Abstract
Genome-wide association study (GWAS) meta-analyses identified thousands of loci. Given genetic variant effect sizes vary across contributing studies due to environmental or genetic differences, study-specific features might contribute to biological mechanism elucidation. We contrasted a GWAS of estimated glomerular filtration rate (eGFR) in a population-based study from the Alps, where thyroid disease is common, against a meta-analysis from the CKDGen Consortium (n>1 million) that previously identified 147 kidney function relevant loci.
In the Cooperative Health Research in South Tyrol (CHRIS) study (n=10,146), we fitted mixed models on ln(eGFR), accounting for relatedness (genomic inflation λ=1.02), to assess direction-consistent replication of CKDGen results. Replicated variants underwent phenome-wide mediation analysis across 72 biomarkers.
We replicated 10 loci, totaling 162 variants correlated with the locus lead variant (r2≥0.8). In CHRIS, replicated variants had 1.3-to-5.8 times larger effects than in CKDGen, at similar minor allele frequencies. Free triiodothyronine and thyroxine resulted as potential mediators for most loci. Variant-by-hyperthyroidism interaction testing P-values were shifted towards small values, compatibly with a systematic effect modification.
The potential modifier role of hyperthyroidism on kidney function genetic associations warrants independent replication.