Genetic signature to provide robust risk assessment of psoriatic arthritis development in psoriasis patients

• Type of Material: Article
• Authors: Matthew T. Patrick, University of Michigan; Philip E. Stuart, University of Michigan; Kalpana Raja, University of Michigan; Johann E. Gudjonsson, University of Michigan; Trilokraj Tejasvi, University of Michigan; Jingjing Yang, Emory University; Vinod Chandran, University of Toronto; Sayantan Das, University of Michigan; Kristina Callis-Duffin, University of Utah; Eva Ellinghaus, Christian-Albrechts-Universitat zu Kiel; Charlotta Enerbäck, Linkopings Universitet; Tonu Esko, University of Tartu; Andre Franke, Christian-Albrechts-Universitat zu Kiel; Hyun M. Kang, University of Michigan; Gerald G. Krueger, University of Utah; Henry W. Lim, Henry Ford Hospital; Proton Rahman, Memorial University of Newfoundland; Cheryl F. Rosen, University of Toronto; Stephan Weidinger, Universitatsklinikum Schleswig-Holstein Campus Kiel; Michael Weichenthal, Universitatsklinikum Schleswig-Holstein Campus Kiel; Xiaoquan Wen, University of Michigan; John J. Voorhees, University of Michigan; Goncalo R. Abecasis, University of Michigan; Dafna D. Gladman, University of Toronto; Rajan P. Nair, University of Michigan; James T. Elder, University of Michigan; Lam C. Tsoi, University of Michigan
• Language: English
• Date: 2018-12-01
• Publisher: Nature Research (part of Springer Nature): Fully open access journals
• Publication Version: Final Published Version
• Copyright Statement: © 2018, The Author(s).
• License: Creative Commons Attribution 4.0 International
• Final Published Version (URL): http://dx.doi.org/10.1038/s41467-018-06672-6
• Title of Journal or Parent Work: Nature Communications
• ISSN: 2041-1723
• Volume: 9
• Issue: 1
• Start Page: 4178
• End Page: 4178
• Grant/Funding Information: This work was supported by the Arthritis National Research Foundation and the National Psoriasis Foundation (L.C.T., M.T.P., and K.R.), and awards from the National Institutes of Health (R01AR042742, R01AR050511, R01AR054966, R01AR063611, and R01AR065183 to J.T.E.; K01AR072129 to L.C.T.), as well as a GAIN award from the Foundation for the National Institutes of Health to GRA.; J.T.E. is supported by the Ann Arbor Veterans Affairs Hospital.; L.C.T., P.E.S., T.T., J.E.G., J.J.V., R.P.N., and J.T.E. are supported by the Dawn and Dudley Holmes Foundation and the Babcock Memorial Trust.; D.D.G., V.C., and C.R. are supported by the Krembil Foundation.; L.C.T. was also supported by the Dermatology Foundation.; J.E.G. was supported by Doris Duke Foundation (Grant #:2013106) and the National Institute of Health (K08AR060802 and R01AR06907) and the Taubman Medical Research Institute as the Frances and Kenneth Eisenberg Emerging Scholar.
• Keywords: machine learning; genome-wide association studies; psoriatic arthritis; predictive medicine
• Research Categories: Biology, Biostatistics; Biology, Molecular; Biology, Genetics

Abstract

Psoriatic arthritis (PsA) is a complex chronic musculoskeletal condition that occurs in ~30% of psoriasis patients. Currently, no systematic strategy is available that utilizes the differences in genetic architecture between PsA and cutaneous-only psoriasis (PsC) to assess PsA risk before symptoms appear. Here, we introduce a computational pipeline for predicting PsA among psoriasis patients using data from six cohorts with >7000 genotyped PsA and PsC patients. We identify 9 new loci for psoriasis or its subtypes and achieve 0.82 area under the receiver operator curve in distinguishing PsA vs. PsC when using 200 genetic markers. Among the top 5% of our PsA prediction we achieve >90% precision with 100% specificity and 16% recall for predicting PsA among psoriatic patients, using conditional inference forest or shrinkage discriminant analysis. Combining statistical and machine-learning techniques, we show that the underlying genetic differences between psoriasis subtypes can be used for individualized subtype risk assessment.