Key Takeaways

• Anti-VEGF treatment for DME was associated with reduced PDR risk primarily during the first 4 weeks following an injection, with the protective effect declining rapidly thereafter
• Severe baseline NPDR, younger age, and type 1 diabetes were among the strongest predictors of progression despite anti-VEGF treatment
• Continued dilated retinal surveillance remains important for patients receiving anti-VEGF therapy because repeated injections do not eliminate the risk of progression to PDR

Real-world data from more than 2800 eyes suggest that the protective effect of anti-VEGF therapy against progression to proliferative diabetic retinopathy (PDR) is greatest soon after treatment and declines rapidly after approximately 4 weeks.¹

The findings, published in BMJ Open Ophthalmology, also indicate that patients with diabetic macular edema (DME) may continue to progress to PDR despite receiving repeated anti-VEGF injections, underscoring the importance of continued diabetic retinopathy surveillance during treatment.

Abraham Olvera-Barrios, MD, and colleagues conducted the retrospective multicenter study using routine clinical data from 27 UK National Health Service centers. The investigators sought to determine how repeated anti-VEGF exposure for DME modifies the risk of progression from nonproliferative diabetic retinopathy (NPDR) to PDR in clinical practice.

The primary analysis included 2858 eyes from 2858 patients with NPDR and DME treated with anti-VEGF therapy. Investigators modeled repeated injections as time-dependent exposures using Cox regression and a weighted cumulative exposure approach. Models were adjusted for baseline diabetic retinopathy severity, age, sex, ethnicity, diabetes type, and deprivation.

During a median follow-up of 1.2 years, 209 eyes in the anti-VEGF-treated cohort developed PDR. Patients received a median of 11 intravitreal injections during follow-up. At baseline, 67% of treated eyes received ranibizumab, 31% received aflibercept, and 2% received bevacizumab.

The weighted cumulative exposure model indicated that current and recent anti-VEGF injections had the strongest protective association with PDR risk. That effect declined rapidly as the interval from the most recent injection increased and reached approximately zero at 4 weeks.

More remote injections did not appear to substantially affect the current risk of PDR, according to the investigators. The finding suggests that the retinopathy-modifying effects of repeated anti-VEGF treatment should not necessarily be considered persistent or cumulative over extended periods.

The investigators noted that decreasing treatment frequency to less than monthly or discontinuing therapy would therefore be expected to increase PDR risk relative to continued monthly treatment, particularly among patients already at high risk of progression. 

Anti-VEGF exposure did not eliminate the importance of established risk factors for progression. Compared with eyes with mild NPDR at baseline, those with moderate NPDR had nearly twice the hazard of developing PDR (HR, 1.99; 95% CI, 1.13-3.51), whereas severe NPDR was associated with more than a 4-fold increase in hazard (HR, 4.63; 95% CI, 2.55-8.41).

Diabetes type and age also were associated with progression. Patients with type 1 diabetes had approximately twice the PDR hazard of those with type 2 diabetes (HR, 2.08; 95% CI, 1.35-3.21). Conversely, every 5-year increase in age was associated with a 9% reduction in PDR hazard.

Among patients with severe NPDR and DME, intraretinal microvascular abnormalities were associated with an increased risk of progression compared with dot-blot hemorrhages in 4 quadrants (HR, 2.55; 95% CI, 1.12-5.79).

Overall, the cumulative incidence rate of PDR among anti-VEGF-treated eyes with DME was 4.45 per 100 person-years. By 3 years, approximately 4% of treated eyes with mild NPDR had progressed to PDR compared with 25% of treated eyes with severe NPDR. Younger patients also demonstrated higher progression rates; by year 3, PDR had developed in approximately 22% of treated patients younger than 55 years compared with 9% of those aged 65 years or older.

The results have implications for patients managed in injection-focused clinical pathways, in which peripheral retinal examination may not occur at every visit.

According to the investigators, repeated anti-VEGF therapy for DME should not be considered a substitute for surveillance for PDR. They recommended that dilated assessment of NPDR in patients receiving anti-VEGF therapy should not extend beyond established monitoring intervals based on retinopathy severity—approximately every 12 months for mild NPDR, 6 months for moderate NPDR, and 3 to 4 months for severe NPDR—with patient age also considered when determining follow-up.

The authors acknowledged limitations inherent to the retrospective electronic health record analysis, including potential inaccuracies and missing clinical information. The dataset also lacked systemic factors such as glycemic control and blood pressure, which could confound PDR risk. In addition, data were extracted through December 2018, and the investigators did not evaluate PDR risk according to individual anti-VEGF agents because treatment switching was common.

Reference

1. Olvera-Barrios A, Lilaonitkul W, Heeren TFC, et al. Impact of anti-VEGF treatment for diabetic macular oedema on progression to proliferative diabetic retinopathy: data-driven insights from a multicentre study. BMJ Open Ophthalmol. 2025;10(1):e002234. doi:10.1136/bmjophth-2025-002234.