Complementary body weight and cardiometabolic benefits of higher GLP-1 and lower GIP: genetic evidence from large-scale phenomic analyses

作者：Lijuan Wang, Fangyuan Jiang, Shuai Yuan, Jing Sun, Jianhui Zhao, Siyun Zhou, Jingyi Liang, Haiqing Li, Peige Song, Shengfeng Wang, Jianzeng Dong, Siyan Zhan, Susanna C. Larsson, Yao Xie, Yuan Ding, Xue Li, Christos S. Mantzoros · 发表于：Metabolism · 年份：2026 · DOI：10.1016/j.metabol.2026.156711 · 研究领域：Diabetes Treatment and Management、Genetic Associations and Epidemiology、Genomics and Rare Diseases

BACKGROUND: The long-term health effects of glucagon-like peptide-1 (GLP-1) and glucose-dependent insulinotropic polypeptide (GIP) remain unclear. Most importantly, it is highly controversial whether GIP agonists vs. antagonists should be developed and what their effects and/or potential side effects would be in the clinic. Further investigation is needed to explore the effects of GIP lowering alone or in combination with GLP-1 enhancing. Tree-structured phenotypic modeling (TreeWAS) and factorial Mendelian randomization (MR) analyses are valuable in this context, even though the highest circulating GLP-1 levels may reflect only a minimum activation i.e. levels achieved with pharmacological intervention with dipeptidyl peptidase 4 (DPP-4) inhibitors and not GLP-1 receptor agonists (GLP-1 RAs) which achieve much higher levels. METHODS: TreeWAS was first conducted to map associations of GLP-1 or GIP levels with a broad range of disease outcomes among 385,917 UK Biobank participants. Disease-trajectory analysis was applied to characterize temporal patterns of co-occurrence of multiple comorbidities related to GLP-1 or GIP or both. Subsequently, factorial MR was performed to investigate the joint effects of GLP-1 and GIP on the associated health outcomes. Finally, mediation analysis was employed to explore the underlying mechanisms through which GLP-1 and GIP exert their effects. RESULTS: TreeWAS analysis revealed associations of elevated genetically predicted GLP-1 levels, even ...