Shared genetic architecture between adhesive capsulitis and type 2 diabetes: a phenotype-specific cross-trait analysis

Adhesive capsulitis (AC) is over-represented in people with diabetes, but whether AC and type 2 diabetes (T2D) share inherited susceptibility remains uncertain. We analysed GWAS summary statistics for phenotype-specific FinnGen R12 M13_ADHCAPSULITIS (ICD-10 M75.0; 7372 cases and 357,549 controls), FinnGen T2D, and Pan-UK Biobank T2D using cross-trait LD score regression, MiXeR, and the official pleioFDR toolbox. AC was positively genetically correlated with T2D (Pan-UK Biobank: rg = 0.334, SE = 0.049, P = 1.4 × 10⁻¹¹; FinnGen: rg = 0.189, SE = 0.042, P = 7.0 × 10⁻⁶). Five bivariate MiXeR fits were compatible with partial overlap, but variation in the shared component and disagreement between the Akaike information criterion (AIC) and the Bayesian information criterion (BIC) for the minimum-overlap comparison precluded a precise quantitative overlap claim. Conditional and conjunctional FDR analyses statistically prioritised AC-associated and jointly associated regions, respectively, without establishing shared mechanisms, causal variants, or effector genes. An exploratory cross-resource contrast of broad Pan-UK Biobank M75 with phenotype-specific FinnGen M75.0 produced similar genome-wide correlations but different polygenicity and locus-level results; because the resource, cohort, ascertainment, GWAS construction, case count, and sample-overlap structure changed simultaneously, these differences cannot be attributed to phenotype breadth alone. These results support consistent population-level genetic sharing, not causality or individual-level clinical utility; fine-mapping, colocalisation, and functional validation remain required.