Could long-term effects of cross-linkage be a concern and a potential source of problems?

Reversible cross-linking methods are needed

John Marshall

It was not until recently that we have been able to measure objectively the biomechanical properties of the cornea. One of the most important parameters is the measurement of stress over strain, ie, the Young’s modulus, which tells us how rigid or elastic the cornea is.

We have been using the Young’s modulus test in normal human corneas aged between 20 years and 104 years and have been able to demonstrate that there is a linear relationship between age and increased rigidity. We then looked at corneas in which the mechanical properties were changed by cross-linkage. On a linear scale, they were equivalent to 600-year-old corneas.

While it may be helpful to stiffen the cornea by cross-linking in conditions such as keratoconus, I ask the question: Do you really want to convert corneal collagen into a very aged system? And having done it, what will be the consequences of this when your patients grow older? The cross-linking procedure is normally performed in young eyes, and long-term consequences are unpredictable. The follow-up we have may appear fairly long, but is not long enough to guarantee safety over a lifetime.

I believe we should do more research on alternative ways of cross-linking. In biological tissues such as the cornea, cross-linking is induced via disulfide bonds between collagen fibrils. This is part of the normal process of aging. What we do by riboflavin and UV irradiation is to dramatically accelerate the accumulation of disulfide bonds. There are less aggressive ways of cross-linking corneal collagen, and they have the significant additional advantage of being reversible.

In my group at St. Thomas’ Hospital, we have an interest in cornea, but our research is mainly focused on the retina. In the Bruch’s membrane, you often deal with a situation that is exactly the opposite to keratoconic cornea. In keratoconus, you have an excessively elastic cornea that you want to make stiffer; in the retina, you have a progressively stiffening membrane that you want to make more elastic. So we have been experimenting ways of undoing cross-linkage, and as a consequence, we have investigated other ways of cross-linking. We have found a number of chemicals that actually produce reversible cross-linkage, so that we can cross-link with one chemical and unlink with a second chemical. This is certainly an area that is worth investigating. Many age-related diseases are the consequence of a cross-linking process. The hardening of arteries is just one of many examples. So this research will have the support of many branches of medicine.

Reversible cross-linking methods are the answer we should be looking for if we want a safe procedure.

John Marshall, PhD, FMedSci, FRCPath, MRCOphth, is the Frost Professor of Ophthalmology and chairman of the Academic Department of Ophthalmology at Kings College, St. Thomas’ Hospital, London.

Best is yet to come

Paolo Vinciguerra

Corneal cross-linking has shown no complications over 10 years. My series amounts to more than 500 patients in 4 years, and apart from a few cases of delayed epithelial healing, no adverse event was reported at any time. We operated on patients with severe keratoconus. Most were or would soon have been candidates for corneal transplantation. To date, none of them has actually undergone keratoplasty. It sounds like a miracle and leaves little doubt that cross-linking should be the first option we offer to keratoconus patients.

The best is yet to come. We expected the corneal stiffening and stabilization of the disease but found that the treatment also reduced astigmatism, coma and refraction, improving visual acuity quite significantly. This was the first surprise. Then, we expected this visual improvement to be circumscribed within the first 7 to 8 months, but reality was actually much beyond expectations. Patients improved from year zero to 1, from 1 to 2, from 2 to 3 and from 3 to 4. They are constantly improving. Between year 3 and 4, we are also beginning to find an increase in corneal thickness and changes in the pachymetry pattern. The thinnest point tends to move more centrally. This was the second surprise.

I think we have triggered a virtuous cycle that is still unknown to us in its mechanisms — an exciting new world to explore in years to come.

So I am not at all concerned about potential long-term damage. What I see is exactly the opposite.

In addition, worrying about the future of these corneas makes no sense at all. If they were not cross-linked, these corneas would no longer be there in a few years. The alternative would be transplantation. If some unexpected, sudden U-turn ever occurred after so many years of positive outcomes of cross-linking, the worse that could happen is that we would have to do corneal transplantation. Nothing worse, nothing different from what would happen naturally much earlier in non-cross-linked corneas.

Paolo Vinciguerra, MD, is Director of the Ophthalmology Department of the Humanitas Clinic in Milan, Italy.