Cutaneous factors and environmental exposures in the development of FA
Early AD is implicated in the subsequent development of allergic diseases, including FA, asthma, allergic rhinitis and is termed the “atopic march”.2-4 In the “outside-in” hypothesis, skin barrier defect allows penetration of allergens and microbes leading to atopic sensitization whereas, in the “inside-out” paradigm, a polarized immune response leads to a defective skin barrier (Figure 4).25
Experimental models and clinical observations in humans support the concept of epicutaneous food allergen sensitization.20,26The epidermis plays a key role in preventing allergens, irritants and microbes from penetrating the skin and eliciting the host immune response. These events are facilitated by skin barrier dysfunction in AD, promoting the penetration of food allergens from topical application or the environment. Lack et al.23 first reported that peanut allergy was associated with the topical application of skin creams containing peanut protein. Subsequently, Fox et al.24 reported increased FA in households that ate peanuts. In addition, Brough et al.27 found house dust dose-dependent peanut sensitization in patients with FLG mutations, and the impact of developing allergy was greater in children with AD28 and in children with egg allergy.29These observations supported a role for the “outside-in” process of food sensitization where exposure to environmental peanut in an individual with skin barrier dysfunction leads to enhanced FA.
The “inside-out” process implicates the immune response in making the skin barrier more susceptible to skin epithelial dysfunction, development of AD, and allergen entry. The current understanding of ’AD’s pathogenesis is centered on the robust activation of Type 2 (IL-4, IL-13, IL-31) and Type 22 (IL-22) cytokine axes in both skin and serum.25,30-34Model systems showed that type 2 cytokine activation inhibits keratinocyte terminal differentiation products (i.e., filaggrin, loricrin), tight junctions (i.e., claudins), and lipid products.35-38 Recent findings show that Th2 cytokines decrease antimicrobial peptides, causing AD skin to be more prone to colonization of infectious organisms, such as S. aureus . Thus, IL-4 and IL-13 play a hallmark role in the Th2 immune response in AD, contributing to both immune activation and skin barrier dysfunction. IL-31, another Th2 cytokine, has been shown to interact synergistically with IL-4, driving pruritus and contributing to the inflammatory and barrier defects of AD.39-44 The Th22 axis also plays a role in suppressing the epidermal barrier and the lichenification and increase of S100As in chronic AD lesions.45,46Additional proinflammatory axes, including Th17, are preferentially upregulated in certain AD populations, such as Asians and children, revealing the heterogeneous nature of AD across its subtypes.47-51Recently, minimally invasive studies of the skin using tape strips, performed in infants, children and adults with moderate-to-severe AD, show robust upregulation of type 2 and 22 T-cell immune cytokines in both lesional and non-lesional AD skin.52-54 The upregulation of immune markers in involved and uninvolved skin showed high correlations with disease severity scores and the functional barrier measure trans-epidermal-water-loss (TEWL).55-57
Allergic disease development is associated with a Th2 cell-mediated inflammatory response58,59described above. Allergic disease is preceded by the formation of specific IgE (sIgE) antibodies against environmental and food allergens, also known as the sensitization phase. In epicutaneous sensitization, specific resident dendritic cell (DC) subsets residing in the skin60 sample antigens and present to naïve CD4+ T cells in draining lymph nodes This promotes differentiation into allergen-specific CD4+ T cells favouring B cell isotype class switching to sIgE cells further driving the production of IgE memory B cells61. Through the maturation and production of plasma cells, large amounts of sIgE antibodies are produced. The sensitization phase drives the production of a large memory pool of allergen-specific B cells and Th2 cells.
The sensitization phase is followed by the effector phase, which is triggered by subsequent exposure to previously sensitized allergens. This causes cross-linking of sIgE bound to receptor FcεRI on sensitized mast cells and basophils. Activation of these cells leads to the release of inflammatory mediators triggering an allergic reaction62. The immune mechanisms linking the skin and gut have their origins in skin injury-induced release of IL-33 from keratinocytes, leading to intestinal mast cell hyperplasia and food-induced anaphylaxis in mice.63 IL-33 blocking antibodies have also been shown to prevent peanut allergy induced anaphylaxis.64
Interestingly, skin sampling in patients with peanut allergy but not AD reveals low filaggrin levels but increased long-chained lipid species, which may protect the skin from dryness and AD.65 Other risk factors have been associated with peanut allergy, including filaggrin mutations, severe infantile AD, environmental irritant exposures such as detergents and S. aureus colonization on the skin.66-68
Skin dysbiosis, often observed among individuals with AD, is often characterized by reduced microbial diversity and the presence of one or few dominant microbes. The loss of commensal microbes is likely due to several factors including host genetics, local immune response, environmental factors such as pH, temperature, humidity, hygiene practice and exposure to antibiotics. It is estimated that 30% to 100% of individuals with AD are colonized by S. aureus , a dominant pathogen implicated in this disease (Figure 5a).69 S. aureusaffects the development of both innate and adaptive immune responses. It can lead to uncontrolled inflammation by inducing lymphocyte and macrophage activation. The increased presence of S. aureus in the dermis directly correlates with a Th2 response evident by increased expression of IL-4, IL-13, IL-31 and TSLP.70 These Th2 cytokines in turn suppress the production of antimicrobial peptides (AMPs) by the skin that inhibits S. aureusproliferation.71Therefore, it is not surprising that colonization by S. aureus  is associated with increased AD severity and treatment thereof has been shown to decrease disease severity.72,73
Malassezia spp., previously known as Pityrosporum, is a genus of lipophilic yeast. Its role in AD’s pathogenesis was initially speculated when some AD patients responded to topical and systemic antifungal therapies.74-78 A large population study showed more than 40% of children with seborrheic dermatitis during early childhood will develop AD later on, suggesting early sensitization of seborrheic skin may result in the onset of AD.79 Most of the Malassezia species lack fatty acid synthases genes, therefore relying on exogenous fatty acid sources that are abundant at certain cutaneous sites such as the head, neck and skin folds (Figure 5b).80 Although the pathogenesis of Malassezia spp in AD is not entirely clear, yeast is known to trigger a multitude of immune responses. It is estimated that 80% of adults with AD have detectable Malassezia IgE antibodies.81-83Malassezia spp. in the epidermis and dermis, can be recognized by keratinocytes and Langerhans cells as well as dermal DCs. These antigen presenting cells in turn activate downstream immunologic cascades that lead to the release of proinflammatory cytokines such as TNF-alpha, IL6, IL-8, IL-10, and IL-12p70. Induced expression of TLR2 and TLR4 on human keratinocytes and DCs upon exposure to Malassezia spp. have been observed, suggesting direct activation of innate immune response.84,85In addition, the NLRP3 inflammasome in skin DCs can also be activated byMalassezia spp with subsequent release of Th2 cytokines (e.g., IL-1beta, IL-4, 5, 13,) likely directly contributing to AD pathogenesis.86,87
Lamellar bilayer structural integrity is highly organized in normal skin, seen under electron microscopy, but very abnormal in those with AD and peanut allergy. The epidermis in AD with peanut allergy is associated with high TEWL, high type 2 immune activation, S. aureus colonization, reduced filaggrin breakdown products, and a reduced proportion of long-chained lipid products. These observations suggest that a defective skin barrier in patients with AD and peanut allergy may predispose affected individuals to epicutaneous allergen sensitization. The availability of minimally invasive skin tape sampling techniques may play an important role in identifying infants with early epidermal barrier dysfunction who may benefit from timely initiation of novel therapies for skin barrier dysfunction, non-lesional immune activation, and microbial dysbiosis. Using this technique epidermal profiling of lipids, proteins, and transcriptome identifies differences in the epidermis between patients with peanut allergy and AD versus AD alone.65,88
Barrier protection is the cornerstone of AD management. Skin hydration and prevention of TEWL are keys in maintaining skin barrier homeostasis. Animal studies also suggest that changes in hydration and corneocyte adhesion within stratum corneum affect the development and maturation of epidermis.89 Although there has been considerable controversy about whether early application of skin emollients can prevent AD and FA,20 these studies have often not targeted high-risk infants with pre-existing evidence for skin barrier dysfunction, or the ingredients of emollients has not been optimized for infant skin barrier repair. The use of topical steroids to prevent AD flares and control subclinical inflammation is being evaluated as a potential strategy to prevent FA in AD.20 Other novel pathogenesis-based topical and systemic therapies targeting inflammation of the skin have also been investigated for their roles in preventing FA.90
Petrolatum, a non-physiologic mineral lipid, is often considered a gold standard ointment-based emollient that can prevent TEWL effectively for 4-6 hours. Therefore, to maintain optimal skin hydration, ointment-based emollients should be applied 3 to 4 times daily to provide complete protection. However, ointment-based emollients can also exacerbate AD; therefore, alternatives must be considered. Lipids including ceramide, fatty acids and cholesterol are mixed in appropriate ratio within stratum corneum to maintain its integrity.91,92Atopic skin is known to be deficient in lipids especially ceramide and hygroscopic amino acids that are the result of filaggrin breakdown products.93,94Newer generations of emollients containing these lipids have been developed in recent years.95,96A recent study demonstrated a trilipid cream was more effective than a paraffin-based emollient in reducing TEWL and sIgE levels.20,97However, efficacy in AD or FA prevention is yet to be proven in a randomized clinical trial.98 While treating AD patients with a barrier-based approach, a liver X receptor agonist upregulated terminal differentiation and lipid products in the skin of patients with AD, consistent with its mechanism of action;99 however, it was not associated with clinical benefit or suppression of immune products (Th17/Th22/IL6). This suggests that although barrier-based approaches may be valuable for disease prevention, the immune abnormalities perpetuate the AD disease phenotypes and should be targeted to resolve active AD.
The discovery of cytokine dysregulation in non-lesional skin from AD patients suggest the role of systemic therapy especially for individuals with severe disease. The increased understanding of AD’s immune pathogenesis led to the development of immune-based treatments targeting Th2 cytokines.100-105Downregulation of immune markers in the skin of patients treated with such agents highly correlated with reductions in disease severity scores, demonstrating clinical improvement.33,106-111Furthermore, the Th2-targeting anti-IL-4R mAb dupilumab was shown to induce significant changes in the microbiome of skin lesions, again supporting the key role of the Th2 cytokines in inducing the disease pathogenesis.112