Onychomycosis is difficult to treat topically due to the deep location of the infection under the densely keratinized nail plate. In order to obtain an in vitro index that is relevant to the clinical efficacy of topical anti-onychomycosis drugs, we profiled five topical drugs: amorolfine, ciclopirox, efinaconazole, luliconazole, and terbinafine, for their nail permeabilities, keratin affinities, and anti-dermatophytic activities in the presence of keratin. Efinaconazole and ciclopirox permeated full-thickness human nails more deeply than luliconazole. Amorolfine and terbinafine did not show any detectable permeation. The free-drug concentration of efinaconazole in a 5% human nail keratin suspension was 24.9%, which was significantly higher than those of the other drugs (1.1-3.9%). Additionally, efinaconazole was released from human nail keratin at a greater proportion than the other drugs. The MICs of the five drugs for Trichophyton rubrum were determined at various concentrations of keratin (0-20%) in RPMI 1640 medium. The MICs of ciclopirox were not affected by keratin, whereas those of efinaconazole were slightly increased and those of luliconazole and terbinafine were markedly increased in the presence of 20% keratin. Efficacy coefficients were calculated using the nail permeation flux and MIC in media without or with keratin. Efinaconazole showed the highest efficacy coefficient, which was determined using MIC in media with keratin. The order of efficacy coefficients determined using MIC in keratin-containing media rather than keratin-free media was consistent with that of complete cure rates in previously reported clinical trials. The present study revealed that efficacy coefficients determined using MIC in keratin-containing media are useful for predicting the clinical efficacies of topical drugs. In order to be more effective, topical drugs have to possess higher efficacy coefficients.