Biomarkers capable of evaluating diabetic retinopathy progression across disease stages remain limited. An analysis published in Diabetologia evaluated intraocular and plasma proteomic signatures associated with diabetic retinopathy progression and assessed the clinical relevance of circulating biomarkers.

The discovery phase included aqueous humour proteomic profiling from 32 participants in the Guangdong Diabetic Retinopathy Multiple-Omics Study using high-throughput proteomics. Temporal trajectory analysis identified 40 candidate proteins whose concentrations changed progressively during diabetic retinopathy development. Among these candidates, 25 proteins demonstrated consistent directional changes in an independent validation dataset, including 17 proteins with increasing concentrations and eight with decreasing concentrations (all p < 0.05).

Single-cell RNA sequencing localized 15 candidate biomarkers, including neurofilament light chain (NFL), to retinal neurons and glial cells. The clinical relevance of plasma-detectable markers was then evaluated in 2,495 individuals with diabetes from the UK Biobank.

Higher baseline plasma NFL differentiated individuals with diabetic retinopathy from control participants (OR 1.98; 95% CI 1.61-2.42). Elevated plasma NFL also predicted incident diabetic retinopathy (HR 2.01; 95% CI 1.48-2.73) and vascular complications during a median follow-up of 12 years, including microvascular events (HR 2.28; 95% CI 1.94-2.69) and macrovascular events (HR 1.49; 95% CI 1.26-1.77).

Incorporation of plasma NFL into conventional risk models improved prediction of diabetic retinopathy (NRI 0.194; 95% CI 0.042-0.297; IDI 0.015; 95% CI 0.003-0.047). These findings indicate that plasma NFL represents a minimally invasive biomarker associated with diabetic retinopathy progression and diabetes-related vascular complications.