Article Text

Download PDFPDF

Gene therapy: new “magic bullets” to prevent ocular scarring
  1. P T Khaw,
  2. A D Cambrey,
  3. G A Limb,
  4. J T Daniels
  1. Wound Healing Research Unit, Epithelial Repair and Regeneration Group, Divisions of Pathology, Cell Biology and Glaucoma, Moorfields Eye Hospital and the Institute of Ophthalmology, University College London, UK
  1. Correspondence to: P T Khaw, Wound Healing Research Unit, Glaucoma Unit and Divisions of Pathology and Cell Biology, Moorfields Eye Hospital and Institute of Ophthalmology, Bath Street, London EC1V 9EL, UK; p.khaw{at}

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Will the advances in modern molecular biology open doors to new therapies?

The processes involved in ocular scarring play a part in either the pathogenesis or failure of treatment of most of the major blinding diseases in the world. These processes include capsular opacification and contraction after cataract surgery. Although posterior capsular opacification is relatively easily treated with laser, this biological process poses great problems following cataract surgery in developing countries, and will inhibit the development of a true accommodating lens replacement. The retinal scarring that occurs in proliferative vitreoretinopathy (PVR) and macular degeneration is also an important example of the blinding scarring process.

The scarring process following glaucoma filtration surgery is one of the best examples of the importance of being able to control healing in virtually all patients having a particular procedure. Recent data from the NIH advanced glaucoma intervention study (AGIS)1 have shown that individuals with the lowest intraocular pressures (average 12.3 mm Hg) had virtually no overall glaucomatous progression over nearly a decade. The healing response after surgery is the main long term determinant of long term intraocular pressure. Therefore, if we are able to control the healing response in all patients after glaucoma surgery, it offers us the tantalising prospect of minimal or no disease progression in the vast majority of our glaucoma patients, even those with advanced disease.

The advent of anticancer agents has revolutionised glaucoma surgery in patients who have a high risk of failure following surgery. Their use has now been extended to patients with a lower risk of surgical failure in an attempt to achieve lower final intraocular pressures. However, these agents are relatively non-specific and exert their action by causing cellular growth arrest and widespread cell death. It is therefore still difficult to titrate the effects, and many side effects occur. These include corneal toxicity associated with 5-fluorouracil injections because this drug is non-specific and kills epithelial cells as well. Other side effects include hypotony, and thin cystic drainage blebs that are associated with blinding side effects such as hypotony and endophthalmitis. Therefore the search continues for novel treatments that can control the wound healing response without the side effect profile seen with current anticancer agents.

Will the advances in modern molecular biology open doors to new therapies? The molecular knowledge of enzyme systems in non-mammalian systems allows us to transfect human cells with these enzymes. This allows us to specifically target only cells that have been transfected with these genes. Drugs that require these enzymes to be activated target the transfected cells—hence the term “magic bullet.”

if we are able to control the healing response in all patients after glaucoma surgery, it offers us the tantalising prospect of minimal or no disease progression in the vast majority

In this issue of the BJO (581), Akimoto et al report an innovative combination of existing anticancer agents and the new gene therapies. Tenon's fibroblasts were transfected with an enzyme called cytosine deaminase, which is only found in bacteria and fungi. 5-Fluorocytosine is a non-toxic prodrug that is converted to 5-fluorouracil by cytosine deaminase. This gene was then inserted into genetically marked Tenon's fibroblasts, which were implanted subconjunctivally. Cell death appeared to occur only in cells transfected with the enzyme gene, without any corneal toxicity. In theory, if cells in a trabeculectomy area could be transfected with this gene, then topical drops could be applied affecting only the cells in the wound area. This moves us closer to the magic bullet so coveted in cancer chemotherapy.

There are still many hurdles to overcome. Adenovirus transfer to the subconjunctival space can be achieved successfully, but transfection is only transient with a peak at 7 days and elimination by 14 days.2 However, inhibition of scarring in the early period of scarring may be all that is necessary in most patients. After all, this is part of the basis of single intraoperative applications of anticancer agents.3 Longer lasting treatments may be required for more aggressive prolonged scarring which occurs in higher risk patients.

Adenovirus vectors do have disadvantages that include a host immune reaction which increases with repeated use, and non-specific transfection of all cells. Modifications of the adenovirus vector may overcome many of these problems.4 Other virus vectors may have advantages,5 but the fact that they integrate with the host genome may make them less desirable, particularly in less permanent situations such as scarring after surgery.6 Advances in the type of vectors will be very important in determining how much this type of therapy becomes clinical reality in the near future.7

Advances in our understanding of molecular biology also offer us other exciting forms of gene therapy. Certain growth factors found in damaged tissues stimulate healing and scarring. One growth factor, transforming growth factor β (TGF-β), stimulates more ocular fibroblast scarring activity than other growth factors.8 Fetal wound healing, which is associated with scarless healing, has an environment that is lacking in TGF-β. The local production of these growth factors can be inhibited by blocking or destroying the RNA molecules that encode the production of this growth factor. This achieves highly specific control of one arm of wound healing leaving others intact. This can be done using short protected chains of DNA (antisense oligonucleotides) or by inserting the genes encoding for enzymes called ribozymes which destroy precise sequences of RNA.9 Advanced molecular techniques also permit the creation of another type of “magic bullet” by facilitating the selection of immunoglobulin genes and the synthesis of highly specific human antibodies to TGF-β2. These antibodies have been shown to be effective in preventing scarring in a model of filtration surgery10 and resulted in reduced final intraocular pressures in a pilot human study without the thin cystic blebs seen in eyes treated with anticancer agents.11 Future advances in our understanding of genotype, perhaps helped by gene microarrays, may help us to identify groups of patients that scar more aggressively and also identify subgroups that may respond better to certain treatments.

Finally, it is appropriate that the current paper by Akinoto et al combines both old and new technologies. In our rush to embrace modern molecular medicine, we must not forget that existing treatments may still have much to offer. A simple change in the technique of antimetabolite application has reduced our long term bleb related complications in a high risk group from 15% to 0%.12 Based on simple cell culture modelling,13 the use of a inexpensive continuous infusion of 5-fluorouracil combined with heparin has more than halved the incidence of PVR in a high risk group from 26% to 11%,14 the first randomised clinical trial to show that PVR could be significantly reduced. It is likely that there are many more “hidden treasures” in combinations of existing and new treatments, an analogy being the 90+% “cure” rates seen in some previously untreatable cancers with combinations of old and new treatments. There are literally millions of patients undergoing surgical treatments that could benefit if these treasure chests can be unlocked. The keys lie in the commitment and support for future basic and clinical research.


The authors are supported by the Guide Dogs for the Blind, RNIB, Wellcome Trust 062290, and the Medical Research Council (UK) grant G9330070.


Linked Articles