Inherited retinal diseases (IRDs) are a clinically and genetically heterogeneous spectrum of disorders that lead to progressive and irreversible vision loss. Gene therapy is the most promising emerging treatment for IRDs. While gene augmentation strategies have demonstrated clinical benefit and results within the first approved ocular gene therapy, their application is restricted by adeno-associated virus (AAV) packaging capacity and limited efficacy for dominant mutations. Recent breakthroughs in precision genome editing, particularly base editing (BE) and prime editing (PE), have provided alternatives capable of directly correcting pathogenic variants. BE enables targeted single-nucleotide conversions, whereas PE further allows for precise insertions and deletions, both circumventing the double-strand DNA cleavage or repair processes typically induced by conventional CRISPR–Cas editing systems, thereby offering advantages in post-mitotic retinal cells. Preclinical investigations across murine and non-human primate models have demonstrated the feasibility, molecular accuracy, and preliminary safety profiles of these platforms in targeting IRD-associated mutations. However, critical challenges remain before clinical application can be realized, including limited editing efficiency in photoreceptors, interspecies variability in therapeutic response, potential risks of off-target effects, and barriers in large-scale vector manufacturing. Moreover, the delivery of genome editors to the outer retina remains suboptimal, prompting intensive efforts in capsid engineering and the development of non-viral delivery systems. This review synthesizes the current progress in BE and PE optimization, highlights innovations in delivery platforms that encompass viral and emerging non-viral systems and summarizes the major barriers to clinical translation. We further discuss AI-driven strategies for the rational design of BE/PE systems, thereby outlining their future potential and perspectives in the treatment of IRDs.

Age-related macular degeneration (AMD), particularly non-exudative AMD, requires experimental models that better replicate human pathology. Current in vivo models remain technically demanding and time-intensive, whereas conventional in vitro systems fail to recapitulate disease-specific pathological triggers. Here, we present a method to establish a retinal pigment epithelial (RPE) degeneration model using primary mouse RPE cells transfected with Alu RNA, a retrotransposon directly implicated in geographic atrophy pathology. This protocol details the enzymatic isolation of primary mouse RPE cells, followed by Alu RNA transfection to induce RPE degeneration. Validation integrates morphological (hexagonal architecture), functional (polarity loss and mouse protein ZO-1 disruption), and molecular analysis (quantitative PCR). As a result, we observed multifactorial changes triggered by Alu RNA transfection: inflammatory cytokine secretion (mouse genes Ifn-β, Il-6, Tnf-α; p < 0.05) and cellular senescence (mouse genes p21 and p53 upregulation; p < 0.05). Compared to traditional acute stress models, this system recapitulates chronic inflammatory-degenerative cascades of AMD through standardized techniques, ensuring reproducibility. By combining aspects of simplified in vitro assays and complex in vivo models, this approach could serve as a preliminary platform for exploring retrotransposon-driven mechanisms and screening potential therapeutic candidates.

Background: Retinal vascular occlusion (RVO) and retinal artery occlusion (RAO) have been reported as rare adverse events following COVID-19 vaccination, raising concerns about vaccine safety. This review synthesizes cohort and case–control studies assessing the association between COVID-19 vaccines and RVO/RAO, while exploring potential pathophysiological mechanisms. Methods: We analyzed large-scale population-based studies from South Korea, Europe, and the TriNetX database, focusing on odds ratios (OR), hazard ratios (HR), and relative risks (RR) across mRNA and adenoviral vector vaccines. Pathological processes were hypothesized based on molecular and clinical evidence. Results: Studies investigating the association between COVID-19 vaccination and retinal vascular occlusion show conflicting results; some studies report no association (e.g., OR 0.93, 95% CI 0.60–1.45), others suggest reduced risk (e.g., OR 0.80, 95% CI 0.64–0.99), and one indicates increased risk over two years (HR 2.19, 95% CI 2.00–2.39). Adenoviral vector vaccines, particularly ChAdOx1, show higher RAO incidence in specific cohorts. Proposed mechanisms include vaccine-induced immune thrombotic thrombocytopenia (VITT) via anti-PF4 antibodies, spike protein-mediated endothelial dysfunction, and adjuvant-driven inflammation. Conclusions: While causality remains unproven, temporal heterogeneity and vaccine type-specific risks warrant further investigation. Longitudinal studies with robust controls are needed to clarify these associations in the post-pandemic context.

Age-related macular degeneration (AMD), particularly non-exudative AMD, requires experimental models that better replicate human pathology. Current in vivo models remain technically demanding and time-intensive, whereas conventional in vitro systems fail to recapitulate disease-specific pathological triggers. Here, we present a method to establish a retinal pigment epithelial (RPE) degeneration model using primary mouse RPE cells transfected with Alu RNA, a retrotransposon directly implicated in geographic atrophy pathology. This protocol details the enzymatic isolation of primary mouse RPE cells, followed by Alu RNA transfection to induce RPE degeneration. Validation integrates morphological (hexagonal architecture), functional (polarity loss and mouse protein ZO-1 disruption), and molecular analysis (quantitative PCR). As a result, we observed multifactorial changes triggered by Alu RNA transfection: inflammatory cytokine secretion (mouse genes Ifn-β, Il-6, Tnf-α; p < 0.05) and cellular senescence (mouse genes p21 and p53 upregulation; p < 0.05). Compared to traditional acute stress models, this system recapitulates chronic inflammatory-degenerative cascades of AMD through standardized techniques, ensuring reproducibility. By combining aspects of simplified in vitro assays and complex in vivo models, this approach could serve as a preliminary platform for exploring retrotransposon-driven mechanisms and screening potential therapeutic candidates.