Inclusion and exclusion criteria
Client owned dogs of any sex and breed, aged between 1 and 12 years old, with documented history of atopic dermatitis and previously diagnosed according to accepted criteria (Favrot et al. 2010) were recruited from 12 veterinary clinics in France, Spain and the USA. To be included in the study, dogs had to present with a skin flare, defined as a Canine Atopic Dermatitis Extent and Severity Index (Olivry et al. 2014) (CADESI-04)>10/180 and a pruritus score on a 10 cm visual analogue scale (Rybníček et al. 2009) (PVAS) between 2 and 8 (score given by the pet owner). The dogs also had to be clear of ectoparasites as well as secondary bacterial and fungal infections.
Exclusion criteria comprised of pregnant and lactating dogs, dogs with flea allergic dermatitis, skin infections, dermatophytosis, endocrine or immune-mediated disease, and dogs with evidence of demodicosis within the past 6 months. Any major disease or clinical signs that may interact with the study were also forbidden. To avoid recurrence of clinical signs linked to a food intolerance, dogs with any change in their food regimen (including fatty acid supplementation) within the last 8 weeks were also excluded. Dogs with lesions in the auricular canal that required beginning the application of a topical treatment containing antibiotics and/or anti-fungals and/or glucocorticoids were also excluded from the study.
Concurrent treatment was allowed on the condition that it had been stable for several weeks and remained unchanged during the whole study. Thus, dogs fulfilling at least one of the following conditions were excluded from the study:
- Initiation or modification (including dose change) to long-term immunomodulatory treatment (cyclosporine, oclacitinib) within 8 weeks before inclusion
- Initiation or modification of ASIT (Allergen Specific Immunotherapy) protocols within the last 9 months
- Initiation or modification (including dose change) of oral glucocorticoid treatment within 4 weeks before inclusion
- Initiation or modification (including dose change) of topical corticoids or calcineurin inhibitor treatments or ear cleaner within 4 weeks prior to inclusion
- Initiation or modification of anti-histamine treatment within 1 week prior to inclusion
- Lokivetmab injection within 2 months prior to inclusion and during the study.
To avoid a residual effect during the study, some other treatments were also subject to restrictions:
- Treatment with antimicrobial or antifungal therapies within 4 weeks prior to inclusion
- Injection with parenteral long-acting glucocorticoids within 8 weeks prior to inclusion
- Application of topical antiseptic(s) (e.g. chlorhexidine, diluted bleach bath) within 2 weeks prior to inclusion
- Application of any of the following topical products: lotions, sprays, shampoos within 7 days prior to inclusion.
Application protocol and study schedule
All included dogs received the same application protocol combining applications of Ophytrium-based Shampoo and Mousse (DOUXO® S3 CALM Shampoo and Mousse, Ceva Santé Animale, Libourne, France). The products were provided in a neutral packaging. The application protocol consisted in one Shampoo application and 2 Mousse applications during the first week and 3 Mousse applications weekly during the 2 following weeks (Figure 1). A minimum interval of 2 days and a maximum of 3 days had to be respected between each topical application. The Shampoo and Mousse were applied on the whole body according to the manufacturer’s recommendations. The number of pumps of the Mousse to be applied to the dog was estimated on Day 0 according to the dog’s bodyweight, hair length and hair density. If necessary, this dose could be adjusted by the owner at home in order to cover the whole body surface.
Each dog was examined four times by the veterinarian: at the inclusion visit on day 0, and at follow-up visits on day 7, day 14 and day 21 (Figure 1).
During each weekly visit, veterinarians evaluated dogs’ skin health using the CADESI-04 (Olivry et al. 2014). The effect of the protocol on this parameter was the primary outcome. The percentage of dogs with a final ≥50% or ≥70% CADESI-04 decrease and the percentage of dogs with a final CADESI-04 <10 were secondary outcomes. As suggested by a recent publication for short-term studies (shorter than 6 weeks) (Olivry 2019), CADESI4-E score (erythema from CADESI-04 score) was also evaluated. Also the veterinarian evaluated the status of the skin condition from back to normal to very severe.
At each veterinary visit, owners evaluated their dog’s pruritus using a validated 10-cm visual analogue scale (PVAS) , taking into account their dog’s pruritus in the previous 24 hours. Mean evolution of this parameter was assessed, as well as the percentage of dogs with a final ≥30% or ≥50% PVAS decrease and the percentage of dogs with a final PVAS <2. A decrease of at least 2 cm on the PVAS between the beginning and the end of study was already used as a success criterion(Cosgrove et al. 2013).
At the end of the study, both pet owner and veterinarian overall assessment and satisfaction were collected through questionnaires. The veterinarian was asked to appraise the clinical evolution at D21 compared to D0 (absent or worse, weak, satisfactory, good, excellent). The owners were asked to assess the performance and the practicality of the protocol using a 4-point scale: totally disagree, somewhat disagree, somewhat agree, totally agree. They were also asked to evaluate the overall response to the protocol (no response, a poor response, a fair response, a good response, an excellent response) (OGATE) (Olivry et al. 2018).
As suggested by Olivry et al. in 2018 to allow some standardization of clinical trial results, the COSCAD’18 was evaluated, considering the OGATE, the number of dogs with a PVAS<2, and number of dogs with a CADESI-04 <10 at the end of the study.
Data were analysed using the SAS software (V.9.4; SAS Institute) and were two-tailed with a type one error of 5%. Parameters were analysed using descriptive statistics. Quantitative variables were analysed using mean and standard deviations, categorical variables were analysed using frequency counts and percentages.
The dogs which were prematurely withdrawn for “lack of efficacy” were used in the analysis. The “LOCF” (Last Observation Carried Forward) imputation method was used to use to not advantage the protocol. Other premature terminations which were not linked to AD were excluded from efficacy analyses and kept in the safety analysis.
The evolution of the CADESI-04 score and PVAS were assessed with linear mixed models. The covariates were the study visit and the baseline as fixed effects. Additional covariates like coat length, ectoparasite treatments, coat density or treatment for atopic dermatitis were tested in the statistical model and removed if p-values were above 5%.
An order 1 autoregressive was used as the covariance matrix in the CADESI-04 model and the dog id was used as a random effect in the PVAS analysis.
Least-square means with their 95% confidence intervals were calculated to see significant differences between visits.
A linear mixed model was used to analyze the evolution through time of the CADESI-04 and the VAS. The time and the CADESI-04 at D0 or PVAS at D0 were used as fixed effects. To take into account the repeated measures, different correlation structures (No correlation, component symmetry, variance component and order 1 autoregressive) were tested with the AIC. The best correlation structure retained was an order 1 autoregressive for the CADESI-04 analysis and no correlation in the PVAS analysis. A random effect was also tested but retained only on the PVAS analysis. To compare the different visits (D14 vs D0 and D21 vs D0), the Least Square means (LS-Means) from the linear mixed model were computed and tested if they were significantly different from 0.