The Antiangiogenic Effect and Ocular Pharmacology of Novel Modified Nonsteroidal Anti-Inflammatory Drugs in the Treatment of Oxygen-Induced Retinopathy

**Background:** Abnormal vessel growth is a hallmark of retinal diseases such as retinopathy of prematurity (ROP), wet age-related macular degeneration, and proliferative diabetic retinopathy (DR). The oxygen-induced retinopathy (OIR) mouse model recapitulates key features of these conditions. Nonsteroidal anti-inflammatory drugs (NSAIDs) have shown antiproliferative and antiangiogenic properties, but conventional NSAIDs have limitations. Three novel modified NSAIDs—OXT-328 (phospho-sulindac), Q-922 (a heavily modified aspirin derivative), and CL-717 (a modified sulindac derivative)—were synthesized to overcome esterase-mediated hydrolysis and improve bioavailability. This study evaluated their efficacy via different routes of administration in the OIR model and in the chicken chorioallantoic membrane (CAM) angiogenesis model.

**Methods:** The OIR model was induced in C57BL/6 mice by exposure to 75% oxygen from postnatal day 7 (P7) to P12, followed by return to room air from P12 to P17. Drugs were administered as follows: OXT-328 as a single 1-μL intravitreal injection of 1% solution on P12; Q-922 as a single 1-μL intravitreal injection of 1% solution on P12, or intraperitoneally at 100 or 150 mg/kg/day from P7 to P17; CL-717 as 5-μL eye drops of 10% solution five times daily from P12 to P17. Retinal flat mounts were stained with isolectin-B4 on P17, and avascular area (AVA) and neovascular area (NVA) were quantified as percent of total retinal area using ImageJ. Pharmacokinetics and biodistribution were assessed in adult mice and explanted human eyes by HPLC. The CAM assay was performed with filter paper discs containing 50 μM OXT-328, 50 μM Q-922, or 100 μM CL-717 applied twice daily for 3 days; microvessel density was measured.

**Key Results:** In the OIR model, intravitreal OXT-328 significantly reduced AVA from 13.9%±4.2% (vehicle) to 6.9%±2.4% (P<0.001) and NVA from 11.5%±1.7% to 5.6%±1.9% (P<0.03). OXT-328 eye drops had no effect. Intravitreal Q-922 reduced AVA from 13.9%±4.2% to 3.5%±0.5% (75% reduction) and NVA from 10.8%±1.8% to 5.2%±1.7% (52% reduction). Intraperitoneal Q-922 at 150 mg/kg reduced AVA to 8.4%±0.5% vs. vehicle 11.7%±1.0% (28% reduction, P<0.01) and NVA to 9.5%±2.6% vs. vehicle 24.1%±5.6% (61% reduction, P<0.001); the 100 mg/kg dose had no significant effect. CL-717 eye drops reduced AVA from 16.8%±1.1% to 5.8%±0.8% (65% reduction, P<0.001) and NVA from 22.7%±3.4% to 8.3%±1.6% (63% reduction, P<0.01). In the CAM assay, OXT-328 50 μM reduced new vessel formation by 29.3%, Q-922 50 μM by 41.4%, and CL-717 100 μM by 34.2% (all P<0.05 vs. control). Pharmacokinetics: Intravitreal OXT-328 achieved retinal Cmax=516.2±64.4 μM, Tmax=2h, t1/2=4.8h. Topical CL-717 in mice reached retinal Cmax=2.5 μM, Tmax=30 min, t1/2=4.7h. In explanted human eyes, CL-717 reached retinal levels of 12.6 μM (1-min exposure) and 46.6 μM (10-min exposure) after 1h incubation. No systemic side effects were observed except mild ocular surface irritation with topical Q-922.

**Clinical Implications:** These findings demonstrate that modified NSAIDs can effectively suppress retinal angiogenesis through non-injectable routes. CL-717 eye drops and systemic Q-922 offer potential alternatives to intravitreal anti-VEGF injections, which require repeated intraocular administration and carry risks of endophthalmitis and retinal detachment. The reduction of both AVA and NVA distinguishes these compounds from anti-VEGF agents, which may increase AVA. The rapid uptake of CL-717 into human retinal tissue supports clinical translation. Further development is warranted to assess safety and efficacy in human proliferative retinopathies.