Comment on "psoriasis and COVID-19: A Bidirectional Mendelian Randomization Study".

Xiaoyu Gu,Xiang Chen,Minxue Shen
DOI: https://doi.org/10.1016/j.jaad.2022.11.038
IF: 15.487
2022-01-01
Journal of the American Academy of Dermatology
Abstract:To the Editor: We read with interest the study “Psoriasis and COVID-19: A bidirectional Mendelian randomization study” by Chalitsios et al.1Chalitsios C.V. Tsilidis K.K. Tzoulaki I. Psoriasis and COVID-19: a bidirectional Mendelian randomization study.J Am Acad Dermatol. 2022; ([Epub ahead of print]. https://doi.org/10.1016/j.jaad.2022.10.019)Google Scholar Their results did not support the genetic association of psoriasis with COVID-19 that was reported by our study.2Gu X. Chen X. Shen M. Association of psoriasis with risk of COVID-19: a 2-sample Mendelian randomization study.J Am Acad Dermatol. 2022; 87: 715-717Abstract Full Text Full Text PDF PubMed Scopus (7) Google Scholar The discrepancy between our findings and the current study can be explained by the differences with respect to the diagnostic method, study populations, and sample size. Compared with self-reported psoriasis in UK biobank, Chalitsios et al1Chalitsios C.V. Tsilidis K.K. Tzoulaki I. Psoriasis and COVID-19: a bidirectional Mendelian randomization study.J Am Acad Dermatol. 2022; ([Epub ahead of print]. https://doi.org/10.1016/j.jaad.2022.10.019)Google Scholar stated that they included 13,229 doctor-diagnosed psoriasis cases and elaborated a larger sample size to increase the power of test. Besides, the severity of COVID-19 was additionally evaluated in their study. The methodologies and data used by Chalitsios et al1Chalitsios C.V. Tsilidis K.K. Tzoulaki I. Psoriasis and COVID-19: a bidirectional Mendelian randomization study.J Am Acad Dermatol. 2022; ([Epub ahead of print]. https://doi.org/10.1016/j.jaad.2022.10.019)Google Scholar showed some advantages. However, 2 Genome-wide association studies6Tsoi L.C. Spain S.L. Knight J. et al.Identification of 15 new psoriasis susceptibility loci highlights the role of innate immunity.Nat Genet. 2012; 44: 1341-1348Crossref PubMed Scopus (720) Google Scholar, 7Skrivankova V.W. Richmond R.C. Woolf B.A.R. et al.Strengthening the reporting of observational studies in Epidemiology using Mendelian randomization: the STROBE-MR statement.JAMA. 2021; 326: 1614-1621Crossref PubMed Scopus (205) Google Scholar included in the original study did not mention that psoriasis was diagnosed by the dermatologist (about 4094 psoriasis cases).3Nair R.P. Duffin K.C. Helms C. et al.Genome-wide scan reveals association of psoriasis with IL-23 and NF-kappaB pathways.Nat Genet. 2009; 41: 199-204Crossref PubMed Scopus (1089) Google Scholar And the samples from Tsoi et al4Strange A. Capon F. Spencer C.C. et al.A genome-wide association study identifies new psoriasis susceptibility loci and an interaction between HLA-C and ERAP1.Nat Genet. 2010; 42: 985-990Crossref PubMed Scopus (805) Google Scholar overlapped with datasets of Strange et al.7Skrivankova V.W. Richmond R.C. Woolf B.A.R. et al.Strengthening the reporting of observational studies in Epidemiology using Mendelian randomization: the STROBE-MR statement.JAMA. 2021; 326: 1614-1621Crossref PubMed Scopus (205) Google Scholar Therefore, the classification of psoriasis was in fact heterogeneous in their study. In addition, sensitivity analysis is required for MR to assess the robustness of the results.5Tsoi L.C. Stuart P.E. Tian C. et al.Large scale meta-analysis characterizes genetic architecture for common psoriasis associated variants.Nat Commun. 2017; 8: 15382Crossref PubMed Scopus (181) Google Scholar In the study by Chalitsios et al1Chalitsios C.V. Tsilidis K.K. Tzoulaki I. Psoriasis and COVID-19: a bidirectional Mendelian randomization study.J Am Acad Dermatol. 2022; ([Epub ahead of print]. https://doi.org/10.1016/j.jaad.2022.10.019)Google Scholar, they did not report the results of sensitivity analysis such as leave-one-out analysis. Last, the severity of a disease is generally used to evaluate the dose-response relationship. The authors did not evaluate whether psoriasis susceptibility was associated with the severity of COVID-19 by comparing mild, moderate, and severe cases. The limitations of Mendelian randomization (MR) are noteworthy. Constructing a valid instrument variable is challenging, the power of test in MR is low because genetic variance only explains a small proportion of the exposure. The Beavis effect may exist in the results of MR because the associations can be biased by unmeasured confounders. Additionally, MR is a simulation of randomized controlled trials; however, the results of MR are confined to assumptions rather than real-world observations. The inconsistencies between MR and observational studies are commonly observed. Therefore, the association of psoriasis with the risk of COVID-19 warrants further evaluation in large-scale epidemiologic studies. None disclosed. The authors would like to thank all the funders. Response to comment on “Psoriasis and COVID-19: A bidirectional Mendelian randomization study”Journal of the American Academy of DermatologyVol. 88Issue 3PreviewTo the Editor: We want to thank Gu et al1 for their interest in our article2 and comments. We appreciate that Gu et al1 recognized that our data showed some advantages. Full-Text PDF
What problem does this paper attempt to address?