To replicate the association of genetic variants with estimated glomerular filtration rate (GFR) and albuminuria, which has been found in recent genome‐wide studies in patients with Type 2 diabetes.

We evaluated 16 candidate single nucleotide polymorphisms for estimated GFR in 3028 patients with Type 2 diabetes sampled from clinics across Tayside, Scotland, UK, who were included in the Genetics of Diabetes Audit and Research Tayside (GoDARTs) study. These single nucleotide polymorphisms were tested for their association with estimated GFR at entry to the study, with albuminuria, and with time to stage 3B chronic kidney disease (estimated GFR<45 ml/min/1.73 m²). We also stratified the effects on estimated GFR in patients with (n = 2096) and without albuminuria (n = 613).

rs1260326 in GCKR (β=1.30, P = 3.23E‐03), rs17319721 in SHROOM3 (β = −1.28, P‐value = 3.18E‐03) and rs12917707 in UMOD (β = 2.0, P‐value = 8.84E‐04) were significantly associated with baseline estimated GFR. Analysis of effects on estimated GFR, stratified by albuminuria status, showed that in those without albuminuria (normoalbuminura; n = 613), UMOD had a significantly stronger effect on estimated GFR (β_normo = 4.03 ± 1.23 vs β_albuminuria = 1.72 ± 0.76, P = 0.002) compared with those with albuminuria, while GCKR (β_normo = 0.45 ± 0.89 vs β_albuminuria = 1.12 ± 0.55, P = 0.08) and SHROOM3 (β_normo = −0.07 ± 0.89 vs β_albuminuria = −1.43 ± 0.53, P = 0.003) had a stronger effect on estimated GFR in those with albuminuria. UMOD was also associated with a lower rate of transition to stage 3B chronic kidney disease (hazard ratio = 0.83[0.70, 0.99], P = 0.03).

The genetic variants that regulate estimated GFR in the general population tend to have similar effects in patients with Type 2 diabetes and in this latter population, it is important to adjust for albuminuria status while investigating the genetic determinants of renal function.