Current Understanding in Pathogenesis of Atopic Dermatitis.

There have been advances in our understanding of the complex pathogenesis of atopic eczema over the past few decades. This article examines the multiple factors which are implicated in this process.

What was known?

Previously eczema was thought to be primarily caused by immune dysfunction in susceptible children. Our understanding of the complexities of how barrier function plays a key role has changed over the past decades.

Introduction

Atopic dermatitis (AD) or eczema is a chronic inflammatory itchy disease which starts early in life. Our understanding of the complex pathogenesis of AD has advanced considerably over the past few decades. The emerging picture is a complex disorder with genetics, barrier function, immunity, and environmental factors all playing key roles. Many of these pathological mechanisms interact and can be seen to work synergistically to maintain and progress AD. Despite an increasing knowledge of the various components of pathogenesis, there are still many aspects of this complex process that remains poorly understood. This article will summarize the current understanding of the pathogenesis of atopic eczema and attempts to look at the relevance to key clinical questions.

Why do Children Get Eczema?

This is the question that patients or parents of children with eczema or AD ask when they come to clinic. The answer is not simple, and there are many aspects to pathogenesis and why a single child will develop AD. Our current understanding is that many factors are inherited and therefore cannot easily be avoided. Genetics explain part of the risk, but the increase of AD over the past few decades is faster than any possible change in gene pool and shows that there must be a role for exogenous factors. These are likely to be multiple and also largely unavoidable. How these multiple factors interrelate and influence the expression of key genes is critical to the processes that cause AD.

Parents should be reassured that they are not responsible for their child's eczema as it is unlikely that they would have been able to change the factors that led to the development of AD. Furthermore, it should be stressed that while AD, once established, is likely to be chronic, although there is no “cure”, there are effective treatments that can control and possibly reduce further atopic disease.[1] It is also to be hoped that increasing understanding of the pathogenesis of AD will lead to further improvements in prevention and management strategies.

Genetics of Eczema

There has long been known to be a genetic component to AD. The monozygotic twin concordance rate is much higher than for dizygotic twins. Heritability studies combined with family-based linkage studies supported heritability of AD as a complex trait with interactions between genes and environmental factors. It is likely that the interplay between multiple genes contributes to disease manifestation. There are more than 100 published reports on genetic association studies; these have identified over 80 possible genes with a demonstrated association with AD. Most candidate gene studies to date have focused on adaptive and innate immune response genes; however, over the past few decades, there has been increasing interest in the important role of skin barrier dysfunction genes.[2]

There are several genetic syndromes with known single-gene defects that have atopic eczema as a feature. Examples include ichthyosis vulgaris (filaggrin; FLG), Netherton syndrome (SPINK-5), SAM severe dermatitis, allergies, metabolic wasting (Desmoglein 1), Wiskott–Aldrich syndrome (WASp), Omenn syndrome (SCID RAG1/2), hyper IgE (Jobs syndrome - STAT3, DOCK8), and IPEX (Foxp3). These show increased risk from genetic abnormalities in both structural proteins in the skin and in components in the immune system. This would fit with our understanding of the complex pathological processes seen in AD.[2]

Although there are still a significant amount of children with no known genetic defect; there are now known abnormalities in genes important in skin barrier function and immune function components. The last few decades have really advanced our understanding of the genetic basis of eczema and in particular, our growing understanding of mutations in genes encoding an important epidermal protein, FLG.[3]

Barrier Function in Pathogenesis of Atopic Dermatitis

Barrier function has long been known to be reduced in the skin of patients with AD. The studies 50 years ago showed increased transepidermal water loss in patients with eczema.[4] Whether this is a primary or secondary phenomenon has previously been of some debate. With the discovery of the importance of FLG, a protein important in the structure of the epidermis, we now know that in many cases of AD, this is likely to be a primary inherited problem.[3] Secondary phenomena, for example inflammation and trauma from scratching (itch remains an area poorly understood in eczema), will then lead to further barrier dysfunction. Furthermore, inflammation and cellular responses in AD have been shown to downregulate the expression of FLG and likely other key proteins, which will then even further disrupt the ability of the skin to effect barrier dysfunction.[56]

The Filaggrin Story

Of all the multitude of target genes, the gene encoding FLG has been most consistently replicated as a risk factor for AD. FLG mutations (heterozygous) are common to varying degrees and with different mutations worldwide. They are present in around 9% of the UK population. FLG is important for the formation of the corneocyte, as well as the generation of its intracellular metabolites, which contribute to stratum corneum hydration and pH. A well-functioning epidermis protects humans from exogenous stressors and helps to maintain internal fluid and electrolyte homeostasis. Loss-of-function mutations in the FLG gene result in either a reduction (heterozygous) or complete absence (homozygous) of epidermal FLG and its degradation products, which lead to dysfunction of the epidermis and reduced barrier function. Null mutations in FLG cause the condition ichthyosis vulgaris which is the most common inherited disorder of keratinization. Ichthyosis vulgaris is an autosomal semi-dominant condition and disease severity can vary considerably even within affected families as there are incomplete penetrance and variable expressivity.[7] Groundbreaking work by groups in Dublin and Dundee showed that FLG mutations are more prevalent in children with atopic eczema; being present in 25%–50% of those with AD.[8] In particular, the risk seen with FLG mutations is for early onset and severe disease,[9] allergen sensitization, development of IgE-mediated food allergies,[10] and progression to asthma and allergic rhinitis.[11] Clinically, FLG null mutations can be recognized by the presence of hyperlinearity on the palms in particular over the thenar eminence.

Work since this discovery has significantly advanced and changed our understanding of eczema pathogenesis. Before this, most focus had been on abnormalities and dysregulation in the immune system. The finding of the importance of FLG and epidermal integrity has shifted focus to the structure of the skin and barrier function. Rather than being a primary immune problem, it is clear that AD in association with FLG mutations is due to a primary barrier defect with resulting secondary inflammation.

As mentioned, FLG null alleles are a significant risk factor not only for AD but also for all aspects of atopy (AD, allergic rhinitis, asthma, and food allergies). This supports the role of skin barrier dysfunction as a key driver of allergic disease and AD as the first point in this inflammatory cascade. This leads to a model of disease where initial barrier problems lead to exposure of skin to allergens which can then drive inflammation and a skewed immune response.[121314]

It is likely that FLG plays other roles besides barrier function which may be relevant in AD and other diseases. However, although it is clearly very important, it is still the case that 40% of people with FLG null mutations (ichythyosis vulgaris) and 60% of heterozygotes do not develop AD. It is therefore clearly not the only pathogenic mechanism. Overall, FLG mutations account for 50% of moderate-to-severe cases and up to 20% of mild-to-moderate. There is a suggested role for additional genetic factors and gene–gene interaction.[315] Increasing understanding of how genes are controlled and epigenetic factors shows FLG expression is dynamic, and even in the absence of somatic mutations, several factors affect the expression of FLG including TH2 cytokines, pH, and bacterial infections which have all been shown to downregulate expression.[1617]

Of additional interest is why there is high persistence of FLG mutation in the population. It is thought that FLG deficiency may confer some heterozygote advantage. One theory is that it may induce “natural vaccination” against microbial pathogens via the skin barrier, which could have driven natural selection due to survival protection in pandemics.[15]

Immune Response

There are abnormalities in both adaptive and innate immunity in children with AD. The altered immune responses and increased development of atopic disease seen in AD are at least partially due to barrier dysfunction and exposure to allergens. However, there are also primary immunological factors that also drive pathogenesis. In certain cases, an underlying genetic cause of immunodeficiency (e.g., Jobs syndrome) is the cause of dermatitis. The “atopic” trait that mounts an immune response (TH2) to common ubiquitous allergens is likely a complex trait with a genetic predisposition. In AD, total IgE is often elevated, and many children have specific IgE elevated to ubiquitous environmental allergens, for example, house dust mite, grass, cat, dog, pollen.[1318]

Barrier Dysfunction and Immune Response - The “Atopic March”

The “atopic march” describes the tendency for atopic eczema to precede the development of food allergies, asthma, and allergic rhinitis in a temporal sequence.[18]

As discussed, FLG mutations have been reported as a risk factor for each step in the atopic march showing the importance of barrier function in this process. However, even without FLG mutations, children with AD will have reduced barrier function due to inflamed skin and secondary factors such as itching. The role of barrier dysfunction in pathogenesis in AD and the development of further atopic disease is key. Mechanisms that sense skin barrier disruption direct local inflammation and these synergistically drive a strong TH2 inflammatory response.[16]

Whether genetic or acquired, once established, inflamed skin and defective barrier in AD permit penetration of allergens that further trigger immune response and can lead to IgE-mediated allergies. The fact that AD is the first presentation of atopy in many cases and likely plays a key role in the development and progression of the atopic march clearly makes it an important condition to treat; by targeting AD, we can potentially modify this process.

Allergies: Do these Cause Eczema?

Many patients will be adamant that an allergy is “causing” their child's eczema and therefore that allergy testing and subsequent avoidance may offer a cure. The role of allergy and eczema is complex. The studies over the past decade suggest that the role of allergy in AD might have been overemphasized and although allergic sensitization can be seen in children with AD, this probably develops as a consequence of a primary barrier defect and inflammatory eczema.[1920]

Knowledge of FLG has increased our understanding of how a structural abnormality in the skin can increase the risk of atopic march and allowed for a working model of pathogenesis to be developed. This is illustrated in the case of peanut allergy in the UK. Peanut proteins are highly immunogenic and if IgE-mediated allergy develops, it tends to be persistent. Advice until recently was to avoid eating peanuts in high-risk children. Over the past few decades, there has been increased peanut allergy in atopic children in the UK. However, peanut allergy was found to be rare in Israel where peanuts are weaning food.[21] Research has shown an increased risk of peanut sensitization and allergy in children who carry an FLG mutation.[10] This suggests that barrier dysfunction confers this increased risk. The association between AD and peanut allergy remained significant even after adjustment for FLG status. This suggests that inflammatory eczematous skin acts as a mediator of food sensitization regardless of inherited abnormalities in barrier function. This is confirmed with work that shows AD severity is seen to be an independent risk factor for peanut allergy. In breastfed infants, the risk of allergen sensitization relates to the severity of their eczema.[22] Contrary to the previous advice in the UK, it has now been shown that oral avoidance of peanuts increases risk of developing IgE-mediated allergy. This is likely due to the risk of cutaneous sensitization with early-life environmental peanut exposure. A recent trial has confirmed that in fact, early oral exposure provides immune tolerance and reduces the risk of peanut allergy.[23]

This and other evidence, including longitudinal studies, reveal allergen sensitization is probably a consequence, not a cause, of AD. When managing patients with eczema, this enforces the importance of the focus being on controlling eczema to prevent allergy rather than allergy testing to prevent AD.

Exogenous Factors: Environment

The prevalence of AD has increased faster than any changes in gene pool would explain. Inherited factors cannot explain this increase, and there must be exogenous reasons for this. There is much interest in possible trigger factors, in particular ones that are potentially modifiable. There are epidemiological studies attempting to identify reasons for differences in prevalence but, to date, no definitive causation has been identified. In some cases, specific risk factors have been suggested and include house dust mites, exposure to allergens, infections, breastfeeding, use of antibiotics, and irritants. There are conflicting results for some of these factors (e.g., breast feeding) and difficulty with dealing with confounding factors. This complex work supports our knowledge of AD as a complex process and that therefore the etiology is unlikely to be simple or unidirectional.

The studies of environmental exposures are notoriously hard to study due to multiple confounding factors and problems with analyse design. However, there are some environmental factors, which do have likely mechanisms and evidence to support a role in AD pathogenesis. There is growing evidence for exposure to certain factors causing an initial switch toward AD early in life. Once established, this may be difficult to reverse increasing the need to identify key trigger points which are modifiable. Some of the best evidence is coming from intervention studies, for example, the emollient study in atopic risk infants.

Hygiene Hypothesis

The question of why AD has increased to some extent in line with development and “sanitization” of society was first raised in 1989 with the “hygiene hypothesis.”[24] This used epidemiological factors – such as larger family size and attendance at nursery with a reduced risk of AD and postulated that this was due to exposure to infections. The “hygiene hypothesis” may go some way to explain the rapid rise in prevalence, and the observation that AD seems to be more common in wealthier, smaller, and more educated families.[25] Lack of stimulation of the immune system by microbes at an early age because children are “too clean” or failure of a balanced gut flora to mature may well be relevant factors in the pathogenesis of AD.[26]

Immunological models suggest this lack of exposure to certain infections may drive a “skewed” immune TH2 system. There is certainly some evidence that parasitic or helminth infections play a role in this process.[272829] There is ongoing evidence which supports some of these theories, for example, a recent study showing parents who clean pacifiers by sucking rather than wiping (therefore exposing them to human pathogens) can reduce the risk of allergy development.[30]

The hygiene hypothesis may be helpful to explain some features of AD, but the area is complex and it seems likely that the type of infection and the timing of exposure are critical to how this relates to the pathogenesis of AD. There are likely some infections which may offer some protection against the development of atopic disease, but certain infections seem to increase risk, for example, measles.[26]

Exogenous Factors: Irritants

Irritant exposure has been shown to increase AD. The use of soap as a cleanser acts as an irritant, which has been shown to cause skin dryness and increased transepidermal water loss. The alkaline nature of soap also increases skin pH. The studies in the 1990s showed that increased soap use was associated with increased AD in some households.[31] Recently, it has been shown that removal of such irritants or avoidance from birth can reduce AD.

The pH of water exposed to the skin and whether this confers a risk of AD has also been an area of scrutiny in terms of water “hardness.” The studies have shown that children born in a hard water area in the UK have an increased risk of developing AD compared with children born in a soft water area.[3233] However, a recent study in Spain has not supported these findings and questions the validity of the previous studies.[34] The effect may be through bystander factors including both pH and soap use. As discussed, soap use increases skin pH and additionally hard water requires increased soap use to lather on skin. This demonstrates how many risk factors can become interrelated. Increasing pH of the skin can further increase the risk of AD as this regulates FLG expression down. However, once AD is established, buying a water softener does not change the severity of disease.[35] This supports the model that, once established, AD may be difficult to reverse. Early factors likely drive the initial switch to AD and interventions should be put in place as early as possible from birth to prevent AD and the subsequent atopic march.

An exciting recent study that tracks a cohort from birth has confirmed a role of soap avoidance. Emollient use from birth is shown to reduce the risk of development of AD and prevented eczema in children with a strong family history of atopy. This is a really important step toward being able to modify and prevent AD in high-risk children.[36] Future work, which would increase understanding of the relationships between the disease and factors such as behavior, lifestyle, and home and external environments, is crucial.[3337]

Exogenous Factors: Cutaneous Infections

Certain infective organisms play an important role in triggering and aggravating AD. Patients with AD show an altered skin microbiome. They have a reduced ability to fight certain cutaneous infections. These infections can then stimulate inflammation and trigger further disease. Children who have AD are particularly prone to skin infections with Staphylococcus aureus. This is likely to be due to multiple factors including breaks in dry, split skin from scratching, and from diminished barrier function. The colonization of lesional and nonlesional skin in AD with S. aureus represents an important trigger factor for severity and frequency of skin symptoms.[38] Bacteria that cause problematic infections in AD are also commonly found on healthy skin. However, the environment on healthy skin protects against invasive infection due to increased oils as bacteria are lipophobic. Staphylococcal and streptococcal bacteria thrive and invade dry, atopic skin. Antimicrobial peptides on the skin surface normally fight these bacteria in healthy skin. Bacteria (staphylococci and streptococci) and yeasts (Malassezia) on the skin provide stimulation to the immune system resulting in ongoing chronic inflammation. There are synergistic processes to drive AD, for example, S. aureus produces enterotoxin which has been shown to induce the production of enterotoxin-specific IgE resulting in proliferation and recruitment of more T-cells and aggravation of dermatitis.[39]

A particular pattern of eczema on the face and neck, known as seborrhoeic eczema, is related to skin colonization by yeast called Malassezia. In patients with this condition, prick tests and Malassezia-specific IgE may be positive and targeted anti-yeast treatment can improve AD outcomes. Malassezia sensitization is believed to be a phenomenon specific to atopic eczema, which reflects a tendency to Th2 immune response.[40]

Other infections are likely to be secondary problems, reflecting active disease rather than involved in the primary pathogenesis. Cutaneous viral infections such as warts from HPV virus and molluscum contagiosum are more numerous and can be resistant in AD. Candida (thrush) is also more likely to thrive in poorly controlled, moist areas of eczema. Eczema herpeticum can be a serious problem in AD and has been shown to relate to the severity of AD and more recently seen to be more common in those patients with FLG mutations demonstrating the role of multiple factors in infection and pathogenesis.[41] Dermatophyte infections (tinea) however do not appear to be more prevalent in AD.

Antibiotics may be required to eliminate infections and can help control eczema or treat infective flares; however, there are growing problems with resistance. Importantly, active inflammatory eczema can be misdiagnosed as “impetiginized or infected” by nonspecialists. This is relevant to the WHO target to reduce inappropriate use of antibiotics. Inflammatory eczema should be treated with appropriate anti-inflammatory medications (e.g., topical corticosteroids). Use of antiseptics (“bleach baths”) has been found to be a safer and very effective way of reducing infection rates, and subsequent inflammation in AD.[42]

Why does Eczema Usually Get Better?

This is an important question, which may offer key insights into pathogenesis and treatment strategies. Up to 60% of children with AD are seen to “clear” by age 10.[43] The reasons for the improvement seen in some children are not understood and are likely to be complex; current theories include age-specific changes to immunity or skin barrier. Unfortunately, when looked at more closely, the idea of resolution is attractive but may not actually be true. Although an improvement to skin may be perceived, and for example, flexural eczema may clear up, many of these people will have other atopic problems ongoing through life, for example, rhinitis or asthma. They are more likely to develop hand dermatitis in adult life. Increasingly, AD is thought of more accurately as a “life-long” illness.[43]

What is the Current Model of Atopic Eczema?

Current evidence shows that children with AD have abnormalities in barrier, innate immunity, acquired immunity, and microbial flora. These can be genetic or acquired and are interrelated and synergistic. Other factors further drive disease, for example, itching and stress. The model of AD that is now emerging portrays AD as a complex disease with a genetic predisposition and secondary complex environmental factors, which exist in a viscous cycle of inflammation. The trigger factors likely occur soon after birth with alterations in skin barrier leading to early life alterations in skin colonization and impaired immune defense, uptake of allergens, irritants, and microbes which all trigger an inflammatory immune response which can further affect gene expression. The complexities of epigenetics (or how exogenous factors drive gene expression) may explain the processes of heritability and how this links with environmental exposures.

Can we Prevent Eczema and What does all this Mean for Treatment?

There is growing evidence that addressing known understanding of pathogenesis means some AD can be prevented with potentially modifiable factors that have been elicited over past few decades. Changing the environment of a genetically predisposed baby from birth may not only influence the development of AD but also the progression of some babies along the “atopic march.”[44]

One recent study has shown that strict soap avoidance from birth can dramatically reduce the risk of development of AD in at-risk children.[36] Probiotics for a mother in late pregnancy and during breastfeeding can reduce risk presumed to be due to improving the microbiome.[45] In the UK, we are no longer advising avoidance of allergenic foods orally as this has clearly added to the risk of developing food allergy and is likely to increase atopic march. Instead, the immune system should be exposed orally to allergenic foods earlier.[23]

Once AD is established, it is difficult to reverse the process. Recognition that there is ongoing inflammation in the skin of children with eczema which may be subclinical has resulted in “proactive” treatment using topical corticosteroids and calcineurin inhibitors which have been shown to be extremely effective for keeping control of eczema and flare prevention.[1] Other measures that focus on pathological mechanisms have also been shown to reduce disease activity, for example, staph reduction by regular antiseptics cleansing/bleach baths which not only reduce infections but seems to reduce inflammation in AD.[4246]

Finally, there is the tempting theory that early aggressive treatment of eczema may help prevent establishment of chronic disease and further atopic disease by preventing the cycle of barrier/immune interactions causing cutaneous inflammation.[47] Current studies are ongoing and based on our current understanding of pathogenesis and allergen exposure, this is an area of high potential future interest.

In conclusion, the pathogenesis of AD is a complex interplay between impaired barrier, infective agents, microbiome, epigenetics, and dysfunctional immune response. Our growing understanding of these factors is leading to better preventive and management strategies of eczema.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.

What is new?

Understanding the complexities of inflammation in atopic eczema, in particular the role of filaggrin in barrier function and how this plays a role in many processes in atopic eczema

How this understanding may start to improve our management of patients.