Narrative review on the management of moderate-severe atopic dermatitis in pediatric age of the Italian Society of Pediatric Allergology and Immunology (SIAIP), of the Italian Society of Pediatric Dermatology (SIDerP) and of the Italian Society of Pediatrics (SIP)

1. Topical Corticosteroids

2. Topical calcineurin inhibitors

3. Antibiotic, antiseptic, antiviral, and antifungal therapy

4. New topical therapies: Topical phosphodiesterase 4 inhibitors and Jak kinase inhibitors

5. Wet Wrap dressing

6. Special fabrics

7. Systemic Corticosteroids

8. Cyclosporin A

9. Methotrexate

10. Azathioprine

11. Mycophenolate mofetil

12. Antihistamines

13. Probiotics

14. Phototherapy

15. Dupilumab

16. New Drugs

17. Allergen Specific Immunotherapy

18. Thermal Therapy

19. Therapeutic education

Topical corticosteroids (TCSs) are currently recommended as a first-line anti-inflammatory therapy for AD [3, 8‐10]. The efficacy and availability of molecules with different potency and in different formulations have contributed to the widespread use of TCSs. TCSs exert an anti-inflammatory, antiproliferative, immunosuppressive, and vasoconstrictive action [11, 12]. TCSs have direct regulatory effects at the cellular level by binding to their receptor to form a corticosteroid-receptor complex capable of translocating within the nucleus and stimulating or inhibiting the synthesis of proteins. In addition, they can indirectly regulate transcription by blocking the effect of other transcription factors [12]. TCSs have been shown to inhibit the transcription of various pro-inflammatory cytokines involved in skin diseases (e.g. interleukin (IL)-1, IL-2, IL-6, TNF-α, and interferon-gamma (INF-γ) and stimulate the expression of genes that encode anti-inflammatory cytokines such as TGF-beta and IL-10. Through this immunomodulatory activity, TCSs probably play an effective role in balancing the ratio of Th1 to Th2 lymphocytes at the level of skin lesions [12]. Other anti-inflammatory effects include acceleration of apoptosis of eosinophils and T lymphocytes, suppression of the function of endothelial cells and lymphocytes, inhibition of dermal edema and dilation of capillaries, and reduction of vascular permeability. Finally, TCS showed anti-proliferative effects on several cells, including T lymphocytes [11].

In choosing the TCS to use in a child with AD, several factors must be taken into consideration such as the potency of the drug, the delivery vehicle, the age of the patient, the site and the body surface to be treated. Based on their potency, TCSs are classified in various ways. In Europe, they are divided into four groups, where group 1 contains the low potent TCSs, while group 4 contains the very potent TCSs (Table 1).

In the US, the classification includes seven groups, where group 1 contains the most potent TCSs, while group 7 contains the least potent TCSs [8]. In Japan, the classification includes five groups and the potency of TCSs decreases as the number of the group increases [9]. The potency of the TCS to be used should be chosen based on the degree of severity of AD, favoring moderately potent and low potent TCSs for mild cases of eczema, and reserving potent TCSs for severe and moderate cases of AD [9]. Very potent TCSs (e.g. clobetasol propionate) should be avoided in children as their use carries a significantly higher risk of adverse effects, both local and systemic, than TCSs from other groups [14]. The early use of a TCS of adequate potency, in sufficient quantity, and at the beginning of the flare-up of AD, increases the likelihood of controlling inflammation, restoring the skin barrier and reducing the subsequent need for TCSs [15, 16].

When prescribing TCSs, it is also important to choose the adequate delivery vehicle, which must be selected based on the characteristics of the lesions and the skin site, and the patient's preferences and tolerability, in order to ensure adherence to therapy.

The delivery vehicle, in addition to being the carrier of the pharmacological active principle, plays an important role in determining its bioavailability. To be active, a vehicle must be able to maintain the solubility and stability of the pharmacological active principle, release the drug and distribute it adequately in the skin, allow penetration through the skin barrier, guarantee the pharmacological effect of the active ingredient and limit its systemic absorption [17]. Depending on the type of vehicle used, the effects of a given active ingredient can vary in terms of potency and clinical efficacy [17]. Creams, which are often enriched with humectant molecules to enhance the moisturizing effect, are indicated for the treatment of acute or subacute lesions; ointments are instead indicated to treat areas of chronic lesions (e.g. lichenified and particularly xerotic lesions) and with a thick corneal layer (e.g. palm / plantar regions), given the high lipid content and high occlusive properties [17].

The Fingertip Unit (FTU) is the unit of measurement that has been in use for years to determine the right amount of TCS to be applied [18]. This amount corresponds to approximately 0.5 g and is sufficient to cover an area of skin equal to two palms of an adult's hand. Table 2 shows the various FTUs needed to treat the different skin surfaces according to the age of the child.

The use of this method allows the healthcare professional to have a more precise idea of the dose of TCS prescribed and the parent to overcome the increasingly widespread concern about the risk of an overdose of corticosteroids. This helps to counteract so-called corticophobia, promote therapeutic adherence and avoid under-dosage of TCSs.

The appropriate use of TCSs, even in the long term, may only rarely cause the adverse effects reported in the past (e.g. skin atrophy, striae rubre, telangiectasias). A cross-sectional observational study involving 70 children with a mean age of 3.2 years, did not document any degree of skin atrophy following an average use of about 11 months of a combination of potent TCS (betamethasone dipropionate 0.05% ointment, methylprednisolone aceponate 0.1% ointment or mometasone furoate 0.01%), moderate potency (betamethasone valerate 0.02% ointment) and low potency (hydrocortisone acetate 1% ointment). In this study, the mean monthly dose of TCS was 79 g, 128 g, and 34 g for potent, moderate potency, and low potency TCS, respectively [19]. Even in the short term, the use of TCS appears safe in terms of local adverse effects. In another study involving 174 children with AD treated with a 3-day course of a potent TCS (betamethasone valerate 0.1%), no skin thinning was documented on skin ultrasound [20]. Among the various molecules, mometasone furoate, methylprednisolone aceponate, hydrocortisone aceponate and betamethasone valerate have little or no atrophying effects [20, 21].

A few studies have evaluated the systemic effects of TCSs, such as suppression of the hypothalamic-pituitary-adrenal axis. In a meta-analysis that analyzed a total of 12 trials and 522 children, evidence of biochemical suppression of the hypothalamic-pituitary-adrenal axis, assessed at 2-4 weeks, was documented in 2%, 3.1% and 6.6% of children who used low, medium and high potency TCSs, respectively [22]. However, the normal endocrine function was restored 1-10 weeks after discontinuation of therapy. Another study showed that using a potent TCS (fluticasone dipropionate 0.05% cream) twice a day for 3-4 weeks over a large body surface area (on average 64%) in two groups of patients (3 months-3 years and 3-6 years of age) and with a mean dose over the study period of 96.7 g in the first group and 209.1 g in the second group, had a low risk of suppression of the hypothalamic-pituitary-adrenal axis [23]. In particular, of the 43 children studied, only 2 (4.7%) showed a suppression of the axis, also in this case asymptomatic and reversible.

On the other hand, currently available data on the effects of long-term therapy with TCS are scarce. The Petite Study [13], is one of the few studies in which a large follow-up period was used. In this randomized trial, the authors compared the safety of using topical pimecrolimus vs TCS on 2439 children divided into two groups (1210 were treated with pimecrolimus and 1229 with TCS). Participants were followed up to 5-6 years of life with a particular focus on the effect on the trend of stature growth and the immune system. The TCSs used in this study were low potency (e.g. hydrocortisone 1% cream/ointment) and/or medium potency (e.g. hydrocortisone butyrate 0.1% cream/ointment) TCSs, with exposure to TCS therapy ranging from 77- 396 days (although the average dose taken per time interval is not specified). The effect on stature growth and the development and maturation of the immune system was the same for the two groups of patients. The analysis of the available data suggests that TCS are sufficiently safe drugs in terms of systemic adverse effects in long-term therapy, but to date, it is not possible to establish a "safe" recommendable dose (for example as a monthly cumulative dose), due to the scarcity and poor homogeneity of data. Likewise, it is difficult to recommend the "best" TCS. However, the newer generation of TCSs (i.e. the non-halogenated double esters) has a better risk/benefit ratio, as it can balance a powerful anti-inflammatory effect with reduced systemic toxicity and a low atrophying potential [24]. Furthermore, it is not clear that two administrations per day are more effective than a single application 10). Therefore, the number of daily administrations may be decided according to the severity of the lesions: twice a day in the severe acute phase, and once a day in the mild forms or remission [9, 10].

Once remission of skin lesions is achieved, TCSs should be gradually decreased or discontinued. In this regard, a useful therapeutic strategy is termed proactive therapy. It consists of the application of TCSs twice a week (eg weekend therapy) in the skin areas of most frequent recurrence, despite the absence of active lesions, in order to maintain remission [8, 9, 14].

This therapeutic strategy is particularly effective in patients who have frequent flare-ups, which are reduced [8, 25]. The choice of potency, vehicle and amount of TCS should also be based on the skin region of application. The sites with the greatest absorption of TCSs, such as the eyelids, genitals, face and skin folds [17], require careful monitoring of possible side effects. Prolonged use of TCSs on these skin areas should be avoided, especially if they are of moderate-high potency [9]. Also, adverse intra-ocular effects (eg glaucoma and cataracts) which can occur with the use of TCS in the periorbital region, are rare and may be limited by the use of potent or moderately potent TCSs only for short periods. If maintenance therapy is required in this skin region, the use of low-power TCS or topical calcineurin inhibitors (TCIs) is recommended.

Regarding the risk of osteopenia/osteoporosis, the available data suggest that, in children treated with TCSs, the expected prevalence of low bone mineral density is the same as in the general population [26].

Despite the robust safety profile of TCSs, corticophobia - that is excessive worry, fear and reluctance about the use of TCSs - is increasing among caregivers. It is estimated that corticophobia affects up to 60-73% of patients or parents of children with AD [27] and that it represents one of the most important causes of non-adherence to therapy, partly due to insufficient explanations by healthcare professionals [28]. Given the magnitude of the problem, it is important to promote the practice of a standardized assessment of corticophobia and strengthen the therapeutic education strategies of the patient and families. A tool that can be used in the evaluation of corticophobia is the Topical Corticosteroid Phobia (TOPICOP) score, conceived by Aubert-Wastiaux in 2013 [29] which consists of 12 items organized in the form of a questionnaire aimed at evaluating two different areas: fears and beliefs about TCSs. A standardized evaluation of corticophobia, ideally to be performed at the first visit of a new patient with AD, can be an important tool to improve adherence to future therapy with TCSs [28].

In the years 2000-2001, two TCIs, tacrolimus (tTAC) and pimecrolimus (tPIM), were approved for the treatment of AD in patients ≥ 2 years of age. TCIs are non-steroidal anti-inflammatory agents that selectively inhibit the phosphatase activity of calcineurin, which leads to reduced transcription of various pro-inflammatory cytokines involved in the development and maintenance of inflammation in AD, such as IL-2, IL-3, IL-4, IL-5, INF-γ, and granulocyte-macrophage colony-stimulating factor (GM-CSF) (Table 3) [30‐32].

TCIs have also been shown to down-regulate the expression of the high-affinity receptor for immunoglobulin E class (FcεRI) on Langerhans cells, suppress the T cell proliferation induced by S. aureus toxins and the expression of IL-33 mRNA, triggered by enterotoxin B, and correct the altered balance of Toll-like receptors [31, 33]. Some data suggest that tTAC can also directly act on sensory nerves and inhibit their activation by inducing sensory desensitization and suppression of the release of substance P [34]. A 10-year follow-up study reports that tTAC significantly reduces the body surface area affected by AD and serum IgE levels, and in those individuals with active asthma and rhinitis, tTAC can also reduce respiratory symptoms and bronchial hyper-reactivity [35].

Compared to tTAC, tPIM shows lower skin penetration, higher affinity to epithelial structures, lower affinity to lymphoid structures and lower immunosuppressive effects [36]. Regarding pharmacokinetics, the cutaneous absorption of TCIs is minimal due to their large molecular size, respectively 822 Dalton (Da) for tTAC and 810 Da for tPIM. The transepidermal penetration of TCIs is 70-100 times lower than that of topical corticosteroids (TCSs), with tPIM in cream showing approximately five times lower transepidermal flow than tTAC in ointment [37].Conversely, CSTs have a molecular weight <500 Da and are therefore more absorbed even by healthy skin [36, 37]. The maximum absorption of TCIs is observed in the initial stages of treatment, when the inflammation is greater, while, when the inflammation is reduced and the skin barrier begins to improve, their penetration further decreases [36, 37]. There is conflicting evidence regarding TCIs effects on the skin barrier, with recent studies indicating the superiority of tTAC over CSTs in restoring skin barrier integrity [38, 39].

The role of Malassezia spp has been considered in the pathogenesis of a clinical variant of AD that presents in adolescence or adulthood with eczematous lesions of the head, neck (head and neck dermatitis) and upper trunk associated with intense pruritus and resistance to therapy with TCSs and/or TCIs [93‐96].

Malassezia spp has been hypothesized to be the cause of this particular form of dermatitis. Malassezia furfur is a family of yeasts, mainly lipophilic, which normally colonize human skin from puberty, especially in the head and neck area and in the intertriginous folds. The close contact between Malassezia spp and the cutaneous immune system induces both a humoral and cell-mediated immune response. Specific IgE and positivity of the prick tests against specific antigens of Malassezia spp and sometimes also of the specific patch test have been demonstrated [93‐96].

Treatment with itraconazole at (200 mg/day) was found to be effective. The duration of treatment is variable, from one week to a month, and is sometimes continued with 1 dose per week for several weeks. TCSs can be associated locally. Alternatively, fluconazole can be used. Cyclopyroxolamine is also indicated as a topical therapy in the treatment of "head and neck" AD [95‐97].

Candida albicans species, frequently found in both healthy and damaged skin, may play a role in exacerbating skin lesions [98, 99].

The treatment consists of the application of topical azoles (clotrimazole 1% or miconazole 2% twice a day for 1–2 weeks); in refractory cases, a systemic azole such as single dose fluconazole is used.

The role of various dermatophytes (e.g. Trichophyton Epidermophyton and Specie Microsporum) in the pathology of AD is not yet fully understood but may be suspected in patients resistant to standard therapies (e.g., tinea incognita, variant of modified skin dermatophyte infection. incorrect use of a topical or systemic steroid). Topical antifungal drugs e.g. azoles, allylamines, butenafine, ciclopirox, and tolnaftate once or twice a day for 1–3 weeks are indicated for therapy.

In recalcitrant AD with dermatophyte infection, systemic treatment with azoles (200 mg/day of itraconazole) is used [100].

In recent years, the improved knowledge of the complex immunological mechanisms underlying AD, together with the progress of the pharmaceutical sector, is profoundly changing the therapeutic approach and has allowed the development of molecules capable of interfering with various intracellular pathways. In this regard, phosphodiesterase-4 (PDE4i) inhibitors and Janus kinase (JAKi) inhibitors are gaining increasing interest [101].

Crisaborole is the only topical PDE4 inhibitor to be approved in the United States by the FDA for the topical treatment of mild and moderate AD from three months of age. On March 27, 2020, the European Medicine Agency (EMA) authorized its marketing also in Europe, in patients with AD from two years of age. Crisaborole is a non-steroidal molecule capable of selectively inhibiting PDE4, resulting in an increase in intracellular levels of cyclic adenosine monophosphate (AMP) and preventing the release of inflammatory mediators, such as interleukins IL-2, IL-4 and IL- 5, tumor necrosis factor-alpha (TNF-α) and IFNγ, thus improving the protective function of the skin barrier [101, 102]. The demonstration of the efficacy of crisaborole is based on some important randomized trials, including the one conducted double-blind by Paller et al. [103] on 1,522 patients aged between 2 and 79 years, suffering from mild-moderate AD. Significant and more rapid improvement in pruritus and clinical signs of lesions was documented in the 960 patients treated with crisaborole who completed the study, compared with 438 who used the vehicle alone. Among the most common side effects were reactions at the application site, such as tingling and burning. However, the high cost of this drug currently limits its use to patients who cannot use TCSs and TCIs.

JAKi is a promising new class of topical drugs currently under study. Janus kinases (JAKs) are a family of intracellular tyrosine kinases that transduce signals mediated by cytokines and growth factors [104‐106]. The results obtained from the phase I and II studies are extremely encouraging, as most patients quickly reach the primary outcome of the study, with a high safety profile. In these studies, there is also a rapid improvement in itching, which patients report easing from the first day of application of this drug.

Delgocitinib 0.5% (Corectim ®) ointment is the only JAKi that received approval in Japan in January 2020 for the treatment of AD in adults. It is a non-selective inhibitor of JAK, able to inhibit JAK1, JAK2, JAK3 and TYK2, which has been shown to improve the alterations at the level of the skin barrier, to favor the terminal differentiation of keratinocytes, to reduce and suppress itching induced by IL-13 [107]. In mouse models of AD, its topical application has been associated with an increase in NMF levels (natural moisturizing factors) and an improvement in skin inflammation and alterations in the skin barrier [107].

An important role in the pathogenesis of AD is represented by the increase in transepidermal water loss from the altered skin barrier. Wet-wrap therapy (WWT) is an ancient remedy that is considered an effective and safe second-line treatment, to be used in severe or refractory forms of AD, in patients older than 6 months of life [45, 108, 109]. This therapy must always be started and followed by experienced personnel.

There are two main methods for WWT. In the first method, after a short bath of 5-15 minutes with warm water, the skin is dried by dabbing without rubbing, then topical therapy is applied directly to the skin. Immediately afterward, the skin is bandaged with a double layer of gauze or tubular, of which the first layer (the internal one) is moistened with warm water, while the second (the external one) remains dry. Whenever possible, the humidification of the first layer is renewed every 2-3 hours during the day but not at night, with a steamer and warm water, after removing the dry layer of gauze or tubular. In the second method, a skin bandage is performed with two layers of gauze, the first moistened with diluted steroid or water directly on the injured skin and the second dry layer. In both methods, the use of an external state results in decreasing evaporation of water from the inner to the outer layer, thus resulting in cooling and prolongation of the moisturizing effect.

The TCSs most suitable for WWT are fluticasone propionate, methylprednisolone aceponate, mometasone furoate, hydrocortisone acetate and prednicarbate. TCSs are to be combined with an emollient with a hydrophilic base at a dilution of 10% (1 part of steroid and 9 parts of emollient) or 5% when treating the face [110‐112].

Latex-free and washable viscose elastic bandages can also be used, which can be applied for 3-24 hours, although daytime bandages are always preferred despite the difficulty in being accepted by children. The duration of treatment varies from 2 to 14 days and generally, the best results are obtained during the first week. Recently, a study was published using a 100% nanopolyester fabric [113]. This less expensive, more acceptable, and the longer-lasting fabric has given excellent results.

WWT is an effective treatment thanks to its anti-inflammatory and cooling action [45, 108, 109]. Side effects vary according to the patient's age, the TCS used, the occlusion time, and the total duration of treatment [109] The most frequent adverse effects are related to poor acceptance and possible chills during the application of the wet layer. Bacterial and/or viral over-infections are also described. The most fearful side effect is the transient increase in cortisol levels due to systemic absorption of the TCSs used. Therefore, the WWT requires specialized personnel and can only be carried out at the patients' homes if they receive adequate and specific training.

Clothes, or rather fabrics in direct contact with the skin, can play an important role in patients with AD. It is widely demonstrated that some fabrics can be irritating factors (wool) or cause allergic contact dermatitis (colored fibers). On the contrary, some fabrics can have a protective role, by constituting a defensive barrier against exogenous irritants and favoring the formation of the physiological skin microbiome through antibacterial activities.

The ideal tissue for the skin of the child with AD should reduce transepidermal water loss, promote hydration, limit inflammation and itching. A fabric, therefore, made up of smooth fibers with a small diameter (the diameter of the fibers seems to be correlated with the ability to evoke itching) [114] non-occlusive and with antimicrobial capabilities that persist after several washing cycles. Occlusive fabrics such as polyester or nylon temporarily promote the trapping of water in the skin but when removed, they cause significant evaporation and transepidermal water loss.

The fibers used first for better control of AD are cotton and silk. Subsequently, natural cellulose-based fibers, such as Lyocell, were built. Recently, fibers with associated silver, quaternary ammonium, or other antibacterials have been developed to obtain an antibacterial action, in particular against S. aureus. The latter products are those that have been particularly studied in recent years. Table 6 lists the various fibers currently on the market.

Two clinical trials, one open-label [115] and one double-blind controlled [116] have documented the efficacy of silver-coated cotton fabric in reducing SCORAD and S. aureus colonies in patients with moderate AD. The use of a seaweed-based fabric with associated silver (Sea Cell Active fibers®, Smart Fiber AG, Thuringia, Germany) resulted in an improvement in transepidermal water loss [117] and a reduction in the number of S. aureus colonies in patients with mild-moderate AD [118]. A randomized controlled study with a silver-associated cellulose-based tissue (SkinDoctor® Ventex Co., Ltd., Korea) in patients with mild to moderate AD also described an improvement in SCORAD, a reduction in transepidermal water loss and a reduction of S. aureus colonies, compared to cotton tissue [119]. A double-blind randomized controlled study with a cellulose-based fabric with an associated silver (Skintoskin® New Textiles, Ltd, London) observed an improvement in SCORAD, sleep quality and itching as early as 7 days, with a lasting effect for 90 days, compared to cotton fabrics [120]. The possible percutaneous absorption of silver, which is greater on damaged skin, should be considered. However, a recent study showed no increase in silver in urine in the group who wore silver clothing for at least 8 hours a day for 5 days [121] Chitosan-based fabrics (a biopolymer derived from chitin) also show antibacterial activity. A double-blind controlled trial in adolescents and adults who wore cotton pajamas with chitosan, showed an improvement in SCORAD with a reduction in the use of topical therapies compared to the group who wore only cotton [122].

Several but similar randomized controlled trials have compared fabrics enriched with quaternary ammonium (MICROAIR DermaSilk ® AlPreTec Srl, San Donà di Piave, Italy) vs untreated cotton or silk fabrics, in pediatric patients with mild to moderate AD [123‐127]. These studies have shown encouraging results, reporting improved clinical AD scores and reduced exacerbations associated with the use of such fabrics compared to the control tissue.

A randomized controlled trial, conducted on 300 children with moderate-severe AD, compared the clinical efficacy of two silk garments made from antimicrobially protected knitted sericin-free silk, Dermasilk TM and DreamSkin TM (DreamSkin Health Ltd, Hatfield, UK), with standard therapy, and did not show significant clinical improvement after 6 months [128]. However, it should be noted that in this study, most of the children presented a severe clinical picture.

Recently, a study evaluated the persistence of silver and quaternary ammonium in fabrics after several washing cycles [129]. It should be noted that the fabrics in which silver is added do not behave in the same way. Only Padycare® and Binamed® show persistence of silver even after 150 cycles. Instead, less relevant results are found with other fabrics added with silver. Even less persistent is the quaternary ammonium, which is no longer detected after 30 washing cycles. While there is a rationale in the possible use of fabrics and/or garments that have structural characteristics to counteract the inflammation of AD, there is currently insufficient robust data to suggest their routine use, as claimed in a recent review of evidence-based literature [130]. New controlled studies will be able to provide more precise indications in this regard.

Oral corticosteroids act rapidly on acute lesions of AD by reducing inflammation. However, despite their widespread use, few randomized controlled clinical trials have been performed to confirm their superiority compared to other drugs [131, 132]. Besides, they have a limited role in itching and pain [133].

Despite their effectiveness, the risk of systemic side effects strongly limits their use. When used in the long term, they can lead to suppression of the hypothalamic-pituitary-adrenal axis, immunosuppression, hypertension, weight gain, osteoporosis, and stature growth retardation in children [131, 134]. Upon discontinuation, a significant rebound of acute AD has frequently been described [131, 132]. Therefore, the use of oral corticosteroids should only be recommended in exceptional cases, for a limited time (e.g. in case of major exacerbations in patients with severe AD), and with close monitoring for possible side effects, especially in pediatric patients [45, 132]. Oral corticosteroids could be used for a 1-2 week cycle according to the following scheme: methylprednisolone at a dose of 0.5 mg/kg/day for 1-2 weeks, tapering the drug in about a month [45, 135].

Cyclosporin A (CsA) is an oral immunosuppressive agent belonging to the calcineurin inhibitor family, used in the prevention of organ transplant rejection and to treat various inflammatory skin diseases, including psoriasis and AD [136].

CsA acquires activity upon binding with cytoplasmic proteins known as cyclophilins and this complex competitively binds and inhibits calcineurin. This reduces the transcription of genes encoding IL-2, TNF ‐ α, IL-3, IL-4, INF-γ, GM-CSF, and leads to the reduction of lymphocytes T [137]. In patients with AD, CsA can reduce eosinophil counts, E-selectin, and soluble CD30 levels and correct the Th1 / Th2 imbalance [138]. The dosage range is between 2.5-5 mg/kg/day.

Based on a systematic review of 34 randomized clinical trials [139] and a meta-analysis and review of 15 randomized trials [140], CsA is currently recommended as a conventional therapeutic option in the short-term treatment of moderate-to-severe AD in adults. The majority of patients experience a significant reduction in disease activity within 2-6 weeks of starting therapy and its estimated efficacy, expressed as a reduction in AD severity scores, is between 53 and 95% [139].

In the first clinical trial on the use of CsA in pediatric AD, 27 children aged between 2 and 16 years of age were treated with CsA for 6 weeks at a dose of 5 mg/kg/day [141, 142]. In this open study with short-term observation, a significant improvement was observed in all measures of disease activity in the majority of treated patients. Treatment was well tolerated and there were no significant changes in serum creatinine and/or blood pressure. However, after discontinuation of the drug, the majority of patients experienced a relapse of AD within a few weeks.

In a subsequent prospective, open, parallel-group, multicenter study, 40 children aged 2 to 16 years with severe AD unresponsive to TCSs therapy, were randomized to receive either CsA in a continuous long-term mode (up to 12 months) or an intermittent short-term schedule (multiple cycles of 12 weeks) [143]. The starting and maximum dose of CsA for all patients in the 2 groups was 5 mg/kg/day. CsA was effective with both therapeutic regimens. The continuous therapy scheme showed a more consistent improvement, although the short-term regimen proved also to induce a prolonged remission of AD in some patients, with a reduction in the cumulative exposure to CsA. Therefore, the possibility of using tailored dosages on an individual basis was suggested [143].

CsA has also been compared to methotrexate in a randomized clinical trial involving 40 children aged 8 to 14 years with severe and TCSs refractory AD and poor compliance or poor response to phototherapy [144]. The drugs were administered in low doses, 2.5 mg/kg/day for CsA and 7.5 mg/week for methotrexate, respectively. Both groups of patients showed a statistically comparable reduction in SCORAD. CsA showed greater rapidity of action (2–3 weeks), but also greater relapse of AD (mean 14 weeks) after discontinuation. A randomized, controlled trial on the efficacy and safety of methotrexate versus cyclosporine in severe AD in children is ongoing [145].

Currently, the use of CsA is off-label in children and adolescents, while it is approved by the European Medicines Agency (EMA) for severe AD in adults [146]. However, its use is widely accepted in children with moderate-to-severe AD [147]. In such cases, most patients exhibit a rapid and good to excellent improvement in AD, with a very low incidence of serious side effects [148]. A 2013 survey conducted by the European Treatment of Severe Atopic Eczema in Children Taskforce (TREAT) indicated CsA as the first-line systemic therapy for severe AD in the pediatric age [149].

CsA is usually administered with intermittent treatment regimens lasting up to 12 months. A break of 2 weeks before and again 4-6 weeks after vaccination is suggested, although there is no evidence to support such management [146]. Potential adverse events include infections, nephrotoxicity, hypertension, tremor, hypertrichosis, headache, gingival hyperplasia, and increased risk of skin cancer and/or lymphoma. Evaluation of side effects should include physical examination, blood pressure measurement, regular monitoring of complete blood counts, blood urea nitrogen, creatinine, electrolytes, magnesium, uric acid, bilirubin, cholesterol, triglycerides and urinalysis. Caution is advised in patients taking other medications due to the many possible drug interactions [15, 150].

European guidelines suggest that CsA should be administered in patients with SCORAD index> 50 and/or with persistent AD [15]. A very recent systematic review has shown that CsA and dupilumab are equally effective, compared to placebo, in inducing remission of clinical signs of AD and are more effective than methotrexate and azathioprine in the short term [150].

Methotrexate (MTX) is an antimetabolite that inhibits the synthesis of folate, thereby blocking the synthesis of purines, DNA and RNA. Its therapeutic action is probably linked to the inhibition of the function of T lymphocytes [151]. In recent years, numerous studies have shown the efficacy of MTX in adults with AD in the absence of serious adverse events [134, 152]. A randomized study comparing MTX vs azathioprine showed similar effects in severe AD [153]. In a randomized, controlled pediatric study comparing MTX vs CsA, the MTX-treated group achieved a similar reduction in SCORAD to that seen in the CsA-treated group [144]. A retrospective study of children undergoing MTX treatment for severe AD demonstrated that MTX is an effective and safe drug with long-term lasting therapeutic effects [154]. A randomized and controlled study on the efficacy, safety and cost/benefit ratio of MTX vs CsA in the treatment of severe AD in children is currently underway [145].

In a retrospective study of 55 pediatric patients, the clinical severity score of AD significantly improved in most patients, demonstrating that MTX is an important option for long-term control in severe AD with a favorable adverse effect profile and costs [155].

However, in a recent systematic review and meta-analysis on adult patients, CsA and dupilumab showed greater short-term efficacy than MTX and azathioprine in patients with severe AD [150].

Safety data for MTX are derived from studies of other diseases treated with MTX. Stomatitis, nausea and vomiting are reversible adverse effects with dose reduction and/or discontinuation of treatment. Liver toxicity and teratogenicity, in both sexes, are the major side effects. The blood count and liver enzymes must therefore be monitored during treatment. Screening for hepatitis B and C and chest x-rays should be performed before starting treatment with MTX. Folic acid supplementation is always recommended during treatment with MTX to reduce hematological and gastrointestinal toxicity.

MTX appears to have a slow response (8-12 weeks) but is maintained over the long term. The recommended therapeutic dose in children is between 0.2 and 0.5 mg/kg/week (maximum 25 mg/week) and the duration of treatment is 10–16 weeks [45]. The drug can be increased by 2.5–5 mg/week and decreased by 2.5 mg/week, up to the lowest effective dose [15]. In conclusion, the off-label use of MTX can be considered as a second-line systemic treatment option, in particular when prolonged treatment is needed.

Azathioprine is a purine analog that has immunosuppressive activity by inhibiting DNA production, therefore affecting cells with high proliferation rates, such as T and B lymphocytes during the inflammatory disease phases [139, 156‐158].

Azathioprine can be considered a second-choice therapy for moderate-severe AD in adults when CsA has not shown efficacy or has given side effects or is contraindicated [139, 156]. Azathioprine has lower efficacy than CsA, but comparable to that of methotrexate. The recommended dosage is 2-3 mg/kg/day and it generally takes at least 1-2 months to manifest the beneficial effects [139, 156]. The low thiopurine-methyltransferase activity has been associated with increased myelotoxicity, but at-risk patients can be identified before starting azathioprine therapy by testing the activity of this enzyme. If this test is not available, one can start with a low dose of azathioprine and monitor blood tests (i.e. complete blood count, liver, and kidney function) after 7-15 days to gradually increase the dose. It is advisable to continue to monitor blood tests regularly throughout azathioprine therapy [139]. Because of its slowness of action, azathioprine can initially be used in association with TCSs. The most reported side effects associated with the use of azathioprine are gastrointestinal symptoms, myalgia, fever, skin reactions, headache, lymphopenia, and neutropenia.

In conclusion, azathioprine is a drug rarely used in the treatment of AD in adults and even less in the pediatric age, in which azathioprine is to be considered an off-label therapy [139, 156‐158].

Mycophenolate mofetil is rarely used in the treatment of AD. It is an immunosuppressive drug that acts by inhibiting the enzyme inosine monophosphate dehydrogenase, thus blocking the synthesis of purines, and reducing the proliferation of T lymphocytes [159‐162].

Mycophenolate mofetil is proposed as a third-choice treatment in patients with severe AD in whom cyclosporine has not shown efficacy, has produced side effects and/or is contraindicated [159‐162].

It can be prescribed at an initial dosage of 10-40 mg/kg/day, increasing the dosage by 500 mg every 2-4 weeks until the effective dose of 20-50 mg/kg/day is reached. Mycophenolate mofetil appears to offer a good efficacy and safety profile in the treatment of adult patients with severe AD and the most common adverse effects are gastrointestinal disturbances (nausea, diarrhea, and abdominal pain). It can be teratogenic in both sexes [159‐162].

Oral antihistamines have always been included in guidelines for the treatment of AD, although their usefulness is much debated [163, 164]. Itching is the most common feature of AD and the one that most affects patients' quality of life. Itching triggers scratching which perpetuates skin inflammation through the release of various mediators, such as thymic stromal lymphopoietin (TSLP), thus feeding the "itch-scratching" cycle [134‐165]. Through a competitive mechanism on H1 receptors, oral antihistamines act by blocking the effects produced by histamine (vasodilation, itching). Most guidelines recognize a possible role of first-generation sedating H1 antihistamines in the treatment, for short periods, of sleep disturbances associated with AD pruritus, due to their sedative effect [134, 166‐168]. Some authors recognize a rationale in the use of oral antihistamines only in patients with concomitant AD and allergic rhino-conjunctivitis [134, 149, 169].

In 2015, the EMA has issued a warning on the safety of the first-generation H1 antihistamines under 2 years, in particular for hydroxyzine, due to the possible undesirable effects on the electrical activity of the heart (low but defined risk of QT prolongation and torsades de pointes) [170]

In 2018, a Cochrane review assessed 25 clinical trials, including 8 studies conducted in pediatric populations, to investigate the efficacy of oral second-generation H1 antihistamine as an adjunct to topical therapy in AD [171]. While reporting that the evidence in this regard was qualitatively limited, this review did not show a consistent efficacy of the “add-on” therapy with second-generation H1 antihistamine on AD. No clinical efficacy emerged for cetirizine and loratadine compared to placebo, although it confirmed their safety in use [172].

The most common side effects of antihistamine therapy are excessive and unwanted sedation (even for non-sedating antihistamines), and anticholinergic symptoms (e.g. dry mouth, blurred vision, tachycardia [165]. Furthermore, since first-generation antihistamines with sedative properties can influence daytime wakefulness, particular attention is required in the dosage and administration schedule, especially in school-age children due to possible negative interference on school performance, and in adolescents who are allowed to drive motorcycles and automobiles [172‐174].

Lastly, it should be noted that the use of topical antihistamines is not recommended for the risk of absorption and systemic toxicity (e.g. reported with diphenhydramine) [175] and for the possibility of promoting the onset of allergic contact dermatitis [15, 176].

The possible use of probiotics in the prevention and treatment of AD has recently been the subject of numerous studies. Recent data show that patients with AD have a dysbiosis of the intestinal microbiota, with a reduction in Bifidobacterium spp and an increase in Staphylococcus spp, Escherichia coli and Clostridium difficile, compared to healthy individuals [177]. This dysbiosis is associated with a reduction in the activity of regulatory T lymphocytes, which favors the increase in intestinal permeability and loss of immunological tolerance [178, 179]. These observations support the hypothesis that a specific composition of the intestinal microbiota may modify the imbalance towards Th2 responses observed in AD, thus favoring an immune-regulatory response [180, 181].

In recent decades, several clinical trials have examined the possible efficacy of probiotics in treating AD, often with conflicting results [180, 182]. A recent Cochrane review of 39 randomized controlled trials analyzed data from 2,599 individuals with AD who took probiotics of the Lactobacillus or Bifidobacterium species, alone or in combination with other probiotics, with or without supplementation with prebiotics. Different doses and concentrations were used in the trials examined and the duration of active therapy ranged from 6 weeks to 3 months, thus resulting in an extreme heterogeneity of results. Overall, the comparison between patients treated with probiotics versus untreated ones revealed little or no efficacy of the therapy on skin symptoms, as reported by the patient and/or parents, and on the severity of the skin condition, documented by the investigators. Furthermore, there was also no evidence of benefit in terms of quality of life. Therefore, the authors conclude that, according to the available evidence, the use of probiotics for the treatment of AD is not an evidence-based approach [183]. Further studies are needed to clarify and define which bacterial strains are effective, the doses, the schedule and the duration of therapy.

Finally, in recent years, the in vitro demonstration that some strains of probiotics can exert an anti-inflammatory, antimicrobial and barrier activity has led to the hypothesis of possible topical use of both probiotics and their derivatives (e.g. bacterial lysates) in the treatment of AD [184‐186]. Recently, the use of a topical emollient containing lysates of the Gram-negative bacterium Vitreoscilla filiformis grown in thermal water (LRP-TSW), has been shown to improve SCORAD and normalize the microbial flora with reduction of S. aureus in patients with AD [187, 188], Lactobacillus reuteri is another highly studied probiotic strain, which has shown a good tolerability and safety profile and could therefore be a promising topical therapy [189].

The definition of phototherapy comes from the Greek "light therapy" and refers to a technique that uses the beneficial effect of light waves in various pathological conditions. The technique can be performed using different artificial light sources, which include broadband UVA (315-400 nm) and oral psoralens (PUVA), UVA and topical psoralens (bath-PUVA), UVA-1 or long UVA (340–400 nm), broad‐spectrum UVB (280–315 nm) and narrowband (NB‐) UVB (311–313 nm) [45].

Phototherapy is indicated as a second-choice therapy for moderate and/or relapsing forms of AD not responsive to TCSs and/or TCIs [45]. This treatment, used in both adults and children (typically> 12 years of age), induces improvement in skin lesions, pruritus, and sleep (Strength of recommendation B and level II of evidence [45, 190‐193].

PUVA phototherapy is rarely used, due to the important side effects such as cataracts, nausea, headache, itching, skin hyperpigmentation and, in the long term, also the increased carcinogenic risk [45, 190‐192193].

UVA1 phototherapy is effective in the acute phase of AD because it favors T-lymphocyte apoptosis, lower expression of pro-inflammatory cytokines, including IL-5, IL-13, IL-31, and reduction of dendritic cells. Currently, medium doses (65 J/cm2) are preferred to high doses (130 J/cm2), which produce excessive heat and intense sweating. The use of the so-called "UVA1 cold light", which uses lamps that filter infrared rays, has shown a better tolerance than traditional UVA-1 or UVA/UVB. The therapeutic scheme of UVA1 phototherapy includes 3-5 sessions per week for 3-8 weeks with a maximum dose of 80 J/cm2. The duration of the single session can vary from 10 to 60 minutes. Although effective, UVA1 therapy is only available in a few centers, and its use is limited by possible long-term side effects, such as photo-damage and carcinogenic risk. Therefore, UVA1 phototherapy is currently only recommended for adult AD and its use in pediatric age is rare [45].

NB-UVB or UVBTL01 phototherapy is indicated for chronic moderate forms of AD due to the limited penetration of the UVB used [45, 194]. The treatment has a good anti-inflammatory activity, reduces the colonization of S. aureus, promotes the improvement of the skin barrier function, and the thickening of the stratum corneum, with a consequently increased resistance to environmental irritants [9, 194].The starting dose is calculated from the minimum erythematogenic dose or according to the phototype according to Fitzpatrick. The number of weekly sessions is variable as well as the duration of the treatment, as it can be used in short cycles until remission or in longer cycles as maintenance therapy. In general, the scheme includes 6-12 weeks of treatment, with 2-3 sessions/week with subsequent reduction to 1-2 sessions/week once the remission of the skin lesions is achieved. A 50% improvement in SCORAD with 3 sessions per week of NB-UVB for 12 weeks has been reported and the result may persist for up to 6 months beyond the end of treatment [9, 194].

In the pediatric age, NB-UVB phototherapy is considered the best therapeutic option for efficacy, good tolerability, low risk, and availability. Sometimes NB-UVB phototherapy can cause excessive heat and increase sweating with the possible flare-up of the AD. Air-conditioned NB-UVB regimens improve tolerability and have been shown to be more effective. The most common acute side effects of NB-UVB phototherapy are erythema, sunburn, xerosis. NB-UVB phototherapy can be used in monotherapy or combination with emollients and TCSs. Due to the possible carcinogenic risk, it is recommended to avoid combination therapy with TCIs [45]. In adults with psoriasis, NB-UVB phototherapy was performed in addition to biologics and this combination therapy resulted in better and faster responses than monotherapy [195].

The safety of long-term NB-UVB treatment has been well documented in psoriasis studies, however, short courses are recommended in children [154, 190]. The main limitations of NB-UVB phototherapy are the only hospital availability and the poor response of some sites, such as the scalp, folds, and eyelids which must be covered with protective goggles. Age-specific factors, such as fear of devices, poor compliance of the pediatric patient, and concerns of caregivers, should not be underestimated in the treatment decision. Therefore, it is necessary to be cautious and not to use phototherapy in prepubertal age. Subjects with light skin, phototype I and II, require additional attention as they tend to easily burn [154, 190]. It should be noted that the European Task Force on AD does not recommend the use of phototherapy in children under the age of 12 due to potential long-term side effects, although it is not contraindicated in children and properly selected cases [45].

Home phototherapy is little used in AD and the data available refer to the treatment of psoriasis. Cameron et al. [196] reported observations from a 13-year follow-up to a UK home service with NB-UVB and concluded that home phototherapy is just as effective as hospital phototherapy. However, home phototherapy requires motivated patients who adhere to the instructions and the supervision of an expert (dermatologist). Home phototherapy can be considered an option for patients in whom phototherapy cannot be used in conventional settings [197].

The main immunological mechanism in AD is an excessive type 2 inflammatory response [8, 9]. Eczematous skin lesions have a complex and varied inflammatory pattern but are dominantly characterized by the expression of Th2 CD4 + and innate type 2 lymphoid cells (ILC2) [45, 64]. The release of alarmins (DAMPs), determined primarily by the alteration of the skin barrier, activates the resident dendritic cells to promote a type 2 response with the release of IL-4 and IL-13 and other Type 2 cytokines, as chemokines related to thymic activation (TARC), TSLP and IL-33. IL-4 and IL-13 can block the production of some proteins that contribute to the integrity of the skin barrier. These two cytokines contribute to the differentiation of keratinocytes and down-regulate the production of filaggrin, loricrin and involucrin, adhesion molecules, antimicrobial peptides (beta-difensins and cathelicin LL-37) and ceramides, amplifying the skin barrier damage [198]. IL-4, IL-13, TSLP, and IL-33 contribute to pruritus, the predominant symptom of AD. The discovery of the IL-4Rα rceptor for IL-4 on afferent neurons and the control of pruritus itself through inhibition of the IL-4Rα receptor and JAK, has further reinforced the clinical relevance of the interaction between Type 2 immune response and neuronal pathways of pruritus [199].

Dupilumab is a human IgG4 monoclonal antibody, with molecular weight 147 Kda, directed against the alpha subunit of the IL-4 receptor (IL4Rα), an essential subunit for inducing IL-4 and IL-13 signaling. The biological function of IL-4 and IL-13 is mediated by the binding of two receptor subtypes (IL4R), which share IL-4Rα. The type I receptor, IL-4R, composed of the heterodimer IL-4Rα / gc, binds exclusively to IL-4 while the type II, composed of IL-4Rα / IL-13Rα1, binds both IL-4 and IL- 13 [199, 200].

The first studies on the clinical use of dupilumab in AD were performed in adults. Promising results have been reported in phase III, randomized, double-blind registration studies [201, 202] that evaluated 2119 patients aged 18 years or older with moderate to severe AD not adequately controlled by topical therapies (SOLO 1: 671 patients; SOLO 2: 708 patients; CHRONOS 740 patients). Based on the results observed in these studies, dupilumab was approved by the FDA in March 2017 and by the EMA in September 2017 for the treatment of adult patients with moderate to severe AD who are candidates for systemic therapy (Table 7).

In 2020, a Phase 3 study was published in 251 adolescents (12-17 years of age with moderate-severe AD, not adequately controlled by topical therapies) which also demonstrated the efficacy and safety of dupilumab in this age group [203]. The treatment period was 16 weeks. Three groups were considered: a first group (85 patients) received placebo; a second group (82 patients) received subcutaneous dupilumab every 2 weeks with a weight-dependent dose, i.e. 200 mg if baseline weight was <60 kg [loading dose of 400 mg (43 patients)] and 300 mg if baseline weight ≥ 60 kg [loading dose of 600 mg(39 patients)]; the third group (84 patients) received dupilumab 300 mg every 4 weeks (600 mg loading dose) [203]. TCSs, TCIs, and crisaborole could be used as rescue therapy. Dupilumab showed rapid and significant efficacy in reducing signs and symptoms of AD, including pruritus, and in improving quality of life, even in patients who only achieved incomplete lesion remission. In particular, at week 16, the mean EASI improvement from baseline was approximately 66% compared to 24% achieved by placebo. In addition, 42% of patients on dupilumab achieved a ≥ 75% improvement in the EASI score (EASI-75), compared with 8% in the placebo group. Finally, 24% of patients on dupilumab achieved complete or near-complete remission compared to 2% on placebo, as measured by an IGA score of 0 or 1. Rescue therapy was primarily required in placebo-treated patients. The main side effects were seen in adolescents, i.e. allergic and infectious conjunctivitis and keratoconjunctivitis (10.8 and 9.8% respectively in treated every 4 or 2 weeks of cases versus 4.7% in placebo) and injection site reactions (6 and 8.5% versus 3.5% of placebo), were similar to those found in adults. Conversely, skin infections, observed in 20% of placebo patients, were found only in 13.3 and 11% of treated patients [203]. In conclusion, the efficacy and safety data of dupilumab in adolescents were consistent with those observed in adults (Table 8).

Dupilumab was therefore approved on March 11, 2019 by the FDA and on August 6, 2019 by the EMA for the treatment of adolescents with moderate to severe AD inadequately controlled with topical therapies or when such therapies are not advisable.

The “Redefine Atopic Dermatitis in Adolescent: an italian report” (RADAR) is a consensus project conducted on adolescents with AD [204]. The results of this project confirmed the high levels of efficacy and tolerability of dupilumab in this patient group, underlining the role of this biologic as a first-line systemic agent in the treatment of moderate to severe AD.

Even more recently, a randomized, double-blind, placebo-controlled phase 3 study (LIBERTY AD PEDS) was conducted in children aged 6 to 11 years with AD diagnosed more than 1 year prior to screening [205]. In total, 367 patients were randomized to be treated for 16 weeks either with dupilumab every two weeks (100 mg for weight 15-30 kg and 200 mg for weight> 30 kg), or placebo or dupilumab every four weeks (300 mg independently weight). Concomitant use of medium strength TCSs once daily was allowed in all groups. The primary endpoint was the achievement of IGA 0-1 at week 16 and the co-primary endpoint in the EU reference countries was EASI-75. Secondary endpoints were the percent changes in EASI and pruritus NRS scale from baseline to week 16. Both groups of patients treated with dupilumab + TCSs showed statistically significant improvement in signs and symptoms of AD, and in quality of life compared to patients treated with placebo + TCSs. Specifically, at week 16, the results showed total or near-total clearance of skin lesions (IGA 0 or 1) in 33 and 30% of patients treated with dupilumab every four weeks (300 mg) and every two weeks ( 100 mg or 200 mg) respectively, compared with 11% of patients treated with placebo (p <0.0001 and p = 0.0004, respectively). 70% of patients treated with dupilumab every four weeks and 67% of patients treated with dupilumab every two weeks achieved EASI-75, versus only 27% of patients treated with placebo (p <0.0001 in both cases). In addition, the biological drug showed significant relief from itching and improved anxiety and/or depression reported by both patients and their parents, and quality of life. The main side effects found in the study were: both allergic and infectious conjunctivitis and atopic keratoconjunctivitis (6.7 and 14.8% respectively in treated every 4 or 2 weeks of cases versus 4.2% in placebo), keratitis (observed in one patient in the dupilumab group every two weeks 100 mg or 200 mg) and injection site reactions (10 and 10.7% dupilumab group versus 5.8% for placebo). Conversely, skin infections, observed in 13.3% of patients on placebo, were found only in 5.8 and 8.2% of treated patients [205].

On May 26, 2020, the FDA approved dupilumab for the treatment of patients 6 years of age or older with moderate to severe AD whose disease is not adequately controlled with topical therapies or when such therapies are not advisable. Dupilumab can be used with or without TCSs. The EMA approved the use of dupilumab from 6 years of age on November 30, 2020, but only recently (January 29, 2022) the Italian Medicines Agency (AIFA) approved it for this age group (Table 7).

Table 9 reports the recommended posology of dupilumab for the treatment of pediatric AD [206].

Therapeutic efficacy is assessed after 16 weeks of treatment, which must be discontinued if there is no improvement. Live and live attenuated vaccines should not be administered concomitantly with dupilumab as safety and clinical efficacy have not been established. Immune responses to diphtheria-tetanus-acellular pertussis (DTPA) vaccine and meningococcal polysaccharide vaccine were evaluated. It is recommended that patients repeat immunizations with live and live attenuated vaccines, per current immunization guidelines, before treatment with dupilumab [206].

Recent advances in the knowledge of the underlying mechanisms of AD have contributed to the development of new systemic therapies, with high efficacy and safety profiles, which target specific molecules involved in the pathogenetic pathways of AD. Promising biological therapies with monoclonal antibodies targeting IL-31, IL-13, IL-22, JAK and other cytokines are under development. [207]. None of the drugs described below are currently prescribable for pediatric AD.

A percentage of patients with AD may have an increase in total serum IgE and be sensitized to aeroallergens such as house dust mites (HDM) [223].

The efficacy of allergen-specific immunotherapy (AIT) is now well documented both in allergy to Hymenoptera venom and in allergic rhinitis with or without asthma. However, the role of AIT in the therapy of AD is still debated [223].

Some clinical trials have evaluated the clinical efficacy of AIT, administered subcutaneously (SCIT) and sublingually (SLIT), in adults and children with AD. These studies, conducted on small populations, have shown contradictory results regarding actual efficacy [224‐227]. More recently, Liu et al. [228] evaluated 239 patients treated with HDM SLIT for 36 weeks demonstrating clinical improvement in subjects with moderate to severe AD. Galli et al. [229] conducted the first controlled study of 60 children with AD and found no significant benefit after 3 years of HDM SLIT. In 2007 Pajno et al. [230] treated 56 patients between 5 and 16 years of age in a double-blind for 18 months, obtaining significant results in patients with mild to moderate AD. Subsequent studies have confirmed the safety and efficacy of SLIT, although conducted on selected populations and with different AIT administration schedules. In recent years, several systematic reviews and meta-analysis studies on both routes of administration, SCIT and SLIT, have come to controversial conclusions and no official recommendations could be made [228].

In 2017, the guidelines of the American Academy of Dermatology [231] confirmed that the available data did not allow to support the use of AIT in AD. However, as in 2018, a Joint Task Force of the European Academy of Dermatology [8] suggested, instead, to consider AIT in selected patients with moderate-severe AD, allergic sensitization to HDM and clinical exacerbation after the exposure to the causative allergen.

It should be noted that to date there are no studies to support the use of AIT in individuals with AD and allergic sensitization to allergens other than HDM, such as dogs and cats. Currently, in the literature, there is only one observational uncontrolled study on a population of 19 adults with AD and sensitization to cat and/or dog dander, in which a marked improvement in skin symptoms, a reduction in specific IgE, and an increase in IgG4 after 2-5 years of SCIT are documented [232]. AIT could be a potential treatment option in patients with severe AD, associated with other allergic comorbidities, and allergy to dogs and cats, if exposure to the allergen cannot be avoided, although further studies are needed to confirm this therapeutic indication. Also, the identification of specific biomarkers capable of predicting the actual clinical efficacy of AIT in different patients is highly needed, as these could represent a decisive element for an adequate selection of the subjects to be treated.

Although the literature data suggest the possible beneficial role of thermal treatments in dermatological diseases, this is nevertheless still debated due to the difficulty of standardizing adequate methods of investigation [233].

Bicarbonate-calcium-magnesium waters are more suitable for the treatment of AD. Retrospective studies have shown beneficial effects of climate treatment and Dead Sea salts in patients with AD, in the absence of major side effects [234]. An open-label, randomized controlled study evaluated the clinical efficacy of the combination of balneotherapy in a 10% Dead Sea salt solution and NBUVB phototherapy compared to phototherapy alone, finding an improvement in AD in patients undergoing balneotherapy [235].

Several studies have reported clinical and quality of life improvement in AD patients after 3 weeks of treatment at the Avène spa (Avène-les-Bains, France) [236‐239]. Another study, which involved AD patients treated with 3 weeks of balneotherapy at the La Roche-Posay spa in France, showed a significant improvement in AD in terms of EASI score, quality of life, itching and xerosis [240].

An open-label, randomized controlled study on the efficacy and safety of thermal balneotherapy in children was conducted at the Terme di Comano (Comano Terme, Trento, Italy) on 104 children (1-14 years) with mild-moderate AD. Patients were alternatively assigned to balneotherapy or TCSs for 2 weeks. Four months after the end of treatment, a significant difference in the number and duration of exacerbations was found in favor of the balneotherapy group, with a significant reduction in TCS consumption and days of acute AD [241].

More recently, studies have focused on the microbiological properties of thermal waters and the anti-inflammatory and immunomodulating effects of the new microbial entities discovered in them [242‐247]. A new microorganism, called Aquaphilus dolomiae (Neisseriaceae), was found in the thermal water of Avène. The Aquaphilus dolomiae extract seems to counteract the effect of the cutaneous S. aureus secretome isolated from atopic children on the proliferation of CD4 + T cells [243]. The regulatory activity of the ES0 extract of Aquaphilus dolomiae on inflammation associated with itching, on alteration of the skin barrier and the innate and adaptive immune response was also described in vitro [244]. These effects suggest the potential role of this extract inserted in topical preparations for the treatment of AD. The thermal water of La Roche-Posay also contains specific non-pathogenic minerals and microbes that can affect the human microbiota. Its microbial composition was characterized by 16S next-generation sequencing metagenomic techniques. The main characteristics are a high bacterial diversity, a low bacterial concentration, and a predominance of Gram-negative bacteria. These bacteria could affect the skin microbiota and therefore the water itself could be considered a probiotic [245]. The effect of thermal balneotherapy on the skin microbiota was evaluated in 31 patients with AD, both on lesional and non-lesional skin sites, at the beginning and end of treatment. Biodiversity of the skin microbiota (measured by Shannon index) improved after 21 days of thermal balneotherapy. Furthermore, a reduction of Firmicutes in particular of Staphilococci spp with a significant increase of Gram-negative bacteria was found [246]. The Microbial Genomics Laboratory of the CIBIO Department of the University of Trento (Trento, Italy) has carried out the first characterization study of the groundwater microbiome (Antica Fonte di Comano Terme) with high-resolution sequencing technology (NGS shotgun sequencing). The study found a stable microbiome over time, made up of more than 250 species of which about half had never been described before. The genomes of these bacteria contain thousands of genes and their functions are yet to be characterized. Furthermore, more than 100 bacterial strains have been isolated in pure culture and characterized for various bioactivities [247]. The anti-inflammatory activity found in various strains has proved to be of particular interest for potential future applications in the dermo-cosmetic sector to support the treatment of skin diseases. Although further studies are needed to further clarify these initial data, one bacterial strain with potential immunomodulatory properties has been described as a new bacterial species, the Mesorhizobium comanense [247].

In conclusion, emerging evidence seems to support the hypothesis that the therapeutic effects of different thermal waters are due to the concomitance of physical and chemical effects, in conjunction with the immunological and microbiological properties of the waters [247]. Further evidence is needed to explain the role of non-pathogenic bacteria present in thermal waters in promoting skin microbiota diversity and any beneficial effects in AD patients.

The management of skin manifestations and pruritus in children with AD, especially the early-onset phenotype, would require multidisciplinary support for the patient's family, covering bio-pharmacological, educational and psychological/psychotherapeutic aspects. Therapeutic education, as defined by the WHO, also includes a personalized plan to be built in collaboration with the patient and his family.

The anxiety generated by the complex therapeutic management of skin lesions and pruritus and the emotional and economic burden of AD therapies can compromise the quality of life of the entire family and affect compliance with treatment.

An ideal model of integrated management should include the coordinated intervention of a multidisciplinary team composed of the medical specialist (Pediatrician, Allergist, Dermatologist), the psychologist/psychotherapist and the Nurse, and be oriented towards improving the disease and the quality of life of the children and their families [248].

The educational component plays an equally fundamental role within the therapeutic plan [249] (Ministero della Salute-Direzione Generale della Programmazione Sanitaria “Piano Nazionale della Cronicità”. Febbraio 2016)

The role of education is important for any degree of severity of AD since the impact on the quality of life of this disease can be relevant even in non-severe forms [250, 251]. Any educational intervention must be preceded by an unhurried listening to the convictions, difficulties, expectations and prospects of the patient and the family. Indeed, therapeutic education puts the patient and caregivers at the center of the educational process and, in an ideal model, puts in place an integrated and multidisciplinary intervention. The educational approach must be modulated especially towards patients and families with special needs (poverty, social hardship, language difficulties).

Although in many clinical settings such integration of professional figures is a utopian model, this model can still be used as a guide for the specialists, albeit in a simplified form [248]. In general, a synergy must be established at least between the general practitioner/pediatrician and the specialist pediatrician, who recognizes the following as fundamental in the therapeutic approach:

The 2018 European Guidelines on the management of AD [7] emphasize that poor adherence to therapies is the most important factor limiting treatment outcomes [252]. Poor adherence to therapeutic indications may be due to various factors, including stress [253] and intra-family psychological and psychodynamic influences, known triggers capable of directly influencing the clinical course of AD and favoring severe exacerbations of eczematous skin lesions [254]. A 2014 Cochrane review assessed ten randomized controlled clinical trials of psychological or educational interventions, in addition to conventional therapy, for AD in children, and reported that educational interventions (such as multi professional eczema interventions and nurse-led clinics) may lead to improvements in AD severity and quality of life [255]. Based on such conclusions, the 2018 European Guidelines concluded that “patient education programs are recommended as an adjunct to conventional therapy of AE” (grade of recommendation 1a, A) [15]. The Guidelines report the following grades of evidence regarding patient educational programs:

The topic of patient educational programs was also addressed in the Position Paper of 2020 [38] which still recommends participation in a therapeutic education program for AD for all patients with moderate and severe form, underlining how poor adherence to the prescribed treatment is the most frequent condition leading to a therapeutic failure [256]. There is therefore a significant and urgent need for physicians to ensure that their patients are educated and confident in the use of prescribed drugs to gain control of the disease and of itching. The educational program, in group or individual, should be started as early as possible and should consider both the physical and emotional aspects of the patient and family, providing practical guidance for the home management of AD [257]. To improve the quality of life, it is also crucial to teach distraction techniques to control itching, which has a strong negative impact on patients and the entire family [258]. The choice of distraction techniques, best known for pain management, depends on several factors: age, personality, patient preferences, motivation, and emotions of the reference figures.

Because AD ​​is a complex disease, both from pathogenetic and clinical standpoints, management is also affected by this complexity. The presence of different clinical phenotypes of AD implies the need for an individualized and multidisciplinary approach in which the interaction between the primary care pediatrician, the pediatric allergist, and the pediatric dermatologist is crucial to finding the best management strategy. In the manuscript, the latest shreds of evidence on pharmacological and non-pharmacological therapies for AD are reviewed. This narrative review aims to define a pathway to appropriately managing children and adolescents with moderate-severe atopic dermatitis (AD). A flow chart for the possible multidisciplinary management of pediatric moderate-severe AD in the Italian Health Care System is also reported (Table 10).