Susceptibility testing of amorolfine, bifonazole and ciclopiroxolamine against Trichophyton rubrum in an in vitro model of dermatophyte nail infection

Transcription

1 Medical Mycology November 2009, 47, Susceptibility testing of amorolfine, bifonazole and ciclopiroxolamine against Trichophyton rubrum in an in vitro model of dermatophyte nail infection MARTIN SCHALLER*, CLAUDIA BORELLI, URSULA BERGER, BIRGIT WALKER*, SYBILLE SCHMIDT *, GÜNTHER WEINDL* & ANDREAS JÄCKEL *Department of Dermatology, University of Tuebingen, Department of Dermatology, University of Munich, Institute of Statistics (StatBeCe), University of Bielefeld, and Galderma Laboratorium GmbH, Medical Department, Düsseldorf, Germany Introduction Nail infection is a common and chronic condition that is caused in 90% of cases by dermatophytes and most often, in about two-third of all cases, by Trichophyton rubrum ( T. rubrum ). In about three-quarters of patients, less than 75% of the nail surface area is affected without matrix involvement [1 ]. Received 28 June 2008; Final revision received 26 September 2008; Accepted 24 October 2008 Correspondence: Martin Schaller, Department of Dermatology, Eberhard Karls University Tuebingen, Liebermeisterstrasse 25, Tuebingen, Germany. Tel: ; fax: ; Martin.Schaller@med.uni-tuebingen.de Antimycotic nail lacquers are effective and safe for the treatment of onychomycosis. To assess the efficacy of three topical agents we studied the minimum inhibitory and fungicidal concentration of amorolfine, bifonazole and ciclopiroxolamine. Amorolfine showed the most effective fungistatic and fungicidal activity in vitro against seven clinical Trichophyton rubrum nail isolates, followed in descending order by ciclopiroxolamine and bifonazole. To mimic a nail infection more appropriately, the nail minimum fungicidal concentration (Nail-MFC) was determined in an onychomycosis model. Amorolfine and ciclopiroxolamine had Nail-MFCs ranging from 2 32 μg/ml and μg/ml, respectively. In contrast, bifonazole was unable to kill T. rubrum in this model. Statistical analyses of the results show a significant difference between the two treatments with amorolfine and ciclopiroxolamine ( P 0.001). For amorolfine a mean concentration of μg/ml (95%-CI [8.66, 17.41]) was sufficient to kill all strains, while for ciclopiroxolamine about twice that concentration was needed, i.e., μg/ml (95%-CI = [17.06, 34.13]). The individual sensitivity of six of the seven T. rubrum strains was higher for amorolfine. These data demonstrate that both amorolfine and ciclopiroxolamine effectively kill T. rubrum growing on nail powder and suggest a better cidal action for amorolfine. Further investigation would be required to determine if these in vitro data can partially explain the clinical observation of significantly higher cure rates in onychomycosis following a therapy with an amorolfine-containing nail lacquer formulation. Keywords Amorolfine, bifonazole, ciclopiroxolamine, Trichophyton rubrum, onychomycosis, nail infection, in vitro Oral antifungal drugs, such as terbinafine, itraconazole, and fluconazole are widely used, based on their well-established antifungal activity [ 2 ]. Because of their drug interactions and side effects, a clear need exists for antifungal drugs that can be administered topically, either in monotherapy or in combination with oral antifungals, so as to achieve higher mycological and clinical cure rates. A European consensus treatment guideline recommends topical monotherapy in the absence of matrix area involvement and when less than half of the nail surface is affected and combination therapy in more severe cases [ 3 ]. Three different topical nail lacquer formulations are available in Germany: amorolfine 5% (Loceryl Nail 2009 ISHAM DOI: /

2 754 Schaller et al. lacquer, Galderma Laboratorium GmbH) [ 4 ], ciclopirox 8% in a hydroxypropyl chitosan (HPCH) based, water soluble vehicle (Ciclopoli 8% Nail lacquer, Taurus Pharma) [ 5 ] and ciclopirox 8% (Nagel Batrafen, Sanofi- Aventis or Ciclopirox Winthrop Nail lacquer, Winthrop) [ 6 ]. The two latter mentioned are the same ciclopirox drugs pharmaceutically, containing the same galenic formulation, but marked under two different brand names by different pharmaceutical companies. A third antifungal substance, bifonazole, is available as a cream formulation containing 40% urea additionally (Canesten Extra Nail set, Bayer Vital GmbH) aimed at ensuring the non-traumatic removal of the nail plate. After the removal of the nail plate, a bifonazole cream needs to be applied subsequently for another 4 weeks in monotherapy [ 7 ]. These three antifungal agents have substantial activity against T. rubrum using the established CLSI (Clinical and Laboratory Standards Institute) in vitro procedures, which allows determination of antifungal susceptibility using an enriched growth medium. We also used in this study a recently established nail keratin-based culture onychomycosis model that was developed for testing antifungal compounds by presenting the test drug with the challenge of an established dermatophyte mycelium growing on and metabolising its natural substrate, the nail keratin [ 8 ]. The purpose of this study was to assess the activity of the topical antifungals that are the most commonly used, i.e., amorolfine, bifonazole and ciclopiroxolamine, in an in vitro test model of onychomycosis. Materials and methods Trichophyton rubrum strains The seven T. rubrum strains used in this study were isolates from infected nails that were collected at the mycological laboratory of the Department of Dermatology, University of Tuebingen, Germany. The strains have been recently isolated from nail material of untreated onychomycosis patients. A part of each sample was processed using conventional mycological techniques: Direct microscopic examination with tetramethylammonium hydroxide solution, followed by culture on Kimmig and Sabouraud Dextrose agar at 25 C for 4 weeks. Direct microscopic examination revealed septate hyphae. Downy white-coloured colonies with crimson pigment on the reverse were observed after 2 3 weeks incubation. Teardrop-shaped microconidias along septate hyphae were established after staining with lactophenol cotton blue on microscopic examination. The urease test was negative. Isolated species were classified as T. rubrum which was confirmed by PCR-ELISA (M. Schaller, unpublished results). To prepare stock inocula, cultures were grown at 30 C for 5 weeks on potato dextrose agar (PDA) (Oxioid LTD., Basingstoke, UK). The conidia and mycelium were then harvested, dispersed in Sabouraud 2% dextrose broth (Oxioid), and stored at 80 C after addition of a cryoprotectant, 5% (vol/vol) dimethyl sulfoxide (DMSO) (Sigma Chemical Co, Munich, Germany). Colony forming units (cfu) were determined, after rapid thawing, by spreading 50 μl from 10-fold serial dilutions in a physiological saline solution on PDA plates and counting the colonies after a 1-week incubation at 30 C. Antifungal drugs Amorolfine was a gift from Galderma Laboratorium GmbH, Düsseldorf, Germany. Bifonazole and ciclopiroxolamine were purchased from Sigma. The free acid derivative ciclopirox is the active agent in two identical topical nail lacquer formulations, marketed under different brand names (see introduction). Ciclopirox have the same antifungal activity as its chemical salt derivative ciclopiroxolamine [ 9 ], the active ingredient in other topical formulations such as cream, lotion, gel and solution. The antifungal activity is attributed to the hydroxypyridone group present in both molecules. Therefore, because ciclopirox is not commercially available as a pure chemical substance for experimental investigations, we used ciclopiroxolamine in our assays. All drugs were dissolved in DMSO. The tested concentrations were μg/ml for amorolfine and μg/ml for both bifonazole and ciclopiroxolamine in the onychomycosis model. For the determination of the minimum inhibitory concentration and fungicidal concentration the tested range was μg/ml for bifonazole, μg/ml for ciclopiroxolamine and μg/ml for amorolfine. Minimum Inhibitory Concentration (MIC) and Minimum Fungicidal Concentration (MFC) determination MIC was determined in 96-well flat-bottom assay plates with a final volume of 200 μl using a slight modification of the CLSI microdilution procedure M38-A [ 10 ]. A small volume of drug (10 μl), initially dissolved in DMSO was added to the desired concentration (final concentration of DMSO 1%). The size of the inoculum was 1.25 /10 3 cfu/ 200 μl. Plates were incubated for 4 5 days at 35 C. Growth inhibition was scored visually with the aid of an inverted magnifying mirror. The MIC for ciclopiroxolamine and for amorolfine corresponded to the lowest concentration inducing a 100% growth inhibition. A value of 80% growth inhibition was defined for the azole agent bifonazole ISHAM, Medical Mycology, 47,

3 Susceptibility testing in a model of dermatophyte nail infection 755 After MIC determination, starting from the last well in which growth was observed up to the highest drug concentration tested, 100 μl was transferred in duplicate onto potato dextrose agar (PDA) medium (Oxioid) in a 20 ml/9-cm-diameter plate. Plates were incubated for 1 week at 35 C followed by visual inspection of growth. MFC corresponded to the lowest drug concentration (in the assay microtitre plate) at which no more than one colony subsequently grew in the 9-cm plate, corresponding to 99.9 % killing activity. All strains were tested in duplicate. Preparation and collection of nail powder Human nail clippings were collected from healthy volunteers within Department of Dermatology. Finger and toe nails were pooled and ground to a powder using a stainless coffee grinder followed by a nail micronizer (Delasco, Council Bluffs, USA). The nail powder was washed with chloroform and then ethanol, prior to being autoclaved at 120 C for 20 min. Nail model culture system and determination of Nail-MFC Autoclaved nail powder (approximately 2 mg per well) was added to 96-well flat-bottomed plates. T. rubrum, 5 μl of cfu/ml, was added directly onto the nail powder, which was left at 35 C for 1 h. Then 200 μl RPMI select amine medium (Biochrom AG, Berlin, Germany) buffered with mol/l MOPS (Biochrom AG) was added to the wells containing the nail powder but also to control wells without nail material. The plate was returned back to 35 C for 5 days. A small volume of drug (10 μl), initially dissolved in DMSO, and then in medium, was added to the desired concentration (final concentration of DMSO 1%) and the culture was returned to 35 C. Drug incubation period of 4 weeks was investigated. After incubation with the drug, the content of each well was washed twice with RPMI 1640 (Biochrom AG) before being transferred via an inoculating loop to a PDA plate (containing no drug) for nail MFC (Nail-MFC) determination. Plates were incubated at 30 C, examined and photographed after 1, 2, 3 and 4 weeks. Nail-MFC was defined as the lowest drug concentration (in the assay microtitre plate) at which no more than two colonies subsequently grew in the 9-cm plate, corresponding to a 99.9% killing activity. Nail-MFC tests were performed at least six times for each T. rubrum strain with each antifungal drug. Statistical methods To compare the efficacy of the different treatments in the onychomycosis model we analysed the logarithm at basis 2 of dose, i.e., the power of the dose. Note that the average mean of this transformed outcome (log2(dose)) is equal to the logarithm of the geometric mean of the original outcome dose. As bifonazole treatment failed completely we focussed our analyses on amorolfine and ciclopiroxolamine. We calculated a linear mixed model which allowed us to take the intra-subject correlation into account, i.e., the correlation between the outcomes of the different experiments, which were performed for each strain. All analyses where performed in SPSS Results MIC and MFC susceptibility testing The MIC and MFC values of amorolfine, bifonazole and ciclopiroxolamine against the seven T. rubrum strains are given in Table 1. MIC and MFC were tested twice, with similar results. The observed MIC of amorolfine and ciclopiroxolamine showed a range of variability against the different T. rubrum strains while all strains presented an identical MIC μg/ml for bifonazole. Overall, the dermatophyte MIC values (geometric means) of all tested strains for the three antifungals were as follows: μg/ml for amorolfine, μg/ml for bifonazole and μg/ml for ciclopiroxolamine. As expected, the MFC of all tested antifungal drugs were higher than the MIC values. MFC values varied between 1 and 2 μg/ml for amorolfine and between 64 and 256 μg/ml for ciclopiroxolamine. No fungicidal activity at the highest test concentration of 256 μg/ml could be shown for bifonazole. Nail-MFC susceptibility testing The observed Nail-MFCs for all three tested antifungals are given in Table 2. The individual T. rubrum strains showed variations in their susceptibility against amorolfine and ciclopiroxolamine. The Nail-MFCs of the seven tested strains (6 12 independent experiments) varied between 2 and 32 μg/ml for amorolfine and between 16 and 32 μg/ml for ciclopiroxolamine and showed a significant difference between the two treatments ( P 0.001). For amorolfine a mean concentration of μg/ml (95%-CI = [8.66, 17.41]) was sufficient to kill all strains, while for ciclopiroxolamine about twice the concentration ( = 1.96) was needed, i.e., μg/ml (95%-CI = [17.06, 34.13]). The subject specific heterogeneity was estimated with a variance of In contrast to amorolfine and ciclopiroxolamine, bifonazole was unable to effectively kill T. rubrum growing on nail powder. In all cases, prominent fungal growth was observed on PDA plates also at the highest concentrations 2009 ISHAM, Medical Mycology 47,

4 756 Schaller et al. Table 1 Minimum inhibitory concentration (100% MIC for amorolfine and ciclopiroxolamine, 80% MIC for bifonazole) and minimum fungicidal concentration (MFC) values (n 2) of amorolfine, bifonazole and ciclopiroxolamine against seven Trichophyton rubrum strains Amorolfine (μg/ml) Bifonazole (μg/ml) Ciclopiroxolamine (μg/ml) Strain MIC MFC MIC MFC MIC MFC tested (256 μg/ml) after a 4-week incubation period with this drug ( Table 2 ). In the control wells without nail powder the dermatophyte failed to grow in the presence RPMI 1640 select medium, demonstrating that nail was an essential nutrient source in this assay. To exclude an inhibitory effect of chloroform and ethanol on fungal growth we compared the fungal growth on chloroform and ethanol washed nail powder with that on untreated nail material. Prominent nail growth could be seen in both samples. Discussion The model established by Osborne et al. [ 8 ] might be more useful for evaluating the therapeutic efficacy of antifungal agents against tinea unguium than the conventional in vitro methods such as CLSI, which has been modified to enhance the antifungal susceptibility testings of dermatophytes [ 11 ]. Moreover, the Nail-MFC results obtained in the present study showed a good reproducibility for most of the tested strains but it should be mentioned that the long incubation time of the proposed model results in a delay of the data outcome as compared to conventional testing according to the CLSI procedure. In the present study, the MIC (100%) values of amorolfine against all strains were 0.5 μg/ml whereas those of cicopiroxolamine were 16 μg/ml and those of bifonazole (MIC 80%) were μg/ml, which confirms the previous results that bifonazole is effective on proliferating dermatophytes in ng concentrations [ 12 ]. In contrast, the MFC results clearly demonstrate that bifonazole has no fungicidal effect on T. rubrum. A comparison of the MFC values for amorolfine and ciclopiroxolamine indicates a better cidal action for amorolfine. These results support the earlier findings that amorolfine appears to be the most effective topical agent to kill T. rubrum in vitro, followed by ciclopiroxolamine, while bifonazole is less effective [10,13,14 ]. Santos and Hamdan also tested the in vitro susceptibility of T. rubrum isolates to ciclopiroxolamine and found lower MIC levels ranging from μg/ml [ 15 ]. There are some variations in both studies concerning the inocula preparation and the determination of the MIC endpoints which might be responsible for the different results. The CLSI or other in-vitro methodologies for antifungal testing (MIC, MFC) using conidial suspensions growing in a rich medium, is not fully representative for evaluating the efficacy of antimycotics against T. rubrum infection of the human nail plate. We therefore used a recently described testing procedure in which the drug must overcome an established infection of nail powder [ 8 ]. Osborne et al. (2004) compared the Nail-MFC of terbinafine (2 μg/ml) with that of amorolfine, ciclopiroxolamine, clotrimazole, fluconazole, griseofulvin, itraconazole, and naftifine and Table 2 Nail-minimum fungicidal concentration (Nail-MFC) of amorolfine, bifonazole and ciclopiroxolamine against seven Trichophyton rubrum strains. Geometric mean and variability (n 6 12) after 4-week incubation with drug followed by 4-week incubation on PDA plates without drug Amorolfine (μg/ml) Ciclopiroxolamine (μg/ml) Strain Mean Variability Bifonazole (μg/ml) Mean Variability ISHAM, Medical Mycology, 47,

5 Susceptibility testing in a model of dermatophyte nail infection 757 found that all these antifungal agents were unable to effectively kill T. rubrum growing on nail powder at the highest concentrations tested. In the case of amorolfine and ciclopiroxolamine the highest concentrations investigated were 1 μg/ml and 128 μg/ml respectively. After application of the corresponding nail lacquer formulations on mycotic nails, concentrations detectable are μg/ml for amorolfine even in subungual tissue and about 10 μg/ml for ciclopirox in relevant nail layer depths (after one single application). Final drug concentrations in bovine hoof membranes are the same in transungual permeation experiments in vitro after applying ciclopirox either in a HPCH based vehicle (Ciclopoli Nail laquer) or in a non-hpch based formulation (Nagel Batrafen /Ciclopirox Winthrop Nail laquer) [ 5 ]. These data demonstrate that at least for amorolfine, the concentrations tested by Osborne et al. (2004) are below values detectable in onychomycosis patients [16 19 ]. Using this nail model culture system we therefore tested the cidal action of various concentrations ranging from μg/ml for amorolfine and μg/ml for both bifonazole and ciclopiroxolamine. A Nail-MFC 32 μg/ml of amorolfine (range: 2 32 μg/ml) or ciclopiroxolamine (range: μg/ml) was necessary to effectively kill all seven T. rubrum strains growing on nail powder. The individual sensitivity of all tested strains against the two antifungal agents was clearly higher for amorolfine than for ciclopiroxolamine. T. rubrum nail strains seemed to be more sensitive to lower concentrations of amorolfine as compared to those of ciclopiroxolamine. This may suggest a better mycological and clinical efficacy of amorolfine against T. rubrum nail isolates. In contrast to amorolfine and ciclopiroxolamine, bifonazole was unable to effectively kill T. rubrum growing on nail powder. As expected, amorolfine and bifonazole Nail-MFC obtained in the onychomycosis model are higher than the MFC values of the conventional assays. Interestingly, this was not the case for ciclopiroxolamine, which cannot be fully explained at the moment. One possible explanation might be the different pharmacokinetic properties of the agent in keratinized tissue or the different molecular mode of antimicrobial action of ciclopiroxolamine not affecting sterol biosynthesis [ 19 ]. To exclude deterioration of the antimycotic compounds after a 4-week incubation period we investigated the influence of incubation time on Nail-MFC (data not shown). Corresponding to the results of Osborne et al. we observed a decrease of the Nail-MFC values with longer exposure to the drug. These observations clearly demonstrate the activity and stability of the antimycotic drugs during and at the end of the 4-week exposure. Other interesting models have been cited in the literature in an attempt to better mimic the clinical course of onychomycoses in human nails and thus, treat more appropriately. One group reported binding affinity rates of some antifungals to keratin material and their in vivo efficacy directly on dermatophyte infected nail tissue of guinea pigs [ 20 ]. Another group used a different experimental in vitro Tinea unguium model to determine the antifungal efficacy of agents [ 21 ]. All these experiments confirmed a relevant and potent inhibition of dermatophyte growth in nail keratin by amorolfine. Increased concentrations and/or longer incubation times were necessary to effectively kill dermatophytes compared with classical standard in vitro media and was confirmed in our tests. We also want to mention that for the antimycotic drugs investigated, various therapeutic regimes have been recommended that will result in nail concentrations of the drugs different from those achieved in the in vitro assays. For this reason, it is not possible to draw any direct conclusions regarding the efficacy of the different drugs preparations used in clinical treatment from the in vitro data obtained in this study. For evaluation of the clinical relevance of Nail-MFC data for amorolfine presented here, it is relevant to notice, that after regular topical application of the amorolfine-containing lacquer on nails of patients with onychomycosis, even in subungual keratin material, comparable and even higher concentrations have been determined in vivo [18 ]. Whether these new in vitro data can reflect the significant higher clinical and mycological cure rates in onychomycosis following therapy with the amorolfine nail lacquer in monoor combination therapy compared to one ciclopirox nail lacquer formulation [22,23 ], requires further investigation. Acknowledgements The authors thank Prof. Walter Burgdorf (Department of Dermatology and Allergology, University of Munich) for critical reading of the manuscript. The experiments have been financially supported by Galderma Laboratorium GmbH, Germany. A. Jäckel is an employee of Galderma Laboratorium GmbH, Germany. M. Schaller and C. Borelli contributed equally to the work. Declaration of interest: The authors report no other conflicts of interest. The authors alone are responsible for the content and writing of the paper. References 1 Effendy I, Lecha M, Feuilhade de Chauvin M, Di Chiacchio N, Baran R. Epidemiology and clinical classification of onychomycosis. J Eur Acad Dermatol Venereol 2005; 19 (Suppl. 1 ): ISHAM, Medical Mycology 47,

6 758 Schaller et al. 2 Gupta AK, Ryder JE, Skinner AR. Treatment of onychomycosis: pros and cons of antifungal agents. J Cutan Med Surg 2004; 8: Lecha M, Effendy I, Feuilhade de Chauvin M, Di Chiacchio N, Baran R (2005) Treatment options development of consensus guidelines. J Eur Acad Dermatol Venereol 2005;19 (Suppl. 1 ): Flagothier C, Pierard-Franchimont C, Pierard GE. New insights into the effect of amorolfine nail lacquer. Mycoses 2005; 48: Monti D, Saccomani L, Chetoni P, et al. In vitro transungual permeation of ciclopirox from a hydroxypropyl chitosan-based, watersoluble nail lacquer. Drug Dev Ind Pharm 2005; 31: Gupta AK, Baran R. Ciclopirox nail lacquer solution 8% in the 21st century. J Am Acad Dermatol 2000; 43: Bonifaz A, Guzman A, Garcia C, Sosa J, Saul A. Efficacy and safety of bifonazole urea in the two-phase treatment of onychomycosis. Int J Dermatol 1995; 34: Osborne CS, Leitner I, Favre B, Ryder NS. Antifungal drug response in an in vitro model of dermatophyte nail infection. Med Mycol 2004; 42: Summary of product characteristics (Fachinformation) Batrafen Nail lacquer (Nagel Batrafen ); Chapter 5 Pharmacodynamic properties ; Status Aug Li RY, Wan Z, Wang AP, et al. In vitro susceptibility testing of amorolfine in pathogenic fungi isolated from dermatomycosis patients in China. Mycoses 2004; 47: Jessup CJ, Warner J, Isham N, Hasan I, Ghannoum MA. Antifungal susceptibility testing of dermatophytes: establishing a medium for inducing conidial growth and evaluation of susceptibility of clinical isolates. J Clin Microbiol 2000; 38: Plempel M, Regel E, Buchel KH. Antimycotic efficacy of bifonazole in vitro and in vivo. Arzneimittelforschung 1983; 33: Feuilhade De Chauvin M, Marie M-D, Claire L. In vitro antifungal activity of amorolfine, bifonazole, ciclopirox and terbinafine against Trichophyton rubrum and Trichophyton mentagrophytes This paper was first published online on ifirst on 16 April var. interdigitale. J Eur Acad Dermatol Venereol 2003 (Suppl. 3 ); 17: S Favre B, Hofbauer B, Hildering KS, Ryder NS. Comparison of in vitro activities of 17 antifungal drugs against a panel of 20 dermato phytes by using a microdilution assay. J Clin Microbiol 2003; 41: Santos DH, Hamdan JS. In vitro antifungal oral drug and drugcombination activity against onychomycosis causative dermatophytes. Med Mycol 2006; 44: Hui X, Wester RC, Barbadillo S, et al. Ciclopirox delivery into the human nail plate. J Pharm Sci 2004; 93: Polak A. Kinetics of amorolfine in human nails. Mycoses 1993; 36: Mensing H, Polak-Wyss A, Splanemann V. Determination of the subungual antifungal activity of amorolfine after 1 month s treatment in patients with onychomycosis: comparison of two nail lacquer formulations. Clin Exp Dermatol 1992; 17 (Suppl. 1 ): Bohn M, Kraemer KT. Dermatopharmacology of ciclopirox nail lacquer topical solution 8% in the treatment of onychomycosis. J Am Acad Dermatol 2000; 43 (Suppl. 4 ): Tatsumi Y, Yokoo M, Senda H, Kakehi K. Therapeutic efficacy of topically applied KP-103 against experimental tinea unguium in guinea pigs in comparison with amorolfine and terbinafine. Antimicrob Agents Chemother 2002; 46: Nakashima T, Nozawa A, Ito T, Majima T. Experimental tinea unguium model to assess topical antifungal agents using the infected human nail with dermatophyte in vitro. J Infect Chemother 2002; 8: Halmy K. Experience with nail lacquers containing amorolfine 5% and ciclopirox 8% in patients with onychomycosis. J Eur Acad Dermatol Venereol 2004; 18 (Suppl. 1 ): Baran R, Kaoukhov A. Topical antifungal drugs for the treatment of onychomycosis: an overview of current strategies for monotherapy and combination therapy. J Eur Acad Dermatol Venereol 2005; 19: ISHAM, Medical Mycology, 47,