Proliferative diabetic retinopathy remains one of the most challenging complications of diabetes, but specialists still lack the tools to predict how the condition will evolve, ophthalmology experts say.
A team led by clinicians at the Centre for Ophthalmology and Visual Science at the University of Western Australia outlined current evidence on pathophysiology and determinants of disease persistence in a comprehensive review, arguing that major knowledge gaps needed to be closed before prevention became realistic.
Writing in Survey of Ophthalmology [link here], UWA’s Dr Brendan Luu and colleagues said proliferative diabetic retinopathy (PDR) was a “complex, incompletely understood neurovascular and inflammatory cascade”. They said available therapies largely stabilised rather than regenerated the retina, leaving residual ischaemia and functional deficit.
The authors highlighted four key unknowns.
PDR pathogenesis
Dr Luu and colleagues said PDR was understood as a multifactorial neurovascular and inflammatory disorder, but the contribution of each element and the sequence of progression remained unclear. They said hyperglycaemia, oxidative stress, inflammation and glial activation all played roles, but how these intersected needed further work.
They said key questions included why some ischaemic eyes mounted “exuberant neovascular responses” while others did not, and what drove the shift from angiogenic to fibrotic phenotypes.
The authors argued that longitudinal human studies integrating advanced imaging with single cell and spatial transcriptomics, proteomics and metabolomics of ocular fluids and retinal tissue were needed.
Biomarker evidence
Dr Luu and colleagues said clinicians still could not identify which patients with diabetes were at risk of PDR, or which patients would progress despite optimal treatment. They said existing biomarkers such as HbA1c, duration of diabetes and ETDRS severity levels explained only part of the variation in outcomes.
They noted that most biomarker studies had been small and single centre, while inconsistent endpoints limited the development of standardised risk models. They said artificial intelligence had been used mainly to identify referable diabetic retinopathy rather than forecast outcomes.
The authors said future research should focus on prospective, imaging rich cohorts of patients with high risk non proliferative and early proliferative disease, using standardised, clinically meaningful progression endpoints. They said systematic collection of blood, aqueous and vitreous samples would support multi omic analyses and help identify “high trajectory” phenotypes early.
Restorative therapies
Dr Luu and colleagues said current PDR treatments such as PRP, intravitreal anti VEGF, PPV and corticosteroids were fundamentally reactive. They said treatment began only once extensive microvascular damage had occurred, and therapies did not restore normal capillary architecture or fully reverse retinal ischaemia.
They said research needed to shift towards restorative vascular therapies, interventions that preserved or restored retinal neurons, and cell based approaches. They noted that evidence for regenerative strategies remained sparse and largely conceptual or pre clinical, and said robust translational pathways would be required to move from mechanistic insights to early phase trials.
Health system gaps
The authors said implementation barriers would remain even with restorative therapies and validated risk models. They said literature on predictors of PDR progression relied on small samples and did not include real world factors such as adherence and treatment interruptions.
They said there was no accepted core outcome for PDR progression, and consensus was needed on what clinicians should aim for. They said predictors identified in studies might not perform similarly in practice, with missed appointments, resource constraints and systemic comorbidities strongly influencing outcomes. They noted that younger age and socioeconomic disadvantage were already linked to worse outcomes for anti VEGF treated PDR.
Dr Luu and colleagues said implementation focused research was needed to test risk stratified recall systems, tele ophthalmology and integrated diabetes eye care pathways.
They argued that deeper mechanistic work, large biomarker studies and new treatment strategies would be crucial for achieving a future where PDR was largely preventable and well controlled. They said emerging biomarkers and therapies needed to be evaluated in low and middle income settings, where the burden of PDR was greatest.