Skip to main content
NCBI home page
PLOS One logoLink to PLOS One
. 2025 Sep 8;20(9):e0322602. doi: 10.1371/journal.pone.0322602

Causal associations between psoriasis and atopic dermatitis: A bidirectional Mendelian randomization study

Zixia Wang 1,2, Hao Nie 2, Xinyu Fan 2, Muyao Wang 2, Weili Chen 3, Quncai Huang 3, Junchao Xiao 3, Wenhui Wang 1,*
Editor: Giuseppe Novelli4
PMCID: PMC12416699  PMID: 40920623

Abstract

Background

Although previous studies suggested associations between psoriasis and atopic dermatitis (AD), the directionality and causality of these relationships remain controversial. This study employed bidirectional Mendelian randomization to investigate the potential causal relationships between these two inflammatory skin conditions.

Methods

Genome-wide association statistics were obtained for psoriasis and AD from large-scale consortia and meta-analyses of genome-wide association studies. Inverse-variance weighting, as the primary analysis, was combined with five complementary sensitivity analyses to evaluate the robustness and potential pleiotropy of the data. Additionally, we performed gene mapping of psoriasis-associated single-nucleotide polymorphisms and subsequent pathway analysis to further elucidate the potential relationships.

Results

Genetic predisposition to psoriasis was significantly associated with a decreased risk of AD (odds ratio = 0.876; 95% confidence interval = 0.834–0.921; p = 1.6 × 10−7). Conversely, genetic predisposition to AD did not affect the risk of psoriasis. The associations remained consistent across multiple sensitivity analyses, and no evidence of horizontal pleiotropy was observed. Gene mapping identified eight key genes (ENSG00000249738, ENSG00000291336, ENSG00000291338, ENSG00000285703, OR2W1-AS1, HLA-DQA1, FBXL18, and nitric oxide synthase 2 (NOS2) located on chromosomes 5, 6, 7 and 17. Notably, NOS2 emerged as a core gene involved in key biological processes, including the TCR signaling pathway and protein metabolism.

Conclusions

This comprehensive MR study provided evidence of the protective causal effect of psoriasis on the risk of AD, whereas no reverse causal relationship was noted. These findings enhanced our understanding of the relationship between psoriasis and Ad and identified potential implications for their clinical management.

Introduction

Psoriasis is a common debilitating disease of the skin and/or joints. Approximately 90% of all cases involve chronic plaque psoriasis, also termed psoriasis vulgaris, making it the most prevalent of the four types of the disease [1]. Chronic plaque psoriasis manifests as thick patches of erythematous skin with overlying white or silvery scales, with patients regularly experiencing symptoms including pain, itching, and bleeding [2]. Atopic dermatitis (AD), also called eczema, is a chronic autoimmune pruritic inflammatory skin condition that affects skin creases and flexure surfaces. Psoriasis and AD are the two most prevalent chronic inflammatory skin conditions globally, and both diseases are characterized by immune dysfunction and inflammatory cell infiltration in skin lesions [3]. Multifactorial interactions between hereditary and environmental factors are the main causes of both diseases, and these diseases markedly decrease the quality of life and emotional well-being of patients, in addition to various comorbidities closely related to their occurrence.

Psoriasis and atopic dermatitis (AD) represent two distinct chronic inflammatory skin conditions that share some clinical manifestations but differ fundamentally in their pathogenic mechanisms [4]. While both conditions present with erythema and pruritus, their cutaneous lesions exhibit distinct morphological and distributional patterns [5]. Psoriasis typically manifests as well-demarcated, thick, scaly plaques predominantly affecting extensor surfaces, whereas AD characteristically presents with poorly defined, excoriated patches and plaques in flexural areas. These clinical distinctions reflect underlying differences in their immunological signatures, with psoriasis being predominantly driven by Th1/Th17 responses and AD by Th2-mediated inflammation [6]. Despite their apparent differences, the potential interplay between these conditions has intrigued researchers, particularly given their shared inflammatory nature and occasional clinical overlap.

Previous observational studies reported both the coexistence and mutual exclusivity of psoriasis and AD, with both diseases caused by T cells activated by specific antigens [4,7,8]. Ivert et al.[9] found that AD was related to several autoimmune illnesses, especially those involving the skin such as psoriasis, with stronger relationships being reported for individuals with multiple autoimmune comorbidities [10]. However, some studies suggested that psoriasis and AD can coexist, although no clear consensus has been reached, making the causal direction between the risks of these diseases unclear [11].

Conventional observational epidemiological studies on psoriasis and AD face several challenges in elucidating disease etiology and inferring causality, such as the effects of potential confounders and the presence of reverse causality bias [12]. Therefore, more reliable methods are urgently needed to eliminate these interfering factors and confirm the potential causal relationships between psoriasis and AD. Mendelian randomization (MR), which has recently garnered increasing attention, utilizes genetic variants as the instrumental variables (IVs). MR is an effective strategy for determining the causal effects of exposure to genetic variants with single-nucleotide polymorphisms (SNPs) on outcomes independent of residual reverse or confounding causality bias [13]. MR has three core assumptions: genetic variants should exhibit a robust association with exposure; the association between the genetic variants and exposure should be independent of the potential confounders; and the influence of the genetic variants on the risk of the outcome should only occur through a risk factor without any potential effect from alternative pathways.

This study represents a significant addition to understanding of the causal relationship between psoriasis and atopic AD by employing bidirectional Mendelian Randomization (MR)—a methodological innovation that resolves longstanding controversies regarding the directionality and causality of their association. Unlike previous studies that predominantly relied on observational data, this research leverages genetic instruments to infer causality, effectively minimizing biases from confounding factors and reverse causation. Besides, this study could improve our understanding of the pathogenesis of AD, especially in the context of immune dysfunction.

Materials and methods

Study design

All of the genome-wide association statistics obtained from the corresponding consortia are publicly available. The flowchart of the study design is presented in Fig 1.

Fig 1. Overview of the study design.

Fig 1

Open in a new tab

The orange boxes and arrows represent the forward MR analyses, with psoriasis as the exposure and AD as the outcome, while the cyan boxes and arrows represent the reverse MR analyses, with AD as the exposure and psoriasis as the outcome. Abbreviations: MR, Mendelian randomization; AD, atopic dermatitis; SNPs, single-nucleotide polymorphisms.

Data sources

The genome-wide association statistics for psoriasis and AD used in this study were both obtained from genome-wide association studies (GWASs), which were limited to research that only included individuals of European ancestry to prevent population stratification bias [14]. Publicly available summary statistics (β, standard error, and p-value) for psoriasis reported by the Neale laboratory in the IEU Open GWAS project (ID:ukb-a-100) were assessed. These data included 3871 patients with psoriasis and 333,288 controls in the UK Biobank. The summary-level data for AD (ID: ebi-a-GCST90018784) in our study were obtained from a cross-population atlas study conducted by Sakaue et al.[15]. The study also incorporated data from the GWAS Catalog, which included 24,185,642 SNPs across 481,299 participants (6224 patients and 475,075 controls) of European ancestry. The main characteristics of these GWAS data are presented in Table 1.

Table 1. Primary information of the utilized consortia and data.

Phenotypes GWAS ID Sample size SNPs Author Consortium Link Year
Psoriasis ukb-a-100 337,159 10,894,596 Neale Neale Lab https://gwas.mrcieu.ac.uk/datasets/ukb-b-6134/ 2017
Atopic dermatitis ebi-a-GCST90018784 481,299 24,185,642 Saori Sakaue NA https://gwas.mrcieu.ac.uk/datasets/ebi-a-GCST90018784/ 2021
Open in a new tab

Selection of genetic instruments

For selection, the IVs required a threshold of genome-wide significance (P < 5 × 10−8) to ensure a robust association between the SNPs and exposure. We performed linkage disequilibrium clumping to include independent SNPs (r2 < 0.001; clumping window size = 1000 kb) [16]. Palindromic SNPs with intermediate allele frequencies were excluded from the IVs. To detect pleiotropic pathways, the PhenoScanner database was searched for correlations between specific genetic variants such as IVs and other phenotypes. Specific examples were phenotypes associated with inflammatory bowel disease or body mass index (BMI), which are recognized as risk factors for AD [17,18]. Variants correlated with these and other phenotypes were classified as confounders and excluded from MR.

Statistical analysis

The study employed bidirectional two-sample MR to examine the cause-and-effect relationships between psoriasis and AD. The principal statistical method was inverse-variance weighting (IVW), which was complemented by five sensitivity analyses: weighted median, simple mode, weighted mode, MR-Egger regression, and MR Pleiotropy RESidual Sum and Outlier (MR-PRESSO). These six approaches identified IVs with horizontal pleiotropy effects based on different assumptions. IVW is characterized by the assumption that all included SNPs are effective genetic instruments, and a comprehensive process of Wald estimates and meta-analysis was used to evaluate the overall effect of exposure on outcome [19]. In circumstances in which horizontal pleiotropy and heterogeneity were absent, the IVW result was considered the most reliable estimation [20]. In addition, the weighted median method generates a better casual estimate when >50% of the instruments are derived from valid IVs [21]. Although less powerful than IVW, simple mode is recognized as a robust method for detecting pleiotropy [22]. Based on the instrument strength being independent of the direct effect (InSIDE) hypothesis, MR-Egger regression facilitates judgment of the potential impact of pleiotropy on the intercept. Specifically, an intercept of zero indicates the absence of horizontal pleiotropy. When the InSIDE assumption is violated, the weighted mode estimate technique outperforms MR-Egger regression for detecting causal effects, as indicated by lower type I error rates and smaller deviation [23]. The weighted mode estimation method can detect causal effects. Compared with MR-Egger regression, it has less deviation and a lower type I error rate when the InSIDE hypothesis is violated [23]. MR-PRESSO can detect significant outliers and statistics including MR-Egger regression and horizontal pleiotropy effects after correction to remove corresponding outliers [24].

Cochrane’s Q value was used to further determine the heterogeneity among all SNPs, and “leave-one-out” analysis was conducted to determine whether the results were influenced by any independent potentially heterogeneous SNPs. Further, we calculated the F-statistic separately to quantify the strength of combined SNPs and each individual SNP. To classify the phenotypic variations, F-value > 10 for the corresponding SNP(s) was considered sufficiently strong, whereas F-value ≤ 10 denoted a weak instrument [25]. Finally, we performed reverse MR analyses of psoriasis and AD that were similar to those of the forward MR analyses.

All analyses were performed with R software version 4.2.2 using the TwoSampleMR (version 0.5.6) and MR-PRESSO (version 1.0) packages.

SNP-to-gene mapping and pathway analysis

To further investigate the potential biological mechanisms underlying our findings, we utilized g:Profiler online tools (https://biit.cs.ut.ee/gprofiler/gost) to match all harmonized psoriasis-associated SNPs to the corresponding genes. The specific positions of identifiable genes were then mapped on the chromosome image using the “Rcircos” package in R software. In addition, we identified the signaling pathways involving nitric oxide synthase 2 (NOS2, located on 17q11.2), the identified core gene, through the Reactome (https://reactome.org/) and performed pathway analysis.

Results

Directional relationship from psoriasis to AD

To investigate the causal influence of psoriasis on the risk of AD, we identified 14 independent SNPs as IVs in psoriasis. The detailed characteristics of these IVs are listed in S1 Table. As the primary method, IVW suggested a significant causal relationship between psoriasis and AD. Specifically, each one standard deviation (SD) decrease in psoriasis morbidity was associated with an approximately 12.4% higher risk of developing AD (odds ratio [OR] = 0.876, 95% confidence interval [CI] = 0.834–0.921, p = 1.6 × 10−7; Fig 2A, 2D).

Fig 2. MR to identify the causal impact of genetically predicted psoriasis on the risk of AD.

Fig 2

Open in a new tab

(A) Scatter plots of the genetic effects of IVs on the genetic predisposition to psoriasis and the subsequent risk of AD. The slope of the straight lines with different colors indicates the causal estimate in the corresponding methods of MR. Each black dot corresponds to an IV included in MR, whereas the gray bars represent the SDs of the genetic effect estimations. (B) Forest map of the visualized causal influences of each included SNP on the risk of AD risk. (C) “Leave-one-out” plots of the causal relationship between psoriasis and AD. (D) Causal effect of psoriasis on AD visualized in a forest map using five methods. Abbreviations: AD, atopic dermatitis; MR, Mendelian randomization; IVs, instrument variables; SD, standard deviation.

The robustness of this finding was supported by consistent results across three other estimation methods (Fig 2D), namely MR Egger (OR = 0.820; 95% CI = 0.743–0.906; p = 0.002), weighted median (OR = 0.846; 95% CI = 0.794–0.901; p = 2.14 × 10−7), and weighted mode (OR = 0.838; 95% CI = 0.780–0.901; p = 3.39 × 10−4). As presented in S2 Table, comprehensive sensitivity analyses revealed no heterogeneity among these IVs in both the Cochrane’s Q test (PQ > 0.05) and the MR-Egger test (Pintercept > 0.05) and no horizontal pleiotropy in the MR-PRESSO global test (Pglobal > 0.05). Additionally, no noticeable alterations in the estimated causal effect were detected in individual SNP removed in the “leave-one-out” analysis (Fig 2B–2C).

The combined F-statistic of the included 14 SNPs was 12.8, indicating sufficient instrument strength. However, the F-values of the individual SNP varied considerably (range: 0.9–179.3), suggesting that the individual SNPs had varied strengths of association with psoriasis. We therefore conducted a subgroup analysis and categorized the 14 SNPs as strong or weak instruments. Importantly, the causal association remained consistent in direction and significance in both subgroups (S3–S4 Tables), further supporting the reliability and robustness of our MR findings.

SNP-to-gene mapping and pathways analysis

In this study, we mapped the casual SNPs to genes to identify the genes potentially influenced by SNPs associated with both psoriasis and AD. In the aggregate, the 14 SNPs were detected in eight genes located on chromosomes 5, 6, 7, and 17, namely ENSG00000249738 (5q33.3), ENSG00000291336 (6p22.2), ENSG00000291338 (6p 22.2), ENSG00000285703 (6p22.1), OR2W1-AS1 (6p22.1), HLA-DQA1 (6p21.32), FBXL18 (7p22.1), and NOS2 (Fig 3A). Based on previous research [26,27], we recognized NOS2 as the core gene because of its close connections to both psoriasis and AD. The findings of pathway analysis illustrated the significant enrichment of NOS2 in several biological processes, including protein metabolism, post-translational protein modification, neddylation, and protein localization. The findings also indicated that NOS2 is involved in several immune-related processes such as class I MHC-mediated antigen processing and presentation, antigen processing through ubiquitination and proteasome degradation, and costimulation by the CD28 family (Fig 3B).

Fig 3. SNP-mapped genes and NOS2-related molecular pathways. (A) SNP-to-gene mapping displaying the genes associated with the forward SNPs, both of which are labeled in the figure. (B) Pathway analysis demonstrating the biological mechanism related to the NOS2 gene. Abbreviations: SNP, single-nucleotide polymorphism; NOS2, nitric oxide synthase 2.

Fig 3

Open in a new tab

Directional relationship from AD to psoriasis

The comprehensive meta-analysis of large-scale GWASs identified eight SNPs with robust associations with AD. Detailed information on each SNP is provided in Summary Table 5. The total strength (F-statistic) of the integrated eight SNPs was 14.9, with the F-values of the individual SNPs ranging 10.1–28.7. These results indicated that either the incorporated SNPs or each separate SNP had sufficient strength to explain the phenotypic variations of AD.

As detailed in Fig 4D, no evidence providing a positive or reverse causal relationship between genetic predisposition to AD and the risk of psoriasis was obtained according to the results of IVW, MR-Egger, weighted median, simple mode, and weighted mode using the full set of eight SNPs (p > 0.05). Coincidently, no outlying SNP was identified by the outlier-corrected MR-PRESSO method (causal estimate = −0.0014, SD = −0.0009, T-statistic = −1.5356, p = 0.1685; S6 Table). No noticeable heterogeneity or pleiotropic effects for the MR estimate of the causal effect of AD on psoriasis were detected using either the MR-Egger, Cochrane’s Q, or MR-PRESSO global tests (all p > 0.05; Fig 4A–4B, S6 Table). The “leave-one-out” analysis also identified no significant alterations in the MR estimates when each specific SNP was discarded or returned (Fig 4C). This finding confirmed the robustness and effectiveness of our results.

Fig 4. MR of the causal estimates of the effects of genetic predisposition to AD on the risk of psoriasis. (A) Scatter plots of the effects of IVs on the genetic predisposition to AD and the subsequent risk of psoriasis. The slope of the straight lines in different colors indicates the causal estimation in the corresponding methods of MR. Each black dot corresponds to an IV included in MR, whereas the gray bars represent the SD of the estimation of the genetic effect. (B) Forest map of the visualized causal influence of each included SNP on the risk of psoriasis. (C) “Leave-one-out” plots of the causality between AD and psoriasis. (D) Causal impact of AD on the risk of psoriasis visualized as a forest map of the five methods. Abbreviations: MR, Mendelian randomization; AD, atopic dermatitis; IVs, instrument variables; SD, standard deviation.

Fig 4

Open in a new tab

Discussion

To the best of our knowledge, this is the first study to investigate the reciprocal impact between psoriasis and AD. This study identified a significant causal relationship between increased psoriasis morbidity (genetically proxied) and a decreased risk of AD (genetically proxied), whereas a genetic predisposition to AD did not influence the later risk of psoriasis.

In an earlier epidemiological study of a European population, we reported that individuals with psoriasis had a 25-fold lower prevalence of AD [28]. An observational study involving 29,159 German patients by Christophers et al.[29] confirmed this finding and suggested that the concurrence of psoriasis and AD was extremely rare. Another cross-sectional study also recorded a lower prevalence of AD in individuals with psoriasis than in healthy controls [30]. The findings of this MR study were consistent with those of previous studies, supporting the existence of a causal effect of psoriasis on AD. Nonetheless, based on the shared immunopathogenesis of psoriasis and AD, the accumulated clinical studies suggest that the two diseases can coexist. For example, a retrospective review by Barry et al.[11] found that the rate of concomitant AD and psoriasis was 1.5% with notable hand involvement. A recent nationwide cohort study in Korea also reported a bidirectional association between psoriasis and AD [31]. These findings contradict the results of our study, which found that psoriasis was closely correlated with a lower risk of AD. Because of these inconsistent findings, the causal relationship between psoriasis and AD remains to be confirmed.

Our findings of a protective effect of psoriasis on AD risk align well with the distinct immunological profiles of these conditions. The inverse relationship we observed may be explained by the antagonistic nature of their predominant T-helper cell responses. The Th1/Th17-dominated environment in psoriasis might suppress the development of Th2-driven allergic inflammation characteristic of AD. This molecular antagonism is reflected in their contrasting clinical presentations: while both conditions manifest with inflammation and pruritus, psoriatic lesions typically present as well-defined, thick, erythematous plaques with silvery scales, predominantly affecting extensor surfaces. In contrast, AD lesions are generally poorly demarcated, with a predilection for flexural areas and accompanied by intense pruritus leading to excoriation. These distinct clinical and immunological features support our genetic findings and suggest that the pathogenic mechanisms underlying psoriasis may actually protect against the development of AD through complex immunological interactions.

Our gene mapping analysis revealed several key genes that provide mechanistic insights into the relationship between psoriasis and AD. Most notably, NOS2 emerged as a core gene, showing elevated expression in psoriatic lesions while maintaining lower levels in AD-affected skin, which aligns with our observed protective effect of psoriasis on AD risk [32]. The identification of HLA-DQA1 is particularly significant, as its variants demonstrate distinct immunological effects in these conditions. While psoriasis-associated variants promote Th1/Th17 responses, AD-linked variants predominantly drive Th2-mediated inflammation [10]. This differential immune programming may explain the inverse relationship we observed between these conditions. Our pathway analysis revealed that several identified genes, including FBXL18 and novel transcripts (ENSG00000249738, ENSG00000291336), are involved in protein metabolism and TCR signaling pathways. These pathways are crucial in determining T-cell fate and function, potentially explaining the antagonistic relationship between psoriasis and AD [33]. Additionally, the identification of OR2W1-AS1, located near the HLA complex, suggests potential regulatory functions in immune responses, though further functional studies are needed to elucidate its precise role. Indeed, there is evidence of directional associations between psoriasis and AD, such as their shared immune pathways and histopathological features [34]. Both diseases are characterized by activated T cell infiltration in the skin accompanied by enhanced keratinocyte proliferation, resulting in skin thickening [35]. In addition, the immune pathways of psoriasis and AD both involve Th1 and Th22 cells, with Th17 cells specifically present in the pediatric, intrinsic, and Asian types of both conditions [35–37]. However, the development of psoriatic lesions is mainly attributed to chronic inflammation through the TNFα/interleukin (IL)-17/Il-23/Th17 axis, whereas AD is associated with skewed correlations among Th2 cells, IL-4, and IL-13 [35,38,39]. In psoriasis, damaged keratinocytes release antimicrobial peptides such as β-defensins and LL-37 in the initiation phase, resulting in the formation of LL-37–DNA complexes that further amplify Toll-like receptor 9 signaling to activate plasmacytoid dendritic cells (pDCs) [40]. These activated pDCs produce interferons (IFNs), including IFNα and IFNβ, which subsequently expedite the maturation of myeloid dendritic cells (mDCs). MDCs in turn release IL-12, IL-23, and TNF-α to stimulate Th1 and Th17 cells [36,41]. IFNα is also involved in the secretion of IFNγ and IL-17 and the differentiation of Th1 and Th17 cells. Generally, the IL-17 family consists of six members (IL17A–F), among which IL-17A displays a strong correlation with the pathogenesis of psoriasis [42]. Other members of the IL-17 family, such as IL-17C and IL-17F, are expressed prominently in psoriasiform dermatitis, and they enhance innate defenses in epithelial cells [43]. IL-23 dominates in plaque psoriasis and plays a major role in modulating JAK2/TYK2 and STAT signaling, thereby increasing IL-17 production by immune cells [42,44,45]. Taken together, these findings demonstrate that the central mechanism of psoriasis is related closely to adaptive immune pathways comprising IL-17 and IL-23, together with the pivotal roles of Th17 cells and TNF-α.

Among the genes identified in this study, NOS2 is considered the core gene because of its strong associations with both diseases. According to previous genome-wide association analyses, NOS2 has been postulated to be a susceptibility gene in psoriasis based on its increased expression [46,47]. Pathological manifestations in patients with psoriasis, including inflammation and vascular and keratinocyte proliferation, are caused by cellular NOS2 activation (Fig 5). In contrast to low NOS2 expression in AD lesions, Köhler et al. identified significantly increased NOS2 mRNA expression in psoriatic epidermis and proved that NOS2-derived low levels of NO mediated downstream psoriatic effects of IL-17 [48]. Recently, NOS2/iNOS (inducible nitric oxide synthase) was proposed as a novel gene classifier to diagnose psoriasis and AD with 100% accuracy by quantitative PCR [49]. Other articles similarly recognized the potential of NOS2 as a specific molecular disease classifier for both diseases [26,27,50].

Fig 5. A possible model of the mechanism linking NOS2 expression to the development of psoriasis.

Fig 5

Open in a new tab

A variety of activators such as cytokines, microbial products, and immune complexes, stimulate cellular NOS2 expression in keratinocytes. The formed •NO then induces inflammation, angiogenesis stimulation, lipid peroxidation, keratinocyte proliferation, and other effects through interactions with other superoxide moieties. In this process, NOS2 protein expression is not fully correlated with its mRNA expression because NOS2 protein translation can be inhibited by various factors (e.g., miR-31, arginase 1).

Regarding the strong genetic backgrounds of psoriasis and AD, several different chromosomal loci have been identified in each disease with minimal overlap [34]. Studies have implied that HLA-C*06:02 (on PSORS1 6p21.33) is a major susceptibility gene locus for psoriasis, whereas null mutations of the FLG gene (1q21.3) have been suggested to increase the risk of AD [5,36,41,51]. Particularly in European populations, HLA-C06:02 demonstrates a strong association with psoriasis, with an odds ratio of approximately 3.5, while FLG null mutations are found in up to 50% of moderate-to-severe AD patients of European descent [52]. Current genome-wide comparative analyses have confirmed the presence of FLG gene mutations in the epidermal differentiation complex on chromosome 1q21.3 in AD but not in psoriasis [10,53]. Similarly, removal of the late cornified envelope genes 3B/3C on chromosome 1q21 has been linked to psoriasis but not AD [10,54]. Nevertheless, other FLG genetic variants have been confirmed to confer susceptibility to psoriasis, especially in the Chinese population, indicating a shared region on the genome between AD and psoriasis [55,56]. In general, the majority of genetic analyses supported the mutual exclusivity of psoriasis and AD, although several overlapping loci have been discovered recently [10,57,58].

Strengths and limitations of the study

The study had several strengths. The major advantage was its bidirectional MR design, which strengthened the causal inference between psoriasis and AD. The bidirectional analysis was also used to determine the existence of a two-way association between the diseases when compared to standard MR analysis. In addition, the data utilized in this study were derived from consortia with adequate sample sizes, with the identified SNPs divided into two groups according to the strength of statistical power, thereby ensuring the robustness of our findings. Moreover, we detected and excluded potential outliers after implementing MR-PRESSO and MR-Egger regression intercept tests, which contributed to low heterogeneity and bias in our study. Ultimately, no duplicate SNP was detected in the GWAS summary data related to psoriasis and AD.

However, some limitations requiring further consideration must be acknowledged. First, the conflicting results between our study and other studies might reflect the influence of unmeasured potential confounding factors on causal inferences. Second, the data used in our study were exclusively related to individuals of European ancestry, thereby constraining the generalizability of our findings to other demographic cohorts. For example, studies have identified specific genetic and manifestation characteristics of psoriasis or AD in Asian patients, distinguishing them from other ethnic groups [4,34]. In addition, the depth of our study was constrained because of the lack of detailed classified data, precluding hierarchical analyses to further investigate the causal associations between different subtypes of the two diseases, such as the four subtypes of psoriasis (psoriasis vulgaris, erythrodermic psoriasis, pustular psoriasis, and arthritic psoriasis). Although no pleiotropic effect was detected in our analysis, we could not dismiss the possibility that genetic instruments proxied for psoriasis and AD were related to other traits that could affect the final outcome. Furthermore, the negative association between psoriasis and AD might have been induced by overlapping genes rather than a causal link. However, the limited shared genetic background between psoriasis and AD minimized the potential of this bias [34].

Conclusion

This study identified a causal association between genetically predicted psoriasis and a lower risk of AD, whereas the hypothesis of reverse causality was not confirmed. However, we cannot dismiss the possibility that AD influences the incidence and progression of psoriasis. Further experimental and clinical studies are warranted to validate and extend our findings and discover the molecular mechanisms underlying the observed causal associations.

Supporting information

S1 Table. Instrumental variables used in MR analysis of the association between psoriasis and AD.

Detailed information of SNPs used in MR analysis: From psoriasis to AD.

(XLSX)

pone.0322602.s001.xlsx (12.5KB, xlsx)
S2 Table. Detailed information of the sensitive analyses used in MR analysis of the association between psoriasis and AD.

Cochran’s Q test, MR Egger test and MR PRESSO global test used in this study: From psoriasis to AD.

(XLSX)

pone.0322602.s002.xlsx (11.6KB, xlsx)
S3 Table. Strong instrumental variables correlated sensitive analyses used in MR analysis of the association between psoriasis and AD.

Detailed information of strong instruments group in subgroup analyses: From psoriasis to AD.

(XLSX)

pone.0322602.s003.xlsx (13.9KB, xlsx)
S4 Table. Weak instrumental variables correlated sensitive analyses used in MR analysis of the association between psoriasis and AD.

Detailed information of weak instruments group in subgroup analyses: From psoriasis to AD.

(XLSX)

pone.0322602.s004.xlsx (14.3KB, xlsx)
S5 Table. Instrumental variables used in MR analysis of the association between AD and psoriasis.

Detailed information of SNPs used in MR analysis: From AD to psoriasis.

(XLSX)

pone.0322602.s005.xlsx (12KB, xlsx)
S6 Table. Detailed information of the sensitive analyses used in MR analysis of the association between AD and psoriasis.

Cochran’s Q test, MR Egger test and MR PRESSO global test used in this study: From AD to psoriasis.

(XLSX)

pone.0322602.s006.xlsx (11.7KB, xlsx)

Data Availability

All relevant data are within the manuscript and its Supporting information files.

Funding Statement

This work was supported by Wenhui Wang Dongguan Famous TCM Expert Inheritance Studio, with grant number (no.5000164), awarded to WH, and Higher Education Research Key Fund Project of Guangdong University of Technology, with grant number (no.GXLX20210106), awarded to WC. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

References

  • 1.Griffiths CE, Barker JN. Pathogenesis and clinical features of psoriasis. Lancet. 2007;370(9583):263–71. doi: 10.1016/S0140-6736(07)61128-3 [DOI] [PubMed] [Google Scholar]
  • 2.Boehncke WH, Schön MP. Psoriasis. Lancet. 2015;386(9997):983–94. [DOI] [PubMed] [Google Scholar]
  • 3.Ren Y, Liu J, Li W. Causal associations between vitamin D levels and psoriasis, atopic dermatitis, and vitiligo: A bidirectional two-sample Mendelian randomization analysis. Nutrients. 2022;14(24). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Guttman-Yassky E, Krueger JG. Atopic dermatitis and psoriasis: two different immune diseases or one spectrum?. Curr Opin Immunol. 2017;48:68–73. doi: 10.1016/j.coi.2017.08.008 [DOI] [PubMed] [Google Scholar]
  • 5.Langan SM, Irvine AD, Weidinger S. Atopic dermatitis. Lancet. 2020;396(10247):345–60. [DOI] [PubMed] [Google Scholar]
  • 6.Siegfried EC, Hebert AA. Diagnosis of atopic dermatitis: mimics, overlaps, and complications. J Clin Med. 2015;4(5):884–917. doi: 10.3390/jcm4050884 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Eyerich S, Onken AT, Weidinger S, Franke A, Nasorri F, Pennino D, et al. Mutual antagonism of T cells causing psoriasis and atopic eczema. N Engl J Med. 2011;365(3):231–8. doi: 10.1056/NEJMoa1104200 [DOI] [PubMed] [Google Scholar]
  • 8.Cunliffe A, Gran S, Ali U, Grindlay D, Lax SJ, Williams HC, et al. Can atopic eczema and psoriasis coexist? A systematic review and meta-analysis. Skin Health Dis. 2021;1(2):e29. doi: 10.1002/ski2.29 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Ivert LU, Wahlgren CF, LINDELÖF B. Association between atopic dermatitis and autoimmune diseases: a population-based case-control study. Br J Dermatol. 2021;185(2):335–42. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Baurecht H, Hotze M, Brand S, Büning C, Cormican P, Corvin A, et al. Genome-wide comparative analysis of atopic dermatitis and psoriasis gives insight into opposing genetic mechanisms. Am J Hum Genet. 2015;96(1):104–20. doi: 10.1016/j.ajhg.2014.12.004 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Barry K, Zancanaro P, Casseres R, Abdat R, Dumont N, Rosmarin D. Concomitant atopic dermatitis and psoriasis - a retrospective review. J Dermatolog Treat. 2021;32(7):716–20. doi: 10.1080/09546634.2019.1702147 [DOI] [PubMed] [Google Scholar]
  • 12.Zhao N, Guo P, Tang M, Yang F, Zhang T, Mao R. Evidence for a causal relationship between psoriasis and cutaneous melanoma: a bidirectional two-sample Mendelian randomized study. Front Immunol. 2023;14:1201167. doi: 10.3389/fimmu.2023.1201167 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Davey Smith G, Hemani G. Mendelian randomization: genetic anchors for causal inference in epidemiological studies. Hum Mol Genet. 2014;23(R1):R89-98. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Dashti HS, Jones SE, Wood AR. Genome-wide association study identifies genetic loci for self-reported habitual sleep duration supported by accelerometer-derived estimates. Nat Commun. 2019;10(1):1100. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Sakaue S, Kanai M, Tanigawa Y, Karjalainen J, Kurki M, Koshiba S, et al. A cross-population atlas of genetic associations for 220 human phenotypes. Nat Genet. 2021;53(10):1415–24. doi: 10.1038/s41588-021-00931-x [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.1000 Genomes Project Consortium, Auton A, Brooks LD, Durbin RM, Garrison EP, Kang HM, et al. A global reference for human genetic variation. Nature. 2015;526(7571):68–74. doi: 10.1038/nature15393 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Soh H, Lee HJ, Han K, Park S, Hong SW, Moon JM, et al. Atopic diseases are associated with development of inflammatory bowel diseases in Korea: a nationwide population-based study. Clin Gastroenterol Hepatol. 2021;19(10):2072–2081.e6. doi: 10.1016/j.cgh.2020.07.049 [DOI] [PubMed] [Google Scholar]
  • 18.Nicholas MN, Keown-Stoneman CDG, Maguire JL. Association between atopic dermatitis and height, body mass index, and weight in children. JAMA Dermatol. 2022;158(1):26–32. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Burgess S, Dudbridge F, Thompson SG. Combining information on multiple instrumental variables in Mendelian randomization: comparison of allele score and summarized data methods. Stat Med. 2016;35(11):1880–906. doi: 10.1002/sim.6835 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Burgess S, Butterworth A, Thompson SG. Mendelian randomization analysis with multiple genetic variants using summarized data. Genet Epidemiol. 2013;37(7):658–65. doi: 10.1002/gepi.21758 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Bowden J, Davey Smith G, Haycock PC, Burgess S. Consistent estimation in mendelian randomization with some invalid instruments using a weighted median estimator. Genet Epidemiol. 2016;40(4):304–14. doi: 10.1002/gepi.21965 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Milne RL, Kuchenbaecker KB, Michailidou K. Identification of ten variants associated with risk of estrogen-receptor-negative breast cancer. Nat Genet. 2017;49(12):1767–78. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Bowden J, Davey Smith SM, Burgess S. Mendelian randomization with invalid instruments: effect estimation and bias detection through Egger regression. Int J Epidemiol. 2015;44(2):512–25. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Verbanck M, Chen C-Y, Neale B, Do R. Detection of widespread horizontal pleiotropy in causal relationships inferred from Mendelian randomization between complex traits and diseases. Nat Genet. 2018;50(5):693–8. doi: 10.1038/s41588-018-0099-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Burgess S, Davies NM, Thompson SG. Bias due to participant overlap in two-sample Mendelian randomization. Genet Epidemiol. 2016;40(7):597–608. doi: 10.1002/gepi.21998 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Garzorz-Stark N, Krause L, Lauffer F, Atenhan A, Thomas J, Stark SP, et al. A novel molecular disease classifier for psoriasis and eczema. Exp Dermatol. 2016;25(10):767–74. doi: 10.1111/exd.13077 [DOI] [PubMed] [Google Scholar]
  • 27.Coimbra S, Santos-Silva A. A specific molecular signature for psoriasis and eczema. Ann Transl Med. 2015;3(6):76. doi: 10.3978/j.issn.2305-5839.2015.01.24 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Henseler T, Christophers E. Disease concomitance in psoriasis. J Am Acad Dermatol. 1995;32(6):982–6. doi: 10.1016/0190-9622(95)91336-x [DOI] [PubMed] [Google Scholar]
  • 29.Christophers E, Henseler T. Contrasting disease patterns in psoriasis and atopic dermatitis. Arch Dermatol Res. 1987;279 Suppl:S48–51. doi: 10.1007/BF00585919 [DOI] [PubMed] [Google Scholar]
  • 30.Kirsten N, Mohr N, Maul J-T, Augustin M. Incidence of atopic conditions in people with psoriasis: a population-based analysis. Eur J Dermatol. 2021;31(1):60–4. doi: 10.1684/ejd.2021.3963 [DOI] [PubMed] [Google Scholar]
  • 31.Dai Y-X, Tai Y-H, Chang Y-T, Chen T-J, Chen M-H. Bidirectional association between psoriasis and atopic dermatitis: a nationwide population-based cohort study. Dermatology. 2021;237(4):521–7. doi: 10.1159/000514581 [DOI] [PubMed] [Google Scholar]
  • 32.Patrick MT, Stuart PE, Raja K, Gudjonsson JE, Tejasvi T, Yang J, et al. Genetic signature to provide robust risk assessment of psoriatic arthritis development in psoriasis patients. Nat Commun. 2018;9(1):4178. doi: 10.1038/s41467-018-06672-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Tsoi LC, Stuart PE, Tian C, Gudjonsson JE, Das S, Zawistowski M, et al. Large scale meta-analysis characterizes genetic architecture for common psoriasis associated variants. Nat Commun. 2017;8:15382. doi: 10.1038/ncomms15382 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Tsai YC, Tsai TF. Overlapping features of psoriasis and atopic dermatitis: from genetics to immunopathogenesis to phenotypes. Int J Mol Sci. 2022;23(10). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Kemény L, Varga E, Novak Z. Advances in phototherapy for psoriasis and atopic dermatitis. Expert Rev Clin Immunol. 2019;15(11):1205–14. doi: 10.1080/1744666X.2020.1672537 [DOI] [PubMed] [Google Scholar]
  • 36.Schäbitz A, Eyerich K, Garzorz-Stark N. So close, and yet so far away: The dichotomy of the specific immune response and inflammation in psoriasis and atopic dermatitis. J Intern Med. 2021;290(1):27–39. doi: 10.1111/joim.13235 [DOI] [PubMed] [Google Scholar]
  • 37.Noda S, Suárez-Fariñas M, Ungar B, Kim SJ, de Guzman Strong C, Xu H, et al. The Asian atopic dermatitis phenotype combines features of atopic dermatitis and psoriasis with increased TH17 polarization. J Allergy Clin Immunol. 2015;136(5):1254–64. doi: 10.1016/j.jaci.2015.08.015 [DOI] [PubMed] [Google Scholar]
  • 38.von Csiky-Sessoms S, Lebwohl M. What’s new in psoriasis. Dermatol Clin. 2019;37(2):129–36. doi: 10.1016/j.det.2018.11.001 [DOI] [PubMed] [Google Scholar]
  • 39.Chen WY, Chen SC, Hsu SY. Annoying psoriasis and atopic dermatitis: a narrative review. Int J Mol Sci. 2022;23(9). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Prinz I, Sandrock I, Mrowietz U. Interleukin-17 cytokines: effectors and targets in psoriasis-a breakthrough in understanding and treatment. J Exp Med. 2020;217(1). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Griffiths CEM, Armstrong AW, Gudjonsson JE. Lancet. 2021;397(10281):1301–15. [DOI] [PubMed] [Google Scholar]
  • 42.Mcgeachy MJ, Cua DJ, Gaffen SL. The IL-17 family of cytokines in health and disease. Immunity, 2019, 50(4): 892–906. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Johnston A, Fritz Y, Dawes SM, Diaconu D, Al-Attar PM, Guzman AM, et al. Keratinocyte overexpression of IL-17C promotes psoriasiform skin inflammation. J Immunol. 2013;190(5):2252–62. doi: 10.4049/jimmunol.1201505 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Xing X, Liang Y, Sarkar MK. IL-17 responses are the dominant inflammatory signal linking inverse, erythrodermic, and chronic plaque psoriasis. J Invest Dermatol. 2016;136(12):2498–501. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Mahil SK, Capon F, Barker JN. Update on psoriasis immunopathogenesis and targeted immunotherapy. Semin Immunopathol. 2016;38(1):11–27. doi: 10.1007/s00281-015-0539-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46.Stuart PE, Nair RP, Tsoi LC. Genome-wide association analysis of psoriatic arthritis and cutaneous psoriasis reveals differences in their genetic architecture. Am J Hum Genet. 2015;97(6):816–36. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 47.Stuart PE, Nair RP, Ellinghaus E, Ding J, Tejasvi T, Gudjonsson JE, et al. Genome-wide association analysis identifies three psoriasis susceptibility loci. Nat Genet. 2010;42(11):1000–4. doi: 10.1038/ng.693 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48.Köhler I, Bivik EC, Kasic NK. NOS2-derived low levels of NO drive psoriasis pathogenesis. Cell Death Dis. 2024;15(6):449. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49.He H, Bissonnette R, Wu J, Diaz A, Saint-Cyr Proulx E, Maari C, et al. Tape strips detect distinct immune and barrier profiles in atopic dermatitis and psoriasis. J Allergy Clin Immunol. 2021;147(1):199–212. doi: 10.1016/j.jaci.2020.05.048 [DOI] [PubMed] [Google Scholar]
  • 50.Garzorz N, Eyerich K. NOS2 and CCL27: clinical implications for psoriasis and eczema diagnosis and management. Expert Rev Clin Immunol. 2015;11(2):167–9. doi: 10.1586/1744666X.2015.996549 [DOI] [PubMed] [Google Scholar]
  • 51.Rendon A, Schäkel K. Psoriasis pathogenesis and treatment. Int J Mol Sci. 2019;20(6). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 52.Paternoster L, Standl M, Waage J, Baurecht H, Hotze M, Strachan DP, et al. Multi-ancestry genome-wide association study of 21,000 cases and 95,000 controls identifies new risk loci for atopic dermatitis. Nat Genet. 2015;47(12):1449–56. doi: 10.1038/ng.3424 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 53.Zhao Y, Terron-Kwiatkowski A, Liao H, Lee SP, Allen MH, Hull PR, et al. Filaggrin null alleles are not associated with psoriasis. J Invest Dermatol. 2007;127(8):1878–82. doi: 10.1038/sj.jid.5700817 [DOI] [PubMed] [Google Scholar]
  • 54.Bergboer JGM, Zeeuwen PLJM, Irvine AD, Weidinger S, Giardina E, Novelli G, et al. Deletion of late cornified envelope 3B and 3C genes is not associated with atopic dermatitis. J Invest Dermatol. 2010;130(8):2057–61. doi: 10.1038/jid.2010.88 [DOI] [PubMed] [Google Scholar]
  • 55.Hu Z, Xiong Z, Xu X, Li F, Lu L, Li W, et al. Loss-of-function mutations in filaggrin gene associate with psoriasis vulgaris in Chinese population. Hum Genet. 2012;131(7):1269–74. doi: 10.1007/s00439-012-1155-5 [DOI] [PubMed] [Google Scholar]
  • 56.Chang Y-C, Wu W-M, Chen C-H, Hu C-F, Hsu L-A. Association between P478S polymorphism of the filaggrin gene and risk of psoriasis in a Chinese population in Taiwan. Arch Dermatol Res. 2008;300(3):133–7. doi: 10.1007/s00403-007-0821-2 [DOI] [PubMed] [Google Scholar]
  • 57.Bowcock AM, Cookson WOCM. The genetics of psoriasis, psoriatic arthritis and atopic dermatitis. Hum Mol Genet. 2004;13 Spec No 1:R43–55. doi: 10.1093/hmg/ddh094 [DOI] [PubMed] [Google Scholar]
  • 58.Tamari M, Saeki H, Hayashi M, Umezawa Y, Ito T, Fukuchi O, et al. An association study of 36 psoriasis susceptibility loci for psoriasis vulgaris and atopic dermatitis in a Japanese population. J Dermatol Sci. 2014;76(2):156–7. doi: 10.1016/j.jdermsci.2014.08.005 [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

S1 Table. Instrumental variables used in MR analysis of the association between psoriasis and AD.

Detailed information of SNPs used in MR analysis: From psoriasis to AD.

(XLSX)

pone.0322602.s001.xlsx (12.5KB, xlsx)
S2 Table. Detailed information of the sensitive analyses used in MR analysis of the association between psoriasis and AD.

Cochran’s Q test, MR Egger test and MR PRESSO global test used in this study: From psoriasis to AD.

(XLSX)

pone.0322602.s002.xlsx (11.6KB, xlsx)
S3 Table. Strong instrumental variables correlated sensitive analyses used in MR analysis of the association between psoriasis and AD.

Detailed information of strong instruments group in subgroup analyses: From psoriasis to AD.

(XLSX)

pone.0322602.s003.xlsx (13.9KB, xlsx)
S4 Table. Weak instrumental variables correlated sensitive analyses used in MR analysis of the association between psoriasis and AD.

Detailed information of weak instruments group in subgroup analyses: From psoriasis to AD.

(XLSX)

pone.0322602.s004.xlsx (14.3KB, xlsx)
S5 Table. Instrumental variables used in MR analysis of the association between AD and psoriasis.

Detailed information of SNPs used in MR analysis: From AD to psoriasis.

(XLSX)

pone.0322602.s005.xlsx (12KB, xlsx)
S6 Table. Detailed information of the sensitive analyses used in MR analysis of the association between AD and psoriasis.

Cochran’s Q test, MR Egger test and MR PRESSO global test used in this study: From AD to psoriasis.

(XLSX)

pone.0322602.s006.xlsx (11.7KB, xlsx)

Data Availability Statement

All relevant data are within the manuscript and its Supporting information files.

Articles from PLOS One are provided here courtesy of PLOS

RESOURCES