Advances in Corneal Research at ARVO 2009

Our coverage focuses on several areas that have seen considerable innovation in the last five years.

BY LISA BUTLER, CONTRIBUTING EDITOR

Each spring, the annual ARVO conference generates a bumper crop of new research from some of the brightest minds in ophthalmology, who seek to advance the basic science of ocular pathology and to improve therapeutic options for management. Though it would probably take until next year's meeting to fully assimilate all their contributions into our existing body of knowledge, we've identified some of the more thought-provoking findings on cornea topics that may impact clinical practice now or in the near future.

Below are summaries of selected papers on dry eye, wound healing and corneal transplants from this year's ARVO. For those readers who'd like more detail on the studies, presentation numbers are provided in brackets throughout the discussion. Abstracts can be found at www.arvo.org.

Corneal Transplantation

The ongoing evolution of corneal transplant techniques was given much podium time at this year's ARVO.

Full-thickness corneal transplantation by penetrating keratoplasty (PKP), the standard of care for 50 years, was superseded in 2000 by deep lamellar endothelial keratoplasty (DLEK), which replaces a pocket of tissue in the endothelium through a limbal scleral tunnel. Now, deep anterior lamellar keratoplasty (DALK) and Descemet's stripping endothelial keratoplasty/Descemet's stripping automated endothelial keratoplasty (DSEK/DSAEK) account for most transplants being performed.

DALK replaces the top 80% to 90% of the cornea but preserves Descemet's membrane and the endothelium, lessening the possibility of graft rejection since the donor tissue is not recognized as foreign. DSEK and DSAEK, by contrast, replace only the endothelium and Descemet's membrane. PKP still has a place, according to Mark Terry, M.D., of Devers Eye Institute in Portland, Ore., who pioneered the DLEK technique. "PKP is done on any eye that has both endothelial dysfunction and surface or stromal disease and for regraft procedures with patients who had unacceptable levels of astigmatism before the primary graft."

Above, a DSEK patient at one week postop. Below, the same patient illustrated with Scheimpflug photography showing excellent tissue detergencence and donor adhesion. Images courtesy of Michael W. Belin, M.D.

Dr. Terry [1236] discussed the survival of endothelial keratoplasty grafts at 5 years and beyond, drawing from a patient population of ~900 cases. He reported dislocation rates of 2.8% for DLEK and 1.8% for DSAEK with 0% graft failure at 3 years. Endothelial cell loss after DLEK seems to plateau at 5 years at 60% for 9.0 mm incisions and 72% for 5.0 mm incisions. Endothelial cell loss after DSAEK plateaus between 12 to 24 months at 32%, based on data from patients followed for 2 years.

Clements and colleagues [1237] described factors related to survival of 97 DSEK regrafts after primary PKP. Although 31% of regrafts dislocated and 97% required rebubble for attachment, 98% of grafts were attached after 6 months of follow-up. The most significant risk factor for dislocation was a glaucoma drainage device. He concluded that DSEK is an effective option for failed PKP.

DelMonte [1238] provided the Duke experience with risk factors for complete graft dislocation after DSEK, which occurred in 5% of cases. Statistically significant factors included donor tissue trauma (during insertion and unfolding or from frequent eye rubbing) and anterior chamber characteristics such as a shallow anterior chamber precluding visibility or presence of an IOL.

Comparative outcomes for PKP vs. DALK for keratoconus were presented by Cohen et al. [1241]. Endothelial rejection occurred in approximately 40% of PKP procedures as opposed to 0% of DALK procedures, but interface haze was present more often after DALK.

Two investigators described experimental procedures using the femtosecond laser to prepare the donor tissue for DALK. Pogorelov [1239] used the laser in rabbit eyes to create a tunnel to precisely position the needle for air insertion, whereas Muftuoglu [1240] used eye bank corneas to compare the proportion of perforations in femtosecond-assisted ring cuts or zigzag side cuts to manual trephination, concluding that the zigzag side cut was promising. Dr. Terry noted that "femtosecond laser-assisted keratoplasty offers significant advantages in terms of wound strength and early suture removal," but cautioned that "there is no evidence it produces better visual outcomes."

Ocular immunology research may someday offer a different means of improving graft survival. After keratoplasty, migration of dendritic cells from the cornea to the draining regional lymph nodes triggers an immune response. The chemokine receptor CCR7 mediates this migration, guided by the inflammatory protein CCL19. Bachmann et al. [3036] showed that blockade of CCR7 with CCL19-IgG significantly reduced dendritic cell migration and studied the effect of preop intraperitoneal dosing and postop topical dosing of CCL19-IgG on graft survival in mice. Survival proportions in the treated group were 76% compared to 38% in controls (IgG) 8 weeks after transplantation.

Research reported by Eghrari et al. [3888] in large families with Fuchs corneal dystrophy (the #1 indication for DSEK/DSAEK) not only verifies that females are more likely to be affected than males, but also shows that females are over-represented in these families. It is possible a genetic factor may exist that increases both the number of females and the development of disease.

Total limbal stem cell deficiency is a relatively recently recognized condition in which conjunctivalization and neovascularization of the corneal epithelium causes repeated, persistent surface breakdown. Management usually involves transplantation of stem cells from the corneoscleral limbus using carrier tissue from the conjunctiva or cornea from the fellow eye (limbal autograft) or from a cadaver donor (limbal allograft). Zheng [6178] emphasized the importance of harvesting tissue from stem cell niches in order to assure the re-epithelialization of the corneal surface. Nakamura [6178] described long-term results of autologous oral mucosal epithelial transplantation, reporting that after 36 to 72 months of follow-up nearly all the transplanted epithelium survived on the ocular surface and 38% of eyes showed more than 2 lines of improvement in visual acuity.

Dry Eye Diagnosis And Potential Therapeutic Targets

The advantages of conventional in-office methods of dry eye diagnosis (Schirmer's test, fluorescein staining, etc.) are convenience and low cost, but they come at the expense of accuracy, as these methods interact with the tear film and thus alter the result to some extent. To investigate noninvasive alternatives, objective measurement of tear film dynamics was the focus of several presentations.

Koh et al. [1230] used two high precision methods — wavefront analysis and OCT scanning — to simultaneously measure optical aberrations caused by the tear film and the quantity of the lower tear meniscus in normal subjects and dry eye patients. Correlations found between these measurements suggest that the tear quantity before the blink has a role in maintaining optical quality of the tear film. Yuan and colleagues [1233] also studied the tear menisci with realtime OCT, concluding that it is a versatile technique.

Using the Optical Quality Analysis System, or OQAS (Visiometrics, Spain), Perez et al. [1232] measured ocular scattering characteristics of the tear film in healthy subjects and dry eye patients. The OQAS records images from a point source after reflection in the retina and a double pass through ocular media to derive an objective scatter index (OSI), a measure of optical aberrations. Dry eye patients exhibited increased average OSI and higher fluctuations than controls. The authors suggest the test could be used to estimate tear film quality, aid diagnosis and quantify response to therapy. Given the increasing deployment of wavefront and OCT technologies in ophthalmic practices, they may one day be used for objective differentiation between normal and dry eyes in the clinic.

Research investigating the components of the tear film may bring us one step closer to understanding distinctions between subsets of dry eye patients — and perhaps identifying targets for drug therapy. Grus et al. [2546] used high performance liquid chromatography/mass spectrometry to analyze protein patterns in the tear films of 39 subjects subdivided into those with aqueous deficiency, lipid abnormality, a combination of both characteristics, and healthy controls. Each subgroup presented a very distinct profile of tear film proteins. The authors concluded that the presence of clear distinctions between groups and identification of biomarkers such as calgranulin, proline-rich protein 4 and alpha anti-trypsin could lead to a diagnostic tool for dry eye and innovative targets for drug development.

Calonge and associates [2548] studied the tear cytokine levels of patients with Meibomian gland disease, determining that inflammatory mediators, identified in their research by Luminex bead assay, are a more sensitive and reliable indicator of dry eye than ocular surface staining. A third methodology, the micro plate array, was modified by Sack and colleagues [2547] to increase its sensitivity to 40+ low-abundance proteins (LAP) that modulate inflammation and wound healing. They hoped to identify biomarkers to help refractive surgeons screen candidates for surgery based on their propensity to develop postoperative dry eye. Nine cases of postop dry eye and nine healthy controls were studied.

Results suggest that 4% of the general population may have a LAP profile that would suggest reduced capacity to withstand the short-term inflammatory stress of corneal surgery. This research may yield an in-office test capable of identifying patients who will go on to develop dry eye after surgery, allowing the physician to either prescribe a medication regimen to counter it or to advise against surgery.

Steve Pflugfelder, M.D., a key contributor to the understanding of the inflammatory component to dry eye disease, stated that "based on this research and previously published studies, I envision there will be diagnostic tests to measure inflammatory mediators in the tears or ocular surface cells in the near future."

Ocular Surface Health

The toxicity of the preservative BAK (benzalkonium chloride) was investigated by Brignole-Baudoin [4780] and Liang [4781]. Both researchers used a novel in vivo confocal microscopy (IVCM) technique to examine the effect on rabbit eyes of antiglaucoma eye drops containing various concentrations of BAK vs. preparations without BAK. Brignole-Baudouin's study observed live cell trafficking in conjunctiva-associated lymphoid tissue. Latanoprost, containing the highest concentration of BAK (0.02%), induced substantial inflammatory cells whereas bimatoprost, containing the lowest concentration of BAK (0.005%), and BAK-free travoprost showed rare leukocyte infiltration.

Dr. Liang compared BAK-containing antiglaucoma eye drops to BAK-free travoprost and latanoprost in a cationic emulsion formulation, using IVCM, flow cytometry and impression cytology to rate ocular surface toxicity. The BAK-containing formulations caused moderate irritation, hyperemia, and chemosis, with infiltration of inflammatory cells (leukocytes and eosinophils). The non-BAK formulations did not affect the microscopic structure, and normal epithelium and goblet cells were present.

The ensuing discussion was heated; while one researcher insisted that some BAK effect increases drug penetration, others found this demonstration of BAK toxicity, manifested as attraction of inflammatory cell infiltration, compelling. It's worth bearing in mind, however, that these were animal studies rather than human subjects and the results have yet to be replicated in glaucoma patients.

Dr. Liang's study was partially supported by French pharmaceutical manufacturer Novagali Pharma, which has a BAK-free formulation of latanoprost, called catioprost, currently in a Phase 3 FDA trial.

Pharmacologic Targets For Corneal Wound Healing

Currently, there is no therapeutic agent that promotes wound healing. Sosne et al. [3491] described a small compassionate-use study of thymosin β4, a naturally occurring peptide known to promote corneal reepithelialization, to treat neurotrophic corneal ulcers in 4 patients. Treatment for 28 days resulted in significant reduction of ulcer size and increase in comfort while producing no clinically significant adverse events. Dr. Sosne concluded that controlled clinical trials were warranted. Regenerx Biopharmaceuticals terminated a phase 2 trial in corneal wound healing in diabetic patients in February due to slow enrollment.

Aspects of molecular regulation of corneal wound healing in diabetic corneas were highlighted by two research groups. Xu and Yu [5196] determined that high glucose levels impaired cellular antioxidant defense and the PI3K/Akt pathway, leading to the delayed wound healing seen in diabetic patients. In an article in Diabetes, they theorize that antioxidants in combination with EGFR ligands may be a promising therapeutic target.¹ Ljubimov et al. [5197] transduced normal human corneas with matrix metalloproteinase-10 and cathepsin F. Overexpression of these two proteinases resulted in reduction of proteins responsible for the integrity of the basement membrane, and resulted in slower wound healing. The author suggests a potential role for gene therapy in corneal diabetes in which these genes might be silenced using siRNA or antisense technology.

Acceleration of corneal healing through inhibition of the tumor suppressor, PTEN, was the focus of research by Zhao et al. [3493]. In human and rat cornea preclinical models, suppression of PTEN by a vanadium derivative, bpV(pic), doubled the migration rate of cells and provided directional cues to the migrating cells by activating the PI3K/Akt signaling pathway. This significantly increased the healing rate of scratch wounds in both models, and Zhao noted that PTEN appears to be a good therapeutic target.

Gregory and associates [3456] studied the role of CD36 as a barrier to corneal infection. This class B scavenger receptor maintains the corneal epithelium by contributing to migration, adhesion and formation of tight junctions. After epithelial wounding, CD36 KO mice show delayed reepithelialization and persistent inflammation which allow corneal flora to bind.

Although some of these papers raise more questions than they answer, one thing is clear: the accelerated pace of research in cornea will yield more dramatic improvements in the years to come. OM

Reference

1. Xu KP, Li Y, Ljubimov AV, Yu FS. High glucose suppresses epidermal growth factor receptor/phosphatidylinositol 3-kinase/Akt signaling pathway and attenuates corneal epithelial wound healing. Diabetes 2009;58:1077-85. Epub 2009 Feb 2.