Corneal transplantation is often the only treatment option in cases of corneal blindness, with the main challenges being the scarcity of human donors, risk for graft failure and suboptimal visual outcome due to suture-related issues. Alternative therapies are sought that either address the above transplantation issues directly or stimulate the cornea’s repair mechanism and regenerative properties to restore transparency without the need for transplantation.   

The first aim of this thesis was to develop a cell-free substitute for the human corneal stroma made of porcine collagen, a purified byproduct from the food industry that is already approved by FDA as a raw material, used for example in cosmetic surgery or as a medical device in glaucoma surgery. Abundance, cost-effectiveness, low rejection risk due to acellularity, high purity and worldwide availability are among the main advantages of purified porcine collagen compared to human donors and other corneal stromal substitutes. The second aim was to address the risk of graft rejection in cases of neovascularized and inflamed corneas by loading the bioengineered porcine collagen constructs with drugs that can ideally promote corneal regeneration, secure the survival of the graft in cases of high-risk keratoplasties and minimize the need for prolonged topical immunosuppressive therapy following operation, which has drawbacks of low bioavailability and need for good patient compliance. The third aim was to develop alternative techniques of lamellar transplantation, all assisted by femtosecond laser, that are less suture-dependent and enable the intrastromal implantation of biomaterials.  Finally, the fourth aim of this thesis was to evaluate the role of Regenerating Agent (RGTA) eye drops in corneal wound healing following therapeutic laser ablation of the cornea in a randomised, blinded, placebo-controlled preclinical study. Regenerating Agent is a biomimetic of extracellular matrix with well-established favourable outcomes in the treatment of skin wounds and preliminary positive results in the treatment of corneal wounds.    

Different manufacturing protocols were used to enhance the mechanical properties of bioengineered porcine collagen (BPC) and to address different requirements. The combination of both chemical and photochemical crosslinking with riboflavin and ultraviolet A light (to form BPCDX) and reinforcement with cellulose nanofibers extracted from the Ciona intestinalis sea invertebrate (BPCDX-CNF) resulted in stronger biomaterials compared to earlier BPC versions. The biomaterials could be manufactured in different sizes and in core-and-skirt forms with the peripheral skirt degrading faster due to mechanical compression without the addition of any cross-linkers during manufacturing. BPC could be successfully loaded with nerve growth factor (NGF) and BPCDX-CNF with dexamethasone without sacrificing transparency and both drugs could be released from BPC-based materials in vitro up to at least 2 months. The biological activity of dexamethasone released from the drug-loaded BPCDX-CNF could be confirmed by the decreased expression of inflammatory cytokines in human corneal epithelial cells grown on dexamethasone-loaded BPCDX-CNF. A compatible packaging and sterilization process was developed for BPCDX, tested internally and externally by Good Laboratory Practice-certified laboratories, that can enable worldwide distribution and storage at room temperature or in a refrigerator up to two years without the need of extra quality controls before transplantation.  

BPC-based materials could be safely implanted in rabbit, minipig and human corneas with advanced keratoconus using femtosecond-laser assisted intrastromal keratoplasty procedures. A femtosecond laser was used to create intrastromal pockets of different dimensions based on the size of the biomaterials. Through an access cut or by lifting a flap, both created by femtosecond laser, biomaterials could be implanted intrastromally with or without native tissue removal. Additional sutures were used in a surgical inflammatory model to test the biological activity of dexamethasone released by BPCDX-CNF. These minimally invasive surgical procedures required shorter period of immunosuppression following operation and maintained the anatomy of the surrounding host tissue. Apart from the skirt part of the composite BPC that was mechanically compressed and was designed to degrade faster, the crosslinked BPC remained stable in animal models, while no degradation was observed in the BPCDX 2 years after implantation in humans. The biomaterials were biocompatible and native cells were found in the biomaterial-host interface or in the periphery of the biomaterial. The inflammatory response following operation depended on individual response to injury and did not appear to be stimulated by BPC implantation. Neovascularization and haze formation were mainly restricted around the sutures used to secure either the access cut or flap overlying the biomaterials as well as intentionally placed close to the limbus to trigger inflammation in the dexamethasone study. In the human studies, no sutures were used and corneal transparency was maintained at the highest level in all subjects after 2 years without any signs of rejection, inflammation, vascularization or scarring. Topographic indices including mean anterior corneal curvature and maximal corneal apical curvature were significantly reduced in both clinical cohorts resulting in improved best corrected visual acuity. Importantly, following operation all human subjects could tolerate contact lenses and no subject was considered legally blind. The release of dexamethasone from the drug-loaded BPCDX-CNF could be also confirmed in vivo by sustained intraocular pressure increase, tendency to reduce neovascularization and haze formation and sustained suppression of inflammatory cytokines in the aqueous humor of eyes implanted with dexamethasone-loaded BPCDX-CNF compared to non-loaded BPCDX-CNF.   

Finally, RGTA eye drops following excimer laser ablation of the anterior healthy rabbit cornea did not affect the already quick and uneventful epithelial closure. Although there was significantly less haze in the RGTA group compared to placebo as measured by in vivo confocal microscopy, this decrease was not clinically relevant as all corneas in both groups were clinically transparent following laser. The microscopic haze formation and staining problems did not allow quantification of nerve regeneration, but subbasal nerves were found to repopulate the central ablated regions in both RGTA and placebo groups.   

In conclusion, our results indicate that biomaterials made of bioengineered porcine collagen are a safe alternative to human donor tissue for corneal transplantation, offering the advantage of custom-made manufacturing to address different requirements with potential applications to different corneal stromal diseases. Femtosecond laser enables safe intrastromal implantation of biomaterials with less suture-related issues and immunosuppression following operation compared to traditional corneal transplantation methods. Regenerating Agent eye drops do not affect the already rapid wound healing following laser ablation of the healthy cornea.