The Joslin Medalist Study identified Retinol-binding protein 3 (RBP3) as a potential protective factor against diabetic retinopathy (DR) in the vitreous and retina of individuals with type 1 diabetes for 50 years or longer. This study evaluated the association of aqueous RBP3 concentration to vitreous RBP3 concentrations and DR severity. We measured 131 aqueous samples from patients with type 1 and 2 diabetes undergoing cataract surgery at the Joslin Beetham Eye Institute (BEI) and postmortem samples of Medalists. Aqueous RBP3 concentrations in BEI and Medalists samples were similar for eyes with proliferative DR (PDR, P=0.11) and were well correlated between fellow eyes (r=0.65, P<0.0001) . With an average 12-fold decreased concentration compared to vitreous samples, RBP3 concentrations in aqueous and vitreous samples were correlated within the same individual (r=0.32, P=0.003, N=89) . Aqueous RBP3 was not associated with A1c (P=0.60) , however, RBP3 concentration was inversely associated with the presence of pan-retinal laser photocoagulation scars (β estimate −22.0, 95% CI −31.9;−12.0, P<0.0001) . With increasing DR severity, aqueous RBP3 levels decreased from mild DR (median 0.7nM ± 0.2) , moderate-severe (0.65nM ± 0.3, P=0.0092 vs. PDR) , to PDR (0.5nM ± 0.2, P=0.0vs. mild DR) . Individual retinal layer thicknesses measured by optical coherence tomography (Heidelberg v6.0c, Germany, N=45) showed that higher aqueous concentrations were associated with increased thickness of the ganglion cell layer (β estimate 0.002, 95% CI 0.0019;0.0023, P<0.0001) and inner nuclear layer (0.0026, 0.0025;0.0028, P<0.0001) . The retinal pigment epithelium was inversely associated with aqueous RBP3 levels (−0.001, −0.0017;0.0003, P=0.004) . These findings indicate that aqueous RBP3 concentrations are correlated to vitreous fluids and reduced in people with advanced DR, supporting its potential use as a biomarker for DR severity in people with type 1 and 2 diabetes.


T.Chokshi: None. J.Sun: Consultant; American Diabetes Association, American Medical Association, Research Support; Adaptive Sensory Technology, Boehringer Ingelheim International GmbH, Genentech, Inc., Jaeb Center for Health Research, Janssen Pharmaceuticals, Inc., Novartis Pharmaceuticals Corporation, Novo Nordisk, Optovue, Incorporated, Physical Sciences, Inc, Roche Pharmaceuticals. G.L.King: Advisory Panel; Medtronic, Research Support; Janssen Pharmaceuticals, Inc. W.Fickweiler: None. M.G.Mitzner: None. I.Wu: None. T.Boumenna: None. D.B.Robinson: None. H.Park: None. K.Park: None. L.P.Aiello: Consultant; KalVista Pharmaceuticals, Inc., Novo Nordisk, Stock/Shareholder; KalVista Pharmaceuticals, Inc.


American Diabetes Association (7-21-PDF-022) ; National Eye Institute (R01EYE26080-01) , the National Institute of Diabetes and Digestive and Kidney Diseases and National Institutes of Health (DP3-DK-094333-01) ; JDRF (17-2013-310) ; the Dianne Nunnally Hoppes Fund; the Beatson Pledge Fund

Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. More information is available at