Table 3 Univariable, two-sample Mendelian randomization studies using genetic proxies to estimate effects of Cer22:0 on the risk of T2D.

From: Dihydroceramide- and ceramide-profiling provides insights into human cardiometabolic disease etiology

T2D-associations from DIAGRAM (n = 74,124 T2D cases and 824,006 controls)

Cer22:0-associations from

EPIC-POTSDAM

EUROSPAN

FHSOCa

 

(n = 1094)

(n = 4034)

(n = 2217)

 

rs680379

rs680379

rs680379

 WR

0.070

0.259

0.452

 SE

0.032

0.120

0.210

 p-value

0.031

0.031

0.031

  1. Several SNPs near the SPTLC3 gene were associated with Cer22:0 plasma concentrations in GWAS in the EPIC-Potsdam study. We compared our SNP-phenotype associations with data from two published GWAS on Cer22:037,39, and SNP-T2D associations were drawn from DIAGRAM (T2D)41. Among the SNPs available in all these studies, the rs680379 association with Cer22:0 plasma concentrations in EPIC-Potsdam had the lowest p-value (P = 2.3E-07), and rs680379 was therefore used for a univariable, two-sample MR in EPIC-Potsdam. We replicated the MR with SNP-Cer22:0 association from the independent cohorts (EUROSPAN and FHSOC). SNP-ceramide 22:0 associations were reported per standard deviation (EPIC-Potsdam, EUROSPAN) and per µM (FHSOC).
  2. aThe beta estimates for the association of rs680379 with Cer22:0 in FHSOC included in the MR were extracted from Table 2 in ‘Genetic Architecture of Circulating Very-Long-Chain (C24:0 and C22:0) Ceramide Concentrations’ by Cresci et al.39.
  3. WR Wald ratio, SE standard error, T2D type 2 diabetes, CAD Coronary artery diseases, MR Mendelian randomization, FHSOC Framingham Heart Study Offspring Cohort.
Back to article page