CXL – Epi-on or off?

Debate continues as corneal cross-linking tech advances

Howard Larkin

Posted: Sunday, September 1, 2019

There’s no question corneal collagen cross-linking (CXL) stiffens the cornea and is an effective treatment for keratoconus and corneal ectasia. But does CXL’s future lie with the epithelium-off approach developed for the canonical Dresden protocol two decades ago? Or are more recent epithelium-on approaches using various methods to help riboflavin, oxygen and UV radiation penetrate the intact epithelium more promising? Two eminent clinicians debated the question at the 2019 ASCRS ASOA Annual Meeting in San Diego, USA.

Roy Rubinfeld MD, MA

Arguing for epi-on, Roy Rubinfeld MD, MA, of MedStar Georgetown University Hospital, Washington DC, and Medical Director of Re:Vision, Chevy Chase, Maryland and Fairfax, Virginia, USA, challenged the entire debate premise. Given the pain and potential complications of removing the epithelium, he believes the real question should be “why isn’t all standalone cross-linking epi-on?”

The answer goes back to Theo Seiler MD, PhD, and colleagues who invented the procedure in Dresden, Germany, in 1998.

“They did not have a formulation that would go through the epithelium so they just took it off,” Dr Rubinfeld said.

Major potential complications of epi-off CXL include corneal oedema, sterile infiltrates, infectious keratitis, delayed epithelial healing, corneal haze, persistent ocular surface issues as well as stromal scars in nearly 3% of cases – and these are all related to removal of the protective epithelium (Koller et al. JCRS 2009;35:1358).

“We prefer not to have any of these,” Dr Rubinfeld emphasised.

After numerous attempts and failures, one epi-on approach using a novel riboflavin formulation and system has had notable success in multiple studies, Dr Rubinfeld said. This epi-on CXL technology was tested in 592 eyes including 49 paediatric cases with keratoconus and postoperative ectasia. Progression was arrested in all eyes including a consistent cohort over a full two years of follow-up in the consistent cohort (Stulting D et. al. JCRS 2018;44:1363-1370).

In published animal research, this sodium iodide riboflavin formulation known as RiboStat™, CXLO, LLC (Encinitas, California, USA), adequately penetrated corneal epithelium and loaded corneal stroma in rabbit eye studies Dr Rubinfeld and colleagues conducted (Quantitative analysis of corneal stromal riboflavin concentration without epithelial removal. Roy Rubinfeld, MD, R. Doyle Stulting, MD, Glenwood Gum, PhD, Jonathan Talamo, MD. J Cataract Refract Surg. 2018 Feb;44(2):237-242. Erratum in: J Cataract Refract Surg. 2018 Apr;44(4):523. and the sodium iodide protected the riboflavin in the stroma from UV photodegradation (Gum, Rubinfeld, Parsons. ARVO 2019).

CXLO, in which Dr Rubinfeld has a financial interest, is finishing a large phase II clinical trial with interim results on 1,195 keratoconus and 190 post-surgical ectasia cases are largely consistent with Dr Stulting’s findings.

Visual recovery and clear corneas one day after surgery make epi-on the future choice for CXL, Dr Rubinfeld concluded. “The trend in surgery is toward less invasive procedures.”

William J Dupps Jr MD, PhD

In support of epi-off CXL, William J Dupps Jr MD, PhD, of the Cleveland Clinic, Cleveland, Ohio, USA, reviewed criteria for the ideal CXL. In descending order of importance, they are: effective in stabilising the cornea through stiffening; low risk of sight-threatening complications; comfortable for patients; and cost-effective and sustainable in practice.

“So why are we even interested in epi-on? Which of these characteristics could be improved?” Dr Dupps asked.

He allowed that epi-on CXL might lower the procedure’s risk profile and improve patient comfort, but added that the case must be made based on all relevant criteria. He further cautioned that various epi-on approaches, such as varying riboflavin concentrations, admixing various ingredients, mechanical augmentation and iontophoresis to increase stromal penetration of riboflavin, are very different and must be evaluated individually.

Regarding effectiveness, epi-off may be more reliable and predictable, Dr Dupps said. Removing the epithelial barrier helps ensure penetration of riboflavin, oxygen and UV radiation, and reduces the risk of reduced stromal reaction, thus maximising the magnitude of stiffening.

It also may reduce the variability of stiffening, which is essential for developing titratable procedures for safely treating thinner corneas and enhancing reproducibility for future refractive applications of CXL. Several studies also suggest that epi-on CXL induces less clinical flattening, greater late loss of flattening effect and shallower stiffening depth than epi-off.

Regarding avoiding sight-threatening complications and patient comfort, Dr Dupps pointed out that epithelial disruption can still occur with epi-on CXL, and reduced effectiveness could require retreatment.

Epi-on techniques ideally should be equivalent in terms of long-term stability, or show sufficient safety advantages

that repeat treatment is as safe and effective, as epi-off treatment, Dr Dupps concluded.
“Comparative trials large enough to compare complication rates and patient-reported outcomes are needed.”

Roy Rubinfeld:
William J Dupps Jr:

Latest Articles

escrs members advert