TSLP is an IL-7-like cytokine highly expressed in AD skin. TSLP binds to the heterodimeric receptor composed of the TSLP receptor and the IL-7R subunit α (IL-7Rα). TSLP is mainly produced in keratinocytes and other epithelial cells [29, 30]. Various extrinsic and intrinsic factors induce TSLP production; environmental factors such as TLR ligands (pathogen-associated molecular patterns, or PAMPs), danger-associated molecular patterns, virus, allergens, helminthes, and chemicals can trigger TSLP production [31]. Moreover, pro-inflammatory cytokines (TNFα and IL-1α) and type 2 cytokines (IL-4 and IL-13) synergistically enhance TSLP production [32, 33]. We found that periostin, a downstream molecule of IL-4/IL-13 signals, induced TSLP production in keratinocytes by directly binding to integrins on their cell surface [21]. It was reported that basophils directly activated by protease allergens could produce TSLP as well as IL-4, suggesting that basophils are another source of TSLP in addition to keratinocytes/epithelial cells [34]. Moreover, it has been shown that mast cells are indispensable for TSLP production in the mouse model of allergic rhinitis [35].

TSLP is a pleiotropic cytokine exerting its effects on many types of cells—DCs, T cells, B cells, mast cells, eosinophils, macrophages, basophils, and group 2 ILCs. TSLP induces polarization of myeloid DCs (CD11c⁺ cells) that drive differentiation of naïve T cells into T_H2 cells, suggesting that TSLP plays an important role for linking innate immunity and acquired immunity [29, 36]. Expression of OX40 ligand (OX40L) and inhibition of IL-12 production are critical for these effects [37]. In concordance with this finding, administration of neutralizing Abs against OX40L inhibited type 2 inflammation in mice [38]. TSLP acts as a growth factor for T cells; in particular, it has been shown that TSLP directly acts on T_H2 cells, causing their expansion [39]. TSLP acts on innate immune cells such as mast cells, basophils, and NKT cells, inducing production of pro-inflammatory and type 2 cytokines in these cells involved in the initiation of an innate phase in allergic inflammation [29, 36]. It has also been shown that TSLP activates group 2 ILCs in the skin, rather than group 2 ILCs in the lungs [40]. Moreover, it has been demonstrated that TSLP directly acts on a subset of sensory neurons expressing transient receptor potential vanilloid 1 (TRPV1), a calcium-permeable ion channel, to trigger robust itch behaviors [41]. The significance of TSLP in the pathogenesis of AD has been also confirmed using model mice; selective expression of TSLP in the skin caused spontaneous AD-like phenotypes [42], whereas genetic deficiency of TSLP or administration of neutralizing Abs against TSLP inhibited allergen-induced skin inflammation [43].

IL-33 is a member of the IL-1 family and is mainly expressed by epithelial cells, fibroblasts, and endothelial cells [44]. IL-33 binds to the heterodimeric receptor composed of ST2, also known as IL-1RL1, and the IL-1R accessory protein (IL-1RAcP). IL-33 localizes to the nucleus, and cellular necrosis leads to the release of IL-33 followed by activation of IL-33 by cleavage with inflammatory proteases [44]. This suggests that IL-33 acts as an alarmin, sensing danger signals in the body, as well as high mobility group box 1 (HMGB1) and IL-1α.

Like TSLP, IL-33 is also a pleiotropic cytokine exerting its effects on many types of cells—T_H2 cells, mast cells, basophils, eosinophils, macrophages, DCs, and group 2 ILCs. IL-33 causes expansion, activation, and recruitment of T_H2 cells [45, 46]. It induces production of pro-inflammatory and type 2 cytokines in mast cells [47, 48] and basophils [49]. It potently activates eosinophils and enhances their adhesion and survival [50, 51]. In macrophages, IL-33 amplifies polarization of M2 macrophages [52]. It also activates DCs by upregulating expression of MHC class II molecules and co-stimulatory molecules [53]. Moreover, it should be noted that group 2 ILCs are main targets for IL-33 and that they are expanded by administering IL-33 [4–6].

Expression of both IL-33 and its receptor components, ST2 and IL-1RAcP, was upregulated in the lesional skin of AD patients, and expression of IL-33 and ST2 was further enhanced after HDM or staphylococcal enterotoxin B (SEB) exposure [54]. Specifically expressed IL-33 in the skin caused pruritic inflammatory skin similar to the features in AD patients accompanied with expansion of group 2 ILCs [55], suggesting the potential of IL-33 to cause the phenotypes of AD.

IL-25, also known as IL-17E, is a member of the IL-17 family. IL-25 binds to the heterodimeric receptor composed of the IL-17 receptor B (IL-17RB), also known as IL-25R, and IL-17RA. IL-25-positive keratinocytes increased in AD skin [56]. DCs and activated eosinophils and basophils produce IL-25 [56, 57]. IL-25 enhances proliferation and T_H2 polarization as well as production of type 2 cytokines in T_H2 memory cells. Expression of IL-25 and IL-17RB was upregulated in AD patients, suggesting the possibility that this cytokine is involved in the pathogenesis of AD. Moreover, IL-25 targets group 2 ILCs as well as IL-33 expanding these cells and inducing secretion of IL-5 and IL-13 from these cells [4–6].

IFN-γ is a signature cytokine of type 1 immunity. IFN-γ binds to the tetrameric receptor composed of two molecules of IFN-γ receptor 1 (IFN-γ-R1) and IFN-γ receptor 2 (IFN-γ-R2). In the past, it was widely accepted that type 2 inflammation is dominant in the acute phase of AD, and then in the chronic phase, it is switched into type 1 inflammation in which IFN-γ is highly expressed [84]. However, as mentioned earlier, it is now widely accepted that the type 1/type 17 axis is dominant in the pathogenesis of psoriasis, whereas AD showed the dominance in the type 2/type 22 axis, even in the chronic phase [71–73].

IL-17 is a signature cytokine of type 17 immunity. IL-17 binds to the heterodimeric receptor composed of the IL-17 receptor A (IL-17RA) and the IL-17 receptor C (IL-17RC). There are some reports showing that expression of IL-17 is upregulated in AD skin [85, 86]. However, as mentioned earlier, the contradict concept is now widely accepted; the type 1/type 17 axis exists in the pathogenesis of psoriasis, not AD, and IL-17 expression decreases, particularly in the chronic phase [71].

IL-19 is a member of the IL-20 cytokine family, as is IL-22. IL-19 binds to the heterodimeric receptor composed of the IL-20 receptor subunits α and β (IL-20RA and IL-20RB). IL-19 is induced by IL-17 and IL-4/IL-13 and enhances IL-17’s actions on keratinocytes [87, 88]. IL-19 is highly expressed in pediatric AD patients rather than in adult AD patients [89]. Moreover, Asian AD patients showed higher IL-19 expression than European-American AD patients [90]. Based on this finding, it was suggested the Asian AD phenotype is a blend of the European-American AD phenotype and the psoriatic phenotype in which high expression of IL-19 is included. However, the details of the roles of IL-19 in the pathogenesis of AD remain unclear.

Dupilumab is a fully human monoclonal Ab against IL-4Rα shared with type 1 IL-4R and type 2 IL-4R/IL-13R, so that it can inhibit both IL-4 and IL-13 signals. In phase I studies (M4A and M4B), moderate-to-severe AD patients received 75–300 mg of dupilumab once a week for 4 weeks [91]. The patients receiving dupilumab got better in a 50% improvement in the Eczema Area and Severity Index referred to as EASI-50 as well as the pruritus numerical rating scale (NRS) and TARC, but not IgE. The gene expression profiles were dramatically changed by dupilumab treatment; epidermal proliferation markers—K16 and MK167—and type 2-related chemokines, CCL17, CCL18, CCL22, and CCL26, were downregulated, whereas barrier-related genes were upregulated [92]. In the phase IIa study, monotherapy with 300 mg of dupilumab was administered to moderate-to-severe AD patients for 12 weeks (M12), showing improvements of clinical end points—EASI-50, EASI-75, Investigator’s Global Assessment (IGA), the percent change in EASI, and pruritus NRS—and biomarkers such as TARC and IgE [91]. The combined therapy of dupilumab (300 mg) and topical glucocorticoids for 4 weeks (C4) showed improvements in clinical end points—EASI-50, EASI-75, pruritus NRS, IGA, and body surface area (BSA) affected—and biomarkers (TARC and IgE) and decrease of topical glucocorticoid use compared to topical glucocorticoids alone. Moreover, the phase IIb study (100–300 mg of dupilumab once per 2 or 4 weeks for 16 weeks) showed good efficacy again in improved clinical outcomes—EASI scores, SCORAD scores, patient-reported outcomes (PROs) of pruritus, the pruritus NRS, and the Dermatology Life Quality Index (DLQI)—and decreased TARC [93]. Finally, the phase III study (SOLO 1 and SOLO 2) was performed, in which 300 mg of dupilumab was administered once per 2 or 4 weeks for 16 weeks to moderate-to-severe AD patients inadequately controlled by topical treatment [94]. The IGA score, the primary end point, and EASI-75, a key secondary end point, were significantly improved. Other parameters—EASI scores, pruritus NRS, SCORAD, the Hospital Anxiety and Depression Scale (HADS), the Patient-Oriented Eczema Measure (POEM), and DLQI—were also improved. During all the studies, no serious adverse events occurred. Thus, dupilumab has become the first molecularly targeted drug for the treatment of AD to complete clinical trials.

Mepolizumab is a human monoclonal Ab against IL-5. Moderate-to-severe AD patients received two 750 mg of mepolizumab 1 week apart [25]. Peripheral blood eosinophils were significantly decreased by treatment of mepolizumab; however, the Physician Global Assessment (PGA) of improvement, SCORAD, pruritus scoring, and TARC did not show statistically significant improvements.

Two kinds of Abs targeting TSLP/TSLPR have been applied to clinical trials. One is AMG 157 targeting TSLP, and the other is MK-8226 targeting TSLPR. The phase I study for AMG 157 was completed, but the details are unclear. The phase I study for MK-8226 was terminated due to business reasons.

CIM331 is a human monoclonal Ab against IL-31RA, and the results of the phase I study have been reported [95]. A single dose of CIM331 up to 3.0 mg/kg was administered into healthy volunteers and AD patients. No serious adverse events occurred. In AD patients, CIM331 reduced the pruritus visual analogue scale (VAS) score and decreased the amounts of topical corticosteroids used.