Effectiveness and safety of a prevention-of-flare-progression strategy with pimecrolimus cream 1% in the management of paediatric atopic dermatitis

Transcription

1 JEADV ISSN Blackwell Publishing Ltd ORIGINAL ARTICLE Effectiveness and safety of a prevention-of-flare-progression strategy with pimecrolimus cream 1% in the management of paediatric atopic dermatitis B Sigurgeirsson,* V Ho, C Ferrándiz, K Andriano, A Grinienko, P Jimenez, for the Pimecrolimus 1% cream in (paediatric) Eczema: Prevention of Progression multi-centre investigator study group Department of Dermatology, University of Iceland, Reykjavik, Iceland Skin Care Center, Vancouver, BC, Canada Hospital Germans Trias I Pujol, Barcelona, Spain Novartis Pharmaceuticals Corporation, East Hanover, NJ, USA Keywords atopic dermatitis, eczema, Elidel, flare prevention, long-term management, pimecrolimus cream 1% Received: 3 September 2007, accepted 31 March 2008 *Corresponding author, Hudlaeknastodin ehf, Smáratorg 1, 200 Kópavogur, Iceland, tel ; fax ; bsig@hudlaeknastodin.is DOI: /j x Abstract Objective This study was performed to investigate the efficacy and safety of a prevention-of-flare-progression strategy with pimecrolimus cream 1% in children and adolescents with atopic dermatitis (AD). Methods A 26-week multi-centre, randomized, double-blind, vehiclecontrolled study was conducted in 521 patients aged 2 17 years, with a history of mild or moderate AD, who were clear/almost clear of disease before randomization to pimecrolimus cream 1% (n = 256) or vehicle cream (n = 265). Twice-daily treatment with study medication was started at the first signs and/or symptoms of recurring AD. If, despite the application of study medication for at least 3 days, AD worsened (as confirmed by the investigator), treatment with a moderately potent topical corticosteroid (TCS) was allowed in both groups. The primary efficacy end point was the number of days on study without TCS use for a flare. Results The mean number of TCS-free days was significantly higher (P < ) in the pimecrolimus cream 1% group (160.2 days) than in the control group (137.7 days). On average, patients on pimecrolimus cream 1% experienced 50% fewer flares requiring TCSs (0.84) than patients on vehicle cream (1.68) (P < ). Patients on pimecrolimus cream 1% also had fewer unscheduled visits (87) than patients on vehicle cream (246). Conclusions In children and adolescents with a history of mild or moderate AD but free/almost free of signs or symptoms of the disease, early treatment of subsequent AD exacerbations with pimecrolimus cream 1% prevented progression to flares requiring TCS, leading to fewer unscheduled visits and reducing corticosteroid exposure. Introduction Atopic dermatitis (AD) is a common inflammatory skin disorder that usually presents in early infancy and childhood. 1 It is associated with considerable social and financial costs 2,3 and has a negative impact on the quality of life of children and their parents. 3 6 A disturbing characteristic of AD is its chronic or chronically relapsing course, which cannot be modified by traditional therapeutic strategies. 1,7,8 Patients in an acute flare present with intensely pruritic erythematous papules, associated with excoriation and serous exudate. 9 After resolution of a relapse, some skin alterations usually persist, particularly skin dryness and an exaggerated The Authors

2 Sigurgeirsson et al. Prevention-of-flare-progression strategy in eczema response to irritants, as a result of epidermal barrier dysfunction and subclinical inflammation. 1,10,11 The chronic form of the disease is characterized by lichenification and chronic papules. 1,9 Atopic dermatitis relapses are effectively treated with a topical corticosteroid (TCS) of adequate potency until the acute inflammation subsides. 7,12 14 After resolution of a flare, regular applications of emollients are recommended to alleviate dry skin, and low-potency TCSs or twice-weekly applications of a moderately potent TCS have been shown to maintain AD control. 12,13 However, the efficacy of skin care measures recommended during the relapse-free period has not been evaluated in a rigorous way, 12 and the remission period between flares can be as short as 2 weeks. 15 Indeed, there are data from a 1-year trial 16 indicating that the level of exposure to TCSs required to gain control of moderate or severe AD is significantly higher than the current labels for most TCSs would allow. In that double-blind, randomized study, comparing the effectiveness and safety of treatment with pimecrolimus cream 1% to TCSs in adults with moderate or severe AD, 50% of the patients on TCSs used these drugs almost continuously for a year. The major problem with TCS treatment is that prolonged or extensive use exposes the patients to an increased risk of adverse events Local side-effects of TCSs include skin atrophy, reduced epidermal barrier function, rosacea-like dermatitis or acneiform eruption when used on the face 31 and elevated intraocular pressure when used on the eyelids and periorbital area. 31 Adverse systemic effects such as suppression of the hypothalamic pituitary adrenal axis and growth retardation in children are the result of the penetration of topically applied corticosteroids through the skin into the circulation, and are more likely to occur in patients with extensive disease and in infants. 20,23,31 Other limitations of TCS therapy include tachyphylaxis and disease rebound after discontinuation An important potential reason for therapeutic failure with TCSs in AD is non-adherence with the therapeutic regimen because of fear of adverse effects Pimecrolimus cream 1% (Elidel, Novartis Pharma AG, Basel, Switzerland) is a topical calcineurin inhibitor that selectively targets T lymphocytes and blocks the synthesis of inflammatory cytokines involved in the pathogenesis of AD. 39,40 It was specifically developed to provide an effective, non-steroid, anti-inflammatory alternative to TCSs with a more favourable safety profile. 39 Randomized clinical trials in adults and paediatric patients 16,41 48 have demonstrated that twice daily use of pimecrolimus cream 1% induces a rapid and sustained improvement of AD, increases the number of disease-free days and ameliorates the quality of life of patients and their families. Some of these trials have also provided evidence that, at variance with TCSs, pimecrolimus cream 1% retains its clinical effectiveness when used for extended periods 16,41,47,48 and can be discontinued without risk of disease rebound. 47 Pharmacokinetic studies in adult and paediatric patients have demonstrated that treatment with pimecrolimus cream 1% twice daily for up to 1 year results in minimal systemic exposure to pimecrolimus irrespective of patients age, percentage of total body surface area treated or duration of treatment In agreement with these data, evaluation of the safety and tolerability of pimecrolimus cream 1% in patients of different ages, with a wide range of disease severity and treated intermittently for up to 2 years, 16,41 53 has indicated that this treatment is not associated with clinically relevant systemic adverse events. Being a non-steroid drug, pimecrolimus cream 1% does not suppress the hypothalamic pituitary adrenal axis 39 and does not cause skin atrophy 24 or other local adverse events associated with the use of TCSs. 31 The risk of skin infections is similar to that observed with TCSs. 16 The intermittent use of pimecrolimus cream 1% has not been associated with systemic immunosuppression-related malignancies that are known to occur during long-term, sustained immunosuppression with oral tacrolimus, cyclosporin A and oral corticosteroids in transplant patients. 54 Rare cases of malignancies (e.g. skin cancer and lymphoma) have been reported in patients treated with topical calcineurin inhibitors, but a causal relationship has not been established. 55 Actually, the incidence rate of these malignancies in patients treated with pimecrolimus cream 1% in clinical trials has been lower than that observed in patients treated with control treatments, including TCSs. 54 In previous long-term, vehicle-controlled studies in paediatric patients 43,44 and adults 42 with mild to severe AD, pimecrolimus cream 1% was used to treat active lesions and also at the onset of new exacerbations to prevent severe flares (e.g. at least severe erythema and severe infiltration/papulation). Although the data demonstrated that pimecrolimus cream 1% prevents severe disease flare-ups, the effectiveness of a prevention strategy based on the use of pimecrolimus cream 1% at the first signs and/or symptoms of AD recurrence was not evaluated because none of the patients enrolled in the studies mentioned earlier was free of signs and/or symptoms of the disease at treatment start. This study was therefore designed to examine whether intermittent intervention with pimecrolimus cream 1% at the first signs and/or symptoms of a relapse was an effective strategy to maintain AD under control over 26 weeks in children and adolescents who had a history of mild or moderate disease, but no or minimally active skin lesions at the time of study entry The Authors 1291

3 Prevention-of-flare-progression strategy in eczema Sigurgeirsson et al. Patients and methods Patients Eligible patients were enrolled at 44 centres in 10 countries (Belgium, Canada, Finland, France, Iceland, Norway, Poland, Spain, Sweden and Great Britain). The institutional review board at each centre approved the protocol, and written informed consent was obtained from all participants or their legal guardians. Study population Included in the study were patients aged 2 17 years, with a history of mild or moderate AD in the 6 months preceding randomization, as indicated by an Investigator s Global Assessment (IGA; see below) score of 2 or 3, who had required treatment with TCSs, pimecrolimus cream 1% or topical tacrolimus for AD at least twice in the 6 months preceding randomization, and who had received at least one of such treatments in the 3 months preceding randomization. Prior to randomization, patients were to be clear or almost clear of disease as indicated by an IGA score of 1. Patients were excluded for medical history, concomitant illnesses and treatments that could interfere with the evaluation of study treatment, including the use of systemic corticosteroids, immunosuppressants, cytostatics or phototherapy in the month preceding the first visit; topical tacrolimus or pimecrolimus cream 1% within 4 weeks; systemic antibiotics within 2 weeks; and topical therapy for AD other than topical calcineurin inhibitors (e.g. TCSs, tar) within 7 days. Also excluded were patients having a history of malignancy or of inadequate response to tacrolimus ointment or to pimecrolimus cream 1%, and patients with immunocompromised status, concurrent skin disease that could interfere with evaluation of study treatment, active infections and infestations or known hypersensitivity to any ingredient of the study medications. Study design This was a 26-week multi-centre, randomized, doubleblind, vehicle-controlled, parallel-group, comparative study. The study consisted of a 1 2 weeks of screening period, followed by a 26-week treatment period. There were six scheduled visits and two telephone monitoring appointments: patients attended the clinic at the screening visit; at the randomization visit; and at weeks 4, 8, 16 and 26; telephone monitoring was conducted at weeks 12 and 20. The last visit at week 26 could be postponed for a maximum of 7 days. Unscheduled visits during the treatment period were required to confirm a disease flare (see below). Patients were randomly assigned to receive either pimecrolimus cream 1% treatment regimen or the vehicle cream treatment regimen in a 1 : 1 ratio. Treatment assignments were balanced both within and among centres. Randomization was performed using a validated system that automates the random assignment of treatment groups to randomization numbers. Blinding was maintained by using treatments that were identical in appearance. Individual sets of code-breaker scratch cards, containing the actual treatment assignment, were prepared for each container of medication. The scratch cards were distributed to each investigator and were not to be opened unless an emergency occurred. Skin care with emollients was recommended for all patients. Pimecrolimus cream 1% or matching vehicle cream was to be started at the onset of the first signs (e.g. transient erythema, bumps, skin thickening) and/or symptoms (e.g. warmth, itching and tingling) of recurring AD. Patients, or their parents and caregivers, were to apply study medication twice daily to all affected areas until complete clearance of signs and symptoms. In the event of a worsening of AD, confirmed by the investigators as being a flare requiring TCSs, and occurring despite at least 3 days of use of study medication, a moderately potent TCS (or a TCS of lower potency for sensitive areas of the body) was allowed, and the use of study medication could be interrupted. The application of study medication could continue for up to 26 weeks (study duration) if required to treat active lesions or newly occurring exacerbations. After 21 days of continuous TCS use, the patient had to be re-evaluated by the investigator. Other concomitant AD-related medications allowed in the study included anti-histamines, oral and topical antibiotics or antifungals or antivirals to treat cutaneous and systemic infections. Routine vaccinations at sites without active AD lesions were also allowed. Compliance was assessed by examination of the completed diary cards, and through patient/caregiver interviews. The data were entered in the database, and the diaries were kept as source documents. Primary and secondary outcome measures Disease severity was assessed using the IGA, a 6-point ordinal scale ranging from 0 = clear to 5 = very severe disease. 39 Pruritus was scored by means of a 4-point ordinal scale ranging from 0 = absent to 3 = severe. 39 The primary efficacy variable was the number of days on study without TCS use for a flare. Each flare requiring TCSs was defined as an increase in IGA and pruritus scores (AD worsening) that warranted TCS treatment in the investigator s opinion and was confirmed by the investigator during a scheduled or an unscheduled visit The Authors

4 Sigurgeirsson et al. Prevention-of-flare-progression strategy in eczema (the flare requiring TCSs is referred to as flare from here onward). The main secondary efficacy variable was ranked flares over 26 weeks. Patients were ranked according to the number of flares they experienced during the trial (with adjustment for early discontinuations) as previously described. 44 Other secondary efficacy variables included the number of unscheduled visits necessary to confirm and treat flares, the time to first flare, the time between resolution of the first flare and the subsequent flare and the duration of TCS treatment necessary to resolve a flare. Safety assessments consisted of monitoring and recording all adverse events, conducting physical examination and assessing vital signs. Laboratory data were not collected in this study. Statistical analysis All efficacy analyses were performed using the intent-totreat population. This included all patients who returned at least once after randomization. Because the number of days on study without TCS use was obtained from the patient s daily diary, the efficacy analysis for the primary end point was performed on all intent-to-treat patients who entered data at least once in their daily diaries (fig. 1). The primary efficacy variable and the main secondary fig. 1 Diagram showing the flow of participants through each stage of the study The Authors 1293

5 Prevention-of-flare-progression strategy in eczema Sigurgeirsson et al. efficacy variable, ranked flares, were analysed using the Wilcoxon rank sum statistic stratified by centre (van Elteren test). Formal statistical inference on the main secondary efficacy variable was only performed if a statistically significant difference was found between groups for the primary efficacy variable at the 0.05 level of significance. For the analysis of ranked flares, patients who participated for more than 174 days after randomization were ranked according to the number of flares experienced during the study (f), with zero flares constituting the best case. Patients who withdrew from the study and participated for 174 days or less were ranked according to the number of flares experienced before withdrawal f plus the expected number of flares they would have experienced had they completed the 26 weeks (182 days) of participation. Early discontinuation because of unsatisfactory therapeutic effect was also counted provided the last use of TCSs occurred at least 4 days prior to the end-of-study visit. The expected number of flares was calculated as θ (182 t), where the rate of flares per day (θ) was multiplied by the remaining duration (182 days t), and where t was the number of days on study before withdrawal. The value of θ was estimated under the null hypothesis of no treatment difference using the actual total number of flares occurring in the study for all observed double-blind days and all randomized patients. The estimate Θ for θ was calculated using the following formula: Θ = (total number of flares)/(total number of double-blind days on study). All patients were assigned a value for the actual + imputed number of flares according to the formula: f + [Θ (182 t)]. For patients who participated for more than 174 days after randomization, the imputed number of flares [Θ (182 t)] was zero. Patients were then ranked on the basis of these values, ignoring treatment assignment. The best case was assigned a rank of 1 and mid-ranks were used in the event of ties. Descriptive statistics of the ranked flares and the actual + imputed number of flares are presented for each treatment group. A relative difference in the mean ranked flares greater than 20% between the pimecrolimus cream 1% and the vehicle cream groups was considered clinically significant. Cumulative life-table curves investigating the time from randomization until first use of study medication and the time to first flare were constructed by the Kaplan Meier method. 56 A recurrent event analysis was performed in patients who had multiple flare episodes. The time between resolution of last flare and onset of the next flare (gap time) was analysed using the conditional proportional hazards model for gap times. 57 The areas under the curve of individual IGA score obtained at scheduled and unscheduled visits in the two treatment groups over 6 months were compared by a weighted analysis of covariance. The model used was the following: standardized absolute area under the curve = treatment + pooled centre + baseline IGA score + weight error. Error weights were set to the number of days between randomization and the day of the last IGA. The assessment of safety was mainly based on the frequency of adverse events. A Kaplan Meier analysis was used to assess time from start of study treatment to the first occurrence of any adverse events or to the worsening of any adverse events already present at the treatment start. Six-month data from the previous 1-year study 44 suggested that the average TCS-free days would be 173 days for the pimecrolimus cream 1% group and 163 days for the vehicle cream group, with a common standard deviation (SD) of 26 days. Because the data were expected to be somewhat skewed, because of the negligible use of TCSs in many patients, the sample size was estimated using the Wilcoxon two-sample test (Mann Whitney) for continuous data. On the basis of the data from the previous study, the effect size was calculated to be and the associated probability of the alternative hypothesis for a two-sided test was estimated to be Using this estimated probability, a two-sided test, type I and type II error rates of 5% each, it was found that approximately 380 patients (190 patients in each treatment arm) had to be randomized. It was felt that 500 patients (250 patients per treatment arm) would be adequate from a safety perspective. This planned sample size also gave adequate power for the main secondary efficacy parameter, ranked flares. Results Patients A flow diagram of patient accounting and treatment outcome is reported in fig. 1. The demographic characteristics were well matched in the two treatment groups (Table 1). The vast majority of the patients were Caucasian. Both treatment groups included more females than males. These random differences were not considered to have an impact on the validity of treatment comparison, because response to treatment with pimecrolimus cream 1% is not known to be associated with race or gender. 58 Baseline disease characteristics were also well matched between treatment groups, and all randomized patients were clear or almost clear of AD (Table 1). Medical histories and conditions at the start of the study were representative of a population of patients with AD, and were similar in the two treatment groups The Authors

6 Sigurgeirsson et al. Prevention-of-flare-progression strategy in eczema Table 1 Baseline demographic and disease characteristics Pimecrolimus cream 1% (n = 256)* Patient follow-up and exposure to study medication Vehicle cream (n = 265)* Age (years) Mean ± SD 6.6 ± ± 3.8 Range Gender [n (%)] Male 109 (42.6) 104 (39.2) Female 147 (57.4) 161 (60.8) Race [n (%)] Caucasian 244 (95.3) 242 (91.3) Black/Oriental/Other 12 (4.7) 23 (8.7) IGA score at baseline [n (%)] 0 (Clear) 68 (26.6) 75 (28.3) 1 (Almost clear) 188 (73.4) 190 (71.7) Total body surface area involved (%) Mean ± SD 3.9 ± ± 8.8 IGA, Investigator s Global Assessment; SD, standard deviation. *Randomized population. The mean duration of follow-up (± SD) was (± 36.3) days in the pimecrolimus cream 1% group, and (± 49.3) days in the vehicle cream group. Durations of study participation were well matched in the two treatment groups and allowed for valid comparisons. Of the 521 patients (256 in the pimecrolimus cream 1% group and 265 in the vehicle cream group) randomized into double-blind treatment, 80 patients (15.4%) discontinued prematurely. Discontinuations occurred more frequently in the vehicle cream group (21.1%) than in the pimecrolimus cream 1% group (9.4%) (fig. 1). Most discontinuations occurred after first use of study medication (fig. 1). Unsatisfactory therapeutic effect accounted for 23 (8.7%) dropouts in the vehicle cream group and for only 8 (3.1%) dropouts in the pimecrolimus cream 1% group (fig. 1). Discontinuations because of adverse events were also less frequent in the pimecrolimus cream 1% arm (0.4% of patients) than in the vehicle cream arm (2.3% of patients) (fig. 1). Application site reactions were the most common adverse events leading to discontinuation in both treatment groups. More than 80% of patients (85% in the pimecrolimus cream 1% group and 82% in the control group) used study medication within the first 10 days after randomization (fig. 2). The mean number of treatment days was higher in the pimecrolimus cream 1% group (70.2 days) than in the control group (51.0 days). Emollient use was similar in the two treatment groups throughout the study. When TCSs were used to treat flares, the drugs employed by at least 1% of patients in any treatment group are listed in Table 2. The use of other allowed concomitant medications before and after the first use of study medication was comparable in the two treatment groups. Efficacy The primary efficacy analysis showed that the mean number of days on study without TCS use was significantly fig. 2 Kaplan Meier estimates for the percentage of patients who used study medication at least once (randomized population) The Authors 1295

7 Prevention-of-flare-progression strategy in eczema Sigurgeirsson et al. Table 2 Topical corticosteroids (TCSs) employed by 1% of patients in any treatment group after first use of study medication TCS Pimecrolimus cream 1% (n = 246)* [n (%)] [n (%)] Vehicle cream (n = 260)* Fluticasone propionate 33 (13.4) 61 (23.5) Hydrocortisone butyrate 25 (10.2) 65 (25.0) Difluprednate 15 (6.1) 14 (5.4) Hydrocortisone 12 (4.9) 25 (9.6) Mometasone furoate 8 (3.3) 4 (1.5) Prednicarbate 7 (2.8) 18 (6.9) *Patients who used study medication at least once. Table 3 Summary of the results for the primary and secondary efficacy outcomes End points Primary and main secondary efficacy outcomes Pimecrolimus cream 1% Vehicle cream P value Number of TCS-free days n* Mean ± SD ± ± 51.9 < Ranked actual + imputed number of flares n Mean rank ± SD ± ± < Actual + imputed number of flares n Mean ± SD 0.84 ± ± 1.89 < SD, standard deviation; TCS, topical corticosteroid. *Daily diary intent-to-treat population. Intent-to-treat population. higher in the group of patients on pimecrolimus cream 1% (160.2 days) than in the group of patients on vehicle cream (137.7 days) (Table 3). In comparison with vehicle cream, pimecrolimus cream 1% reduced the median number of days of TCS treatment for flares by 91%, being the median values 11.5 days in the vehicle cream group and 1.0 day in the pimecrolimus cream 1% group. The corresponding means were 21.9 days in the vehicle cream group and 12.9 days in the pimecrolimus cream 1% group. Pimecrolimus cream 1% was significantly superior to the vehicle cream in preventing the progression of the first signs and/or symptoms of AD to flares requiring TCSs, as primarily demonstrated by ranking patients with regard to the number of flares they experienced while participating in the 26-week study (Table 3). The relative difference in the mean ranks was 26% and exceeded the pre-defined limit ( 20%) beyond which the difference was to be considered as clinically significant. In terms of actual and imputed number of flares, patients treated with pimecrolimus cream 1% had 50% fewer flares than patients who received the vehicle cream (mean number of flares = 0.84 vs. 1.68) (Table 3). Moreover, the percentage of patients who completed the study without a flare requiring TCS use was substantially higher in the pimecrolimus cream 1% arm (63.1%) than in the control arm (40.7%). At the start of TCS treatment for a flare, patients on pimecrolimus cream 1% tended to have less severe disease than patients on vehicle cream. At the time of the first flare requiring TCS use (unscheduled visit), the percentage of patients who had moderate or severe/very severe AD exacerbations (IGA score 3) was 40.7% (33/ 81) in the pimecrolimus cream 1% group, and 58.1% (86/ 148) in the vehicle cream group. This difference between treatment groups was statistically significant (P = ). Moreover, an exploratory analysis of the area under the curve of individual IGA score obtained at scheduled and unscheduled visits over 6 months demonstrated that patients on pimecrolimus cream 1% had significantly milder disease than patients on vehicle cream (P = 0.005). The least square means (95% confidence intervals) were 1.01 ( ) for the pimecrolimus cream 1% group, and 1.19 ( ) for the vehicle cream group. The number of scheduled visits was similar in the two treatment groups, but patients in the pimecrolimus cream 1% arm had three times fewer unscheduled visits (87) to confirm and treat flares than patients in the vehicle cream arm (246). Further, the percentage of patients without any unscheduled visits was higher in the pimecrolimus cream 1% group (82%) than in the vehicle cream group (53%). During the study, patients on pimecrolimus cream 1% remained flare-free for a longer period of time than those on vehicle cream (P < ). The median (mean) time to first flare was > 190 (138.1) days in the pimecrolimus cream 1% group, and 59 (94.6) days in the vehicle cream group (fig. 3). Only 20.3% of patients in the pimecrolimus cream 1% group vs. 41.4% of patients in the vehicle cream group experienced more than one flare during the study. The time between resolution of the first flare and the start of second flare was significantly longer (P = ) for patients treated with pimecrolimus cream 1% than for patients on vehicle cream. The hazard ratios (95% confidence interval) for time to first flare, and time from resolution of first flare to the second flare, were ( ) and ( ), respectively The Authors

8 Sigurgeirsson et al. Prevention-of-flare-progression strategy in eczema fig. 3 Kaplan Meier estimates for the percentage of patients without a flare (intent-to-treat population). The median time to first flare was > 190 days in the pimecrolimus cream 1% group, and 59 days in the vehicle cream group. Safety Table 4 shows the incidence rate of adverse events reported in 3% of patients in any treatment group after first use of study medication (number and percentage of patients affected). The most frequent adverse events were common childhood infections and ailments such as nasopharyngitis, pyrexia, otitis media and cough. In most cases, nasopharyngitis and pyrexia were of mild or moderate severity, and were not considered to be related to study medication by the investigator. Application site reactions, including pruritus, burning, pain and/or irritation, were each reported in fewer than four patients in each treatment group, with the exception of burning (eight cases in the vehicle cream group vs. three in the pimecrolimus cream 1% group) (Table 4). Most of the reported application site reactions were mild or moderate in severity, and all were considered related to study medication by the investigator. Overall, the incidence rate of treatment-related adverse events was 9.3% in the pimecrolimus cream 1% arm, and 10.0% in the vehicle cream arm. The most frequently reported skin infections were impetigo, varicella and molluscum contagiosum, which occurred with similar incidence in the two treatment groups (Table 4). A possible case of eczema herpeticum was reported in one patient in the pimecrolimus cream 1% arm. The viral culture was negative for herpes simplex virus, and the event was not suspected to be treatment related. Kaplan Meier estimates revealed that the time from the first use of study medication to the occurrence or Table 4 Most frequently reported adverse events after first use of study medication Adverse events Pimecrolimus cream 1% (n = 246)* [n (%)] [n (%)] Vehicle cream (n = 260)* Nasopharyngitis 57 (23.2) 39 (15.0) Pyrexia 19 (7.7) 11 (4.2) Influenza 17 (6.9) 16 (6.2) Otitis media 15 (6.1) 17 (6.5) Cough 14 (5.7) 12 (4.6) Asthma 13 (5.3) 9 (3.5) Bronchitis 12 (4.9) 15 (5.8) Rhinitis 11 (4.5) 14 (5.4) Vomiting 11 (4.5) 6 (2.3) Gastroenteritis 10 (4.1) 10 (3.8) Upper respiratory 10 (4.1) 11 (4.2) tract infection Viral gastroenteritis 9 (3.7) 7 (2.7) Impetigo 9 (3.7) 6 (2.3) Streptococcal pharyngitis 9 (3.7) 2 (0.8) Varicella 9 (3.7) 8 (3.1) Pharyngitis 8 (3.3) 8 (3.1) Urticaria 8 (3.3) 4 (1.5) Headache 7 (2.8) 6 (2.3) Pharyngolaryngeal pain 7 (2.8) 7 (2.7) Tonsillitis 7 (2.8) 8 (3.1) Molluscum contagiosum 6 (2.4) 8 (3.1) Influenza-like illness 5 (2.0) 10 (3.8) Application site burning 3 (1.2) 8 (3.1) *Patients who used study medication at least once The Authors 1297

9 Prevention-of-flare-progression strategy in eczema Sigurgeirsson et al. worsening of any adverse event was similar in the two treatment groups. The mean number of days from the treatment start to the occurrence of any adverse event was 83 days in the pimecrolimus cream 1% group, and 88 days in the vehicle cream group. Moreover, there was no statistically significant difference between treatment groups in the time to first occurrence or worsening of upper respiratory tract infections or nasopharyngitis. Of seven discontinuations because of adverse events, only one occurred in the pimecrolimus cream 1% group (application site pruritus). Discussion This is the first study where the effectiveness of a prevention-of-flare-progression strategy with pimecrolimus cream 1% in the management of AD has been examined in children and adolescents without clinically active disease at baseline. In order to mimic the management of AD in clinical practice, the diagnosis of AD and the assessment of disease severity during the course of the study were essentially based on the clinical evaluation of physicians. The control arm represented the current standard of care for AD, which involves the regular use of emollients and the reactive use of moderately potent TCSs for disease flare-ups. 7,12 14 The pimecrolimus cream 1% treatment arm reflected the envisioned paradigm for the use of pimecrolimus cream 1% as an early intervention in order to prevent progression of initial exacerbations of AD to flares requiring TCS use. This prevention strategy significantly prolonged the flare/tcs-free periods in comparison with control treatment. In particular, treatment with pimecrolimus cream 1% reduced the median number of days of TCS use for flares by 91% in comparison with control treatment. Moreover, 63% of patients in the pimecrolimus cream 1% arm completed the 6-month study without a flare requiring treatment with TCSs in comparison with 41% of patients in the vehicle cream arm. Thus, early intervention with pimecrolimus cream 1% substantially reduced the incidence rate of flares and TCS exposure. The differences between treatment groups, in terms of mean number of actual + imputed number of flares, indicate that the use of pimecrolimus cream 1% at the first signs and/or symptoms of subsequent exacerbations reduced the occurrence of flares requiring TCS by 50% in comparison with the vehicle cream. Considering the ranked number of flares, the ranking method was chosen to address study outcome in a clinically meaningful way and to take into account differences in the discontinuation rate between treatment arms. 59 According to this method, a lower rank indicates a better clinical outcome. The relative difference in mean ranks between the pimecrolimus cream 1% group and the vehicle cream group was 26% and was considered clinically meaningful. From a probabilistic point of view, 60,61 the chance that a patient on pimecrolimus cream 1% will have a better clinical outcome after 6 months of treatment than a patient on conventional regimen (the control group in this study) is about 64%. The median time to first flare requiring TCS use was threefold longer in patients treated with pimecrolimus cream 1% than in patients who received control therapy. On the basis of the hazard ratio of obtained from the Cox regression for time to first flare, it is possible to assume that in a similar study of 6-month duration and equal enrolment rates in both treatment groups, 37 patients in the pimecrolimus cream 1% group would experience their first flare by the time 100 patients in the control group have experienced their first flare. In addition, the analysis of the area under the curve of individual IGA score obtained at scheduled and unscheduled visits over 6 months demonstrated that patients on pimecrolimus cream 1% had significantly milder disease than patients in the control group. Therefore, patients on pimecrolimus cream 1% could keep their disease under good control more effectively and longer than control patients. The beneficial effect of early intervention with pimecrolimus cream 1% is also reflected in the low number of discontinuations because of unsatisfactory therapeutic effect (eight dropouts), which was the most frequent reason for discontinuations in the group of patients receiving control treatment (23 out of 56 dropouts). An important finding of this study is that patients on pimecrolimus cream 1% had a threefold lower number of unscheduled visits necessary to confirm and treat flares than patients on vehicle cream. This implies that control patients used substantially more healthcare resources than patients receiving early intervention with pimecrolimus cream 1%. It should be noted that the regular application of emollients was recommended by protocol throughout the 26-week period of this study, and that there was no difference between treatment groups in terms of emollient use. In both treatment groups, the first signs and/or symptoms of an initial exacerbation appeared within 10 days from randomization in the vast majority of patients, and the mean time from randomization to the first use of study medication was similar in the two treatment groups. Thus, the regular use of emollients alone did not appear to be particularly effective at preventing disease relapses. The mean number of days of treatment with study medication was higher in the pimecrolimus cream 1% group than in the control group. This difference in treatment exposure may be related to the higher discontinuation rate observed in the control group. Another possible The Authors

10 Sigurgeirsson et al. Prevention-of-flare-progression strategy in eczema explanation is that patients in the control group interrupted more frequently the use of study drug than patients in the pimecrolimus cream 1% group because of the more frequent development of flares requiring treatment with TCSs. The use of pimecrolimus cream 1% at the onset of early signs and/or symptoms of AD exacerbations to prevent progression to flares requiring TCSs needs a favourable safety and tolerability profile of the drug. Treatment with topically applied pimecrolimus results in minimal systemic exposure, because the percutaneous absorption of this molecule is negligible Indeed, no clinically relevant systemic adverse events have been reported in the clinical trials mentioned earlier, and there is no indication that treatment with pimecrolimus cream 1% is associated with systemic immunosuppression. 54,62 Being a non-steroid drug, pimecrolimus cream 1% does not possess any atrophogenic potential and does not interfere with the hypothalamic pituitary adrenal axis. 24,39,40,62 These characteristics make pimecrolimus cream 1% more suitable than TCSs for the treatment of delicate areas of the body, such as the face, axilla and groin, where the thin skin is particularly prone to atrophy and may facilitate the systemic absorption of a topically applied drug. 31 Moreover, pimecrolimus cream 1% can be used on the eyelids and in the periorbital area without risking the occurrence of TCSs-related side-effects such as cataracts and glaucoma. 31 In the present study, treatment with pimecrolimus cream 1% was well tolerated as demonstrated by the safety profile based on adverse event reports and the low incidence of treatment discontinuations because of adverse events. As expected in a population of children and adolescents, the most frequently reported adverse events were infections and childhood ailments. There was no significant association between the incidence of infections and the administration of pimecrolimus cream 1%, although the incidence of nasopharyngitis (common cold) tended to be slightly higher in patients on pimecrolimus cream 1% than in those on vehicle cream. Overall, the safety and tolerability results were consistent with those reported in previous long-term paediatric studies. 43,44 Conclusion The results of this trial indicate that in children and adolescents with a history of mild or moderate AD but clear/almost clear of active skin lesions at study inclusion, intervention with pimecrolimus cream 1% at the first signs and/or symptoms of new exacerbations prevents progression to flares requiring treatment with TCSs and prolongs the remission periods between major relapses. Similar findings have been obtained in a trial with identical study design involving adult patients, and the results will be published elsewhere. 63 The advantages of this prevention-of-flare-progression strategy in comparison with the conventional strategy of treating only full AD flare-ups are the following: first, the control of AD in the long term is improved because the periods without flares are substantially increased; second, the reduction in the risk of flare requiring TCSs reduces the number of AD-related office visits; and third, the exposure to TCSs is limited to breakthrough flares of AD, with a consequent reduction in the risk of side-effects from prolonged TCS use. Conflicts of interest Dr. Sigurgeirsson has been a consultant, speaker or researcher for Novartis, Galderma, Stiefel, Prostrakan, Jansen, Vertex and Schering-Plough. Dr. Ho has been a speaker, consultant or researcher for Novartis and Astellas. Dr. Ferrándiz has been a consultant or speaker for Novartis, Serono, Shire, Centocor and Wyeth. Drs. Andriano, Grinienko and Jimenez are employees of Novartis Pharmaceuticals Corporation and hold stock options in Novartis. Dr. Grinienko also acted as a clinical trial leader for this study. Acknowledgements We thank the following investigators, who also contributed to the study through data acquisition: M-A Morren and J-M Naeyaert (Belgium); I Landells, R Langley, D Marcoux and E Pope (Canada); A Barbaud, J-M Debarre, G Guillet, M Ruer-Mulard and J-F Stalder (France); M Anderson, D Burwood, A Darrah, JB Hamling, K Holgate, B Hopwood, DJ Howarth and T Maxwell (Great Britain); S Kristjansson, J Olafsson and J Steinsson (Iceland); P Hagg, O-M Kekki, O Mickelsson and E Valovirta (Finland); K Lund-Hanssen and T Sandnes (Norway); M Kaczmarski, M Kowalski and M Kulus (Poland); M Casado and A Vicente (Spain); and S Andrae, J Christiansen, M Gajecki, G Hattevig, C-A Hederos, M Ingemansson, A Pettersson and H Velander (Sweden). This study was supported by Novartis Pharmaceuticals. References 1 Leung DYM, Bieber T. Atopic dermatitis. Lancet 2003; 361: Kemp AS. Cost of illness of atopic dermatitis in children: a societal perspective. Pharmacoeconomics 2003; 21: Su JC, Kemp AS, Varigos GA, Nolan TM. Atopic eczema: its impact on the family and financial cost. Arch Dis Child 1997; 76: The Authors 1299

11 Prevention-of-flare-progression strategy in eczema Sigurgeirsson et al. 4 Elliot BE, Luker K. The experience of mothers caring for a child with severe atopic eczema. J Clin Nurs 1997; 6: Pauli-Pott U, Darui A, Beckmann D. Infants with atopic eczema: maternal hopelessness, child-rearing attitudes and perceived infant temperament. Psychother Psychosom 1999; 68: Finlay AY. Quality of life in atopic dermatitis. J Am Acad Dermatol 2001; 45: S64 S66. 7 Przybilla B, Eberlein-Konig B, Rueff F. Practical management of atopic eczema. Lancet 1994; 343: Williams HC. Do topical steroids reduce relapses in adults with atopic dermatitis? Arch Dermatol 1999; 135: Leung DYM, Boguniewicz M, Howell MD, Nomura I, Hamid QA. New insights into atopic dermatitis. J Clin Invest 2004; 113: Leung DYM. Atopic dermatitis: new insights and opportunities for therapeutic intervention. J Allergy Clin Immunol 2000; 105: Boguniewicz M, Eichenfield LF, Hultsch T. Current management of atopic dermatitis and interruption of the atopic march. J Allergy Clin Immunol 2003; 112: S140 S Hoare C, Li Wan Po A, Williams H. Systematic review of treatments of atopic eczema. Health Technol Assess 2000; 4: Hanifin JM, Cooper KD, Ho VC et al. Guidelines of care for atopic dermatitis. J Am Acad Dermatol 2004; 50: Raimer SS. Managing pediatric atopic dermatitis. Clin Pediatr 2000; 39: Wolkerstorfer A, Strobos MA, Glazenburg EJ et al. Fluticasone proprionate 0.05% cream once daily versus clobetasone butyrate 0.05% cream twice daily in children with atopic dermatitis. J Am Acad Dermatol 1998; 39: Luger TA, Lahfa M, Fölster-Holst R et al. Long-term safety and tolerability of pimecrolimus cream 1% and topical corticosteroids in adults with moderate to severe atopic dermatitis. J Dermatolog Treat 2004; 15: Hill CJ, Rosenberg A Jr. Adverse effects from topical corticosteroids. Cutis 1978; 21: Stoppolino G, Prisco F, Santinelli R, Sicuranza G, Giordano C. Potential hazards of topical steroid therapy. Am J Dis Child 1983; 137: Lewis-Jones S. Atopic dermatitis in childhood. Hosp Med 2001; 62: Queille C, Pommarede R, Saurat JH. Efficacy versus systemic side effects of six topical steroids in the treatment of atopic dermatitis of childhood. Pediatr Dermatol 1984; 1: Turpeinen M, Salo OP, Leisti S. Effect of percutaneous absorption of hydrocortisone on adrenocortical responsiveness in infants with severe skin disease. Br J Dermatol 1986; 115: Turpeinem M. Influence of age and severity of dermatitis on the percutaneous absorption of hydrocortisone in children. Br J Dermatol 1988; 118: Turpeinen M, Lehtokoski-Lehtiniemi E, Leisti S, Salo OP. Percutaneous absorption of hydrocortisone during and after the acute phase of dermatitis in children. Pediatr Dermatol 1988; 5: Queille-Roussel C, Paul C, Duteil C et al. The new topical ascomycin derivative SDZ ASM 981 does not induce skin atrophy when applied to normal skin for 4 weeks: a randomized, double blind controlled study. Br J Dermatol 2001; 144: Haapasaari KM, Risteli J, Koivukangas V, Oikarinen A. Comparison of the effect of hydrocortisone, hydrocortisone- 17-butyrate and betamethasone on collagen synthesis in human skin in vivo. Acta Derm Venereol 1995; 75: Haapasaari KM, Risteli J, Karvonen J, Oikarinen A. Effect of hydrocortisone, methylprednisolone acetonate and mometasone furoate on collagen synthesis in human skin in vivo. Skin Pharmacol 1997; 10: Oikarinen A, Haapasaari KM, Sutinen M, Tasanen K. The molecular basis of glucocorticoid-induced skin atrophy: topical glucocorticoid apparently decreases both collagen synthesis and the corresponding collagen mrna level in human skin in vivo. Br J Dermatol 1998; 139: Sheu HM, Lee JYY, Chai CY, Kuo K. Depletion of stratum corneum intercellular lipid lamellae and barrier function abnormalities after long-term topical corticosteroids. Br J Dermatol 1997; 136: Kolbe L, Kligman AM, Schreiner V, Stoudemayer T. Corticosteroid-induced atrophy and barrier impairment measured by non-invasive methods in human skin. Skin Res Technol 2001; 7: Kao JS, Fluhr JW, Man MQ et al. Short-term glucocorticoid treatment compromises both permeability barrier homeostasis and stratum corneum integrity: inhibition of epidermal lipid synthesis accounts for functional abnormalities. J Invest Dermatol 2003; 120: Hengge UR, Ruzicka T, Schwartz RA, Cork MJ. Adverse effects of topical glucocorticosteroids. J Am Acad Dermatol 2006; 54: du Vivier A. Tachyphylaxis to topically applied steroids. Arch Dermatol 1976; 112: Kligman AM, Frosch PJ. Steroid addiction. Int J Dermatol 1979; 18: Zheng PS, Lavker RM, Lehmann P et al. Morphologic investigations on the rebound phenomenon after corticosteroid-induced atrophy in human skin. J Invest Dermatol 1984; 82: Singh G, Singh PK. Tachyphylaxis to topical steroids measured by histamine-induced wheal suppression. Int J Dermatol 1986; 25: Fischer G. Compliance problems in pediatric atopic eczema. Aust J Dermatol 1996; 37: The Authors

12 Sigurgeirsson et al. Prevention-of-flare-progression strategy in eczema 37 Charman CR, Morris AD, Williams HC. Topical corticosteroid phobia in patients with atopic eczema. Br J Dermatol 2000; 142: Fukaya M. Why do patients with atopic dermatitis refuse to apply topical corticosteroids? Dermatology 2000; 201: Graham-Brown RAC, Grassberger M. Pimecrolimus: a review of preclinical and clinical data. Int J Clin Pract 2003; 57: Grassberger M, Steinhoff M, Schneider D, Luger TA. Pimecrolimus an anti-inflammatory drug targeting the skin. Exp Dermatol 2004; 13: Eichenfield LF, Lucky AW, Boguniewicz M et al. Safety and efficacy of pimecrolimus (ASM 981) cream 1% in the treatment of mild and moderate atopic dermatitis in children and adolescents. J Am Acad Dermatol 2002; 46: Meurer M, Folster-Holst R, Wozel G, Weidinger G, Jungerf M, Brautigam M. Pimecrolimus cream in the long-term management of atopic dermatitis in adults: a six-month study. Dermatology 2002; 205: Kapp A, Papp K, Bingham A et al. Flare Reduction in Eczema with Elidel (infants) multicentre investigator study group: long-term management of atopic dermatitis in infants with topical pimecrolimus, a nonsteroid anti-inflammatory drug. J Allergy Clin Immunol 2002; 110: Wahn U, Bos JD, Goodfield M et al. Efficacy and safety of pimecrolimus cream in the long-term management of atopic dermatitis in children. Pediatrics 2002; 110: e2. 45 McKenna S, Whalley D, de Prost Y et al. Treatment of paediatric atopic dermatitis with pimecrolimus (Elidel, SDZ ASM 981): impact on quality of life and health-related quality of life. J Eur Acad Dermatol Venereol 2006; 20: Ho VC, Gupta A, Kaufmann R et al. Safety and efficacy of nonsteroid pimecrolimus cream 1% in the treatment of atopic dermatitis in infants. J Pediatr 2003; 142: Kaufmann R, Folster-Holst R, Hoger P et al. Onset of action of pimecrolimus cream 1% in the treatment of atopic eczema in infants. J Allergy Clin Immunol 2004; 114: Papp KA, Werfel T, Fölster-Holst R et al. Long-term control of atopic dermatitis with pimecrolimus cream 1% in infants and young children: a two-year study. J Am Acad Dermatol 2005; 52: Harper J, Green A, Scott G et al. First experience of topical SDZ ASM 981 in children with atopic dermatitis. Br J Dermatol 2001; 144: Harper J, Lakhanpaul M, Wahn U et al. Pimecrolimus (Elidel, SDZ ASM 981) cream 1% blood levels are consistently low in children with extensive atopic dermatitis. J Eur Acad Dermatol Venereol 2001; 15 (Suppl. 2): abstract Lakhanpaul M, Davies T, Allen BR, Schneider D. Low systemic exposure in infants with atopic dermatitis in a 1-year pharmacokinetic study with pimecrolimus cream 1%. Exp Dermatol 2006; 15: Allen BR, Lakhanpaul M, Morris A et al. Systemic exposure, tolerability, and efficacy of pimecrolimus cream 1% in atopic dermatitis patients. Arch Dis Child 2003; 88: Papp KA, Breuer K, Meurer M et al. Long-term treatment of atopic dermatitis with pimecrolimus cream 1% in infants does not interfere with the development of protective antibodies after vaccination. J Am Acad Dermatol 2005; 52: Hultsch T, Kapp A, Spergel J. Immunomodulation and safety of topical calcineurin inhibitors for the treatment of atopic dermatitis. Dermatology 2005; 211: Food and Drug Administration: FDA News. FDA approves updated labeling with box warning and medication guide for two eczema drugs, Elidel and Protopic. [WWW document] URL news/ 2006/NEW01299.html (last accessed: 30 April 2008). 56 Kaplan EL, Meier P. Non-parametric estimation from incomplete samples. J Am Stat Assoc 1958; 36: Prentice RL, Williams BJ, Petersen AV. On the regression of multivariate failure time data. Biometrika 1981; 68: Eichenfield LF, Lucky AW, Langley RGB et al. Use of pimecrolimus cream 1% (Elidel ) in the treatment of atopic dermatitis in infants and children: the effects of ethnic origin and baseline disease severity on treatment outcome. Int J Dermatol 2005; 44: Gould AL. A new approach to the analysis of clinical drug trials with withdrawals. Biometrics 1980; 36: Conover WJ, Iman RL. Rank transformations as a bridge between parametric and nonparametric statistics (with discussion). Am Stat 1981; 35: Noether GE. Sample size determination for some common nonparametric tests. J Am Stat Assoc 1987; 82: Paul C, Cork MJ, Rossi AB, Papp KA, Barbier N, de Prost Y. Safety and tolerability of 1% pimecrolimus cream among infants: experience with 1133 patients treated for up to 2 years. Pediatrics 2006; 117: e118 e Gollnick H, Kaufmann R, Stough D et al. Pimecrolimus cream 1% in the long-term management of adult atopic dermatitis: prevention of flare progression. A randomized controlled trial. Br J Dermatol 2008; 158: The Authors 1301