Targeted anti-staphylococcal therapy with endolysins in atopic dermatitis and the effect on steroid use, disease severity and the microbiome: study protocol for a randomized controlled trial (MAAS trial)

The MAAS trial (Microbiome in atopic dermatitis during anti-staphylococcal therapy and the effect on steroid use), is a multi-centre, randomized, double-blinded, placebo-controlled superiority trial with a parallel group design (Fig. 1). The study aims to evaluate the effect of Staphefekt on the use of corticosteroids, disease severity, QoL and composition of the microbiome in patients with AD. The study was designed by the Department of Dermatology of the Erasmus MC University Medical Centre Rotterdam and will be executed in collaboration with the Havenziekenhuis Rotterdam. Enrolment and follow-up visits take place at these two locations. Participants who comply with the criteria for inclusion and exclusion will start with a 2-week run-in period to standardize the corticosteroid use with triamcinolone acetonide 0.1% cream. After completion of the run-in phase, participants will be randomized to either treatment with Staphefekt or a placebo for 12 weeks, followed by an 8-week follow-up period. An Eczema Area Severity Index (EASI) above 50 after the run-in phase is a contraindication for further participation. During the course of the study, participants visit the outpatient clinic six times (visit 1 through 6) and data will be collected on corticosteroid use, disease severity, QoL, skin microbiome and adverse events. See Additional file 1 for an overview of the Standard Protocol Items: Recommendations for Interventional Trials (SPIRIT) 2013 checklist items. See Table 2 for the SPIRIT diagram of the trial procedures.

This study follows the Dutch Medical Research Involving Human Subjects Act 1998 (WMO) and the principles of the Helsinki Declaration 2008. All study procedures have been reviewed and approved by the Medical Ethics Committee of the Erasmus MC University Medical Centre Rotterdam, The Netherlands (reference 2016-233). Protocol amendments will be submitted for review at the Medical Ethics Committee.

This study will enrol adults (18 years or older) diagnosed with AD according to the UK working party diagnostic criteria for AD [23‐25]. Participants are eligible for enrolment if they have a score between 7.1 and 50.0 on the EASI for disease severity. Topical corticosteroids must have been prescribed before enrolment. All patients must be able to read and understand the patient information and provide written informed consent. Patients are not eligible for enrolment if they used: (1) systemic antibiotics or corticosteroids in the 2 months prior to enrolment, (2) oral immunosuppressive agents or UV therapy in the 3 months before enrolment or (3) local antibiotics or Staphefekt (from commercial sources) 1 week before enrolment. Other criteria for exclusion are known contact allergy to any of the components of the study drug (e.g. propylene glycol), clinically infected AD or the existence of other skin condition(s) that could interfere with the assessment of the AD severity.

Participants with AD will be recruited from the dermatology outpatient clinic of the Erasmus MC and the Havenziekenhuis Rotterdam. Furthermore, Dutch dermatologists are informed about the study via the Dutch Trial Network and via scientific conferences. Patients with AD are informed via the patient support group and via online media, such as DermHome (https://​www.​huidhuis.​nl/​). In addition, recruiting advertisements will be placed on student forums and in local newspapers. Patients who are interested in participation in the trial can contact the researcher directly via email or phone. After a first screening with regard to the inclusion and exclusion criteria via email or phone, potentially eligible participants receive an information letter and will be invited to the dermatology outpatient clinic to further assess eligibility. Patients who fulfil the inclusion criteria and are willing to participate, will be included in the study after providing written informed consent.

The sample size for this study was calculated based on the primary outcome, namely the difference in mean days per week of corticosteroid use over 12 weeks between the Staphefekt arm and the placebo arm, in patients who are positive for S. aureus on the skin at baseline. This is the first study measuring clinical outcomes of Staphefekt in patients with AD. We expect to find a mean topical corticosteroid use of 5 days/week in the placebo group. This was based on a study by Hon et al. that showed decreased use of topical corticosteroids when taking bleach baths, an anti-staphylococcal therapy [26]. Based on the results of this study, we anticipate an effect size of 1.25 day/week reduction in topical corticosteroid use in the Staphefekt arm. A sample size was calculated using the unpaired t test to compare means in a superiority trial design. With power of 0.80, alpha of 0.05 and SD of 2.0, 40 patients are needed per treatment arm. Assuming 10% drop out and 90% of the patients being positive for S. aureus on the skin lesions, 50 patients will be assigned to each of the two treatment arms.

The participants are randomly assigned, in a 1:1 fashion to either treatment with Staphefekt or placebo. Stratified block randomization for AD severity is performed to ensure equal distribution of patients with moderate and severe AD over the treatment arms (EASI 7.1–21 and EASI 21.1–50). Randomization is done by an independent biostatistician of the Erasmus MC, using the statistical software package R version 3.2.2. The participants, the researchers and laboratory analyst are blinded to the intervention. The pharmacy manages the randomization list and provides blinded study medication.

After enrolment, all participants start a run-in phase of 2 weeks in which they receive a standardized dosing regimen of topically applied triamcinolone acetonide 0.1% cream (Table 1). After the run-in period, the patient and the researcher evaluate further participation, with very severe AD (EASI >50) as a contraindication for continuation. The run-in period and randomization is followed by a 12-week treatment period and an 8-week follow-up period. During the treatment period, Staphefekt or placebo cream will be applied on the total skin surface twice daily to reach optimal reduction of S. aureus, as both lesional and non-lesional skin are often colonised [4]. The Staphefekt endolysin is made available in a cetomacrogol-based cream. The placebo is composed of the same cetomacrogol-based cream, without Staphefekt. During the treatment and follow-up period triamcinolone will be used according to the corticosteroid dosing regimen (Table 1). Measurements and assessments will be performed at enrolment (start of the run-in phase, visit 1), baseline (start treatment with Staphefekt/placebo, visit 2a), 0.5 hours after baseline and 2, 6, 12 and 20 weeks after baseline (visit 2b to 6). Table 2 provides an overview of the measurements per visit. Unless it is in the best interest of the patients (for example in the case of an eczema flare), patients are not allowed to use systemic or topical immunosuppressive medication (including calcineurin inhibitors), antibiotics or antiseptics during the study. Escape medication will be prescribed according to current treatment guidelines and its use will be registered. At the start of the study, patients receive an emollient according to the patient’s preference for use during the course of the study. The use of this emollient will be registered by weighing the tubes at each visit.

Sampling procedures are based on the “Manual of Procedures” for microbiome sampling of the Human Microbiome Project [27]. All samples are obtained by one of the researchers wearing gloves (sterile for the skin scrub). Sterile Copan 490CE.A swabs are used to sample the skin, nasal cavity and pharynx. Skin samples are taken from lesional skin, preferably located at the antecubital folds or the popliteal fold. The skin surface is swabbed for 30 seconds. The mucosal surfaces of both the anterior nares are gently rubbed, going round the area for 10 seconds. The rear of the oropharynx is swabbed for 5 seconds using a tongue depressor. For the skin scrub sample, a ring with an internal diameter of 4 cm will be placed on the same skin lesion where the swab was collected, but on a non-overlapping area: 1 ml of swab solution (0,85% NaCl, 0.1% bacteriological peptone, 0.1% Tween 80) is pipetted into the ring. After rubbing over the skin with a Copan 480CE swab for 1 minute, the swab solution will be pipetted out of the ring into an Eppendorf tube. The swabs will be sent to the laboratory by mail on the day of collection. A semi-quantitative culture technique and matrix-assisted laser desorption ionization time-of-flight (MALDI-TOFF) for identification of S. aureus will be performed. The scrub samples will be stored at –80 °C at the Erasmus MC Rotterdam until 16S rRNA-sequencing and quantitative S. aureus analysis.

The primary outcome of this study is the days per week of corticosteroid use, compared between the Staphefekt and the placebo group over 12 weeks. Patients report their triamcinolone use daily in a secured digital platform, “DermHome”.¹ Additionally, the use of triamcinolone cream will be measured in grams by weighing the study medication at the time of issue and return (each visit). Secondary outcomes include clinical efficacy and QoL from baseline through week 12 and week 20, change in the microbial composition (including S. aureus) and safety. Clinical efficacy is measured using the EASI, the Investigator’s Global Assessment (IGA) and registration of the number of flares [28]. A flare is defined as an exacerbation that requires the need to intensify treatment, from a doctor or patient’s perspective. This implies stronger topical therapy or the need for systemic treatment. A 50% increase in the EASI score compared to baseline is used as an indication to intensify treatment. The Pruritus Numerical Rating Scale (Pruritus NRS) and the Patient Orientated Eczema Measure (POEM) are included as patient-reported efficacy outcomes [29, 30]. QoL is measured using the Skindex-29 [31, 32]. Changes in the microbiome are evaluated by comparing the changes in bacterial composition between the treatment groups, determined by 16S rRNA sequencing of the skin scrub samples. Reduction in S. aureus is determined by quantitative PCR (and culture for the comparison between visit 2a and visit 2b). Safety and tolerability is assessed by monitoring the incidence of (serious) adverse device events through the end of the study, evaluated by medical check-ups that include evaluation of vital signs. Reportable adverse events will be reported within the set timelines to the competent authorities. Table 2 gives a detailed overview of the measurements per visit.¹

Data collected during the visits are entered in Open Clinica. This data management system allows direct data entry. Data entry is monitored by an independent researcher according to a predefined monitoring plan. Triamcinolone use and itch scores filled in daily by the patients in “DermHome” will be extracted in an SPSS format and combined with the Open Clinica database. Patient confidentiality will be ensured by using identification numbers. Data will be analysed on an intention-to-treat basis. A mixed linear regression model will be used to examine if there is a significant difference in corticosteroid use over 12 weeks between the intervention and the placebo group [34]. This model accounts for repeated measurements in each patient and is valid in the case of missing data. Covariates that could influence the outcome variable will be included in the model. Subgroup analysis will be performed to analyse patients who are positive for S. aureus on the skin versus patients that are negative for S. aureus before the start of the intervention. Positive patients are defined as having positive cultures both at visit 1 and 2a. Negative patients must have two negative cultures. Patients that have one positive and one negative culture will not be included in the subgroup analysis. Secondary outcomes will also be analysed using a mixed model analysis (linear or logistic according to the type of data). The findings of this study will be published in national and international journals (according CONSORT 2010 Statement) and will be communicated to the relevant patient associations.

The first patient was included in the study in July 2016. Patient recruitment is currently ongoing and the recruitment is expected to be completed by August 2017.