Assessment of biomarkers associated with rapid renal decline in the detection of retinopathy and its progression in type 2 diabetes: The Fremantle Diabetes Study Phase II.

Abstract Aims To determine whether biomarkers for diabetic kidney disease (DKD) can be used to determine the prevalence, progression and/or incidence of diabetic retinopathy (DR) complicating type 2 diabetes. Methods Proteomic biomarkers were measured in baseline fasting plasma from 958 Fremantle Diabetes Study Phase II participants whose baseline and, in those returning for follow-up (n=764), Year 4 fundus photographs were graded for DR presence/severity. The performance of PromarkerD (three biomarkers and readily available clinical variables which identify prevalent DKD and predict incident DKD and estimated glomerular filtration rate decline ≥30% over four years) for detecting DR prevalence, progression and incidence was assessed using the area under the receiver operating curve (AUC). Logistic regression determined whether individual proteins were associated with DR outcomes after adjusting for the most parsimonious model. Results Plasma apolipoprotein A-IV (APOA4) was independently associated with moderate non-proliferative DR at baseline (OR (95% CI): 1.64 (1.01, 2.67), P=0.047). Model discrimination was poor for all PromarkerD predicted probabilities against all DR outcomes (AUC ≤0.681). Conclusions PromarkerD and its other constituent biomarkers were not associated with DR prevalence or temporal change. APOA4 was associated with prevalent DR, but not DR incidence or progression. Distinct pathophysiological mechanisms may underlie DKD and DR.