Hurdles and Prospects for Episcleral Drug Delivery

KARL G. CSAKY, MD, PhD

The landscape for long-term ocular drug delivery for the treatment of retinal diseases has exploded in the last several years. The first positive study on the use of an intravitreal drug delivery device for a retinal disease was published in 1997¹ for a ganciclovir-releasing device for the treatment of cytomegalovirus retinitis. This was followed by positive results using a similar intravitreal device delivering fluocinolone acetonide for the treatment of patients with severe uveitis.²

Since that last publication there are now multiple ongoing trials evaluating sustained drug-delivery devices for retinal diseases ranging from diabetic retino pathy, retinal vein occlusions, degenerative retinopathies, choroidal neovascularization, and geographic atrophy. While these trials are evaluating a wide range of drug delivery devices from bioerodible to reservoir to cell-based encapsulation systems,^3-6 the one thing that they all have in common is that they are all being placed into the vitreous cavity.

This approach clearly has the sizable advantage that the drug is known to be released directly into the eye and that even low release rates of drug can potentially achieve therapeutic levels within the eye. However, this approach also suffers from several disadvantages. These include the need to penetrate the eye with its concomitant issues of administration discomfort, and risks of endophthalmitis, retinal detachment, and bleeding. In addition, should removal of the implant be required, then a more extensive surgery is typically required to retrieve and remove the implant.

USING AN EPISCLERAL IMPLANT

All of these disadvantages would be negated with the use of an episcleral implant. Designed to be placed on the outside surface of the eye, these types of devices have tremendous promise. The risk of serious complications is low, patient discomfort would be minimal, and the ability to retrieve or replace the implant would be easy. These characteristics would ostensibly lower the threshold for when to treat a patient, allowing for the inclusion of patients with an ongoing chronic disease, such as early dry age-related macular degeneration (AMD), but still with good visual acuity.

So if this approach has so many advantages compared with an intravitreal approach, why are not more episcleral devices in clinical trials for the treatment of retinal diseases?

The reason is very simple: our understanding of how and to what degree drugs enter into the eye is very crude. While it clearly has been demonstrated with multiple agents that drugs placed on the surface of the eye, in the form of topical drops, penetrate into the anterior chamber, episcleral implants need to be placed on the scleral surface. Is the simple change in location of where the drug comes into contact with the eye that important? The answer, simply put, is that we do not really know.

This lack of clear understanding of the mechanism of episcleral drug penetration does represent a risk for many companies interested in using an episcleral approach to achieve drug approval.

A CASE IN POINT

The case in point is the history of anecortave acetate. Developed in the 1990s by Abe Clark and other researchers at Alcon, anecortave acetate proved to be a potent antiangiogenic agent in many preclinical models.⁷ For the pivotal clinical trial, the decision was made to administer the drug via a juxtascleral approach (Figure 1), a technique that placed the drug on the surface of the sclera overlying the macula. Preclinical data in several animals demonstrated that therapeutic levels of anecortave acetate were achieved in the choroid and retina for several months with this approach.⁸ Most importantly, there were extensive clinical data that demonstrated this administration route in humans to be extremely safe.⁹ However, the results of the pivotal trial comparing juxtascleral administration of anecortave acetate to photodynamic therapy were disappointing, with no clear strong antiangiogenic affect of anecortave acetate demonstrated in humans with neovascular AMD.¹⁰ Of course, the question that will always remain from this trial is whether the drug failed or the mode of drug application failed.

Figure 1. Drawing of a juxtascleral injection where drug (blue arrow) is placed on the sclera overlying the macula.

Significant research has been done in understanding the barriers to episcleral drug delivery (Figure 2). For example, a pioneer in this research is Hank Endelhauser at Emory University. His research has focused on the role of the sclera as an anatomic barrier to drug diffusion. However, his elegant research in this area resulted in an interesting conclusion: that the sclera for the most part does not represent a barrier. Using isolated human sclera, he studied drug movement from one side of the sclera to the other and found that drugs, even with a molecular weight of 150 kD, easily pass through the sclera, although drugs with greater molecular radii did penetrate through more slowly.^11-13

Figure 2. (A) Magnetic resonance images of episcleral gadolinium demonstrating poor ocular penetration (color resonance on ocular surface) of drug in live animals but (B) excellent penetration (color resonance inside vitreous) in animals following euthanasia.

This lack of a scleral barrier was confirmed in humans who had been given episcleral dexamethasone preoperatively. Dexamethasone was found in moderate levels in the subretinal fluid in patients undergoing retinal detachment surgery.¹⁴ And yet a trial completed in 2008 evaluating the effects of an anterior or posterior episcleral triamcinolone acetonide injection found that neither approach was that effective in treating diabetic macular edema (DME).¹⁵ Therefore, many questions still remain about the ability of episcleral drugs to penetrate the eye.

POSTHUMOUS PENETRATION

Interesting clues to understanding this question have been slowly forthcoming. For example, several groups studying episcleral drug delivery have stumbled upon an interesting observation in animals. When drug is placed in an episcleral location in live rabbits, for example, little of the drug appears to penetrate the eye. But when the animals are then euthanized, a rapid and intense penetration of the drug into the eye occurs.^13,16 How can one explain these results? These investigators came to the conclusion that there are physiologic barriers present in the live animals that quickly disappear in the euthanized animal. These physiologic barriers could include rapid elimination of the drug by the episcleral blood and lymphatic flow, choroidal blood flow, or the tight junctions of the retinal pigment epithelium.

Another interesting clue surfaced in a study of episcleral implants for the treatment of horses with equine uveitis (Figure 3). When drug delivery implants releasing cyclosporine were placed on the sclera in horses with frequent flairs of equine uveitis, no efficacy was seen. But when these same implants were placed in the suprachoroidal space, a group of more than 50 animals were essentially cured with recurrence rates of uveitis falling to almost 0.¹⁷ What could explain these results? One explanation is that rapid clearance of episcleral drug by scleral and conjunctival blood vessels and lymphatics do not allow sustained elevated concentrations on the outside of the eye and therefore drug diffusion into the eye remains low. Clearly, the research community is making headway in this problem. Several approaches are now under active investigation.

Figure 3. A unidirectional episcleral implant that is sutured and pressed onto the sclera, resulting in release of drug from the implant only in the direction of the sclera.

NEW APPROACHES

One approach is purely technical in nature. If episcleral drug concentrations are constantly low due to rapid drug elimination by conjunctival blood flow and/or lymphatics, then placing an implant that delivers drug only in one direction (so-called unidirectional drug devices) tightly on the sclera so that drug is only delivered in one direction toward the sclera, then perhaps drug can be "forced" into the eye. Indeed, preliminary work on this type of implant supports that an increase in the amount of drug that enters the eye using this type of implant can occur (Figure 4).¹⁸ And while these results are encouraging, one hurdle of episcleral drug delivery still remains. Unlike intravitreal drug devices where essentially 100% of the released drug is available inside the eye, even in the best episcleral devices only about 1% of drug enters the eye with this approach. In other words, only about 1% of the drug is bioavailable.

Figure 4. Cyclosporine episcleral implant implant ed into the suprachoroidal space of horse eyes with equine uveitis.

To understand the ramifications of this low drug bioavailability, take for example the following hypothetical drug A. To achieve a desired clinical effect, drug A has to maintain a drug amount inside the eye of 100 μg per day. However, if only 1% of the drug gets into the eye, then an episcleral implant would have to release 10 mg of drug per day. If one would like to place these episcleral types of implants once every 6 months into patients, then such an episcleral implant would have to contain 1600 mg of drug (more than a gram of drug), an approach that clearly is not feasible.

An alternative technical approach is the study of episcleral microneedles (Figure 5).¹⁹ This technology approaches the problem of low bioavailability by proposing to place an episcleral implant containing hundreds of tiny needles that penetrate into the sclera and delivers drug directly into the scleral tissue. Extending the excellent results of the cyclosporine implants that were placed in the suprachoroidal space in horses with equine uveitis, the technology offers the hope that placement of drug, with minimum surgical dissection and risk, into the scleral space in humans, can also generate successful therapeutic results in patients.

Figure 5. Size of a microneedle delivery device for intrascleral delivery shown demonstrating the small (500 μm) size of the needle.

Alternatively, investigators are trying to decipher the exact physiologic barriers that exist. Similar to the world of oral drug absorption, this approach would allow drugs to be "tailored" to circumvent these physiologic barriers and more efficiently penetrate the eye.

The issue of episcleral drug penetration and how the unique properties of the drug may influence its ocular penetration has come to the forefront again with the recent announced results of the phase 1 episcleral rapamycin for diabetic macular edema study.^20,21 Demonstration of a reduction in retinal thickening and an improvement in visual acuity in patients with diabetic macular edema receiving episcleral rapamycin has suggested that this approach may indeed be viable for certain types of drugs. However, caution needs to be expressed as positive results were also demonstrated with the early phase trials of juxtascleral anecortave acetate.²²

CONCLUSION

Episcleral drug delivery is the future for the safe delivery of drug to the retina and choroid. Many challenges are still present, but progress is coming. While there are significant hurdles, there are also opportunities that might allow physicians of the future to be able to offer therapies with minimal risks and side effects for many of the common diseases including age-related macular edema and diabetic retinopathy. RP

REFERENCES

1. Musch DC, Martin DF, Gordon JF, Davis MD, Kuppermann BD. Treatment of cytomegalovirus retinitis with a sustained-release ganciclovir implant. The Ganciclovir Implant Study Group. N Engl J Med. 1997;337:83-90.
2. Callanan DG, Jaffe GJ, Martin DF, Pearson PA, Comstock TL. Treatment of posterior uveitis with a fluocinolone acetonide implant: three-year clinical trial results. Arch Ophthalmol. 2008;126:1191-1201.
3. Fluocinolone acetonide implant for retinal vein occlusion (RVO). Clinicaltrials. gov Web site. http://clinicaltrials.gov/ct2/show/NCT00636493. Accessed June 24, 2009.
4. A study of the safety and efficacy of a new treatment for macular edema resulting from retinal vein ecclusion. Clinicaltrials.gov Web site. http://clinicaltrials.gov/ct2/show/NCT00168324. Accessed June 24, 2009.
5. A study of an encapsulated cell technology (ECT) implant for patients with atrophic macular degeneration. Clinicaltrials.gov Web site. http://clinicaltrials. gov/ct2/show/NCT00447954. Accessed June 24, 2009.
6. Fluocinolone acetonide intravitreal inserts in geographic atrophy. Clinicaltrials.gov Web site. http://clinicaltrials.gov/ct2/show/NCT00695318. Accessed June 24, 2009.
7. Penn JS, Rajaratnam VS, Collier RJ, Clark AF. The effect of an angiostatic steroid on neovascularization in a rat model of retinopathy of prematurity. Invest Ophthalmol Vis Sci. 2001;42:283-290.
8. Vinores SA. Anecortave (Alcon Laboratories). IDrugs. 2005;8:327-334.
9. Augustin AJ, D'Amico DJ, Mieler WF, Schneebaum C, Beasley C. Safety of posterior juxtascleral depot administration of the angiostatic cortisene anecortave acetate for treatment of subfoveal choroidal neovascularization in patients with age-related macular degeneration. Graefes Arch Clin Exp Ophthalmol. 2005;243:9-12.
10. Slakter JS, Bochow TW, D'Amico DJ et al; Anecortave Acetate Clinical Study Group. Anecortave acetate (15 milligrams) versus photodynamic therapy for treatment of subfoveal neovascularization in age-related macular degeneration. Ophthalmology. 2006;113:3-13.
11. Cruysberg LP, Nuijts RM, Geroski DH, Gilbert JA, Hendrikse F, Edelhauser HF. The influence of intraocular pressure on the transscleral diffusion of high-molecular-weight compounds. Invest Ophthalmol Vis Sci. 2005;46:3790-3794.
12. Geroski DH, Edelhauser HF. Transscleral drug delivery for posterior segment disease. Adv Drug Deliv Rev. 2001;52:37-48.
13. Kao JC, Geroski DH, Edelhauser HF. Transscleral permeability of fluorescent-labeled antibiotics. J Ocul Pharmacol Ther. 2005;21:1-10.
14. Weijtens O, Schoemaker RC, Lentjes EG, Romijn FP, Cohen AF, van Meurs JC. Dexamethasone concentration in the subretinal fluid after a subconjunctival injection, a peribulbar injection, or an oral dose. Ophthalmology. 2000;107:1932-1938.
15. Diabetic Retinopathy Clinical Research Network; Chew E, Strauber S, Beck R, et al. Randomized trial of peribulbar triamcinolone acetonide with and without focal photocoagulation for mild diabetic macular edema: a pilot study. Ophthalmology. 2007;114:1190-1196.
16. Robinson MR, Lee SS, Kim H, et al. A rabbit model for assessing the ocular barriers to the transscleral delivery of triamcinolone acetonide. Exp Eye Res. 2006;82:479-487.
17. Gilger BC, Salmon JH, Wilkie DA, et al. A novel bioerodible deep scleral lamellar cyclosporine implant for uveitis. Invest Ophthalmol Vis Sci. 2006;47:2596-2605.
18. Pontes de Carvalho RA, Krausse ML, Murphree AL, Schmitt EE, Campochiaro PA, Maumenee IH. Delivery from episcleral exoplants. Invest Ophthalmol Vis Sci. 2006;47:4532-4539.
19. Jiang J, Gill HS, Ghate D, et al. Coated microneedles for drug delivery to the eye. Invest Ophthalmol Vis Sci. 2007;48:4038-4043.
20. Dose ranging study of an ocular Sirolimus (Rapamycin) formulation in patients with diabetic macular edema. Clinicaltrials.gov Web site. http://clinicaltrials. gov/ct2/show/NCT00656643. Accessed June 24, 2009.
21. Blumenkranz MS. Current and future trends in the treatment of macular edema. Paper presented at: Annual meeting of the Retina Society; September 27-30, 2007; Boston, MA.
22. D'Amico DJ, Goldberg MF, Hudson H, et al; Anecortave Acetate Clinical Study Group. Anecortave acetate as monotherapy for treatment of subfoveal neovascularization in age-related macular degeneration: twelve-month clinical outcomes. Ophthalmology. 2003;110:2372-2383; discussion 2384-2385.