Dupilumab Therapy Improves Stratum Corneum Hydration and Skin Dysbiosis in Patients With Atopic Dermatitis

Abstract

Purpose: We aimed to investigate the effects of dupilumab on 1) the permeability and antimicrobial barrier, 2) the composition of the skin microbiome, and 3) the correlation between changes in skin barrier properties and microbiota in atopic dermatitis (AD) patients.

Methods: Ten patients with severe AD were treated with dupilumab for 12 weeks. Disease severity was assessed using the Eczema Area and Severity Index (EASI). Skin barrier function was evaluated by measuring transepidermal water loss, stratum corneum (SC) hydration, and pH. The following parameters were analyzed in the pre- and post-treatment SC samples; 1) skin microbiota using 16S rRNA gene sequencing, 2) lipid composition using mass spectrometry, and 3) human β-defensin 2 (hBD-2) expression using quantitative reverse transcription polymerase chain reaction.

Results: SC hydration levels in the lesional and non-lesional skin increased after 12-week dupilumab therapy (24.2%, P < 0.001 and 59.9%, P < 0.001, respectively, vs. baseline) and correlated with EASI improvement (r = 0.90, P < 0.001 and r = 0.85, P = 0.003, respectively). Dupilumab increased the long-chain ceramide levels in atopic skin (118.4%, P = 0.028 vs. baseline) that correlated with changes in SC hydration (r = 0.81, P = 0.007) and reduced the elevated hBD-2 messenger RNA levels (-15.4%, P = 0.005 vs. baseline) in the lesional skin. Dupilumab decreased the abundance of Staphylococcus aureus. In contrast, the microbial diversity and the abundance of Cutibacterium and Corynebacterium species increased, which were correlated with an increase in SC hydration levels (Shannon diversity, r = 0.71, P = 0.027; Cutibacterium, r = 0.73, P = 0.017; Corynebacterium, r = 0.75, P = 0.012). Increased abundance of Cutibacterium species was also correlated with EASI improvement (r = 0.68, P = 0.032).

Conclusions: Th2 blockade-induced normalization of skin microbiome in AD patients is associated with increased SC hydration.

Keywords: Atopic dermatitis; Cutibacterium; Staphylococcus aureus; ceramides; dupilumab; microbiota; stratum corneum hydration.

Fig. 1. Changes in skin barrier function and SC ceramide content in patients with AD following dupilumab treatment. (A) Changes in TEWL, SC hydration, and skin surface pH in lesional and non-lesional skin from baseline to week 12. (B) EASI scores before and after treatment. (C) Correlation between the percentage change in SC hydration levels from baseline to week 12 and the percentage decline in EASI scores at week 12. (D) Liquid chromatography electrospray ionization tandem mass spectrometry analysis of the ceramide levels in SC of the lesional and non-lesional skin before and after dupilumab treatment. (E) Correlation between the percentage changes in C26 ceramide contents and SC hydration levels from baseline to week 12 in the lesional skin. Data are expressed as mean ± standard error of mean.

SC, stratum corneum; TEWL, transepidermal water loss; EASI, eczema area and severity index; B, baseline; 3M, 3 months of treatment; AD, atopic dermatitis. ^*P < 0.05; ^†P < 0.01.

Fig. 2. Comparison of hBD-2 mRNA (DEFB4A) levels before and after dupilumab treatment. Expression of hBD-2 mRNA (DEFB4A) is increased in lesional skin at baseline and decreased at 12 weeks of dupilumab treatment. The mRNA expression levels of hBD-2 in lesional and non-lesional stratum corneum samples of patients with AD before and after dupilumab treatment were analyzed using quantitative reverse transcription polymerase chain reaction. Data are expressed as mean ± standard error of mean.

HC, healthy control; mRNA, messenger RNA; L, lesional; NL, non-lesional; hBD-2, human β-defensin 2; AD, atopic dermatitis. ^*P < 0.05.

Fig. 3. Changes in microbiota α diversity (Shannon diversity index) in atopic lesional and non-lesional skin before and after 12 weeks of dupilumab treatment. Data are expressed as mean ± standard error of mean.

HC, healthy control; L, lesional; NL, non-lesional; AD, atopic dermatitis. ^*P < 0.05; ^†P < 0.01.

Fig. 4. Changes in skin microbiome in patients with AD before and after 12 weeks of dupilumab treatment. (A) Relative abundance of major taxa in non-lesional and lesionl atopic skin before and after dupilumab treatment. (B) Relative abundance of S. aureus, Lactobacillus spp., Cutibacterium spp., and Corynebacterium spp. in atopic lesional skin and (C) S. aureus in atopic non-lesional skin before and after dupilumab treatment. The center line in the box plots corresponds to the median.

L, lesional; NL, non-lesional; AD, atopic dermatitis.

Fig. 5. Correlation between changes in EASI, SC hydration, bacterial diversity and relative abundance of major taxa following dupilumab treatment. (A) Heat map of the overall correlation matrix. Colors display the strength of correlation based on Spearman's correlation coefficient in each pair of biomarkers. Significant correlations (corrected P < 0.05) are indicated with yellow boxes. (B) Correlation between the percentage change in EASI scores from baseline to week 12 and changes in Shannon diversity index, relative abundance (%) of S. aureus, and relative abundance (%) of Cutibacterium in lesional skin following 12 weeks of dupilumab treatment. (C) Correlation between the percentage change in SC hydration levels from baseline to week 12 and changes in Shannon diversity index, relative abundance (%) of S. aureus, and relative abundance (%) of Cutibacterium in lesional skin following 12 weeks of dupilumab treatment. (D) Correlation between changes in Shannon diversity index from baseline to week 12 and changes in the relative abundance (%) of Corynebacterium and Cutibacterium in lesional skin following 12 weeks of dupilumab treatment. The linear regression curve fit, Spearman's correlation coefficient r, and P values are shown for each graph.

EASI, eczema area and severity index; SC, stratum corneum; AD, atopic dermatitis.