Modifiable risk factors and inflammation-related proteins in polymyalgia rheumatica: genome-wide meta-analysis and Mendelian randomization

Abstract Objective PMR is an age-related inflammatory disease of unknown cause. We aimed to identify potentially modifiable risk factors and therapeutic targets for preventing or treating PMR. Methods We meta-analysed genetic association data from 8156 cases of PMR (defined using diagnostic codes and self-report) and 416 495 controls of European ancestry from the UK Biobank and FinnGen. We then performed Mendelian randomization analyses to estimate the association between eight modifiable risk factors (using data from up to 1.2 million individuals) and 65 inflammation-related circulating proteins (up to 55 792 individuals), using the inverse variance weighted and pleiotropy robust methods. Results We identified three novel genome-wide significant loci in the IL1R1, NEK6 and CCDC88B genes and confirmation of previously described associations with HLA-DRB1 and ANKRD55. Genetically predicted smoking intensity (OR 1.32; 95%CI 1.08–1.60; P = 0.006) and visceral adiposity (OR 1.22; 95%CI 1.10–1.37; P = 3.10 × 10−4) were associated with PMR susceptibility. Multiple circulating proteins related to IL-1 family signalling were associated with PMR. IL-1 receptor-like 2, also known as IL-36 receptor (OR 1.25; P = 1.89 × 10−32), serum amyloid A2 (OR 1.06, 9.91 × 10−10) and CXCL6 (OR 1.09, P = 4.85 × 10−7) retained significance after correction for multiple testing. Conclusion Reducing smoking and visceral adiposity at a population level might reduce incidence of PMR. We identified proteins that may play causal roles in PMR, potentially suggesting new therapeutic opportunities. Further research is needed before these findings are applied to clinical practice.