Assessment of Dermatophytosis Treatment Studies: Interpreting the Data

assuming an unequivocal pre-study demonstration of fungal disease (preferably with speciation to allow post-hoc efficacy data analysis). The first question should be: exactly what disease was being treated? Most studies of tinea pedis address only the interdigital form, and FDA approval is based upon this single morphologic type.^21-23 Nonetheless, there are other forms of tinea pedis (moccasin-type and vesiculobullous) for which FDA approval is lacking, even though some degree of efficacy was suggested during clinical trials with naftifine.^24,25 Similarly, clinical trials for oncyhomycosis are conducted on the distal and lateral subungual form of disease (DLSO).^26-32 As is true of tinea pedis, there are other types of onychomycosis aside from the most common: proximal subungual, white superficial, total dystrophic, endonyx, and mixed.³³ However, since the clinical trials address only DLSO, and FDA approval only includes DLSO, there is no way for the HCP to know how likely it is that either any of the older or newer agents will be efficacious for these less common presentations.

Another inclusion criteria of major interest, especially with reference to onychomycosis trials, is the subjects’ allowable (and actual) age. Since nails grow more slowly with age, less efficacy might be perceived if a trial enrolled a substantial number of older patients compared with similar studies. As it turns out, although the tavaborole study had the oldest enrollee in any onychomycosis trial (aged 88 years),²⁷ virtually all the onychomycosis studies to date have had a mean age of study participants within a narrow range (43-53),^26-32 essentially negating this potentially confounding factor.

Another potential problem in the inclusion criteria in onychomycosis is the extent of involvement of the target nail(s). Interestingly, the trials involving oral agents have routinely had a higher percentage involvement (50%-75%) compared with the trials of topical antifungal agents (35%-40%). Since “complete cure” rates are reportedly higher for the oral drugs, apparently this difference in percentage of nail plate involvement does not place the oral antifungals at a major disadvantage.

Aside from inclusion criteria, reported antifungal efficacy may need to be interpreted in terms of exclusion criteria. For example, in onychomycosis studies, the HCP should take note if significant concomitant tinea pedis was sought and, if found, served as an exclusion. It is intuitive that a topical therapy (more so than a systemic one) may not work as well in treating onychomycosis if there is concurrent tinea pedis present to serve as a source of reinfection. Conversely, a pragmatic consideration might be to treat the tinea pedis concurrently with the onychomycosis in the hopes of securing a more beneficial outcome.³⁴

It goes without saying that certain studies would be of little value to select audiences. For instance, the American clinician cannot readily use data to facilitate therapeutic choices, when the agents (or one of the agents) being studied is bifonazole, tioconazole, amorolfine, fenticonazole, or flutrimazole, as these drugs are neither FDA-approved nor available over-the-counter in the U.S.

There are many methodological factors that might alter reported efficacy outcomes in antifungal studies. In fact, a classic publication reviewing the general subject of clinical studies provides a quantitative scoring system to assess the quality of randomized controlled trials.³⁵ It is considerably beyond the intent and scope of this article to apply this scoring system to the vast universe of antifungal studies. Moreover, this has already been done previously, in part, for both onychomycosis and other dermatophytosis clinical trials.^17,36 Suffice it to say that clinical studies involving more recently approved antifungal preparations are, according to the Jadad scale criteria,³⁵ acceptable, being of medium to high overall quality.^21-28,37,38 Despite the latter, there is at least one methodological feature which may confound the HCP ability to assess efficacy data, and that feature is study duration.

Consider that during onychomycosis treatment, visible clearance occurs as a healthy nail plate replaces a diseased one. This process requires about 12 to 18 months for a toenail.³⁶ All American trials involving recently approved topical treatments for toenail onychomycosis were conducted over 48 to 52 weeks.^26-28 By not taking into account the fact that it may well take over 70 weeks to grow out a toenail fully, the roughly 1 year of study may introduce uncertainty into the stated efficacy data. For example, if the drug has a reservoir effect, then the efficacy may actually be better than reported at 48 to 52 weeks because the agent continues to work as the healthy nail finally finishes growing out. Conversely, the drug’s effect may be optimized by administration past 48 weeks to insure drug presence as the nail completes its full growth. In the latter case, efficacy results are actually overstated, since residual infection and/or relapse may occur in the time period between study’s conclusion (48-52 weeks) and probable date of complete toenail regrowth (up to 72 weeks).