A multicentre, randomised, parallel‐group, double‐blind, vehicle‐controlled and open‐label, active‐controlled study (versus amorolfine 5%), to evaluate the efficacy and safety of terbinafine 10% nail lacquer in the treatment of onychomycosis

Onychomycosis is a difficult‐to‐treat fungal nail infection whose treatment can involve systemic or topical antifungal approaches.


| INTRODUC TI ON
Onychomycosis is a chronic fungal nail infection that results in thickening, discoloration, splitting of the nails and lifting of the nail plate from the nail bed, mainly affecting toes and due to dermatophytes species. [1][2][3][4] Onychomycosis has a high incidence among the general adult population of western countries, accounting for approximately 50%-60% of all nail disorders, and can have negative consequences for patients as a result of pain or physical impairment. The poor cosmetic appearance of onychomycosis can also result in embarrassment and undermine emotional, social and occupational functioning. [5][6][7] It is a chronic, difficult-to-treat disease characterised by low cure rates and high relapse rates. Moreover, currently available systemic treatments suffer from limitations due to serious adverse events and possible drug-drug interactions. Topical treatments are better tolerated, but their efficacy is lower. 6 Terbinafine hydrochloride, a synthetic allylamine, is a wellknown antifungal agent that exerts its antifungal activity by noncompetitive inhibition of squalene epoxidase, a key enzyme in the biosynthesis of fungal ergosterol which is essential for membrane integrity. Terbinafine has a fungistatic effect due to the depletion of ergosterol in fungal cell membranes, and a fungicidal effect resulting from the toxic accumulation of intracellular squalene. [8][9][10] Terbinafine 10% nail lacquer is an antifungal formulation specifically designed for the treatment of onychomycosis due to dermatophytes and/or other terbinafine-sensitive fungi.
The aim of this randomised double-blind phase III study was to assess the efficacy and safety of terbinafine 10% nail solution in comparison with its vehicle, following 48 weeks of treatment. In addition, following an amendment to the protocol, an open-label third arm of treatment with amorolfine 5% was implemented prior to randomisation.

| Study design and treatment
This was a phase III, multicentre, randomised, parallel-group, vehicle-controlled double-blind study that aimed to evaluate the efficacy and safety of terbinafine 10% nail lacquer administered during 48 weeks. The study also included a comparison with open-label amorolfine 5%. Terbinafine 10% nail lacquer, formulated in a hydroalcoholic solution of hydroxypropyl chitosan, was applied topically once daily for the first 4 weeks and once weekly for the remaining 44 weeks. Terbinafine 10% nail lacquer and vehicle lacquer were applied to dry nails, preferably in the evening, just before retiring to bed, to allow treated nails to remain unwashed and dry for at least 6 h. In addition, amorolfine nail lacquer 5% (Loceryl ® ) was used as the active comparator, being administered for 48 weeks in an openlabel manner due to different posology and method of administration, with the remaining nail lacquer having to be removed and the affected nails filed before each new application. 11 The trial was conducted between August 2015 and September 2018 and recruited a total of 953 patients from 114 sites in 13 European countries. The study was reviewed and approved by relevant Institutional Review Boards/Independent Ethics Committees and was conducted in compliance with the study protocol, the recommendations on biomedical research on human subjects included in the Declaration of Helsinki, International Conference of Harmonization-Good Clinical Practice (ICH-GCP) Guidelines, and all applicable national laws and regulations.
All patients provided written informed consent.

| Patients
Male and female subjects >12 years of age with a mild-to-moderate DLSO in at least one big toenail due to dermatophytes were eligible to be screened for the study. Mild-to-moderate toe onychomycosis was clinically defined as involvement of ≥20% and ≤50% of the target big toenail area without lunula/matrix involvement, spikes/dermatophytoma or onychodystrophy, and with nail thickness not exceeding 2 mm; a positive potassium hydroxide (KOH) microscopy examination and positive culture from nail samples of the target big toenail for dermatophytes or mixed dermatophytes/Candida were prerequisites for randomisation. In addition, adequate reported target big toenail growth was required. Re-screening within two weeks was allowed in cases of negative KOH microscopy or culture results. Moreover, patients having positive mycology findings for the target big toenail, but with an affected area slightly less than that required by inclusion criteria could also be re-screened once within at least 1 month.
The main exclusion criteria included allergy to medications or excipients; use of cosmetic products such as nail polish on diseased nails from 24 h prior to screening until the end of the study; use of systemic antifungal drugs in the 24 weeks prior to the screening visit; nail application of topical antifungal drugs or devices in the 4 weeks prior to the screening visit; presence of nail conditions that could confuse clinical assessment (non-dermatophyte infections, onychodystrophy, dermatophytoma, presence of "yellow spikes" [defined as longitudinal streaks extending from the free edge of the nail to the proximal edge] on the target nail); nail abnormalities due to conditions such as psoriasis, lichen planus, immune dysfunction, collagen-vascular diseases and peripheral vascular disease; women who were pregnant/planning a pregnancy or breastfeeding; use of any investigational drug/device or participation in a previous clinical trial within 4 weeks of the screening visit; use of chemotherapy, immunosuppressive therapy in the 12 weeks prior to the screening visit; use of systemic corticosteroids, antimetabolites and immunostimulants in the 4 weeks prior to the screening visit; HIV infection or any other immunodeficiency; and alcohol or substance abuse.

| Study plan
The study consisted of a screening phase of up to 10 weeks, a treatment phase of 48 weeks and a 12-week follow-up period. Eight clinical examinations were planned for each patient at screening, randomisation (Week 0), treatment (Weeks 4,12,24,36 and 48) and follow-up at Week 60 (Table 1). Eligible patients were randomised to treatment with terbinafine, vehicle or amorolfine according to permutated block randomisation (ratio 3:3:1).

| Efficacy and safety assessments
Patients applied the appropriate nail lacquer (terbinafine, vehicle or amorolfine) to the target big toenail and all other toenails, as well as to any fingernails suspected of fungal infection, as determined by the investigator.
A summary of procedures conducted at each visit is shown in Table 1. Demographic data, medical history/current diseases and prior and concomitant medications were recorded at screening. were "not at all" (score 1) to "extremely bothersome" (score 5).
Responses within each category were summed, and the mean calculated. Mean scores were transformed into a 0-100 scale ranging from 'extremely bothersome' (0) to 'not at all' (100). Medication acceptability was rated by patients at Week 48 using a 4-point scale, and local tolerability at the application site was also evaluated.

| Efficacy and safety endpoints
The primary endpoint was complete cure rate at Week 60 (Visit Secondary efficacy endpoints were responder rate at Week 60, defined as negative KOH microscopy, negative culture for dermatophytes and ≤10% residual involvement of the target toenail; mycological cure rate at Week 60, defined as negative KOH microscopy and negative culture for dermatophytes of the target toenail; and overall safety by recording all adverse events (AEs). AEs were classified by severity (mild, moderate and severe), seriousness, and causality (not related, unlikely, possible and probable) and Medical Dictionary for Regulatory Activities (MedDRA V.18.0) terminology.
Supportive efficacy endpoints were the modified cure rate at Week 60, defined as negative culture for dermatophytes and no residual clinical involvement (nail totally clear) of the target toenail; modified responder rate at Week 60, defined as negative culture for dermatophytes and ≤10% residual involvement of the target toenail; and rate of negative dermatophyte cultures for the target toenail at each visit. Patient-reported outcomes were change in ONYCHO QoL questionnaire responses from baseline to Weeks 48 and 60, and medication acceptability at Week 48.
Exploratory efficacy endpoints were evaluation of each of the three components of the primary endpoint.

| Sample size
Power calculations showed that inclusion of 420 patients in each treatment group would provide ≥94% power, at the 2-sided 5% level of significance, to detect a 6.7% difference in complete cure rates between terbinafine 10% nail lacquer and vehicle at 60 weeks, assuming a rate of 11% for terbinafine 10% nail lacquer and 4.3% for vehicle. In addition, approximately 140 patients were randomised to amorolfine 5% according to the established randomisation ratio of 3:3:1, and only for exploratory purposes. Overall, the total sample size was estimated to be approximately 980 patients.

| Study populations
The main efficacy analyses were performed on the intention-totreat (ITT) population comprising all patients randomised and dispensed study medication. All safety analyses were conducted on the Safety population comprising all randomised patients who received at least one application of the assigned study drug. The per protocol (PP) population comprised all patients in the ITT population with a positive nail sample microscopy examination at screening involving dermatophyte(s) with or without coinfection with Candida spp; with confirmed clinical eligibility criteria as assessed by the IPR; who did not take forbidden medications and who completed the study without any major protocol violations. The PP population analysis was considered supportive to the ITT efficacy analysis.

| Statistical analysis
All statistical analyses were performed using SAS, version 9.4.
All efficacy and safety variables were summarised using descriptive statistics. For the primary and the key secondary endpoints, odds ratios were calculated from the logistic regression model for complete cure adjusted by treatment group, study site and treatment-by-site interaction. In case of convergence failure of the model, the same analysis was performed excluding site factor and treatment-by-site interaction from the model. The three components of the primary endpoint (KOH, dermatophytes culture and residual clinical involvement) were summarised using descriptive statistics. Similar analyses were performed for secondary, supportive and exploratory efficacy endpoints. All efficacy analyses were performed in the ITT population using the last observation carried forward (LOCF) approach for missing values. Primary efficacy and exploratory efficacy analyses (ie the single components of the complete cure rate) were repeated in the PP population to assess the robustness of the results. Furthermore, five sensitivity analyses for handling missing data were performed: complete-, best-and worst-case analysis, discrete repeated measure analyses, and a multiple imputation analysis using a Monte Carlo Markov Chain (MCMC) simulation.
The overall type I family-wise error rate for testing the primary and the two key secondary efficacy endpoints was controlled at the 5% significance level using a two-step serial gatekeeping multiple comparison procedure, and the Holm-Bonferroni method in the second step where the key secondary endpoints were tested.
ONYCHO results were calculated using an analysis of covariance (ANCOVA) model for the change in scores from baseline to Week 60 using the baseline value as a covariate; and treatment group, site and treatment-by-site interaction as factors.
For all inferential analyses, two-sided p-values ≤ .05 were considered statistically significant.

| Patient disposition and demographics
Over a 10-week period, a total of 4944 persons were screened for mild-to-moderate DLSO due to dermatophytes involving ≥20% to ≤50% of a target big toenail, and 953 patients were randomised to receive terbinafine (n = 406), vehicle (n = 410) or amorolfine

| Primary efficacy outcome
Complete cure rates at Week 60 (ITT population, LOCF approach) were higher in the terbinafine group (5.67%) compared with the vehicle (2.20%) and amorolfine (2.92%) groups. Patients treated with terbinafine had a statistically significant higher complete cure rate compared with vehicle (p = .0138), but not amorolfine (p = .2095) ( Table 3). Odds ratios for comparisons of terbinafine with vehicle and with amorolfine are shown in Table 3 and graphically in

TA B L E 2 Demographic and baseline characteristics (ITT population)
Comparable results for the primary endpoint were obtained in the PP population and after using different imputation schemes for missing data in the ITT population.

| Safety
In the safety analysis set, mean treatment compliance was high: A total of 52 patients (5.47%) had serious TEAEs with a similar incidence in the terbinafine (5.17%) and vehicle (5.41%) groups, and a slightly higher incidence in the amorolfine group (6.57%). Most reported TEAEs were mild in severity, but 39 patients (4.11%) reported one or more severe TEAEs: rates in the terbinafine, vehicle and amorolfine groups were 4.19%, 3.44% and 5.84%, respectively. Drugrelated TEAEs, defined strictly as any AEs with a causal relationship to study drug of probable, possible or unlikely in the opinion of the investigator, were reported in 35 patients (3.68%): with similar rates in terbinafine (3.94%), vehicle (3.44%) and amorolfine (3.65%) groups (Table 5). The most commonly reported TEAEs suspected to be related to the study drug were categorised as 'skin and subcutaneous tissue disorders' (n = 18, 1.89%). However, most patients (>94%) in each treatment group had no evidence of irritation of the toenails or the surrounding skin at each study visit. Only one patient (in the vehicle group) had some evidence of fingernail irritation (at Week 24).
The incidence rates of new toenail irritation events were very low (<1%) and similar in all treatment groups.
Five patients (0.53%) discontinued treatment due to a TEAE: 4 (0.99%) in the terbinafine group and 1 (0.25%) in the vehicle group. Two patients (0.21%) discontinued due to a TEAE assessed as 'unlikely' by the investigator, but analysed as drug related: 1 (0.25%) in the terbinafine group (epilepsy, severe) and 1 (0.25%) in the vehicle group (onychomadesis, moderate). One patient in the terbinafine group discontinued treatment due to a non-drug related TEAE (mild arthralgia), and two patients (0.49%) in this group had a fatal TEAE (pancreatic carcinoma and cerebrovascular accident, which were considered not to be related to treatment by the investigator).
No differences between treatment groups in mean haematology, serum chemistry and urinalysis parameters were detected throughout the study.

| DISCUSS ION
Onychomycosis is a chronic fungal infection of the nails that re- Currently available systemic treatments for onychomycosis have low cure rates and high relapse rates and may be limited due to serious adverse events and possible drug-drug interactions.
Topical treatments are better tolerated, but their efficacy is lower.
Approved oral onychomycosis therapies include terbinafine and itraconazole, and in Europe, fluconazole is recommended when other agents are not considered appropriate. 6,13 Preference for oral terbinafine is based on its better cure rates and fewer drug-drug interactions. Topical approved therapies in Europe include ciclopirox 8% (in water-soluble or insoluble formulations), amorolfine 5% and tioconazole nail lacquers. However, there is still no treatment that achieves high mycological and complete cure rates and recurrence following onychomycosis therapy is common. 14 Terbinafine is a non-competitive inhibitor of squalene epoxidase. 9 The consequences of squalene epoxidase inhibition are twofold: a fungistatic effect as a result of ergosterol depletion from fungal cell membranes that contributes to impaired growth of the pathogen, and also a fungicidal effect due to the intracellular accumulation of squalene, which perturbs phospholipid membranes and results in cell death. 2,10 However, oral terbinafine is associated with AEs including hepatotoxicity that may occur in patients with or without pre-existing liver disease. Consequently, monitoring of liver function before initiating terbinafine therapy is recommended and, where indicated by a patient's medical history and/or concomitant medications, potentially also during therapy. 14  The safety profile of topical terbinafine showed that the agent was well-tolerated with no important adverse reactions arising, consistent with the findings of the previous study on the different formulation. 18 Compliance was generally high and acceptance good in all treatment groups. In contrast, adverse drug reactions (ADRs) have been reported for orally administered terbinafine. Most ADRs of oral terbinafine involve the gastrointestinal system and skin, and are generally mild-to-moderate in severity and transient. While gastrointestinal symptoms are very common, rarely serious hepatic dysfunction may occur, although most cases involve asymptomatic and reversible elevation of liver enzymes. [21][22][23] In conclusion, the results of the present study indicate that topical terbinafine is an effective treatment with respect to both clinical and mycological criteria when compared with vehicle for patients with mild-to-moderate onychomycosis of the distal-lateral subungual type. In addition, results suggest that there may be some benefits compared to the currently available topical agent, amorolfine.
Overall, this newer formulation of topical terbinafine was safe to use and well-tolerated, producing very few adverse local effects.

ACK N OWLED G EM ENT
Writing and Editorial assistance was provided by Content Ed Net (Madrid, Spain) with funding from Almirall SA (Barcelona, Spain).
Open access funding enabled and organized by ProjektDEAL.

CO N FLI C T S O F I NTE R E S T
UBP received honorarium and grants from Almirall, Boots Healthcare, Cassiopea, Johnson & Johnson, LEO, Lilly, Neuroderm, Novartis,