Corneal Endothelial Regeneration

The World Health Organization estimates corneal blindness affects 20–30 million people worldwide, with roughly 2 million new cases diagnosed each year, but the supply of donor tissue has never matched the growing demand. This disparity has inspired the search for alternative approaches, some of which have now completed the arduous journey from the laboratory to the clinic.

The pump-barrier function of the single posterior monolayer of corneal endothelial cells (CECs) helps maintain corneal function and transparency. A decrease in the number or functional capacity of these cells leads to corneal oedema and reduced vision. Human corneal endothelial cells (HCECs) do not have functional proliferative capacity in vivo, as they are arrested in the G1 phase of the cell cycle. Regardless of age, the standard response to both physiological and pathological cell loss is through polymegathism and cell migration.

Endothelial keratoplasty has transformed management by offering faster rehabilitation and better optical outcomes, particularly with new alternatives such as pre-Descemet’s endothelial keratoplasty (PEDK), which is feasible even in complex anterior segments, and Descemet’s membrane endothelial keratoplasty (DMEK). However, these procedures still depend on donor availability, eye bank infrastructure, surgeon factors, and long-term graft survival. A global shortage of high-quality transplantable corneas remains a barrier to providing the benefit of surgery to everyone who needs it.

Many strategies

Endothelial regeneration includes several different strategies, such as stimulating the remaining endothelial cells to enlarge, migrate, and possibly proliferate; injecting HCECs or scaffold-based endothelial constructs into the anterior chamber (AC); using stem cell-derived or alternate cell sources; and creating bioengineered and artificial endothelial substitutes. The aim is to restore a functional pump using biological, pharmacological, or engineered means.

One leading strategy involves using pharmacology to increase cell migration, adhesion, proliferation, and survival. Rho kinase (ROCK) inhibitors are currently the most promising pharmacological tool available for endothelial regeneration. Pharmacologically stimulating existing endothelial cells provides functional recovery without the risk of rejection or the need for rigorous surgical skills. The ROCK signalling pathway regulates several key cellular processes, including cell adhesion and migration, stress fibre formation, modulation of actomyosin contraction, cell proliferation, and apoptosis. It accelerates centripetal spreading and cell cycle progression from G1 to S phase. Mild vasodilator activity may cause ocular congestion and inflammation.

Another approach, fibroblast growth factors (FGF), enhances CECs through protective and mitogenic effects. They are a standard component of CEC culture media. Engineered FGF can stimulate CEC proliferation while offering greater stability and improved tissue binding, even in dystrophic corneas. They may be used in isolation or to enhance recovery following Descemet stripping only (DSO). Absence of αSMA (alpha-smooth muscle actin) staining in studies indicates that endothelial–mesenchymal transition and scarring are not induced. Angiogenesis and trabecular meshwork blockage are potential problems. Other traditional mitogens such as epidermal growth factor and platelet-derived growth factor stimulate replication when combined in specialised media. Hepatocyte growth factor has also been shown to enhance CEC viability, migration, and proliferation without triggering fibrotic transitions.

Sulforaphane, found in cruciferous vegetables, is thought to activate endogenous antioxidant pathways. It exerts cytoprotective effects and promotes cell survival, potentially preventing degeneration in Fuchs’ endothelial corneal dystrophy (FECD). Ex vivo studies showed a decrease in CEC apoptosis in both stressed and unstressed FECD corneas.

RTA408, a synthetic compound, activates cytoprotective factors that are downregulated in Fuchs’ dystrophy and is being investigated as a topical treatment for patients. Wnt/β-catenin pathway modulators (e.g., TWS119) overcome contact inhibition, triggering localised cell proliferation and lateral migration.

Niacinamide (vitamin B3) is an NAD+ precursor that increases mitochondrial metabolism and protects CECs from ultraviolet and age-related apoptosis. Other substances under investigation include vasoactive intestinal peptide and ciliary neurotrophic factor, which protect CECs from oxidative stress, and connexin hemichannel blockers, which protect CECs from injury.

Leveraging the eye’s capacity to regenerate

DSO, also known as Descemetorhexis without endothelial keratoplasty (DWEK), is used in cases with localised FECD, confined mostly to central cornea. The central 4.0 mm of the abnormal host Descemet’s membrane with guttae is stripped. Loss of contact inhibition allows relatively healthier peripheral endothelial cells to undergo localised expansion and spreading. Slow centripetal migration and polymegathism allow the central cornea to clear over the ensuing one to two months.

Advantages of this approach include avoiding donor tissue and long-term steroid use and their associated risks. Disadvantages include its limited applicability in advanced or diffuse endothelial disease and in eyes with poor peripheral endothelial reserve. Additionally, the stripped zone must be kept small and the margins of the Descemetorhexis smooth to minimise the failure rate. ROCK inhibitors may be administered in conjunction with DSO to increase the rate of deturgescence and hasten recovery. Sulforaphane may be added to protect the peripheral CEC from the oxidative and inflammatory stress of surgery.

Endothelial cell therapy

Cultured endothelial cell injection therapy is the most visible leap beyond keratoplasty toward cell-based therapies. Pioneered by Shigeru Kinoshita MD, PhD, HCECs can be expanded ex vivo from a single donor cornea to treat up to 100 eyes. Donor tissue selection, cell isolation by enzymatic or non-enzymatic digestion, and cell suspension seeding with culture media and growth factors are key steps. The cell suspension is injected into the anterior chamber, and face-down positioning helps to ensure the injected cells settle onto the posterior stroma.

One approach uses supermagnetic iron oxide nanoparticle (SPION) cell labelling followed by an external magnet to draw the cells towards the posterior stroma. ROCK inhibitors may improve cell adhesion and survival, while agents such as sulforaphane or niacinamide help protect the injected cells from metabolic and inflammatory stress.

Advantages of cell injection include ease of therapy and improved availability, as one donor cornea can be used to treat multiple recipients after cell expansion. In addition, unlike DSO, it is not dependent on a reserve of healthy peripheral cells and can therefore be used for more diffuse disease, such as bullous keratopathy.

Host-related challenges include the presence of guttae and abnormal extracellular matrix if the host Descemet’s membrane (DM) is not removed. The fate of unattached cells remains unknown, as does whether transplanted cells will adhere to intraocular structures or even enter the blood stream. Other considerations include the need to inject a much larger number of transplanted cells than may be needed, uncertainty surrounding optimal dosing and delivery systems, patient selection, and postoperative positioning.

Manufacturing challenges include the difficulty in reliably expanding adult HCECs and an undesirable occurrence of endothelial-to-mesenchymal transition. A recent study by Aouimeur (et al.) defined culture success as obtaining typical endothelial morphology and cell density of at least 1,500 cells/mm² at passage three.¹ Success was determined to be dependent on donor age, with 92% in young donors ≤29 years, 42% in intermediate age (30–49 years), and only 12% in donors aged 50 years or older. For older donors, higher initial endothelial cell density (≥1,700 cells/mm²) and shorter organ culture duration (≤15 days) increased the success rate to 25%. Hypertrophic endothelial cells and mesenchymal-like cells with reduced functional phenotype may be seen and therefore quality checks like identity, purity, morphology, density, viability, sterility, potency, and nuclear markers are needed.

Other approaches include using human adipose-derived mesenchymal stem cells, which are very versatile and can be reprogrammed into an endothelial lineage to reduce reliance on donor eye banks. Human embryonic or induced pluripotent stem cells are induced to differentiate into functional, hexagonal endothelial layers. Alternative cells or precursors that can adapt to perform the pump and barrier functions of native endothelium are also being studied.

Mind the scaffolding

Scaffold-based tissue engineering is another fascinating approach. Here, endothelial cells are not simply injected but cultured on a carrier and transplanted, much like a traditional endothelial keratoplasty. This tissue-engineered posterior lamella replaces both the endothelial cells and the DM. However, challenges include creating a scaffold that is thin, transparent, permeable, biocompatible, and appropriately curved, while also providing flexibility and mechanical strength for surgical handling. At the same time, it must allow a high-density endothelial monolayer and support cellular adhesion.

Natural materials such as ultrathin decellularised porcine corneal stroma, thin human corneal stroma, collagen sheets, cellulose, gelatine, chitosan, cross-linked amniotic membrane, and denuded Descemet’s membrane are under investigation. Synthetic material scaffolds, such as polycaprolactone, polylactic-co-glycolic acid (PLGA), polyethylene glycol (PEG), and biohybrids, are also being studied.

Advantages of scaffold-based therapy include uniformity, ease and rapidity of production, easy donor availability, decreased need for transplanted cell numbers, good cellular adhesion on carrier, better systemic safety profile, and known surgeon comfort. Manufacturing and assessment challenges exist, however. PVEK (PreciseBio Vision endothelial keratoplasty) is a 3D-bioprinted, cell-scaffold transplant.

Artificial endothelial substitutes

The use of artificial tissue substitutes is not true regenerative therapy because it does not create an endothelial layer. However, it acts as a functional posterior barrier, particularly in eyes that are unsuitable for other forms of therapy, such as eyes with multiple rejections or uveitis. EndoArt (EyeYon Medical) is a 50-micron artificial lamella positioned similarly to endothelial keratoplasty that deturgesces the cornea by acting as a fluid barrier. Long-term steroids and rejection are avoided. Other materials are also being evaluated.

Biomarkers are being identified that may serve as potential therapeutic targets for FECD treatment. Gene therapy has been used in mouse models to treat early-onset FECD. Viral and nonviral vectors and techniques to introduce genes into the cornea are being researched. Gene editing and CRISPR technologies may give breakthroughs in FECD.

Ethical and regulatory concerns

Embryonic stem cell therapy raises several ethical issues, so adult stem cells are more commonly used. The efficacy and safety seen in animal studies needs to be validated by human clinical trials, which are subject to further safety standards. Biomimetics also face many regulatory and translational challenges before they can come into regular clinical practice. Manufacturers must meet quality assurance and safety guidelines, obtain manufacturing licences, and adhere to Good Manufacturing Practice standards.

Dr Soosan Jacob is Director and Chief of Dr Agarwal’s Refractive and Cornea Foundation at Dr Agarwal’s Eye Hospital, Chennai, India, and can be reached at dr_soosanj@hotmail.com.

1. Aouimeur I, Sagnial T, Travers G, et al. Sci Rep, 2026; 16, 21576. doi:10.1038/s41598-026-52213-3