LIPID METABOLISM AS A DRIVER OF DIABETIC KIDNEY DISEASE PROGRESSION

Diabetes kidney disease (DKD) is the main cause of kidney failure in Singapore. There remains risks of progression even with existing therapies, hence the impetus to investigate and identify novel biomarkers that may allow for earlier diagnosis, or serve as potential therapeutic targets. Altered lipid metabolism leads to accumulation of toxic intermediates that cause cellular dysfunction and lipoapoptosis. It is an important mechanism that warrants studies. We aim to identify plasma metabolomics associated with estimated glomerular filtration rate (eGFR) decline in DKD.

227 patients were recruited from National University Hospital, Singapore. Patients were followed for three years with data captured via medical records. Blood samples were collected at baseline and 6-monthly for metabolomics and proteonomics profiling respectively. Baseline and subsequent eGFR were measured using the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equation. eGFR slope was calculated using linear mixed effected model. Three models of linear regression were performed, with adjustment for various covariates. The first model adjusted for age, gender, ethnicity, body mass index (BMI), duration of diabetes, systolic blood pressure and baseline eGFR. The second and third models additionally adjusted for HbA1c and urine albumin-creatinine ratio respectively. Categorical and continuous variables were compared using Chi-square and Welch T-test.

Mean cohort age was 61.8 years. 62.9% were male. 62.4% were Chinese. Mean HbA1c was 7.9%. Mean urine albumin-creatinine ratio (uACR) was 121 mg/mmol. Mean baseline eGFR was 41 ml/min/1.73m2. Mean eGFR slope was -3.05 ml/min/1.73m2. The cohort was separated into controls and rapid progressors (defined as eGFR slope <-5ml/min/1.73m2). Rapid progressors had higher HbA1c and uACR. 29 metabolites were significantly positively associated with eGFR slope in Model 3, in which there were adjustments for age, gender, ethnicity, BMI, duration of diabetes, systolic BP, baseline eGFR, HbA1c and uACR. 14 metabolites were significant in all three models, of which 11 were glycerophospholipids and 3 glycerolipids. Glycerophospholipids were mostly found to be positively associated with eGFR slope.

PE 36:5 and PC 34:3  were positively associated with eGFR slope in another local study involving cohorts from Chen et al. (Diabetologia 2025 Mar;68(3):557-575). The positive association between phospholipids and eGFR slope is consistent with existing literature. Acylcarnitines, glycerolipids and sphingolipids are negatively associated with eGFR slope, but some metabolites were found to have positive associations instead in our study. Disrupted pathways causing accumulations of acylcarnitines/triglycerides are implicated in lipotoxicity and renal injury. Limitations from smaller sample sizes may cause differing outcomes in our cohort. Additionally, the cohorts from Chen et al. involved patients with eGFR >60ml/min/1.73m2, wherein our cohort involved patients with more advanced chronic kidney disease. Further trials in larger cohorts may be considered to explore relationships between acylcarnitines, glycerolipids and sphingolipids with DKD, and evaluate potential therapeutic targets in phospholipid metabolism.