Understanding genetics to unravel psoriasis and atopic dermatitis pathogenesis

Recent advances in psoriasis and eczema genetics have not only led to innovative interventions and new drug pipelines, in some cases they also allow us to distinguish which patient responds better to which biologic treatment. Prof. Jonathan Barker (Kings College London, United Kingdom) discussed the direct clinical implications that psoriasis and atopic dermatitis (AD) genetics have realised [1]. Furthermore, the main genetic determinant for psoriasis, HLA-C*06:02, directs an autoimmune response against melanocytes through autoantigen presentation in psoriasis through ADAMTSL5, a melanocyte autoantigen [2].

Heritability is major factor of AD and psoriasis, as demonstrated in twin studies, where identical twins have a 25% concordance, whereas fraternal twins have only slightly higher concordance than the general population [1]. If a cohort has large numbers of adequately phenotyped patients, genome-wide association scanning (GWAS) can be an excellent technique to identify regions of genome that associate with disease. GWAS does not identify causative mutations, but it can provide evidence of association between a given locus and a disease manifestation, suggesting biological clues driving the disease of interest. For psoriasis, 16 loci have been identified by GWAS as being highly significant (see Table); 35 loci associate with AD.

Table: Overview of genetic determinants of psoriasis. Data from Barker (2019).



HLA-C*06:02 (formerly called Cw6) is the major genetic determinant for psoriasis, granting an individual carrying this allele a 5-fold higher risk of developing psoriasis. The normal variant of this gene is important in presenting antigen to CD8 and natural killer T cells, which already speaks to the pathophysiology of psoriasis. About half of all psoriasis patients are HLA-C*06:02 positive, as well as about 10-15% of the general population. The genetic effect of the HLA-C*06:02 allele is greater than all other loci identified to date combined, at least for specific subtypes of psoriasis. For example, HLA-C*06:02 is highly associated with early-onset and guttate psoriasis vulgaris, but not really with any other type such as late-onset psoriasis or psoriatic arthritis.

Filaggrin (encoded by the FLG gene) is the main genetic determinant of AD that came out of GWAS. The epidermal protein filaggrin is essential for regulation for epidermal homeostasis. Mutations lead to reductions in filaggrin expression. Mutations in FLG can also cause ichthyosis vulgaris. Depending on severity, 20-40% of patients with AD have FLG mutations. An individual with FLG mutation has a 3-fold higher risk of developing AD. However, >50% of individuals with mutation do not develop AD, in other words, FLG mutation is neither sufficient nor necessary to develop AD.

Neither HLA-C*06:02 nor filaggrin are amenable to therapeutic intervention. However, GWAS provides evidence of other disease mechanisms. If you look at all of the hits from the large GWAS studies, many of the loci suggest genes regulating signalling pathways already implicated in psoriasis, like skin barrier function, interleukin (IL)-17 and IL-23 signalling, NF-kB signalling, and antigen presentation signalling. In the GWAS data from AD cohorts, the IL-13 pathway stands out, which we now know is so important. In brief, pathways that are amenable for intervention become evident when you overlay the genetics with the known immunology of the disease. For example, the dramatic clinical effect that tweaking IL-23 signalling has in psoriasis or tweaking the IL-13 pathway has in AD. The GWAS data from psoriasis and AD has genuinely contributed to the rich biologics available today, which has considerably improved patient outcomes.

Another approach is to look at functional exonic variants, and to ask whether they associate with a particular phenotype. Exome array is a technique that one can use to detect low frequency and rare coding single nucleotide polymorphisms. When Prof. Barker and colleagues performed exonic arrays on over 11,000 psoriasis patients, only 2 genes strongly emerged. Multiple low frequency and rare protein coding variants in IFIH1 (Burden 0.05, 1.8x10^-19) and TYK2 (Burden 0.05, 1.4x10^-39) associate with psoriasis with exome-wide significance. IFIH1 encodes the MDA5 protein, which plays an important role in innate immunity. In particular, the MDA5 protein targets viral double-stranded RNA, which may be a strong clue that there is an infectious agent involved in the pathogenesis of psoriasis. The second hit was TYK2, which encodes a member of the tyrosine kinase/Janus kinases (JAK) protein families. This protein associates with the cytoplasmic domain of type I and type II cytokine receptors and promulgates cytokine signals including IL-23 and interferon by phosphorylating receptor subunits. It is also a component of both the type I and type III interferon signalling pathways. Accumulating evidence suggest that TYK2 plays a role in anti-viral immunity. A recent New England Journal of Medicine article shows impressive results in selectively inhibiting TYK2 in psoriasis patients with clear clinical benefit [3]. Prof. Barker stated: “This is a wonderful example of drug discovery being driven by genetics.”

Genetics can also answer the question of causality. Psoriasis and obesity are well-known co-morbidities, but what has remained difficult is defining whether one causes the other or not. Mendelian randomisation is a genetic method to determine a causal relationship between phenotypes that minimalise confounding variables. When applied to obesity vs psoriasis, a study by Budu-Aggrey et al. concluded that obesity contributes to the pathogenesis of psoriasis [4]. Their results suggested that higher BMI causally increases the odds of psoriasis (by 9% for each unit increase in BMI). The study also concluded that no evidence suggested a causal effect of psoriasis on BMI. This type of genetic evidence supports a holistic approach to working with psoriasis patients, in a multidisciplinary team like many hospitals are developing now.

One last example Prof. Barker provided about the value of genetics was the recent understanding that HLA-C*06:02 predicts response in psoriasis. Biologic-naive patients who were HLA-C*06:02 positive and did not have psoriatic arthritis exhibited a significantly poorer (3x) response to adalimumab at 12 months [5]. A patient who has HLA-C*06:02 should consider ustekinumab, but the opposite may not be true; there is no need to exclude patients with psoriasis from ustekinumab treatment because of a negative HLA-C*06:02 genotype status. There are ongoing prospective studies to see if this can be validated into a clinically useful, genetics-driven, and practice-changing algorithm.

Prof. Jörg Prinz (Ludwig Maximilians University of Munich, Germany) picked up where Prof. Barker left off [2]. The main psoriasis risk allele, HLA-C*06:02, he explained, confers susceptibility to psoriasis by promoting melanocyte-specific autoimmunity through autoantigen presentation. Together with the identification of ADAMTSL5 as a melanocyte autoantigen, these results now allow for redefining the cascade of pathogenic events in psoriasis, defining an HLA class I-restricted autoimmune response against melanocytes as the central pathogenetic event. Thus, HLA-C*06:02 confers an overall risk for psoriasis by facilitating an autoimmune response against melanocytes through autoantigen presentation, and gene variants related to innate immune activation and the IL-23/IL-17A axis may modify disease expression.

1. 
   
   1. Barker J. O006, SPIN 2019, 25-27 April, Paris, France.
   2. Prinz J. O019, SPIN 2019, 25-27 April, Paris, France.
   3. Papp K et al. N Engl J Med. 2018 Oct 4;379(14):1313-1321.
   4. Budu-Aggrey A et al. PLoS Med. 2019 Jan 31;16(1):e1002739.
   5. Dand N et al. J Allergy Clin Immunol. 2018 Dec 20. pii: S0091-6749(18)32780-5. Doi: 10.1016/j.jaci.2018.11.038.

 



Posted on 21 June, 201924 March, 2021