Enhancing outcomes in minimally invasive corneal endothelial transplantation

Good vision is dependent on the eye's ability to transmit and focus light onto the retina. The first structure the light passes through is the cornea, a clear dome shaped structure that lies in front of the iris and pupil. It is the most powerful lens in the eye. The cornea is comprised of three cellular layers. The innermost layer is a sheet of cells called the endothelium which serves to pump water out of the middle layer, the stroma. If these cells stop working the cornea becomes waterlogged and opaque. This causes the patient's vision to be diminished and the corneal skin layer to form painful blisters. Failure of this pump cell layer (endothelial failure) commonly results from inherited disease (Fuchs dystrophy) or complicated cataract surgery, and typically affects older patients. Over 50% of all corneal transplants are performed for endothelial failure. Corneal transplants were, until recently, a simple full-thickness button of new tissue sewn place after removal of an equal sized disc of damaged corneal tissue. Full thickness transplantation is normally performed under general anaesthetic, weakens the eye wall permanently, and requires a 2 year visual rehabilitation which often remains incomplete because of an irregular final corneal shape (and therefore an irregular focus). A key-hole version of this operation called endothelial keratoplasty (EK) has replaced full-thickness transplantation for endothelial failure. In EK, only the diseased back layer of the cornea is taken away and replaced with a new sheet of pump cells implanted with an injector through a self sealing small incision similar to that used in modern cataract surgery. EK can be performed safely under local anaesthetic, maintains a normal eye wall strength and shape, and is associated with a quick visual rehabilitation. The transplanted endothelial cell sheet in current EK techniques is either supported by a thin back layer of corneal tissue or an even thinner natural anatomical membrane (Descemets membrane) and no other tissue. The thicker grafts, named DSAEK (Descemets stripping automated endothelial keratoplasty) grafts, are easier to handle surgically and stick well. The thinner type of graft, DMEK (Descemets membrane endothelial keratoplasty) grafts, offer better vision but are much harder to prepare as donor tissue is easily damaged, and implantation requires lengthy surgery. Patients having DMEK often need multiple surgeries to ensure complete adhesion of the graft which can easily separate away from the host cornea. But importantly, DMEK grafts appear to be relatively protected from corneal transplant rejection and this may be because they do not include as much supporting tissue.

Our research aims to identify reliable ways to produce grafts that combine the beneficial properties of both DSAEK (ease of surgery and good adhesion) with those of DMEK (good vision and low rejection). We will do this by using lasers and image guidance technologies to produce a hybrid DMEK with a thin supporting tissue rim at the edge to promote ease of handling. This will slot into a reciprocal laser cut groove in the recipient cornea ('Lock & Key DMEK-S). Exploring ways in which lasers and image guidance can improve dimensional control in EK donor tissue preparation, and ways in which pre-prepared donor tissue can be transported safely to the operating theatre, we hope to 'maximise benefit from minimally invasive corneal transplantation,' widening access to DMEK surgery, making results more reproducible and reducing cost.

In this fellowship I will investigate the parameters required to reliably produce 'Lock & Key' DMEK-S (Descemets membrane endothelial keratoplasty) grafts. We hypothesize that optical (OCT) coherence tomography could be used for precise guidance of of the femtosecond laser for laser dissection of the graft and host corneas which will standardize the process. We also hypothesize that reduced manual donor tissue manipulation and enhanced organ culture media can limit iatrogenic endothelial cell loss and improve storage time pre-transplant. Preliminary work will be carried out on human corneoscleral rims (normally discarded after transplantation) or donor tissue that is unsuitable for transplantation. Transplant grade corneas will only be required during the final stages of this research if the techniques are shown to be viable and a patient study is undertaken.

The objectives are:

1) Defining the parameters required for consistent DMEK-S preparation using image guidance
2) Examination of graft smoothness, shape memory and adhesion properties of DMEK-S donors using atomic force microscopy
3) To determine endothelial cell viability before and after cutting, and the effect of returning pre-cut grafts to storage media
4) To investigate whether the addition of culture supplements and/or the ROCK inhibitor Y-27632 can limit iatrogenic cellular loss and enhance graft storage times.

By identifying the necessary strategies required to produce and store 'Lock & Key' DMEK-S grafts we aim to widen the access to enhanced endothelial keratoplasty techniques and disseminate the benefits to patients. By performing this work in an eye bank setting from the outset we hope our work will ready to translate into improvements for patients within a short timescale. Although the trial will be extrinsic to this project, if progress is good, we hope to proceed to a phase 1 clinical trial of Lock & Key DMEK-S in year 3.

Over 150,000 corneal transplants are performed each year and endothelial dysfunction is the most commonly cited cause. It is anticipated that the number will increase as the population grows and ages. Selective replacement of the diseased posterior cornea, a procedure termed endothelial keratoplasty (EK), is now more common than full thickness transplantation for isolated endothelial dysfunction. It has several advantages including better visual outcomes, quicker visual recovery and negatively impacts the structural integrity of the eye less.
Universal adoption of the dominant contemporary EK technique (DSAEK) has been slow and is restricted to surgeons with access to pre-cut material (only one eye bank currently supplies pre-cut EK donors in the UK) or expensive technology usually within specialist centres.
In the last 5 years a new technique named EK technique named Descemets membrane endothelial keratoplasty (DMEK) has been developed. Results from the centres pioneering this technique show DMEK offers further improvements to the patient with 85% achieving better than 0.1 logMAR vision. Visual recovery and patient satisfaction are also higher than in DSAEK. Perhaps most importantly the risk of rejection, which remains the commonest cause of graft failure, is more that 15 times lower than in DSAEK. At present there are 3 major barriers to the adoption of DMEK 1) tissue preparation: Peeling Descemets membrane is a difficult task to master and results in much tissue wastage in the learning phase. At present DMEK has only been adopted by surgeons with extremely high volume practises. They perform several hundred EK's per year and most have an in house eye bank with abundant tissue. 2) Graft insertion: Due to the inherent tendency of the DMEK graft to scroll, insertion, followed by unrolling of the graft in the anterior chamber, can take several hours and result in damage to the endothelial cells being transplanted. This makes DMEK unfeasible in an NHS setting where one operation would occupy the whole list. 3) Graft adhesion: Graft dislocation remains the commonest early complication in EK and is associated with a significant increased risk of graft failure. Dislocation and/or re-injection of air is needed in 60-90% of DMEK cases.

Our aims are to develop and refine a technique for producing hybrid DMEK grafts where a peripheral stromal ring acts to support Descemets membrane. These grafts retain the benefits of DMEK but are easier to handle and implant. We also wish to explore strategies for improving graft adhesion by performing matched countersunk femtosecond trephination in the host stroma. Our goal is to prepare the grafts in a national eye bank and deliver pre-cut tissue to surgeons across the country.

If successful this will help disseminate access to this type of surgery throughout the UK. Femtosecond lasers capable of performing the host dissection required for lock & key DMEK-S are already widely disseminated. Corneal transplantation could be simplified to a short operation under topical anaesthetic. This has an impact on both quality of life and also economic factors such as time off work. As was the case when cataract surgery was modernised with the adoption of phacoemulsification, newer, more reliable procedures with quicker recovery allow patients to be treated at an earlier stage of disease. This reduces the overall burden of disease on society.

Enhanced storage media with improved graft survival life in storage would maximise yield from tissue already in short supply.

Development of these new modalities may benefit policy-makers, governments and government agencies by providing a potentially more cost-effective approach to the treatment of this condition. This is important when considering the ageing demographic and its impact on healthcare. It is anticipated that adoption of this technique from proof of concept to clinical practice would take approximately 2 years