Vision protection therapy for prevention of neovascular age-related macular degeneration

**Background:** Neovascular (wet) age-related macular degeneration (AMD) is the leading cause of irreversible visual loss worldwide. Standard care for dry AMD, including AREDS vitamin supplements and lifestyle modifications, reduces the risk of conversion by only about 4% per year. Vision Protection Therapy (VPT), consisting of regular panmacular subthreshold diode micropulse laser (SDM) treatments, has been proposed as a more effective preventive strategy. This study is the third in a series evaluating VPT for preventing neovascular conversion in dry AMD.

**Methods:** This retrospective real-world data study used the Vestrum Health database, which aggregates unidentified patient data from over 320 US retina specialists. Eyes with dry AMD (ICD10 coded) active between January 2017 and July 2023 were included. Inclusion criteria: age ≥50 years, ICD10 diagnosis of dry AMD. Exclusion criteria: prior or current intravitreal injections of steroids or anti-VEGF agents, neovascular AMD in the fellow eye, diabetes mellitus, retinal vascular occlusion, prior conventional macular photocoagulation, or confounding diagnoses (e.g., ocular histoplasmosis, high myopia). After filtering, approximately 200,000 eyes were identified, including 737 VPT-treated eyes from a single practice employing VPT. Propensity score matching (1:10 ratio, VPT:SCA) was performed using major conversion risk factors: age, AREDS vitamin use, hypertension, smoking, and dry AMD severity (early, intermediate, advanced with extrafoveal or subfoveal geographic atrophy). The primary endpoint was conversion from dry to wet AMD, defined by both a change in ICD10 coding to wet AMD (H35.30) and initiation of anti-VEGF therapy. Visual acuity (VA) was a secondary endpoint, converted to approximate ETDRS letters. Additional post-hoc analyses matched encounter frequency and excluded VPT-untreated eyes from the VPT group.

**Key Results:** After propensity score matching, the VPT group (737 eyes, 406 subjects) and SCA group (7370 eyes, 4652 subjects) were well balanced on demographics (mean age 77.7 vs 77.8 years; 61.3% vs 58.9% female; hypertension 54.4% vs 54.0%; AREDS use 49.3% vs 47.1%; smoking 3.2% vs 3.0%; AMD severity distribution similar). The VPT group had significantly more clinical encounters (mean 13.2 vs 11.4) and longer follow-up (mean 677 vs 542.6 days). Despite this, the rate of neovascular conversion was significantly lower in the VPT group: 4.3% (32/737) vs 16.3% (1202/7370) in the SCA group (hazard ratio 5.98, p<0.0001 with encounter matching; HR 5.73 without encounter matching). Stratified log-rank test showed a highly significant difference (Chi-square 107.7, p<0.0001). Cox proportional hazards model adjusting for severity, age, and AREDS use confirmed a HR of 5.98 (95% CI 4.2–8.5 asymptotic; 4.3–9.3 bootstrap) for SCA vs VPT. All severity subgroups (intermediate, non-central GA, central GA) showed increased hazard vs early AMD (HRs 1.66–1.99). Age increased hazard by 2% per year (HR 1.02, p<0.0001). AREDS use was associated with a 50% increase in hazard (HR 1.49, p<0.0001). Visual acuity was consistently better in VPT eyes throughout the study window (e.g., mean VA in 2022: VPT 66.9 letters vs SCA 63.5 letters). Post-hoc analyses excluding VPT-untreated eyes (9% of VPT cohort with bilateral early AMD) and matching encounter frequency yielded similar HRs (5.65–5.84, all p<0.0001). After conversion, SCA eyes required 3.9–4.5 times more anti-VEGF injections than VPT eyes.

**Clinical Implications:** This large real-world study provides strong evidence that VPT (regular panmacular SDM) markedly reduces the risk of neovascular conversion in dry AMD compared to standard care alone, with hazard ratios consistently around 5.7–6.0 across multiple analyses. The benefit was greatest in higher-risk eyes (older age, more advanced dry AMD). VPT also appeared to preserve better visual acuity and reduce the need for anti-VEGF injections after conversion. The study's limitations include its retrospective, non-randomized design, potential unmeasured confounders (e.g., reticular pseudodrusen, genetic factors), and reliance on ICD10 coding accuracy. However, the consistency of results across four distinct statistical approaches, the large sample size, and the robustness of the propensity score matching support the reliability of the findings. The authors argue that VPT may be substantially more effective than current standard care for preventing neovascular AMD and associated visual loss, and that the results likely underestimate the true benefit due to more frequent monitoring in the VPT group. These findings suggest that VPT could become an important addition to the management of dry AMD, particularly for high-risk patients.