Skip to main content
NCBI home page
Health Science Reports logoLink to Health Science Reports
. 2025 Sep 7;8(9):e71040. doi: 10.1002/hsr2.71040

Genetic Association of Lipids and Lipid‐Lowering Drug Target Genes With Atopic Dermatitis: A Drug Target Mendelian Randomization Study

Min Luo 1, Jinger Lin 1, Qianwei Zhuo 1, Lihang Lin 1,✉, Yue Han 1,✉
PMCID: PMC12415263  PMID: 40927812

ABSTRACT

Background and Aims

Several observational studies have reported inconsistent associations between dyslipidaemia, stains use and atopic dermatitis (AD). Nevertheless, the available data on the effects of –C‐lowering as well as TG‐lowering drugs remain inconclusive and limited. The aim of this study was to evaluate the causal association of lipid traits and long‐term use of lipid‐lowering drugs on AD risk.

Methods

Drug‐targeted Mendelian randomization analyses were performed in European ancestry. Pooled statistics on low‐density lipoprotein cholesterol, triglyceride and AD were extracted from large genome‐wide association studies datasets. Instrumental variables located in and around lipid‐lowering target genes were used as proxies for therapeutic inhibition of these target genes. Inverse‐variance‐weighted approach was applied as the primary analysis.

Results

We explored the role of seven lipid‐lowering target genes in AD, among which genetically proxied Niemann‐Pick C1–like 1 inhibition, a target of LDL‐C‐lowering drugs, was associated with an increased risk of AD (odds ratio, 3.03; 95% CI, 1.36–6.75; p = 0.007). This association was replicated in FinnGen cohort (odds ratio, 1.72; 95% CI, 1.09–2.72; p = 0.020). A series of sensitivity analyses confirmed the robustness of the estimates.

Conclusion

We found no genetic support for the repurposing of seven lipid‐lowering drug targets for the prevention of AD. However, pharmacovigilance of AD risk in Niemann‐Pick C1–like 1 inhibitor users may be warranted.

Keywords: atopic dermatitis, LDL‐C, Mendelian randomization, statins, TG

Abbreviations

AD

atopic dermatitis

APOB

apolipoprotein B‐100

APOC3

apolipoprotein C‐III

CAD

coronary artery disease

CI

confidence interval

EAGLE

Early Genetics and Lifecourse Epidemiology

GLGC

Global Lipids Genetics Consortium

GWAS

Genome‐Wide Association Study

HMGCR

HMG‐CoA reductase

IL

interleukin

IV

instrumental variable

IVW

inverse‐variance weighted

LDLR

LDL receptor

LDL‐C

low‐density lipoprotein cholesterol

LPL

lipoprotein lipase

MR

Mendelian randomization

NPC1L1

Niemann‐Pick C1‐like protein 1

OR

odds ratio

PPARA

peroxisome proliferator‐activated receptor‐a

RCT

randomized controlled trial

SNP

single‐nucleotide polymorphism

TG

triglyceride

Th

T helper

1. Introduction

Atopic dermatitis (AD) is a recurrent and pruritic inflammatory skin disease characterized by skin barrier dysfunction and immune dysregulation in genetically susceptible individuals [1]. Recent studies have suggested a potential link between AD and disordered lipid metabolism, which may contribute to an increased risk of cardiovascular disease [2, 3, 4]. However, the existing evidence regarding these associations remains inconsistent [5, 6].

Statins are commonly used to lower low‐density lipoprotein cholesterol (LDL‐C) levels and are considered the standard treatment for hypercholesterolemia in clinical practice [7]. However, reports of certain dermatologic complications associated with statin use have raised concerns in the dermatological community [8, 9, 10]. However, Cheung et al. noted that their criteria included unspecified dermatitis, a potential confounder that may not accurately reflect eczema. On the other hand, a research has suggested that fluvastatin may alleviate skin inflammation in patients with AD [11]. Nevertheless, the available data on the effects of other classes of LDL‐C‐lowering and triglyceride (TG)‐lowering drugs, such as Ezetimibe, Mipomersen, and Fibrates, remain inconclusive and limited. Understanding the causal pathways related to lipid metabolism could enhance our understanding of the underlying mechanisms driving AD development and facilitate the development of more effective interventions for AD patients in high‐risk individuals with dyslipidemia.

Mendelian randomization (MR) takes advantage of the random inheritance of genotypes at conception, similar to the allocation in a randomized controlled trial (RCT), which reduces susceptibility to confounding and reverse causation biases [12]. MR is a highly sensitive method to assess whether genetically proxied drug targets are causal and is recommended as a preliminary step before conducting RCTs [13]. Moreover, since genetic variants remain fixed throughout a person's lifetime, MR provides valuable insights into the long‐term consequences of therapeutic exposure, aiding in drug target validation [13].

In this study, we utilized a drug‐target MR analysis to explore the association between genetically predicted lipid traits, including LDL‐C and TG, and seven lipid‐modifying targets. These targets comprised LDL‐C‐lowering target genes, namely LDL receptor (LDLR), HMG‐CoA reductase (HMGCR), Niemann‐Pick C1‐like protein 1 (NPC1L1), and apolipoprotein B‐100 (APOB), and TG‐lowering target genes, namely lipoprotein lipase (LPL), peroxisome proliferator‐activated receptor‐a (PPARA), and apolipoprotein C‐III (APOC3), with the risk of AD.

2. Methods

2.1. Study Design

To investigate the relationship between genetically proxied inhibition of lipid‐lowering drug targets, as well as overall LDL‐C and triglyceride (TG) levels on the risk of AD, we conducted a two‐sample MR analysis by extracting summary statistics from large‐scale genome‐wide association studies (GWASs) with European ancestry. The overview of the study design is shown in Figure 1. The summary‐level GWAS data used in this study were publicly available and had obtained approvals from their respective Institutional Review Boards. More detailed information about the data sources can be found in Supporting Information S1: Table S1.

Figure 1.

Figure 1

Open in a new tab

Study overview. AD, atopic dermatitis; APOB, apolipoprotein B‐100; APOC3, apolipoprotein C‐III; EAGLE, Early Genetics and Lifecourse Epidemiology; HMGCR, HMG‐CoA reductase; IVW, inverse‐variance weighted; LDL‐C, low‐density lipoprotein cholesterol; LDLR, LDL receptor; LPL, lipoprotein lipase; MR, Mendelian randomization; NPC1L1, Niemann‐Pick C1‐like protein 1; PPARA, peroxisome proliferator‐activated receptor‐a; TG, triglycerides.

2.2. Genetic Association for Atopic Dermatitis

In our primary analysis, the genetic instrumental variables (IVs) for AD were obtained from a GWAS meta‐analysis conducted by the Early Genetics and Lifecourse Epidemiology (EAGLE) eczema consortium. This GWAS meta‐analysis involved 20 population‐based cohorts comprising 10,788 cases and 30,047 controls, excluding the 23andMe study [14]. To replicate the analyses, we extracted AD summary‐level GWAS data involving 13,473 cases and 336,589 controls from the latest FinnGen study (release 9) [15].

2.3. Genetic Proxies for Lipid‐Lowering Drugs

We selected lipid‐lowering drugs based on recent guidelines for the management of dyslipidaemia [16, 17]. The pharmacological target genes for each class of lipid‐lowering drugs were identified using the DrugBank database (https://go.drugbank.com/) and references to previous reviews [18, 19].

These target genes were further classified into LDL‐C‐lowering target genes, namely LDLR, HMGCR, NPC1L1, and APOB, and TG‐lowering target genes, namely LPL, PPARA, and APOC3, based on their primary pharmacological effects. The description of the selected lipid‐lowering drugs and their corresponding target genes is listed in Table 1.

Table 1.

Information on lipid‐lowering drug targets.

Primary pharmacological action Examples of drugs/Class Drug targets Target genes Gene region (GRCh37/hg19 by Ensembl)
LDL‐C‐lowering LDL Receptor LDLR chr19:11,200,038‐11,244,492
Statins HMG‐coA reductase HMGCR chr5:74,632,154‐74,657,929
Ezetimibe Niemann‐Pick C1‐like protein 1 NPC1L1 chr7:44,552,134‐44,580,914
Mipomersen Apolipoprotein B‐100 APOB chr2:21,224,301‐21,266,945
TG‐lowering Lipoprotein Lipase LPL chr8:19,759,228‐19,824,769
Fibrates Peroxisome proliferator‐activated receptor‐a PPARA chr22:46,546,424‐46,639,653
Volanesorsen Apolipoprotein C‐III APOC3 chr11:116,700,422‐116,703,788
Open in a new tab

Abbreviations: APOB, apolipoprotein B‐100; APOC3, apolipoprotein C‐III; chr, chromosome; HMGCR, HMG‐CoA reductase; LDL‐C, low‐density lipoprotein cholesterol, LDLR, LDL receptor; LPL, lipoprotein lipase; NPC1L1, Niemann‐Pick C1‐like protein 1; PPARA, peroxisome proliferator‐activated receptor‐a; TG, triglyceride.

Genetic IVs for lipid‐lowering drug targets were obtained from the largest and most recent GWAS conducted by the Global Lipids Genetics Consortium (GLGC) on TG levels (n = 1,320,016) and LDL‐C (n = 1,320,016) [20]. To mitigate potential biases arising from sample overlap, we specifically extracted genetic associations for LDL‐C (n = 1,177,987) and TG (n = 1,177,987) from GLGC and excluded participants of Finnish descent, which were then used in our replication analysis.

The IVs for genetically proxied target genes were significantly associated with LDL‐C or TG at the genome‐wide significance level (p < 5 × 10−8) and showed minimal correlation (linkage disequilibrium r 2 < 0.1). We used PLINK and Phase 3 of the European 1000 Genomes Project as a reference panel to locate the single‐nucleotide polymorphisms (SNPs) within a 100 kb region surrounding the respective target genes [21]. Furthermore, we examined whether genetically predicted overall LDL‐C or TG levels were associated with the risk of AD. The IVs for overall LDL‐C or TG levels were identified at the genome‐wide significance level and showed pairwise correlations of r 2 < 0.001.

2.4. Positive Control Analyses

As a positive control, we included an analysis using coronary artery disease (CAD) as the outcome. The summary GWAS data for CAD were obtained from the CARDIoGRAMplusC4D consortium, which included 60,801 cases and 123,504 controls [22]. The inclusion of CAD in the analysis served to validate the instrument validity.

2.5. Mendelian Randomization Analysis

The F‐statistic was calculated to measure the strength of each IV, with values less than 10 indicating that the IV is weak, which may bias the results of the MR [23]. The F‐statistics was calculated by the following formula: F = (beta^2)/(SE^2) [24]. R 2 is the proportion of AD variability explained by each instrument, and R 2 was calculated using the 2 × beta^2 × EAF × (1 − EAF) [25]. To ensure consistency in effect estimates, we harmonized the exposure and outcome SNPs. We also excluded potentially ambiguous SNPs with inconsistent alleles and palindromic SNPs from the analysis.

The multiplicative random‐effect inverse‐variance weighted (IVW) method was used as the main analysis method [26]. In addition, MR‐Egger and weighted median methods were applied as complementary MR methods [27]. The MR‐Egger method provides more robust estimates in the presence of pleiotropy, although with larger confidence intervals (CIs) [28, 29]. It is statistically less efficient, meaning that even if a strong relationship is observed using the IVW method, the estimates provided by the MR‐Egger method are likely to be near‐null.

2.6. Sensitivity Analyses

To verify the robustness of the results, we performed sensitivity analyses. Heterogeneity was assessed using Cochran's Q statistics for the IVW method, with a p‐value < 0.05 indicating significant heterogeneity [26]. The presence of horizontal pleiotropy was assessed by examining the intercept term of the MR‐Egger regression, with deviation from zero indicating the presence of pleiotropy [28]. We also conducted a leave‐one‐out analysis to assess the influence of individual SNPs on the effect estimates.

2.7. Colocalization

For the NPC1L1 inhibitors, which showed a causal relationship with AD risk in the MR analysis, we conducted colocalization analyses using the coloc method. This method quantifies the probability of a shared genetic variant between gene expression and AD (H4) within a range of ±300 kb of the NPC1L1 gene region [30, 31].

2.8. Statistical Methods

All statistical analyses were conducted using the TwoSampleMR and coloc packages in R software (version 4.2.3). To account for multiple testing, the Bonferroni method was applied to adjust the nominal thresholds (p < 0.05). Specifically, a significance threshold of 0.05/7 = 0.007 was set for the seven drug targets. Relationships with p‐values < 0.05 but Bonferroni‐adjusted p‐values > 0.05 were considered suggestive.

3. Results

3.1. Genetic Instruments for Lipid‐Lowering Drugs

We utilized a total of 40 genetic variants as proxies for inhibiting LDLR, 21 for HMGCR, 11 for NPC1L1, 18 for APOB, 33 for LPL, 3 for PPARA, 24 for APOC3, 390 for total LDL‐C levels, and 454 for TG in the primary analysis.

These variants were identified through the latest GWAS conducted by the GLGC. More information on the specific SNPs used in this study could be referred to Tables S6 and S7. The F‐statistics for all the genetic IVs associated with the drug targets, as well as overall LDL‐C and TG were greater than 10. These statistics indicate that our results are unlikely to be biased due to weak IVs. Furthermore, genetically proxied inhibition of all drug targets showed an association with a reduced risk of coronary artery disease, which served as our positive control outcome (Tables S2 and S3).

3.2. Mendelian Randomization and Sensitivity Analysis

The genetically proxied inhibition of NPC1L1 was associated with an increased risk of developing AD (IVW MR odds ratio (OR) = 3.03, 95% confidence interval (95% CI) = 1.36 to 6.75, p = 0.007 per standard deviation reduction in LDL‐C) (Figure 2). This finding was further supported by the replication analysis using data from the FinnGen study (IVW MR odds ratio (OR) = 1.72; 95% CI, 1.09 to 2.72; p = 0.020) (Figure 3). However, we did not observe a significant association between the expression of inhibition of HMGCR, LDLR, APOB, LPL, PPARA, APOC3, and AD (Figures 2 and 3). To provide a comparative perspective, we also examined the impact of overall circulating LDL‐C and TG levels on AD risk. However, there was limited evidence suggesting a genetic effect of LDL‐C levels or TG on the risk of AD. This finding was consistent with the results of validation analysis using AD data from the FinnGen study.

Figure 2.

Figure 2

Open in a new tab

Associations between genetically proxied lipid‐lowering drugs and atopic dermatitis risk in the primary data set. (A) the causal effect of LDL‐C‐lowering drugs on AD. (B) the causal effect of triglycerides‐lowering drugs on atopic dermatitis. AD, atopic dermatitis; APOB, apolipoprotein B‐100; APOC3, apolipoprotein C‐III; CI, confidence interval; HMGCR, HMG‐CoA reductase; IVW, inverse‐variance weighted; LDLR, LDL receptor; MR, Mendelian randomization; NPC1L1, Niemann‐Pick C1‐like protein 1; LDL‐C, low‐density lipoprotein cholesterol; LPL, lipoprotein lipase; OR, odds ratio; PPARA, peroxisome proliferator‐activated receptor‐a; TG, triglycerides.

Figure 3.

Figure 3

Open in a new tab

Associations between genetically proxied lipid‐lowering drugs and atopic dermatitis risk in the replication data set. (A) the causal effect of LDL‐C‐lowering drugs on AD. (B) the causal effect of TG‐lowering drugs on AD. AD, atopic dermatitis; APOB, apolipoprotein B‐100; APOC3, apolipoprotein C‐III; CI, confidence interval; HMGCR, HMG‐CoA reductase; IVW, inverse‐variance weighted; LDLR, LDL receptor; MR, Mendelian randomization; NPC1L1, Niemann‐Pick C1‐like protein 1; LDL‐C, low‐density lipoprotein cholesterol; LPL, lipoprotein lipase; OR, odds ratio; PPARA, peroxisome proliferator‐activated receptor‐a; TG, triglycerides.

The sensitivity analyses conducted in this study provided no statistical evidence indicating bias from horizontal pleiotropy (Tables S4 and S5). Additionally, the leave‐one‐out analyses suggest that the overall estimate was not influenced by a single variant of significant influence (Figures S1 and S2).

3.3. Colocalization Analysis

The probability of a causal variant associated with blood LDL‐C or AD risk within the NPC1L1 gene was found to be 90.8%, and 0.0%, respectively. The probability of distinct causal variants existing between these two traits was 7.6%, while the probability of a shared causal variant was 1.6% (Table S8 and Figure S3).

4. Discussion

In this MR study, inhibition of lipid‐lowering drug targets (HMGCR, LDLR, APOB, LPL, PPARA, and APOC3) was found not to be significantly associated with the risk of AD, possibly through a mechanism other than target effects on LDL‐C or TG, as no relationship was identified between overall LDL or TG levels and AD risk. Additionally, the study suggests a possible increased susceptibility to AD in individuals exposed to NPC1L1 inhibitors.

The available observational data have yielded inconsistent findings regarding the relationship between blood lipid traits and AD [3, 6, 32, 33, 34]. Adipose tissue secretes soluble adipokines, which can influence inflammatory and immune responses. These responses, in turn, stimulate the production of pro‐inflammatory cytokines, contributing to a chronic inflammatory state [35]. Dyslipidemia can shift the immune response from Th1 to a stronger Th2 response, creating an imbalance between anti‐ and pro‐inflammatory cytokines, such as interleukin (IL)‐10, IL‐17, and IL‐23, respectively, and this imbalance may exacerbate persistent inflammation seen in atopic conditions [36, 37]. These findings support the interpretation that dyslipidemia may contribute to chronic skin inflammation, such as AD. Nevertheless, it is important to recognize that observational epidemiology is susceptible to biases from confounding and reverse causation. Based on the available evidence, this study does not establish blood LDL‐C and TG levels as causal risk factors for AD, suggesting that primary dyslipidemia may not play a significant role in the development of AD.

Statins are widely used drugs that effectively lower cholesterol levels by inhibiting the enzyme HMG‐CoA reductase. Recent studies have highlighted their pleiotropic effects, including anti‐inflammatory and immunomodulatory properties [38]. One of the notable effects of statins is their ability to inhibit the differentiation of T cells into the pro‐inflammatory Th17 phenotype while promoting the formation of anti‐inflammatory T regulatory cells [39]. They also regulate Th2 responses mediated by dendritic cells and suppress the infiltration of Th2 cells into skin lesions, thereby reducing skin inflammation [11, 40]. These findings suggest that statins may have therapeutic potential in allergic diseases such as asthma and AD [41].

Glycerophospholipids, sphingolipids, and fatty acids are implicated in inflammation associated with AD. In AD patients, phosphatidylcholine is downregulated in both serum and lesional skin, but its levels increase following dupilumab treatment [42, 43, 44]. Linoleic acid is an omega‐6 PUFA, is found at elevated plasma concentrations in individuals with AD [45].

Considering the mechanism of action of statins, which involves reducing cholesterol, it is important to investigate the relationship between statins and epidermal lipids further. This MR study did not find an increased or decreased risk of AD with long‐term exposure to statins. Consequently, this discovery effectively alleviates previous concerns about the use of statins in AD patients. On the other hand, these MR results do not support the need for RCTs to explore the repurposing of HMGCR inhibitors for the treatment of AD.

Ezetimibe is a drug used to lower lipid levels by blocking the intestinal cholesterol transporter NPC1L1. Its mechanism of action reduces blood cholesterol levels [46]. Studies have shown that ezetimibe has anti‐inflammatory effects independent of its lipid‐lowering properties. It inhibits the differentiation of Th17 cells in a mouse model of ankylosing spondylitis [47]. However, the effect of NPC1L1 inhibitors on patients with AD has not been reported. In our study using MR analysis, we found a causal suggestive relationship between genetic variations in NPC1L1 and AD. Traditionally, AD has been considered a disorder that primarily affects early childhood, but recent studies have revealed that its prevalence in the adult population is considerable, suggesting that AD may be a chronic condition persisting throughout life [48]. Therefore, it is advisable to prioritise other lipid‐lowering therapies for patients susceptible to AD. Although the efficacy of NPC1L1 inhibitors as lipid‐lowering agents is well documented, these preliminary findings must be interpreted with caution until more extensive data are available.

5. Strengths and Limitations

This MR study has several strengths. The main advantages are the robustness of results to replication (using data from EAGLE eczema consortium and the FinnGen study) and the utilization of large‐scale GWAS datasets. MR study is relatively straightforward, cost‐effective, and time‐efficient compared to other study designs [49]. Conducting cohort studies to investigate the association between lipid‐lowering drugs and AD risk and prognosis would be challenging due to indication bias, reverse causation, and limitations in patient population size [50].

It is important to acknowledge some limitations when interpreting the results of this study. Firstly, the genetic variants used in this study reflect the long‐term consequences of lipid levels on AD risk, and their effect size may not directly reflect the short‐term impact of lipid‐lowering drugs [51]. Secondly, the original GWAS data were not stratified by specific AD subtypes or age groups, which limits our ability to perform stratified analyses. However, this remains an area of ongoing research interest, and future studies using specific datasets for these subgroups should consider stratified analysis. Lastly, since our study predominantly includes individuals of European ancestry, caution should be exercised when generalizing our findings to other ethnic groups.

6. Conclusions

In conclusion, our research indicates that levels of circulating LDL‐C and TG levels were not associated with AD risk. Meanwhile, this study did not find an increased or decreased risk of AD with long‐term exposure to six lipid‐lowering drugs, namely LDLR, HMGCR, NPC1L1, APOB, LPL, PPARA, and APOC3. Furthermore, our study identifies a potential link between inhibitors of NPC1L1 and an increased risk of AD, highlighting the importance of pharmacovigilance among users of NPC1L1 inhibitors. In patients with a predisposition to AD, it is recommended that other lipid‐lowering treatments be considered. To validate this finding, larger GWAS datasets and additional potentially associated genetic variants need to be investigated. Future research should prioritize elucidating the underlying mechanisms involved in this potential association.

Author Contributions

Min Luo: conceptualization, data curation, methodology, writing – original draft. Jinger Lin: investigation and methodology. Qianwei Zhuo: investigation and methodology. Lihang Lin: conceptualization, writing – review and editing. Yue Han: conceptualization, validation, writing – review and editing.

Ethics Statement

Data used in this study were obtained from publicly available genome‐wide association studies. All included original studies were approved by the corresponding ethical standards committees. Therefore, no additional ethical approval or informed consent was required for this study.

Conflicts of Interest

The authors declare no conflicts of interest.

Transparency Statement

The lead author Lihang Lin, Yue Han affirms that this manuscript is an honest, accurate, and transparent account of the study being reported; that no important aspects of the study have been omitted; and that any discrepancies from the study as planned (and, if relevant, registered) have been explained.

Supporting information

Figure S1: Leave‐one‐out analysis for genetically proxied lipid‐lowering drugs on AD in the primary analysis. Figure S2: Leave‐one‐out analysis for genetically proxied lipid‐lowering drugs on AD in the replication analysis. Figure S3: Genetic associations with low‐density lipoprotein cholesterol and atopic dermatitis within the NPC1L1 gene. Table S1: Details of data sources used in this study. Table S2: Associations between genetically proxied lipid‐lowering drugs and coronary artery disease in the primary analysis. Table S3: Associations between genetically proxied lipid‐lowering drugs and coronary artery disease in the replication analysis. Table S4: Results of heterogeneity and pleiotropy in the primary analysis. Table S5: Results of heterogeneity and pleiotropy in the replication analysis. Table S6: Final instrumental variables used in the primary analysis. Table S7: Final instrumental variables used in the replication analysis. Table S8: Colocalization analysis of low‐density lipoprotein cholesterol and atopic dermatitis in the NPC1L1 gene.

HSR2-8-e71040-s001.docx (1.2MB, docx)

Acknowledgments

National Natural Science Foundation of China (8220120391) and Fujian Province Natural Science Foundation (2022J01745).

Contributor Information

Lihang Lin, Email: 460879404@qq.com.

Yue Han, Email: dr_hanyue@126.com.

Data Availability Statement

The data that support the findings of this study are publicly available from Global Lipids Genetics Consortium (http://csg.sph.umich.edu/willer/public/glgc-lipids2021/), the IEU OpenGWAS database(https://gwas.mrcieu.ac.uk/datasets/ieu-a-996/), the FinnGen study (https://finngen.gitbook.io/documentation/data-download), and the CARDIoGRAMplusC4D consortium (http://www.cardiogramplusc4d.org/data-downloads/).

References

  • 1. Ständer S., “Atopic Dermatitis,” New England Journal of Medicine 384, no. 12 (2021): 1136–1143. [DOI] [PubMed] [Google Scholar]
  • 2. Lundin S., Wahlgren C. F., Johansson E. K., et al., “Childhood Atopic Dermatitis Is Associated With Cardiovascular Risk Factors in Young Adulthood‐A Population‐Based Cohort Study,” Journal of the European Academy of Dermatology and Venereology 37, no. 9 (2023): 1854–1862. [DOI] [PubMed] [Google Scholar]
  • 3. Kim J. H., Lee S. W., Yon D. K., et al., “Association of Serum Lipid Parameters With the SCORAD Index and Onset of Atopic Dermatitis In Children,” Pediatric Allergy and Immunology 32, no. 2 (2021): 322–330. [DOI] [PubMed] [Google Scholar]
  • 4. Ivert L., Johansson E., Dal H., Lindelöf B., Wahlgren C., and Bradley M., “Association Between Atopic Dermatitis and Cardiovascular Disease: A Nationwide Register‐Based Case‐Control Study From Sweden,” Acta Dermato Venereologica 99, no. 10 (2019): 865–870. [DOI] [PubMed] [Google Scholar]
  • 5. Tang Z., Shen M., Xiao Y., Liu H., and Chen X., “Association Between Atopic Dermatitis, Asthma, and Serum Lipids: A UK Biobank Based Observational Study and Mendelian Randomization Analysis,” Frontiers in Medicine 9 (2022): 810092. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6. Ali Z., Ulrik C. S., Agner T., and Thomsen S. F., “Association Between Atopic Dermatitis and the Metabolic Syndrome: A Systematic Review,” Dermatology 234, no. 3–4 (2018): 79–85. [DOI] [PubMed] [Google Scholar]
  • 7. Wilson P. W. F., Polonsky T. S., Miedema M. D., Khera A., Kosinski A. S., and Kuvin J. T., “Systematic Review for the 2018 AHA/ACC/AACVPR/AAPA/ABC/ACPM/ADA/AGS/APhA/ASPC/NLA/PCNA Guideline on the Management of Blood Cholesterol: A Report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines,” Circulation 139, no. 25 (2019): e1144–e1161. [DOI] [PubMed] [Google Scholar]
  • 8. Frishman W. H., Brosnan B. D., Grossman M., Dasgupta D., and Sun D. K., “Adverse Dermatologic Effects of Cardiovascular Drug Therapy: Part III,” Cardiology in Review 10, no. 6 (2002): 337–348. [DOI] [PubMed] [Google Scholar]
  • 9. Salna M. P., Singer H. M., and Dana A. N., “Pravastatin‐Induced Eczematous Eruption Mimicking Psoriasis,” Case Reports in Dermatological Medicine 2017 (2017): 3418204. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10. Cheung K., Powers E. M., McKillip J., and Powers J. G., “Effect of Statin Use on Incidence of Eczema and Atopic Dermatitis: A Retrospective Cohort Study,” Journal of the American Academy of Dermatology 84, no. 2 (2021): 534–535. [DOI] [PubMed] [Google Scholar]
  • 11. Qi X.‐F., Kim D.‐H., Yoon Y.‐S., et al., “Fluvastatin Inhibits Expression of the Chemokine MDC/CCL22 Induced by Interferon‐γ in HaCaT Cells, a Human Keratinocyte Cell Line,” British Journal of Pharmacology 157, no. 8 (2009): 1441–1450. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12. Gill D., Georgakis M. K., Walker V. M., et al., “Mendelian Randomization for Studying the Effects of Perturbing Drug Targets,” Wellcome Open Research 6 (2021): 16. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13. Roberts R., “Mendelian Randomization Studies Promise to Shorten the Journey to FDA Approval,” JACC: Basic to Translational Science 3, no. 5 (2018): 690–703. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14. Paternoster L., Standl M., Waage J., et al, “Multi‐Ancestry Genome‐Wide Association Study of 21,000 Cases and 95,000 Controls Identifies New Risk Loci for Atopic Dermatitis,” Nature Genetics 47, no. 12 (2015): 1449–1456. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15. Kurki M. I., Karjalainen J., Palta P., et al, “Finngen Provides Genetic Insights From a Well‐Phenotyped Isolated Population,” Nature 613, no. 7944 (2023): 508–518. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16. Mach F., Baigent C., Catapano A. L., et al, “2019 ESC/EAS Guidelines for the Management of Dyslipidaemias: Lipid Modification to Reduce Cardiovascular Risk,” European Heart Journal 41, no. 1 (2020): 111–188. [DOI] [PubMed] [Google Scholar]
  • 17. Grundy S. M., Stone N. J., Bailey A. L., et al., “2018 AHA/ACC/AACVPR/AAPA/ABC/ACPM/ADA/AGS/APhA/ASPC/NLA/PCNA Guideline on the Management of Blood Cholesterol: A Report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines,” Circulation 139, no. 25 (2019): e1082–e1143. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18. Borén J., Taskinen M.‐R., Björnson E., and Packard C. J., “Metabolism of Triglyceride‐Rich Lipoproteins in Health and Dyslipidaemia,” Nature Reviews Cardiology 19, no. 9 (2022): 577–592. [DOI] [PubMed] [Google Scholar]
  • 19. Ridker P. M., “LDL Cholesterol: Controversies and Future Therapeutic Directions,” Lancet 384, no. 9943 (2014): 607–617. [DOI] [PubMed] [Google Scholar]
  • 20. Graham S. E., Clarke S. L., Wu K. H. H., et al, “The Power of Genetic Diversity in Genome‐Wide Association Studies of Lipids,” Nature 600, no. 7890 (2021): 675–679. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21. Auton A., Abecasis G. R., Altshuler D. M., et al, “A Global Reference for Human Genetic Variation,” Nature 526, no. 7571 (2015): 68–74. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22. Nikpay M., Goel A., Won H.‐H., et al, “A Comprehensive 1,000 Genomes‐Based Genome‐Wide Association Meta‐Analysis of Coronary Artery Disease,” Nature Genetics 47, no. 10 (2015): 1121–1130. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23. Burgess S. and Thompson S. G., “Avoiding Bias From Weak Instruments in Mendelian Randomization Studies,” International Journal of Epidemiology 40, no. 3 (2011): 755–764. [DOI] [PubMed] [Google Scholar]
  • 24. Li B. and Martin E. B., “An Approximation to the F Distribution Using the Chi‐Square Distribution,” Computational Statistics & Data Analysis 40, no. 1 (2002): 21–26. [Google Scholar]
  • 25. Papadimitriou N., Dimou N., Tsilidis K. K., et al., “Physical Activity and Risks of Breast and Colorectal Cancer: A Mendelian Randomisation Analysis,” Nature Communications 11, no. 1 (2020): 597. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26. Bowden J., Del Greco M F., Minelli C., Davey Smith G., Sheehan N., and Thompson J., “A Framework for the Investigation of Pleiotropy in Two‐Sample Summary Data Mendelian Randomization,” Statistics in Medicine 36, no. 11 (2017): 1783–1802. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27. Bowden J., Davey Smith G., Haycock P. C., and Burgess S., “Consistent Estimation in Mendelian Randomization With Some Invalid Instruments Using a Weighted Median Estimator,” Genetic Epidemiology 40, no. 4 (2016): 304–314. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28. Bowden J., Davey Smith G., and Burgess S., “Mendelian Randomization With Invalid Instruments: Effect Estimation and Bias Detection Through Egger Regression,” International Journal of Epidemiology 44, no. 2 (2015): 512–525. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29. Burgess S., Davey Smith G., Davies N. M., et al., “Guidelines for Performing Mendelian Randomization Investigations,” Wellcome Open Research 4 (2019): 186. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30. Zuber V., Grinberg N. F., Gill D., et al., “Combining Evidence From Mendelian Randomization and Colocalization: Review and Comparison of Approaches,” American Journal of Human Genetics 109, no. 5 (2022): 767–782. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31. Giambartolomei C., Vukcevic D., Schadt E. E., et al., “Bayesian Test for Colocalisation Between Pairs of Genetic Association Studies Using Summary Statistics,” PLoS Genetics 10, no. 5 (2014): e1004383. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32. Shalom G., Dreiher J., Kridin K., et al., “Atopic Dermatitis and the Metabolic Syndrome: A Cross‐Sectional Study of 116 816 Patients,” Journal of the European Academy of Dermatology and Venereology 33, no. 9 (2019): 1762–1767. [DOI] [PubMed] [Google Scholar]
  • 33. Pavel P., Blunder S., Moosbrugger‐Martinz V., Elias P. M., and Dubrac S., “Atopic Dermatitis: The Fate of the Fat,” International Journal of Molecular Sciences 23, no. 4 (2022): 2121. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34. Seong M. K. and Shin M., “Low‐Density Lipoprotein Cholesterol Is Associated With Atopic Dermatitis in Korean Adolescents,” International Archives of Allergy and Immunology 184 (2023): 1230–1236. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35. De Simoni E., Rizzetto G., Molinelli E., et al, “Metabolic Comorbidities in Pediatric Atopic Dermatitis: A Narrative Review,” Life (Basel) 13, no. 1 (2022): 2, 10.3390/life13010002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36. Robertson A. K. L., Zhou X., Strandvik B., and Hansson G. K., “Severe Hypercholesterolaemia Leads to Strong Th2 Responses to an Exogenous Antigen,” Scandinavian Journal of Immunology 59, no. 3 (2004): 285–293. [DOI] [PubMed] [Google Scholar]
  • 37. Manti S., Leonardi S., Panasiti I., Arrigo T., Salpietro C., and Cuppari C., “Serum IL‐10, IL‐17 and IL‐23 Levels as “Bioumoral Bridges” Between Dyslipidemia and Atopy,” Cytokine 99 (2017): 43–49. [DOI] [PubMed] [Google Scholar]
  • 38. Fessler M. B., “Regulation of Adaptive Immunity in Health and Disease by Cholesterol Metabolism,” Current Allergy and Asthma Reports 15, no. 8 (2015): 48. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39. Ulivieri C. and Baldari C. T., “Statins: From Cholesterol‐Lowering Drugs to Novel Immunomodulators for the Treatment of Th17‐mediated Autoimmune Diseases,” Pharmacological Research 88 (2014): 41–52. [DOI] [PubMed] [Google Scholar]
  • 40. Inagaki‐Katashiba N., Ito T., Inaba M., et al., “Statins Can Suppress DC‐‐Mediated Th2 Responses Through the Repression of OX40‐Ligand and CCL17 Expression,” European Journal of Immunology 49, no. 11 (2019): 2051–2062. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41. Mandal P., Chalmers J. D., Graham C., et al., “Atorvastatin as a Stable Treatment in Bronchiectasis: A Randomised Controlled Trial,” Lancet Respiratory Medicine 2, no. 6 (2014): 455–463. [DOI] [PubMed] [Google Scholar]
  • 42. Zhang L., Wen X., Hou Y., et al., “Integrated Metabolomics and Lipidomics Study of Patients With Atopic Dermatitis in Response to Dupilumab,” Frontiers in Immunology 13 (2022): 1002536. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43. Ottas A., Fishman D., Okas T.‐L., et al., “Blood Serum Metabolome of Atopic Dermatitis: Altered Energy Cycle and the Markers of Systemic Inflammation,” PLoS One 12, no. 11 (2017): e0188580. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44. Yin H., Qiu Z., Zhu R., et al., “Dysregulated Lipidome of Sebum in Patients With Atopic Dermatitis,” Allergy 78, no. 6 (2023): 1524–1537. [DOI] [PubMed] [Google Scholar]
  • 45. Yen C.‐H., Dai Y.‐S., Yang Y.‐H., Wang L.‐C., Lee J.‐H., and Chiang B.‐L., “Linoleic Acid Metabolite Levels and Transepidermal Water Loss in Children With Atopic Dermatitis,” Annals of Allergy, Asthma & Immunology 100, no. 1 (2008): 66–73. [DOI] [PubMed] [Google Scholar]
  • 46. Cannon C. P., Blazing M. A., Giugliano R. P., et al., “Ezetimibe Added to Statin Therapy After Acute Coronary Syndromes,” New England Journal of Medicine 372, no. 25 (2015): 2387–2397. [DOI] [PubMed] [Google Scholar]
  • 47. Moon J., Lee S. Y., Na H. S., et al., “Ezetimibe Ameliorates Clinical Symptoms in a Mouse Model of Ankylosing Spondylitis Associated With Suppression of Th17 Differentiation,” Frontiers in Immunology 13 (2022): 922531. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48. Weidinger S., Beck L. A., Bieber T., Kabashima K., and Irvine A. D., “Atopic Dermatitis,” Nature Reviews Disease Primers 4, no. 1 (2018): 1. [DOI] [PubMed] [Google Scholar]
  • 49. Richmond R. C. and Davey Smith G., “Mendelian Randomization: Concepts and Scope,” Cold Spring Harbor Perspectives in Medicine 12, no. 1 (2022): a040501. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 50. Fewell Z., Davey Smith G., and Sterne J. A. C., “The Impact of Residual and Unmeasured Confounding in Epidemiologic Studies: A Simulation Study,” American Journal of Epidemiology 166, no. 6 (2007): 646–655. [DOI] [PubMed] [Google Scholar]
  • 51. Holmes M. V., Richardson T. G., Ference B. A., Davies N. M., and Davey Smith G., “Integrating Genomics With Biomarkers and Therapeutic Targets to Invigorate Cardiovascular Drug Development,” Nature Reviews Cardiology 18, no. 6 (2021): 435–453. [DOI] [PubMed] [Google Scholar]

Associated Data

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

Supplementary Materials

Figure S1: Leave‐one‐out analysis for genetically proxied lipid‐lowering drugs on AD in the primary analysis. Figure S2: Leave‐one‐out analysis for genetically proxied lipid‐lowering drugs on AD in the replication analysis. Figure S3: Genetic associations with low‐density lipoprotein cholesterol and atopic dermatitis within the NPC1L1 gene. Table S1: Details of data sources used in this study. Table S2: Associations between genetically proxied lipid‐lowering drugs and coronary artery disease in the primary analysis. Table S3: Associations between genetically proxied lipid‐lowering drugs and coronary artery disease in the replication analysis. Table S4: Results of heterogeneity and pleiotropy in the primary analysis. Table S5: Results of heterogeneity and pleiotropy in the replication analysis. Table S6: Final instrumental variables used in the primary analysis. Table S7: Final instrumental variables used in the replication analysis. Table S8: Colocalization analysis of low‐density lipoprotein cholesterol and atopic dermatitis in the NPC1L1 gene.

HSR2-8-e71040-s001.docx (1.2MB, docx)

Data Availability Statement

The data that support the findings of this study are publicly available from Global Lipids Genetics Consortium (http://csg.sph.umich.edu/willer/public/glgc-lipids2021/), the IEU OpenGWAS database(https://gwas.mrcieu.ac.uk/datasets/ieu-a-996/), the FinnGen study (https://finngen.gitbook.io/documentation/data-download), and the CARDIoGRAMplusC4D consortium (http://www.cardiogramplusc4d.org/data-downloads/).

Articles from Health Science Reports are provided here courtesy of Wiley

RESOURCES