**Background:** Retinal and choroidal vascular diseases, including age-related macular degeneration (AMD) and diabetic retinopathy (DR), are leading causes of blindness worldwide. Current treatments, such as anti-VEGF injections, are effective but require frequent administration, leading to patient burden and compliance issues. Gene therapy using adeno-associated virus (AAV) vectors offers the potential for long-lasting, single-dose treatment. This review provides an overview of AAV-based strategies for AMD, DR, and diabetic macular edema (DME), covering both constitutive and regulatable expression platforms, modified capsids, multigenic constructs, and cell-specific promoters.
**Methods:** The authors conducted a narrative review of preclinical and clinical studies, focusing on in vivo data. They searched clinicaltrials.gov for trials using AAV vectors for eye diseases. The review covers studies on AAV-mediated delivery of endogenous inhibitors (e.g., sFLT-1, endostatin), anti-VEGF agents (e.g., aflibercept, antibody fragments), complement cascade inhibitors (e.g., soluble CD59, complement factor I), and molecules targeting inflammation and oxidative stress. Key clinical trials are summarized in Tables 1 and 2.
**Key Results:** For wet AMD, the phase I OPTIC trial of ADVM-022 (AAV2.7m8 encoding aflibercept) reported an 81–98% decrease in annual anti-VEGF injection rate, with 53% of patients receiving the lowest dose remaining injection-free after 2 years. However, dose-dependent inflammation was observed, and severe uveitis and dry AMD were reported. The phase I/IIa trial of subretinal RGX-314 (AAV8 encoding anti-VEGF antibody fragment) showed decreased injection frequency over 4 years, especially at higher doses (6E10 and 1.6E11 vg/eye), with improved visual acuity. Suprachoroidal RGX-314 in the AAVIATE trial demonstrated an 85% reduction in treatment burden at the highest dose (1E12 vg/eye) with good tolerability. The phase I/II PRISM trial of 4D-150 (AAV R100 encoding miRNA targeting VEGF-C and aflibercept) reported a 100% reduction in mean annual anti-VEGF injections at 36 weeks in patients treated with 3E10 vg, with no inflammation or toxicity. For dry AMD, the phase I HMR1001 trial of intravitreal HMR59 (AAV2 encoding soluble CD59) showed decreased lesion growth rate and no progression to neovascular AMD. The phase I/II FOCUS trial of subretinal GT005 (AAV2 encoding complement factor I) reported increased vitreous CFI levels and reduced complement activity, with no serious adverse events. For DR, the phase II INFINITY trial of ADVM-022 for DME was halted due to hypotony at the highest dose (6E11 vg), though the lower dose (2E11 vg) reduced injection needs (61% injection-free at week 24). The phase II ALTITUDE trial of suprachoroidal RGX-314 for DR showed improvement of ≥2 steps in DR severity scale in 20% of patients and any improvement in 54%, with no drug-related severe adverse events. Preclinical studies also showed efficacy of AAV-mediated sFLT-1, endostatin, COMP-Ang1, ACE2, Dp71, and other targets in rodent models.
**Clinical Implications:** AAV-based gene therapies represent a paradigm shift for AMD and DR, offering the possibility of sustained efficacy with a single injection, reducing treatment burden and improving compliance. The success of anti-VEGF approaches (ADVM-022, RGX-314, 4D-150) in wet AMD and the emerging complement-targeting therapies (HMR59, GT005) for dry AMD address major unmet needs. However, safety concerns, particularly intraocular inflammation and dose-limiting toxicity (e.g., hypotony in DME patients), highlight the need for careful dose optimization, prophylactic corticosteroids, and patient selection. The multifactorial nature of these diseases may require multigenic or regulatable expression platforms to achieve optimal outcomes. Further research is needed to translate preclinical findings into clinical practice, especially for DR, where only a few therapies have reached trials.