Abstract
Purpose :
Corneal endothelial cell (CEC) injection therapy is a novel method for the treatment of bullous keratopathy, in which cultured donor CECs supplemented with rho-associated protein kinase (ROCK) inhibitor are injected into the anterior chamber and engrafted (Kinoshita S, et al., N Engl J Med. 2018;378(11):995-1003), and it allows for the treatment of numerous patients via a novel cell culturing method with an expansion of CEC proliferation from one donor cornea. Currently, the primary methods used for the evaluation of cells pre surgery, such as the identification of cell surface antigens and examination by immunostaining, requires a certain number of cells. In this study, we investigated a method to evaluate cells noninvasively by using a physical method for image analysis of cultured CECs before surgery.
Methods :
In the corneal endothelial injection example performed as a clinical study at Kyoto Prefectural University of Medicine from December 2013 to September 2015, a binarized image was obtained from the preoperative cultured cell image in the flask (n = 8) and we created and calculated the spring constant between cells on the assumption that the cells are connected by springs. We examined the correlation between the value and the ratio (χCD44(–)) of cell surface antigen [CD44 (-)] that secured the cell function.
Results :
The spring constant was the minimum value, the maximum value was 0.0048 µm–2, 0.052 µm–2, the minimum value of χCD44(–), the maximum value was 0.1%, 99.2%. Their linear correlation coefficient strongly correlated with r2= 0.86. In the spring constant, the images with good values showed clearer peaks, thus allowing the interaction between nonadjacent cells to be confirmed.
Conclusions :
Evaluation of cultured CECs using physiological method in cell injection therapy had a strong correlation coefficient with conventional biological cell function-guaranteed examinations and could evaluate cells noninvasively. Furthermore, interaction between nonadjacent cells could be confirmed, and there is a possibility of evaluating the healthiness of the tissue.
This abstract was presented at the 2019 ARVO Annual Meeting, held in Vancouver, Canada, April 28 - May 2, 2019.