Mass spectrometry-based metabolomic and lipidomic profiling stratifies stages of diabetic retinopathy
Authors' conclusion
Mass spectrometry-based metabolomics/lipidomics integrated with machine learning accurately stratify diabetic retinopathy and identify novel molecular markers, and may support early diagnosis and targeted intervention of diabetic retinopathy.
Study context
- Population
- Individuals with diabetes with or without DR; serum profiles of 962 molecules (653 lipids, 309 polar small metabolites)
- Participants
- 167 individuals
- Exposure
- No retinopathy vs non-proliferative vs proliferative DR
- Without retinopathyn = 45reference
- Non-proliferative diabetic retinopathyn = 69
- Proliferative diabetic retinopathyn = 53
- Outcome
- Metabolic and lipidomic alterations across DR stages
Methodological detailsmass spectrometry-based metabolomic and lipidomic profiling, machine learning integration
- Mass spectrometry-based metabolomic and lipidomic profiling
- Machine learning integration
Main finding
As reported in the abstract
- Purine and sphingolipid metabolism significantly altered in both NPDR and PDR
- Tyrosine metabolism disrupted in NPDR; glycine-serine-threonine metabolism prominent in PDR
- Progression associated with reduced sphingomyelins, phosphatidylcholines and lysophosphatidylcholines
- Among ten machine learning models, K-nearest neighbor achieved highest classification performance
- Top contributors: lipid ST 24:1;O4 and metabolite beta-hydroxyisovaleric acid
Observational data: associations, not effects.
Original publication
Mass spectrometry-based metabolomic and lipidomic profiling stratifies stages of diabetic retinopathy. Diabetes Res Clin Pract. Published 22 August 2025. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →