Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2024 Apr 2;15(1):2839.
doi: 10.1038/s41467-024-46540-0.

Dupilumab-associated head and neck dermatitis shows a pronounced type 22 immune signature mediated by oligoclonally expanded T cells

Christine Bangert #  1 , Natalia Alkon #  1 , Sumanth Chennareddy  2 , Tamara Arnoldner  1 , Jasmine P Levine  2   3 , Magdalena Pilz  1 , Marco A Medjimorec  1 , John Ruggiero  2 , Emry R Cohenour  2 , Constanze Jonak  1 , William Damsky  4 , Johannes Griss  1 , Patrick M Brunner  5
Affiliations

Dupilumab-associated head and neck dermatitis shows a pronounced type 22 immune signature mediated by oligoclonally expanded T cells

Christine Bangert et al. Nat Commun. .

Abstract

Dupilumab, an IL4R-blocking antibody, has shown clinical efficacy for atopic dermatitis (AD) treatment. In addition to conjunctivitis/blepharitis, the de novo appearance of head/neck dermatitis is now recognized as a distinct side effect, occurring in up to 10% of patients. Histopathological features distinct from AD suggest a drug effect, but exact underlying mechanisms remain unknown. We profiled punch biopsies from dupilumab-associated head and neck dermatitis (DAHND) by using single-cell RNA sequencing and compared data with untreated AD and healthy control skin. We show that dupilumab treatment was accompanied by normalization of IL-4/IL-13 downstream activity markers such as CCL13, CCL17, CCL18 and CCL26. By contrast, we found strong increases in type 22-associated markers (IL22, AHR) especially in oligoclonally expanded T cells, accompanied by enhanced keratinocyte activation and IL-22 receptor upregulation. Taken together, we demonstrate that dupilumab effectively dampens conventional type 2 inflammation in DAHND lesions, with concomitant hyperactivation of IL22-associated responses.

PubMed Disclaimer

Conflict of interest statement

C.B. has received personal fees from ALK, Mylan, LEO Pharma, Pfizer, Sanofi Genzyme, Eli Lilly, Novartis, Almirall and AbbVie. C.B. is an investigator for Novartis, Sanofi, Abbvie, Elli Lilly, LEO Pharma and Galderma. C.J. has received personal fees from LEO Pharma, Pfizer, Recordati Rare Diseases, Eli Lilly and Company, Novartis, Takeda, AbbVie, Janssen, UCB, Sandoz, Kyowa Kirin, Bristol Myers Squibb, and Almirall. C.J. is an investigator for Eli Lilly and Company, Novartis, Innate Pharma, and 4SC (grant paid to her institution). W.D. has received personal fees from Pfizer, Incite, Eli Lilly, TWI Biotechnology, Fresenius Kabi, Epiarx Diagnostics, and Boehringer Ingelheim. W.D. has received research support from Pfizer, Advanced Cell Diagnostics/Bio-techne, Abbvie, Bristol Myers Squibb, and Incite. W.D. receives licensing fees from EMD/Millipore/Sigma. JG has received personal fees from AbbVie and Novartis. P.M.B. has received personal fees from LEO Pharma, Pfizer, Sanofi Genzyme, Eli Lilly, Novartis, Celgene, UCB Pharma, Biotest, Boehringer Ingelheim, AbbVie, Amgen, Arena Pharmaceuticals, GSK, BMS, Almirall and Regeneron. PMB has received research support from Pfizer. N.A., S.C., T.A., J.P.L., M.P., M.A.M., J.R. and E.R.C. declare no competing interests.

Figures

Fig. 1
Fig. 1. scRNA-seq map of cells from DAHND, untreated head/neck AD and trunk AD, as well as HC samples.
A Clinical pictures of the 6 patients that were included in this study with de novo dupilumab-associated head and neck dermatitis (DAHND). B UMAP plot of unsupervised clustering of 196,040 cells integrated from DAHND (n = 6), untreated head/neck AD (n = 5), untreated trunk AD (n = 5) and HC samples (n = 4) according to similarity of transcriptome, resulting in 41 different color-coded clusters. C Relative distribution of samples per condition across cell clusters. D Dot plot displaying average gene expression and frequency of canonical cell type markers for each cluster. Circle size represents the percentage of cells expressing the specific marker within a cluster. Coloring denotes expression levels within each cluster (red is high). E Feature plots of disease groups showing the proliferation marker MKI67. Intensity of expression levels for each cell is color-coded (red) and overlaid onto UMAP plots. UMAP Uniform Manifold Approximation and Projection, T T-cells, NK NK cells, ILC innate lymphoid cells, B cells of the B cell lineage, LC Langerhans cells, DC dendritic cells, MP macrophages, MC mast cells, NC neuron cells, MEL melanocytes, SMC smooth muscle cells, LEC lymphatic endothelial cells, BEC blood endothelial cells, P pericytes, FB fibroblasts, SG sweat gland cells, KC keratinocytes; “pro” indicates proliferating subsets.
Fig. 2
Fig. 2. Subclustering of the larger “B cell” cluster.
A Subclustering into B cells, plasmacytoid dendritic cells (pDCs), and plasma cells. B Absolute cell counts across conditions: HC (n = 4), trunk AD (n = 5), head/neck AD (n = 5), and DAHND (n = 6); each dot represents a single individual; data are presented as median with interquartile range (IQR). C–F Feature plots of selected gene expression of canonical markers. Intensity of expression levels for each cell is color-coded (red is high) and overlaid onto UMAP plots. G, H Representative immunofluorescence stainings of the B cell marker CD79A as well as absolute cell counts of CD79A+ cells per mm epidermis across conditions: HC (n = 3), trunk AD (n = 4), head/neck AD (n = 5), and DAHND (n = 3); each dot represents a single individual; data are presented as median with IQR. Statistical significance in this figure was calculated using a Kruskal–Wallis test for multiple comparisons followed by a Dunn’s post hoc test.
Fig. 3
Fig. 3. Comparative analysis of lymphoid populations in DAHND, untreated head/neck and trunk AD, and HC samples.
A UMAP plot of the lymphoid cluster, consisting of T cells, NK cells, and ILCs. B Feature plots of selected T cell canonical genes. Intensity of expression levels for each cell is color-coded (red) and overlaid onto UMAP plots. C Dot plot with canonical markers indicative of respective lymphoid subsets. Coloring denotes expression levels within each cluster (red is high). D Feature plots of genes associated with so-called “Th2A” cells. Intensity of expression levels for each cell is color-coded (red) and overlaid onto UMAP plots. E Absolute cell counts per cluster across conditions: HC (n = 4), trunk AD (n = 5), head/neck AD (n = 5), and DAHND (n = 6); each dot represents a single individual; data are presented as median with IQR. Statistical significance was calculated using a Kruskal–Wallis test for multiple comparisons followed by a Dunn’s post hoc test. F–I Violin plots of selected T cell lineage genes in T1, T4, T6, and Tpro clusters; y-axis indicates expression levels.
Fig. 4
Fig. 4. Clonal landscape within disease groups.
(A) Total cell counts of the top 10 T cell receptor (TCR) clonotypes summarized for each condition: HC (n = 4), trunk AD (n = 5), head/neck AD (n = 5) and DAHND (n = 6); each dot represents a single individual; data are presented as median with IQR. B Absolute numbers of cells harboring the top 10 TCR clonotypes within each donor as distributed across lymphocyte clusters; data are presented as median with IQR. C Venn diagram of genes differentially up- or downregulated between the top 10 T cell oligoclones and the polyclonal T cell infiltrate, as calculated for each disease group, with expression levels of selected markers within the top 10 expanded clones (oligoclonal) and all other TCR+ cells (polyclonal). Differential gene expression was defined as log2 fold change >∣ ± 0.25∣ and adjusted p < 0.05. Selected genes are presented in a dot plot; coloring denotes expression levels within each group (red is high). D Expression of CD8A, CD8B, and CD4 by each of the top 10 expanded T cell clones per sample; coloring denotes expression levels (red is high). E, F Representative pictures of multicolor immunofluorescence stainings for CD8, CD129 (IL9R), CD314 (KLRK1), with triple-positive cells appearing in white (arrows), as well as absolute cell counts per mm of epidermis across conditions: HC (n = 3), trunk AD (n = 4), head/neck AD (n = 5) and DAHND (n = 3); each dot represents a single donor; data are presented as median with IQR. G Location of top 10 expanded T cell clones in UMAP plots of indicated disease groups. H Feature plots of T cell cytokine gene expression in each disease group. Intensity of expression levels for each cell is color-coded (red) and overlaid onto UMAP plots. I Gene expression dot plots displaying average and frequency of expression of selected markers for each disease group. Circle size represents the percentage of cells expressing the specific marker within a cluster. Coloring denotes expression levels within each group (red is high). J, K Representative pictures of multicolor immunofluorescence stainings for CD8, CD129 (IL9R), and IL-22, with triple-positive cells appearing in white (arrows), as well as quantification of absolute cell numbers per mm of epidermis across conditions: HC (n = 3), trunk AD (n = 4), head/neck AD (n = 5) and DAHND (n = 3); each dot represents a single donor; data are presented as median with IQR. Statistical significance in this figure was calculated using a Kruskal–Wallis test for multiple comparisons followed by a Dunn’s post hoc test.
Fig. 5
Fig. 5. Characterization of myeloid cells in DAHND, untreated head/neck and trunk AD, as well as HC samples.
A UMAP plots of myeloid cell subsets. B Dot plot of canonical markers for respective dendritic cell (DC), Langerhans cell (LC), and macrophage (MP) subsets; coloring denotes expression levels within each cluster (red is high). C Absolute cell counts per cluster across conditions: HC (n = 4), trunk AD (n = 5), head/neck AD (n = 5), and DAHND (n = 6); each dot represents a single donor; data are presented as median with IQR. Statistical significance was calculated using a Kruskal–Wallis test for multiple comparisons followed by a Dunn’s post hoc test. D–H Violin plots of selected genes in myeloid cell clusters; y-axis indicates expression levels. I, J Feature plots of ALOX15 and RARRES2 expression in each disease group. Intensity of expression levels for each cell is color-coded (red) and overlaid onto UMAP plots. K, L Violin plots of selected genes in DC2, DC4, and LC clusters.
Fig. 6
Fig. 6. Langerhans cells express the IL22 inhibitor IL-22BP.
A–D Representative immunofluorescence images of CD207 and IL-22BP (IL22RA2) in HC skin, trunk AD, head/neck AD, and DAHND; dotted lines in left images denote dermo-epidermal junction. E Quantification of epidermal and dermal CD207+ Langerhans cells; absolute cell counts are given per mm of epidermis. HC (n = 3), trunk AD (n = 4), head/neck AD (n = 5), and DAHND (n = 3); each dot represents a single donor; data are presented as median with IQR. F Ratio of IL-22BP+ CD207+ cells vs. all CD207+ Langerhans cells within the dermis in respective disease groups: HC (n = 3), trunk AD (n = 4), head/neck AD (n = 5), and DAHND (n = 3); each dot represents a single donor; data are presented as median with IQR.
Fig. 7
Fig. 7. Comparative analysis of keratinocyte (KC) and sweat gland (SG) populations in DAHND, untreated head/neck and trunk AD, and HC samples.
A UMAP plot of KC and SG subsets. B Dot plot of canonical markers for respective epithelial subsets; coloring denotes expression levels within each cluster (red is high). C Absolute cell counts per KC cluster across conditions: HC (n = 4), trunk AD (n = 5), head/neck AD (n = 5), and DAHND (n = 6); each dot represents a single individual; data are presented as median with IQR. Statistical significance was calculated using a Kruskal–Wallis test for multiple comparisons followed by a Dunn’s post hoc test. D Dot plot of selected genes showing the differences between disease groups in all keratinocytes (KC1, KC2, KC3, KC4, KCpro) combined; coloring denotes expression levels within each group (red is high). E Dot plot of IL-13-associated keratinocyte response genes from a published dataset showing differences between disease groups in all keratinocytes (KC1, KC2, KC3, KC4, KCpro) combined; coloring denotes expression levels within each group (red is high). F Feature plots of IL22-receptor components as expressed in epithelial cells of each disease group. Intensity of expression levels for each cell is color-coded (red) and overlaid onto UMAP plots. G Dot dot plot of IL22-receptor components showing the differences between conditions for all keratinocytes combined; coloring denotes expression levels within each group (red is high).
Fig. 8
Fig. 8. Comparative analysis of endothelial cell populations in DAHND, untreated head/neck and trunk AD, as well as HC samples.
A UMAP plot of endothelial cell subsets. B Dot plot of canonical cell type markers of endothelial cell subpopulations; coloring denotes expression levels within each cluster (red is high). C Absolute cell counts of respective clusters across conditions: HC (n = 4), trunk AD (n = 5), head/neck AD (n = 5), and DAHND (n = 6); each dot represents a single individual; data are presented as median with IQR. Statistical significance was calculated using a Kruskal–Wallis test for multiple comparisons followed by a Dunn’s post hoc test. BEC blood endothelial cells; LEC lymphatic endothelial cells; P pericytes. (D, E) Violin plots of selected genes in arteriolar BEC2 and LEC clusters.
Fig. 9
Fig. 9. Comparative analysis of fibroblast (FB) and smooth muscle cell (SMC) populations in DAHND, untreated head/neck and trunk AD, and HC samples.
A UMAP plot of SMC and FB cell subsets. B Dot plot of canonical cell type markers of FB and SMC; coloring denotes expression levels within each cluster (red is high). C–F Volcano plots of differentially expressed genes between DAHND and untreated head/neck AD in FB1, FB2, FB3, and FB4 based on Supplementary Data 5 for respective FB clusters, i.e., DEGs were defined as log2 fold change >∣ ± 0.25∣ and adjusted p < 0.05 using a two-sided Wilcoxon rank-sum test and Bonferroni correction; x-axis indicates average log fold change, y-axis indicates the −log10 of the adjusted p value. G Violin plots of selected genes depicting differences between conditions in the SMC2 subpopulation.
See this image and copyright information in PMC

Similar articles

See all similar articles

References

    1. Ratchataswan T, et al. Biologics for treatment of atopic dermatitis: current status and future prospect. J. Allergy Clin. Immunol. Pr. 2021;9:1053–1065. doi: 10.1016/j.jaip.2020.11.034. - DOI - PMC - PubMed
    1. Renert-Yuval Y, Guttman-Yassky E. New treatments for atopic dermatitis targeting beyond IL-4/IL-13 cytokines. Ann. Allergy Asthma Immunol. 2020;124:28–35. doi: 10.1016/j.anai.2019.10.005. - DOI - PubMed
    1. Facheris P, Jeffery J, Del Duca E, Guttman-Yassky E. The translational revolution in atopic dermatitis: the paradigm shift from pathogenesis to treatment. Cell Mol. Immunol. 2023;20:448–474. doi: 10.1038/s41423-023-00992-4. - DOI - PMC - PubMed
    1. David E, et al. The evolving landscape of biologic therapies for atopic dermatitis: present and future perspective. Clin. Exp. Allergy. 2023;53:156–172. doi: 10.1111/cea.14263. - DOI - PubMed
    1. Czarnowicki T, He H, Krueger JG, Guttman-Yassky E. Atopic dermatitis endotypes and implications for targeted therapeutics. J. Allergy Clin. Immunol. 2019;143:1–11. doi: 10.1016/j.jaci.2018.10.032. - DOI - PubMed

MeSH terms