Feasibility of a corneal endothelial cell-based injection therapy in a rabbit model: a Canadian perspective

Authors: Delphine Gobert¹, Pascale Charpentier², Anabelle Demers², Mathieu Thériault³, Mathieu Mercier¹, Ralph Kyrillos¹, Marie Eve Légaré¹, Stéphanie Proulx². ¹Université Laval, Department of Ophthalmology & CHU de Québec - CUO Research center, ²Université Laval, Department of Ophthalmology & CHU de Québec - Université Laval Research center, division of regenerative medicine, ³CHU de Québec - Université Laval research center, division of infectious and immune diseases.

Author Disclosures: D. Gobert: None. P. Charpentier: None. A. Demers: None. M. Thériault: None. M. Mercier: None. R. Kyrillos: None. M. Légaré: None. S. Proulx: None.  


Abstract Body: 

Purpose: Corneal endothelial dysfunction can cause edema and have dramatic effects on vision. While advances in corneal endothelial keratoplasty techniques have greatly facilitated the selective replacement of endothelial cells, difficulties accessing donor corneas still limit their use. Injecting cultured human corneal endothelial cells (CECs) into the anterior chamber is an interesting alternative to endothelial keratoplasty that has been developed in Japan by Dr. Shigeru Kinoshita and his colleagues. The goal of our project is to bring this technique to Canada for clinical studies. 

Study Design: Prospective preclinical study in an animal model 

Methods: This project was conducted in two phases. In the first phase, we recreated a corneal endothelial dysfunction ex vivo model while the second phase transposed this model in vivo in the rabbit. Adhesion of injected human cultured CECs on Descemet’s membrane was compared to adhesion to the stroma. Corneal endothelial dysfunction was induced by selectively stripping endothelial cells with or without a mechanical descemetorhexis of the central 4-5 mm of Descemet’s membrane. CECs (1x106 human cultured CECs) were then injected into the anterior chamber and the cornea placed apex down for 3 hours. Histological and immunohistochemical analysis were then performed. The same protocol was used for the in vivo phase and rabbits were examined at the slit lamp up to 14 days after surgery. Subjective grading of corneal transparency and signs of intraocular inflammation were noted. Intraocular pressure and pachymetry were also measured. 

Results: Histological analysis confirmed that in our ex vivo model, injected human cultured CECs adhered both directly onto Descemet’s membrane and on the corneal stroma. Morphometric analysis also demonstrated that CECs injected into the anterior chamber first adopted a round shape before transitioning to a flattened shape 24 hours after injection. In the in vivo model, desendothelialization with or without descemetorhexis induced significant corneal edema (central cornea thickness of 853.5 μm on day 1, compared to 356 μm pre-operatively, n =5). Moreover, in rabbits injected with human cultured CECs, the cornea was significantly less edematous at 7 days post-operatively than rabbits receiving a sham injection in our in vivo model. Injected CECs did not have a significant effect on intraocular pressure (11.3 mmHg pre-operatively compared to 13 mmHg on day 1) or inflammation after surgery. 

Conclusions: Injecting human cultured CECs directly into the anterior chamber is a promising alternative to current corneal grafting techniques. While optimization of this technique is still ongoing, we hope that our work will help implement this new technique in Canada, which could ultimately benefit thousands of patients with corneal endothelial dysfunction.

X