Diabetic kidney disease (DKD) has been an increasingly prevalent cause of death and loss of quality of life in Taiwan. We aim to explore the association between genetic alteration of tricarboxylic acid (TCA) cycle enzyme and DKD. This is a 2 stages study design. The gene association study performed on a base sample designed with matched case-control on age, diabetes duration, and HbA1c. The base sample included 852 patients with type 2 diabetes mellitus (T2DM), 426 among them have DKD. Significant signals detected was replicated in an independent target sample of up to 1170 T2DM patients (378 among them have DKD). In base sample, one SNP is associated with DKD after Bonferroni correction (rs6050911 - P=0.00046; OR=0.663) and other 19 SNPs have P-value <0.05. Polygenic risk score (PRS) is calculated based on these 20 SNPs to stratify DKD risk. The very high-risk group (with PRS >90 percentile) have more than 3 fold the risk of DKD (OR = 3.77; 95%CI = 2.03-7.00) compared to the reference group (with PRS ranged 40-60 percentile). Furthermore, 20 SNPs were replicated and PRS was calculated for each individual in the target sample. None of these 20 SNPs were significantly associated with DKD status and no significant differences in risk between various PRS groups and reference group (95% CIs betwesn different PRS group and reference group all included 1.00) were observed. While the results of genes association testing and disease risk stratification based on PRS cannot be replicated on the target sample, the high level of risk observed in the base sample group with PRS >90 percentile could prove to be significant in the informed clinical decision on prevention of diabetic kidney complications.
