In this open-label, prospective, multicenter, comparative study, male or female patients aged ≥2 years with a facial eczematous disease were recruited. Among them, patients with AD (diagnosed according to the criteria of Hanifin and Rajka) or SD (diagnosed by two professional dermatologists) that was rated at least moderate (≥3) on the investigator global assessment (IGA) scale were eligible for the stabilization phase of the study (phase I). The IGA uses a standard six-point scale of SD or AD severity (0=clear, 1=almost clear, 2=mild, 3=moderate, 4=severe, and 5=very severe). The exclusion criteria were any skin disorders other than AD in the area or areas to be treated, clinically infected skin diseases (bacteria, fungus, and virus), and extensive scarring or pigmented lesions in the area or areas to be treated.

During the study, the patients were prohibited from receiving ultraviolet therapy, nonsteroidal immunosuppressive agents, other TCS or TCI products, systemic corticosteroids, and other investigational drugs. The washout period for these therapies ranged from a minimum of 1 week (for TCSs and TCIs) to a maximum of 6 weeks (for ultraviolet treatments). The washout period for systemic nonsteroidal immunosuppressant and systemic corticosteroids used to treat AD was 2 weeks. None of the above-listed medications were permitted during the study (phases I and II), except as a study medication. Topical nonmedicated moisturizers were permitted throughout the study as long as they were applied at least 2 h before or 2 h after the application of tacrolimus ointment. Sunscreens and makeup were permitted but were not to be applied immediately before or 30 min after the application of any study treatments. The dose of systemic antihistamines was decreased or discontinued but not increased. Systemic antibiotics were allowed for the treatment of infections, as needed. In addition, the patients were allowed to use bath oils and nonmedicated emollients 2 h before and after the application of any study ointments.

This was a 20-week, prospective, multicenter, comparative study conducted at nine centers in Korea. The study was conducted in accordance with the ethical principles described in the Declaration of Helsinki. Adult and pediatric patients with facial SD or AD who were clear of disease after up to 4 weeks of treatment with tacrolimus ointment (a stabilization phase, phase I) entered the maintenance phase (phase II) of twice-weekly treatment with tacrolimus ointment (0.1%) for 20 weeks (Fig. 1). The study protocol was approved by the institutional review board of each hospital, and all patients (or legal caretakers) gave written informed consent before enrollment. All study treatments were applied to the active or previously involved areas of AD (IRB No. 2012-12-83, Kangnam Sacred Heart Hospital).

During phase I, patients with facial SD or AD applied 0.1% tacrolimus ointment (Protopic; Astellas Pharma Korea Inc., Seoul, Korea) to the entire affected facial areas twice daily for a minimum of 7 days and a maximum of 4 weeks. At any time after 7 days, patients with an IGA score of 0 or 1 (clear or almost clear) were eligible to enter phase II. If neither of these scores were achieved by 4 weeks, the patients were excluded from the study.

During phase II, 0.1% tacrolimus ointment was applied twice weekly for 20 weeks to facial areas previously affected by SD or AD. Patients with disease relapse (defined as an IGA score of >2) were excluded from the study.

Patient assessment was conducted at baseline/day 1; weeks 1, 2, and 4/end of phase I; and at weeks 4, 8, 16, and 20/final visit of phase II. At the baseline visit, the investigator confirmed the diagnosis of SD or AD, and then identified and evaluated all areas to be treated. Clinical measurements were done on separate facial regions (forehead, periorbital, nose, chin, cheek, and perioral/periauricular regions; Fig. 2) and on the whole face (the average severity of each region). The presence of disease on each facial part, the visual analogue score (VAS) for itching, and the IGA score were determined, and all adverse events were recorded at all visits during phases I and II. As this study was a comparative study of two different diseases, the IGA instead of both the eczema area and severity index and seborrhea area and severity index was chosen for the evaluation of the severity of facial AD and SD to ensure consistency. Also, for estimating the placebo effect, the results were compared with those of the placebo group of a recent study of proactive treatment with 0.1% tacrolimus ointment twice weekly in Korea24. Raw data were obtained from the authors of the paper24.

The primary endpoint was the time to the first disease exacerbation (DE) during phase I that required a substantial therapeutic intervention, which was defined as the use of 0.1% tacrolimus ointment for >7 days to treat an area with an IGA score of 3~5 as measured on day 1 of the DE treatment period. If a patient had two DEs within <7 days (with or without any type of treatment), the episodes were combined and considered as 1 DE. The secondary study endpoints included the total number of DEs during phase II that required a substantial therapeutic intervention and treatment days for DE. Clinical improvement was assessed by using the IGA score and VAS. Safety assessment during the study included the monitoring of adverse events. An adverse event was defined as any untoward occurrence in a patient, regardless of whether it was related to the study treatment, reported by the patient, or observed by the physician.

The demographic differences between the two treatment groups were analyzed by using Fisher's exact test and the χ² test for categorical data, and analysis of variance for continuous data. The Kaplan-Meier method and log-rank test were used to obtain and compare the corresponding 95% confidence intervals for the time to first disease relapse, and to estimate the median time to first disease relapse in those who experienced at least one relapse. The two treatment groups were compared by using the Wilcoxon rank sum test stratified according to disease severity and patient age. The differences between the treatment groups were analyzed by using Student's t-test for the treatment days for DE episodes and the mean IGA score for relapse episodes. The differences between the treatment groups were analyzed by using the χ² test and one-way ANOVA in the number of disease relapse episodes and the IGA score for the first relapse, and the Mantel-Henzel test for their trend. Statistical analyses were performed with SPSS 18.0 (SPSS Inc., Chicago, IL, USA). A p-value of <0.05 was considered statistically significant.

A total of 123 SD patients and 75 AD patients were enrolled in phase I of the study. The mean age was significantly different between the two groups (43.13±18.80 years for the SD group vs. 23.15±18.21 years for the AD group). Other characteristics, such as sex ratio and disease severity as assessed with IGA, were similar in the two groups.

At the completion of phase I, 83 of the 123 (67.5%) SD patients and 49 of the 75 (65.3%) AD patients showed an IGA score of <2 ("clear" and "almost clear") and completed phase I. The remaining 40 and 26 patients were excluded. The primary reason for exclusion at the completion of phase I was an IGA score ≥2 (28 of the 123 [22.8%] SD patients and 14 of the 75 [18.7%] AD patients). Early exclusion occurred because of changes in blinded study medications (3 of the 40 SD patients and no AD patients), noncompliance (7 of the 40 excluded SD patients and 10 of the 26 excluded AD patients), and prohibited medication (2 of the 40 SD patients and 2 of the 26 AD patients) (Fig. 3).

A total of 27 SD patients (30.0%) and 15 AD patients (20.0%) experienced adverse events that were assessed by the investigators of this study. Burning sensations, erythema, folliculitis and pruritus at application sites were common adverse events, some of which were overlapped each other (Table 1). Among these, six SD patients and five AD patients wanted to be withdrawn from the study because of adverse events.

Of the 198 phase I patients, 132 (83 SD patients and 49 AD patients) were enrolled in phase II (Fig. 1). The demographics of the enrolled patients are presented in Table 2. There were no significant differences between the AD and SD groups in demographic and baseline characteristics except for the mean age (Table 2).

Nine patients (10.8%) in the SD group and 11 patients (22.5%) in the AD group were excluded during phase II (Fig. 1). The reasons for the exclusion were voluntary withdrawal (n=3 in the SD group and n=3 in the AD group), follow-up loss, and the lack of efficacy (Fig. 3).

The recurrence-free curves of the two groups are shown in Fig. 4. Fifty-four patients (65.0%) in the SD group and 32 patients (65.3%) in the AD group had no DE during the 20 weeks of phase II (Fig. 4). In the previous study24, all of the 10 patients of the placebo group had DE (IGA ≥2) in 10 weeks. Eight patients of the placebo group had DE at 2 weeks, one patient at 4 weeks, and the other one patient at 6 weeks.

The recurrence-free curves of the two groups of our study were not significantly different from each other at each visit; however, the curve of the placebo group of the previous study was significantly different (p<0.05 by log rank analysis) from both groups of our study (Fig. 4).

Kaplan-Meier analysis was also performed to evaluate the recurrence-free times of each facial region. The recurrence-free curves were similar between the two groups in the forehead, cheeks, chin, and perioral/periauricular regions. On the other hand, the perioral region tended to have recurrences earlier in facial AD patients and the nose in facial SD patients. The cumulative disease-free rate of patients (IGA≤1) at 20 weeks of phase II was 63.6% for facial SD and 63.5% for facial AD, and that of the placebo group was 0% for facial SD (Fig. 5).

The number of DEs was also not significantly different between the two groups (p=0.98). The results of age-stratified analysis (Mantel-Henzel χ² test; ages, 19, 20~39, 41~59, and ≥60 years) were constant. The median time to first DEs was not significantly different between the two groups (mean±standard deviation: 41.17±26.41 days vs. 40.55±24.84 days; p=0.38). In the previous study, the median time to first DE of the placebo group was 18.20±9.44 days24. It was shown that the median time to the first DE of both groups of our study were longer than that of the placebo group of the previous study, although the maintenance period was different (20 weeks vs. 10 weeks). The mean IGA scores for the first DE (2.12±0.33 vs. 2.17±0.38 vs. 2.90±1.10) were not significantly different between the three groups (p=0.51 by one-way ANOVA). Also, the mean VAS scores for the first DE (2.96±1.68 vs. 3.31±1.11) did not differ significantly between the two groups (p=0.84).

During phase II, there were no significant differences between the two groups in the incidences of application-site adverse events. Fifteen patients (18.1%) in the SD group experienced adverse events during phase II, whereas nine patients (18.4%) in the AD group experienced adverse events, such as pruritus (five SD patients, 6.0% vs. three AD patients, 6.1%), irritation (four SD patients, 4.8% vs. two AD patients, 4.1%), folliculitis (three SD patients, 3.6% vs. two AD patients, 4.1%), and erythema (three SD patients, 3.6% vs. two AD patients, 4.1%) (Table 1). None of the patients expressed a desire to quit the clinical trial because of their adverse events during phase II.