Corneal cross-linking (CXL) remains the only intervention proven to halt keratoconus progression. While the original Dresden protocol established the efficacy of epithelium-off (epi-off) CXL, its associated pain, delayed healing, and infection risk have long motivated attempts to develop effective epithelium-on (epi-on) alternatives. However, achieving equivalent biomechanical outcomes with an intact epithelium – transepithelial corneal cross-linking – has proven technically challenging due to barriers in riboflavin penetration and oxygen diffusion.
In a recently published ex vivo study, our group systematically evaluated a newly developed transepithelial corneal cross-linking protocol. Using porcine corneas as a validated ex vivo model, the authors compared biomechanical outcomes between their novel epi-on protocol and the widely adopted accelerated epi-off protocol (9 mW/cm2, 10 minutes, 5.4 J/cm2 fluence). This study was published in the Journal of Refractive Surgery.
Key to this new approach is a second-generation penetration enhancer solution that facilitates riboflavin diffusion through the intact epithelium without inducing epithelial damage. In addition, the protocol employs accelerated pulsed UV-A irradiation (18 mW/cm2, 1 second on/1 second off, for 15 minutes; total fluence 8.1 J/cm2) to optimise oxygen availability during the cross-linking process.
Biomechanical assessment was conducted via stress-strain extensiometry, evaluating the elastic modulus at 5-10% strain (which is the standard range for such analyses). The results demonstrated no significant difference between the elastic moduli of the new epi-on protocol and the accelerated epi-off reference, while both were significantly stiffer than untreated controls. This suggests that, at least biomechanically, the transepithelial protocol delivers an effect equivalent to accelerated epi-off CXL.
First-authored by Dr. Nanji Lu, this study also carefully addresses known challenges in epi-on CXL. The epithelial barrier presents a dual limitation: riboflavin, a hydrophilic molecule, poorly penetrates intact epithelium, while the epithelium itself consumes oxygen at rates substantially exceeding stromal levels. Earlier attempts, including first-generation chemical enhancers and iontophoresis, either failed to deliver sufficient riboflavin concentrations or added procedural complexity without fully resolving these limitations.
By resolving both the riboflavin and oxygen constraints through chemical enhancement and optimised UV-A dosing, the protocol described by Lu et al. potentially simplifies epi-on CXL into a fully office-based procedure that may obviate the need for a sterile surgical environment.
Nonetheless, the authors acknowledge critical limitations. The ex vivo porcine model, while methodologically robust, cannot fully replicate in vivo responses such as wound healing and keratocyte remodelling. Furthermore, differences in anatomy (particularly the thicker porcine epithelium and absence of Bowman’s layer) require careful consideration when extrapolating findings to clinical practice.
While these results provide compelling biomechanical evidence, clinical validation remains essential. A clinical trial is currently ongoing to assess long-term safety and efficacy. If subsequent clinical data confirm these ex vivo findings, this transepithelial cross-linking approach could mark a meaningful advance in keratoconus management by reducing patient burden without compromising biomechanical efficacy.
Reference:
Lu NJ, Torres-Netto EA, Aydemir EA, et al. A Transepithelial Corneal Cross-linking (CXL) Protocol Providing the Same Biomechanical Strengthening as Accelerated Epithelium-off CXL. J Refract Surg. 2025 ePub ahead of print. doi:10.3928/1081597X-20250515-09
