Ex vivo expansion and transplantation of limbal epithelial stem cells.

OBJECTIVE

To determine, using objective measures, the outcome of ex vivo cultured limbal epithelial stem cell (LESC) transplantation performed in compliance with good manufacturing practice using a novel culture system without 3T3 feeder cells.

DESIGN

Prospective, noncomparative, interventional case series.

PARTICIPANTS

Ten eyes of 10 patients with profound LESC deficiency arising from chemical injury (4 eyes), aniridia (3 eyes), ectodermal dysplasia (1 eye), Reiger's anomaly with Pax6 haploinsufficiency (1 eye), and unknown cause (1 eye).

METHODS

Allogeneic (7 eyes) or autologous (3 eyes) corneal LESCs were cultured on human amniotic membrane. Tissue was transplanted to the recipient eye after superficial keratectomy. Impression cytology and confocal microscopy were performed 6 months after surgery with clinical follow-up to 13 months. Success was defined as an improvement in the defined clinical parameters of LESC deficiency, an improvement in visual acuity, the restoration of a more normal corneal phenotype on impression cytology, and the appearance of a regular hexagonal basal layer of cells on corneal confocal microscopy.

MAIN OUTCOME MEASURES

Clinical parameters of LESC deficiency (loss of epithelial transparency, superficial corneal vascularization, epithelial irregularity, and epithelial breakdown), visual acuity, impression cytology and cytokeratin expression profiles, and in vivo confocal corneal confocal microscopy.

RESULTS

The success rate using this technique was 60% (autografts 33%, allografts 71%). All patients with a successful outcome experienced an improvement in visual acuity of >/=2 lines Snellen acuity. Preoperatively, CK3+ and CK19+ cells accounted for 12+/-2.4% (mean +/- standard error of the mean) and 80+/-2.15% of cells, respectively, whereas postoperatively these accounted for 69+/-6.43% (P<0.0001) and 30+/-6.34% (P<0.0001) of cells, respectively. Goblet cells accounted for 8+/-1.19% of cells preoperatively and 1+/-0.35% of cells postoperatively (P<0.0001).

CONCLUSIONS

These data demonstrate that it is possible to culture LESCs ex vivo in compliance with good manufacturing practice regulations. A set of objective outcome measures that confirm the efficiency of this technique in treating LESC deficiency is described. The widespread use of such standardized and objective outcome measures would facilitate a comparison between the different culture methods in use.