reached already after 2 week daily P-3051 application.12 Thus, the time points of our experiments are justified by the available literature. The presence of the active ingredient at the site of action is considered as a prerequisite of efficacy for all antimicrobial drugs, and antifungals used for the treatment of onychomycosis are no exception to this rule. The major drawback of topical antifungals is the failure to establish and maintain effective drug levels at the site of infection. Although various topical antifungal formulations have been developed, they have modest efficacy on account of their low permeability through the nail.21 Nonetheless, CPX first generation nail lacquer is known to penetrate into the nails22 and superiority over placebo was reported in two pivotal studies in patients with onychomycosis.23
P-3051 is a new hydrosoluble nail lacquer endowed with improved nail penetration properties in bovine hoof membranes as a model of human infected toenails.9,24 This formulation showed superior efficacy in the management of onychomycosis, compared to placebo and to the first generation nail lacquer containing the same active ingredient, namely ciclopirox.10 Conversely, no MRF controlled study vs. placebo in the treatment of onychomycosis is available in the literature. Thus, it is difficult to establish the efficacy of this drug.
The present study may represent a model for predicting the clinical efficacy of antimycotic drug formulations, provided the actual nail concentrations are corrected by the intrinsic antimycotic activity. For this reason, the analysis of the efficiency coefficients appears more suitable to predict efficacy than the concentrations alone. At time point day 15, MRF reference showed a certain predicted efficacy on onychomycosis, though to a lower extent compared to P-3051. The comparison between the data at day 15 and at day 25 showed a non-significant difference of nail concentrations for CPX. On the contrary, MRF concentrations at day 25 dropped significantly by 80% and efficiency coefficients decreased accordingly. This unexpected result is probably due to the fact that the steady state concentration of MRF was not reached at that time point, as at day 15 the nail trimming occurred after 12–20 h, while at day 25 the interval between MRF application and nail collection ranged between 84 h and 92 h. Our data on amorolfine disagree with those previously reported13,19 about the 7-day depot after just one application of commercial amorolfine nail lacquer. The main reason for this difference is due to the fact that previous literature data originate from in vitro investigations. In vitro nail penetration is generally considered as a good model for in vivo nail penetration in humans: this may be true for short time experiments, but in vitro studies cannot obviously account for all the environmental interactions occurring in vivo. Both finger and toe nails are continuously touching tissues and objects during real life, with a consequent loss of medication which may be negligible in the short run, i.e. during the first day after the application, but can become important after repeated applications. In our in vivo investigation, the nail concentrations of amorolfine fell by 80% in about 3.5 days. As this drug is presently approved for once a week or twice a week administration, in the clinical setting it might be opportune to revise the dosage schedule to a more frequent application to fill up this gap in the nail concentrations of the drug.
A possible disadvantage of the water soluble formulation may consist in the loss of medication due to nails exposure to water. Assuming that the volunteers of this study lost a part of the test product, P-3051, e.g. using the bathroom at night, this would simply strengthen the results of this study, because even in this disadvantageous situation, P-3051 penetrated the nails in a far larger amount than amorolfine. In the clinical setting, about 90% of patients are predicted to have onychomycosis of the toenails, and the eventual problems of compliance to the recommendation of refrain from washing during the first 6 h are minimized.
In conclusion, the new hydrolacquer technology of ciclopirox P-3051 confers to this drug a predicted clinical efficacy in the treatment of onychomycosis, which is far superior to that of the commercial amorolfine preparation, taking into account the respective capability to penetrate into the human nails and intrinsic antimycotic potency.