Retinal and choroidal vascular diseases, such as neovascular age-related macular degeneration (nAMD), represent a massive global health burden. These conditions cause vision loss by driving pathological neovascularization (the growth of abnormal, leaky blood vessels). This process is primarily fueled by vascular endothelial growth factor (VEGF). Current treatments involve repeated intravitreal injections (injections into the jelly-like vitreous humor of the eye). These are effective but require frequent, lifelong administration. Existing long-acting "depot" formulations (localized masses of drug that release slowly) often migrate into the anterior chamber (the front part of the eye). This migration can lead to dangerous spikes in intraocular pressure (IOP).
Researchers seek ways to deliver anti-angiogenic drugs—substances that inhibit blood vessel growth—more durably and safely. This paper explores delivering drugs into the suprachoroidal space (SCS). The SCS is the narrow anatomical compartment between the choroid and the sclera (the white outer layer of the eye). By reformulating the anti-angiogenic peptide AXT107 into microparticles, the authors demonstrate a method for sustained therapeutic activity. This approach aims to avoid the complications of traditional delivery routes.
The failure of soluble depots in the eye
Treating retinal vascular disease requires balancing drug residence time with anatomical safety. The current standard relies on intravitreal injections. These are invasive and demand high patient compliance. To extend the interval between injections, scientists have attempted to create depots.
These depots face two critical hurdles. First, soluble peptide formulations can fragment and disperse into the anterior chamber [Figure 1A]. This dispersion is associated with increased IOP. Second, depots may not reach the intended target. In the suprachoroidal space, a soluble AXT107 solution was ineffective at inhibiting laser-induced neovascularization in a rat model [Figure 1B, 1C]. The authors hypothesize that the gel-like mass formed by the soluble peptide creates a transport barrier. This barrier prevents enough drug from traversing the choroid to reach the retina in therapeutic concentrations.
Engineering the microparticle release mechanism
To overcome these transport barriers, researchers moved from a continuous gel to a discrete particulate system: MP-AXT107. The goal was to maximize the surface area available for drug release. This should facilitate better hydration and distribution through the choroid.
The construction of these microparticles follows a specific physicochemical logic: 1. Precipitation: The researchers exploited the limited solubility of AXT107 in ionic environments. They combined an AXT107 solution with a sucrose and NaCl (salt) solution under constant stirring. This caused the peptide to precipitate (solidify) out of the liquid phase. 2. Morphology Control: Unlike the opalescent appearance of the soluble solution, the MP-AXT107 forms an opaque, white suspension [Figure 2A]. Image analysis revealed particles with an average diameter of approximately 5.95 µm [Figure 2B]. 3. Stability Encoding: The formulation was optimized for long-term storage. When kept at 2–8 °C, the microparticles remain chemically stable and physically consistent in size for at least 9 months. This longevity is essential for a viable clinical product.
Breaking the drug into millions of tiny spheres instead of one large mass ensures more ready release. This design promotes more uniform distribution across the target tissue.
Evidence of efficacy and ocular safety
The transition to microparticles yielded significant improvements in therapeutic impact and safety. In the rat CNV model, MP-AXT107 achieved a ~60% reduction in the neovascularized area compared to vehicle controls .
This represents a major improvement over the soluble version. The soluble version failed to show significant inhibition [Figure 1C].
The safety of this approach was tested in a 9-month GLP (Good Laboratory Practice) toxicology study using Göttingen minipigs. The authors measured several key physiological markers: * Intraocular Pressure (IOP): The injection caused an immediate, dose-dependent increase in IOP. However, this effect was transient. It resolved fully within one week [Figure 4B]. * Ocular Inflammation: Researchers observed mild, transient findings like corneal redness and vitreous haze. These occurred immediately post-injection but resolved by Day 5 [Figure 4C]. * Drug Localization: Bioanalysis at the end of the 9-month study confirmed AXT107 stayed within the choroid/RPE (the pigmented layer of the retina) and scleral tissues. There was virtually no systemic exposure in the blood.
The study established the 1.25 mg/eye dose as the No-Observed-Adverse-Effect Level (NOAEL). This provides a safe threshold for future clinical scaling.
Limitations in the current model
Despite these results, the study leaves some questions unanswered. First, the researchers did not capture the very earliest post-injection IOP measurements in the minipig study. Because the rise in pressure is described as "immediate," missing those first few hours makes characterizing the peak difficulty.
Second, while the microparticles localized in the choroid and sclera, the exact mechanism of retinal arrival is inferred. The authors suggest that increased surface area facilitates release. However, they do not directly quantify the flux (rate of movement) of individual peptide molecules across the choroidal barrier in real-time. Finally, while the microparticles showed excellent stability at 2–8 °C, their behavior at 25 °C was more variable. This suggests that strict cold-chain management will be necessary for this therapy.
The verdict on suprachoroidal delivery
The evidence suggests that suprachoroidal microparticle delivery is a superior architecture for sustained ocular therapy. Shifting the delivery site away from the vitreous helps bypass common failures. It avoids both "fragmenting depots" and "insufficient transport."
Achieving a 60% reduction in neovascularization while maintaining safety over nine months is a significant milestone. This approach moves us closer to a highly infrequent dosing regimen for chronic retinal diseases. The next step will be clinical translation. Success will depend on ensuring precise suprachoroidal injections in humans to prevent the mis-injections noted in animal models.
Figures from the paper
How this was made
Model: nvidia/Gemma-4-26B-A4B-NVFP4
Persona: science_essayist
Template: engineering_deepdive
Refinement: 0
Pipeline: forge-1.1
Evaluator: nvidia/Gemma-4-26B-A4B-NVFP4
Score: 96% (passed)
Claims verified: 17 / 17
Model: nvidia/Gemma-4-26B-A4B-NVFP4
NVIDIA GB10 · 128 GB unified · NVFP4 · 100% local · $0 cloud
Tokens: 94,222
Wall-time: 215.4s
Tokens/s: 437.4