Targeted review of IL36RN mutations in patients with generalised pustular psoriasis

Abstract Background Generalised pustular psoriasis (GPP) is a rare and chronic skin disease historically treated with therapies that were originally intended to treat plaque psoriasis (PsO). However, GPP and plaque PsO have distinct pathogeneses and clinical courses. Objectives This study aimed to further characterise the unique genetic background of GPP by summarising evidence on the frequency and type of IL36RN gene mutation, a gene that normally suppresses proinflammatory responses, in patients with GPP compared to patients with GPP and plaque PsO, and patients with plaque PsO only. Methods and Results A targeted literature review was conducted to identify studies reporting IL36RN mutations and/or HLA‐Cw6 allele frequency in patients with GPP. Meta‐analyses showed a significantly higher rate of IL36RN mutations in the GPP‐only population compared to the GPP + plaque PsO population (OR 3.51; 95% CI 2.29, 5.38). Monoallelic mutations of IL36RN were found in up to 33.3%, and biallelic mutations in up to 73.2% of patients with GPP (GPP‐only and GPP + plaque PsO), in contrast with mono‐ and biallelic frequencies of only 0%–11.9% and 0%, respectively, in patients with plaque PsO only. Mean age‐of‐onset ranged from 5.9 to 48.9 years old, with most studies reporting a GPP age‐of‐onset between 20 and 40 years old. Twenty‐one mutations were identified in the biallelic state and three in monoallelic. The most reported mutations were c.115 + 6T > C (p. Arg10ArgfsX1) (18 studies); c.227 C > T (p.Pro76Leu) (10 studies); and c.338 C > T (p.Ser113Leu) (8 studies). Mutations varied depending on geography and ethnicity, with the most frequently reported mutation predominantly reported in East Asian studies and international studies that included Asian patients. Rates of HLA‐Cw6, the risk allele most strongly associated with plaque PsO, were 0%–28.6% for patients with GPP, similar to rates in the general population (10.5%–20%). Conclusion Considering the differences between GPP and plaque PsO in aetiology and disease symptoms, effective, GPP‐specific treatment options are needed, and recent research suggests that blockade of IL‐36 signalling may be an effective target for treatment of GPP.

2][3] Flares can result in potentially life-threatening complications, including sepsis or heart failure. 4,5The prevalence of GPP is not well characterised, but estimates range from 1.76 per million to 124 per million globally. 2,5One of the most common genetic mutations correlated with the development of GPP occurs in the IL36RN gene that encodes the interleukin-36 receptor antagonist (IL-36Ra) protein, which normally suppresses proinflammatory responses. 1,2,6The IL-36 pathway is regulated through the dynamic interaction between IL-36 agonists and antagonists, ensuring a harmonised immune response.Dysregulation, such as the overexpression of IL-36 or impaired functionality of IL-36Ra, leads to an aberrant inflammatory response. 7,8The hallmark pustules of GPP result from overactivation of the IL-36 pathway leading to the secretion of inflammatory mediators that promote skin infiltration of neutrophils and ultimately the development of incredibly painful, sterile, pus-filled pustules that may take weeks to subside. 2,3,5,6,9PP, International Classification of Diseases (ICD) code L40.1, can occur with or without concomitant plaque psoriasis (i.e., plaque PsO, ICD L40.0), the most common form of PsO.Historically, GPP has been diagnosed and treated similarly to plaque PsO despite its different presentation and clinical course.Recently, researchers have acknowledged the distinct pathogenesis of GPP and need for differential diagnoses and treatments. 2,91][12] HLA-Cw6 has been observed to affect age-of-onset, disease course, phenotype, severity, comorbidities, and treatment outcomes for PsO. 10 Immunomodulatory therapies, including biologics, are used mostly to treat GPP based mainly on clinical experience in patients with plaque PsO; however, evidence regarding their efficacy and safety for the treatment of GPP is limited. 4,5Recent research has demonstrated that blockade of IL-36 signalling is an effective target for treatment of GPP. 3,6In the pivotal Effisayil TM 1 trial, the first and largest placebocontrolled study of 53 patients with GPP, treatment with the IL-36-receptor inhibitor spesolimab was associated with complete pustule clearance (Generalised Pustular Psoriasis Physician Global Assessment [GPPGA] subscore of 0) after 1 week of treatment (19  of 35 patients [54%] in the spesolimab arm compared with 1 of 18 patients in the placebo arm [6%]). 13e objective of this study was to highlight differences in the genetic background of GPP compared to plaque PsO by identifying and summarising evidence on the frequency and type of IL36RN mutations and HLA-Cw6 allele prevalence in patients with GPP.To further characterise the unique genetic background of GPP, IL36RN mutation rates for patients with GPP only were compared to those with plaque PsO only and those with GPP and concomitant plaque PsO (GPP þ plaque PsO).By examining these genetic factors, we aimed to better understand the distinct pathogenic mechanisms and contributing factors that differentiate these two forms of PsO.

| METHODS
A targeted literature review was conducted on 8 April 2021 and updated on 28 September 2021 and 4 August 2022 to identify all studies published until that date reporting evidence on the frequency and type of IL36RN mutations and HLA-Cw6 allele prevalence in GPP patients.Records were accessed via Excerpta Medica database (Embase), Medical Literature Analysis and Retrieval System Online (Medline), Medline In-Process, and Cochrane Library (see Tables S1-S6 for search strategies).Conference searches included What's already known about this topic?� Generalised pustular psoriasis (GPP) and plaque psoriasis (PsO) have distinct clinical characteristics and pathogeneses, and evidence suggests that their aetiologies may also be different.While the presence of HLA-Cw6 alleles have been associated with plaque PsO, mutations in IL36RN and overactive IL36 signalling have both been observed in GPP.

What does this study add?
� This review and meta-analysis found that rates of IL36RN mutations were significantly higher in patients with GPP only versus patients with both GPP and plaque PsO or plaque PsO only.Moreover, biallelic mutations IL36RN were more common than monoallelic mutations.In contrast, rates of HLA-Cw6 alleles linked to plaque PsO were similar between the GPP population and the general population.Our study suggests that the genetic aetiology of GPP differs from plaque PsO.
National Organisation for Rare Diseases (NORD), European Conference on Rare Diseases (ECRD), International Congress of Research on Rare and Orphan Diseases (RE[ACT]), and European Academy of Dermatology and Venereology (EADV).American Academy of Dermatology (AAD) was not searched separately as it is captured in Embase.Titles and abstracts were assessed for inclusion against pre-defined population, intervention, comparators, outcomes, study types (PICOS) criteria and relevant full texts were further examined against these eligibility criteria (Table 1).Full texts were independently screened by two reviewers and discrepancies resolved through discussion.For excluded studies, the primary reason for exclusion was documented (Table S7).Study characteristics, patient characteristics (including associated plaque PsO and GPP age-of-onset), IL36RN mutations (rates and types), and HLA-Cw6 allele prevalence were extracted and analysed.

| Statistical analyses
Four analyses were conducted to explore and compare rates of IL36RN mutations.Two analyses assessed IL36RN mutation rates, and two analyses explored GPP age-of-onset.The first set of analyses compared rates of any IL36RN mutations (i.e., monoallelic and biallelic pooled) in patients with GPP only versus patients with GPP þ plaque PsO, while the second set of analyses compared rates of biallelic mutations-only across these two populations.The third and fourth sets of analyses compared the age of disease onset in patients with or without biallelic mutations in the overall GPP population, and in the GPPonly subgroup versus the GPP þ plaque PsO subgroup.Under each analysis, one meta-analysis comparing subgroups and one meta-analysis of overall population were performed.As each analysis contained ≥10 studies, publication bias was assessed using funnel plots and Egger's test for meta-analyses comparing subgroups.When >1 study from the same research group was identified, only the most recent study was included in the metaanalyses to prevent unintentional bias introduced by overlapping patient data across studies.Studies were only included if they reported data (or data could be inferred) for the subgroups of interest.Additionally, studies that selected patients that were only from a certain population age group (e.g., paediatric or elderly) were excluded, as were studies with <2 reported cases of age-of-onset, and familial-only studies.In the meta- In order to efficiently review evidence, reporting of mutations was assessed at the title/abstract stage.Articles that did not mention IL36RN mutations or HLA-Cw6 prevalence in the title/abstract were excluded from the review.It is therefore possible that some publications that mentioned these outcomes in their full text, but not in their abstract or title, may have been missed from this review.
b Bibliographies of included literature reviews and consensus studies were reviewed to determine whether they contained any relevant studies that met the inclusion criteria.The results and data used in the literature reviews and consensus studies themselves were not of interest.
analyses, studies were pooled via the Mantel-Haenszel method.Heterogeneity was assessed using the I 2 statistic.
Regarding missing data, mean age of GPP onset and/or standard deviation were not reported and could not be calculated based on individual patient data for several studies.
Meta-analyses that compared patient subgroups were conducted in RStudio (version 4.2.1) 14 using the 'meta' package (version 5.5-0). 15Both fixed and random-effects models were used; however, due to study heterogeneity in terms of country, study design, and number of included patients, the random-effects model results were considered more appropriate estimates in this analysis.
Meta-analyses of proportions were conducted for IL36RN mutations and biallelic IL36RN mutations in RStudio using the 'metaprop' function (meta package).Data were transformed using a logit function and a random-intercept logistic regression model was used to conduct the meta-analysis.
Meta-analyses of single means were conducted for age of disease onset in RStudio using the 'metamean' function (meta package).Fixed and random-effects models were explored, and inverse-variance weighting was used to pool studies.

| RESULTS
Of a total of 1977 initial records identified, 46 studies were included (Figure 1).Among them, 40 studies reported IL36RN mutations only, two studies reported HLA-Cw6 allele only, and four studies reported both.
About 22 studies included were conducted in East Asia (China, Japan, Malaysia), and 13 were international studies conducted in multiple countries (Table 2).Study sample size varied between 4 and 210 patients.
The age of GPP patients at the time of the study ranged between 8.2 and 74.3 years old; two studies only included paediatric patients, 16,17 while one study included elderly-onset GPP patients. 18Thirty-two studies reported patients' sex and most were female patients.In 33 studies, the authors reported whether associated plaque PsO was present in GPP patients (including previous and/or concomitant plaque PsO), and the age of GPP onset was reported in 25 studies.

| IL36RN mutation types
Twenty-one mutations were identified in the biallelic state, including amino acid substitutions (13 mutations), protein truncations (4 mutations), splicing defects (2 mutations), and amino acid deletions (2 mutations) (Table 5).Six of these mutations were also found in monoallelic state.Three additional mutations were only found in the monoallelic state (3 substitutions) (Table 5).The most reported mutations were c.115 þ 6T > C (p. Arg10ArgfsX1), reported in 18 studies; c.227 C > T (p.Pro76Leu), reported in 10 studies; and c.338 C > T (p.Ser113Leu), reported in 8 studies (Figure 4).When examining the patient populations reporting specific mutations, differences in ethnicity were noted.The c.115 þ 6T > C (p. Arg10ArgfsX1) mutation was primarily reported in East Asian studies and international studies that included Asian patients (17/18 studies), and the c.338 C > T (p.Ser113Leu) mutation was mostly reported in European patients (7/8 studies).The c.28 C > T (p.Arg10X) mutation was exclusively reported in Japanese patients, and the c.80 T > C (p.Leu27Pro) mutation was exclusively reported in North-African patients.
8][29] The c.227 C > T (p.Pro76Leu) mutation is located at a loop position where a leucine could not be accommodated owing to steric clashes, causing structural changes of the respective loop and/or decreased protein stability; online prediction tools predicted this mutation to be damaging. 30The c.338 C > T (p.Ser113Leu) mutation affects an evolutionarily conserved IL36RN residue proximal to critical binding residues and is predicted by the Sorting Intolerant From Tolerant algorithm to be damaging, findings confirmed by in vitro assays. 23,30

| GPP age-of-onset by mutational status
GPP age-of-onset by mutational status was reported in 22 studies, excluding paediatric studies and one elderly-onset study.Mean age-of-onset ranged from 5.9 to 48.9 years old, with most studies reporting GPP onset at 20-40 years.The mean (95% CI) age-of-onset  5b and 6c).

| HLA-Cw6 allele frequency
The HLA-Cw6 allele was discussed in 6 studies, in which the frequency ranged from 0% to 28.6% (Table S8).A few patients who had the HLA-Cw6 allele had a history of plaque PsO, however, history of plaque PsO was inconsistently reported.Arakawa (2018) reported no significant difference in HLA-Cw6 allele frequency between GPP patients and the general population. 31In one study, 7 of 10 GPP patients had the HLA-Cw7 allele instead of HLA-Cw6 that is associated with plaque PsO.

| Spesolimab and the IL36RN mutation
In Effisayil TM 1 of spesolimab, 46 out of 53 patients had genotyping data. 32According to a subgroup analysis evaluating the safety and efficacy of spesolimab on the basis of IL36RN mutation status, a high proportion of patients with an IL36RN mutation achieved both the primary and secondary endpoint. 32Among patients who had an IL36RN mutation (n = 14), complete pustule clearance at week one was achieved by 7 of 8 patients (87.5%) treated with spesolimab versus 1 of 6 patients (16.7%) treated with placebo. 32Similarly, among IL36RN mutation-positive patients, 75.0%who received spesolimab and 16.7% who received placebo experienced clear/almost clear skin (GPPGA total score 0 or 1).In patients without IL36RN mutations, complete pustule clearance and clear/almost clear skin at week one were observed in 42.9% and 28.6% among those treated with spesolimab versus 0.0% and 9.1% of patients who received placebo. 32gardless of mutation status, spesolimab treatment resulted in rapid and sustained pustular clearance and clear or almost clear skin over the course of 12 weeks.Furthermore, the safety profile of spesolimab was similar in patients with and without an IL36RN mutation. 32

| DISCUSSION
The concept of autoinflammatory keratinisation disease (AiKD) is widely acknowledged in the context of inflammatory keratinisation disorders characterised by genetic autoinflammatory mechanisms. 33The recent elucidation of the pathogenesis and pathophysiology of PsO and its related disorders, including its rare subtype, GPP, associated with the hyperactivation of skin innate immunity, suggests they are representative diseases of AiKD. 33he distinctions between GPP and plaque PsO are multifaceted; however, differences in genetic aetiology may be most telling and relevant for treatment approach.The first genetic mutation identified in patients with GPP was the homozygous missense mutation c.80 T > C (p.Leu27Pro), which was associated with a deficiency in IL-36Ra; patients with this mutation exhibited symptoms including high-grade fever and general malaise in addition to erythema and pustules. 34n addition to IL36RN, the mutations/variants detected in genes CARD14, AP1S3, MPO, and SERPINA3 were reported to be genetic causative or predisposing factors for pustular PsO. 33In this study, patients with GPP alone presented with IL36RN mutations more frequently than patients with associated plaque PsO, a finding that aligns with previous findings by Zhou et al. (2021), 6 Liu et al. (2020), 9 and Choon et al. (2023). 350][21][22][23][24][25][26] In clinical practice, a proportion of patients with GPP present with concomitant plaques, but this population has not been well characterised. 5,36PP prevalence and disease characteristics also varied by geography and ethnicity.Chinese studies reported higher mutation rates, as did two small German studies, and some IL36RN mutation types were only present in people of certain ethnicities.Of the ethnicity-specific mutations, c.115 þ 6T > C (p. Arg10ArgfsX1) was the most frequently reported and was predominantly reported in East Asian studies and Ser113Leu), are also homozygous missense mutations that impair IL-36Ra protein expression and inhibit suppression of inflammatory responses. 6A recent study reported a novel heterozygous variant c.96 T > G (p.His32Gln) in two cases. 37nother distinction between GPP and plaque PsO is the presence of HLA-Cw6 alleles, which were previously established as one of the most strongly associated risk alleles for plaque PsO but do not seem to be specifically associated with GPP. 10 Rates of HLA-Cw6 in patients with GPP found in this analysis (0%-28.6%) are similar to reported background rates in the general population (10.5% 11 to 20% 12 ).In contrast, reported rates of HLA-Cw6 are 73% 11 and 100% 12 in patients with guttate PsO and 35.7% 39 in patients with plaque PsO.Muto et al. (2014) reported HLA-Cw6 as the primary HLA allele associated with plaque PsO (N = 200) in the Japanese population. 40ltimately, the clinical manifestations of GPP can be much more severe than other types of psoriasis such as plaque PsO, leading to urgent treatment needs and possibly life-threatening complications. 1,2,4,5These differences between GPP and plaque PsO highlight the need for novel treatment approaches that target the underlying aetiology and provide significant impact on the disease course.Today, patients with GPP are often treated with therapies that do not address the underlying pathology of GPP, do not have robust evidence of safety and efficacy in patients with GPP, and/or may paradoxically induce flares. 1,5hile advancements in GPP-specific treatments have been limited, 1,5 the IL-36 inhibitor spesolimab was recently approved in the US, the EU, China, and Taiwan for the treatment of GPP flares, and for the improvement of acute symptoms of GPP in Japan. 41,42lockade of IL-36 receptor signalling with spesolimab was associated with decreases in proinflammatory-and neutrophil-recruitment mediators as well as reduced keratinocyte activation. 43argeted therapy blocking IL-36 receptor signalling has been demonstrated to be a safe and effective treatment option for GPP, 13 with a high proportion of patients with an IL36RN mutation demonstrating complete pustule clearance and clear/almost clear skin. 32Spesolimab treatment leads to rapid, normalising or downregulating immune cell activation, and inhibiting the inflammatory response, which results in marked clinical improvement. 43Given the earlier onset of GPP in those who have biallelic mutations of IL36RN, genetic testing may support streamlining identification and initiation of treatment for patients with a significant predisposition.

| Limitations
This review is limited by its targeted nature, small sample sizes of many included studies, and heterogeneity of study designs and objectives.Assessment of IL36RN mutations in plaque PsO-only patients was limited to studies of GPP that also enroled patients with plaque PsO and therefore may not be representative of that population.Additionally, studies conducted by the same research group likely recruited some patients who were included in their previous studies, potentially breaking the assumption of independent effect sizes, which is required for meta-analyses.To avoid this issue, only the most recent study from each research group was included in the meta-analyses.Evidence of heterogeneity was considerable for several analyses (I 2 between 75% and 100%), especially age-of-onset.This may be due to high variability in the patient numbers in included studies and inclusion of several studies with small sample sizes.Moreover, differences in study locations and designs could have introduced heterogeneity into the analyses.

| CONCLUSION
Our findings provide further evidence that GPP is biologically distinct from plaque PsO, as characterised by higher rates of genetic mutations of IL36RN and the lack of association with the HLA-Cw6 allele.Reported mutation rates of IL36RN in patients with GPP and the types of mutations varied geographically and in different ethnic groups, respectively.Ultimately, GPP is a serious condition that requires effective and rapid treatment to control symptoms and limit complications of flares.The advent of treatments specific for the IL-36 pathway offers new, targeted options for effective treatment.
Abbreviations: GPP, generalised pustular psoriasis; M/F, male/female; NR, not reported; SD, standard deviation; SLR, systematic literature review; UK, United Kingdom.a Values were calculated based on the individual patient data reported in the studies.b Only paediatric patients were included in the study.

F I G U R E 2
IL36RN carrier rate of biallelic mutation in patients with GPP-only versus GPP þ plaque psoriasis.GPP, generalised pustular psoriasis.

F I G U R E 4
Number of studies reporting specific mutation types.

F I G U R E 5
Mean age-of-onset in GPP patients by associated plaque PsO and IL36RN variants.(a) Mean age-of-onset of GPP in patients with or without associated plaque PsO; (b) Mean age-of-onset of GPP patients with or without IL36RN biallelic mutation.GPP, generalised pustular psoriasis; PsO, psoriasis.GPP patients with no IL36RN biallelic mutation include patients with either monoallelic or no IL36RN mutations.

F
I G U R E 6 Meta-analyses assessing age-of-onset in patients with GPP.(a) Age of GPP onset; (b) Comparison of age-of-onset in patients with GPP only versus patients with GPP þ plaque PsO; (c) Comparison of age-of-onset in patients with biallelic IL36RN mutations versus patients with no biallelic IL36RN mutations.CI, confidence interval; GPP, generalised pustular psoriasis; MD, mean difference; PsO, psoriasis; SD, standard deviation.

T A B L E 1
Population, intervention, comparators, outcomes, study types (PICOS) criteria. a

T A B L E 3 IL36RN mutation rate overview. Country Number of studies % GPP patients with IL36RN biallelic mutations a % GPP patients with IL36RN monoallelic mutations a
Note: Total studies and percentage range are given in bold format.Abbreviations: GPP, generalised pustular psoriasis.aValueswere calculated based on the individual patient data reported in the studies.bBiallelicand monoallelic IL36RN mutations were reported pooled in three studies.cBougacha2016 and Marrakchi 2011 were excluded as they only included familial cases.group;however, the results were not statistically significant (mean difference [MD] [95% CI]: −3.54 [−10.29;3.21];Figures5aand 6b).Age of GPP onset was also compared between biallelic IL36RN mutation-positive and negative patients.The meta-analysis suggested a younger age-of-onset in the biallelic IL36RN mutation-

% GPP patients with associated plaque PsO (n/N) %IL36RN biallelic mutations a Among GPP-only patients Among GPP with plaque PsO patients
Sequencing analysis was performed for six of the seven total patients; only the polymorphism c.115 þ 6T > C was analysed because of its dominant role in Chinese patients with GPP.f This study only reported the plaque PsO status for 42 out of 45 patients of the pooled monoallelic and biallelic GPP.
T A B L E 4 IL36RN carrier rate of biallelic mutation and associated plaque PsO.Abbreviations: GPP, generalised pustular psoriasis; PsO, psoriasis.aValues were calculated based on the individual patient data reported in the studies.bBiallelic and monoallelic mutations were only reported pooled.cValue corresponds to number of patients with at least 1 mutation without plaque PsO.d Value corresponds to number of patients with at least 1 mutation and plaque PsO.e g Plaque PsO status was only reported for the 13 patients bearing IL36RN mutations.h Plaque PsO status was unclear for 1 patient.i Plaque PsO status unknown for 4 patients.KRUEGER ET AL.