Long-Term Results of Cultured Limbal Stem Cell Versus Limbal Tissue Transplantation in Stage III Limbal Deficiency

Abstract

We aimed to evaluate efficiency and safety of transplantation of limbal stem cells (LSC) cultured on human amniotic membrane with no feeders and to compare cultured LSC with limbal tissue transplantation. Thirty eyes with stage III LSC deficiency were treated with autologous (autoLSC) or allogeneic (alloLSC) cultured LSC transplantation (prospective phase II clinical trial; average follow-up time, 72 months) or autologous (autoLT) or allogeneic (alloLT) limbal tissue transplantation (retrospective control group; average follow-up time, 132 months) between 1993 and 2014. The 5-year graft survival defined by absence of recurrence of the clinical signs of limbal deficiency was 71% for autoLSC, 0% for alloLSC, 75% for autoLT, and 33% for alloLT. Visual acuity improved by 9.2 lines for autoLSC and 3.3 lines for autoLT. It decreased by 0.7 lines for alloLSC and 1.9 lines for alloLT. Adverse events were recorded in 1/7 autoLSC, 7/7 alloLSC, 6/8 autoLT, and 8/8 alloLT patients. Corneal epithelial defect was the only adverse event recorded after autoLSC, whereas severe sight-threatening adverse events were recorded in the remaining three groups. Compared with failed grafts, successful grafts featured greater decrease in fluorescein staining, greater superficial vascularization-free corneal area, lower variability of the corneal epithelial thickness, and higher corneal epithelial basal cell density. Autologous cultured LSC transplantation was associated with high long-term survival and dramatic improvement in vision and was very safe. Autologous limbal tissue transplantation resulted in similar efficiency but was less safe. Cadaver allogeneic grafts resulted in low long-term success rate and high prevalence of serious adverse events. Stem Cells Translational Medicine 2019;8:1230&1241.

Keywords: Adverse events; Cornea; Limbal stem cell; Survival; Transplantation; Visual acuity.

Figure 1

Morphometric analysis of slit‐lamp images of a patient with stage III limbal stem cell deficiency before (A–D) and 36 months after transplantation of autologous limbal stem cells cultured on human amniotic membrane (E–H). The corneal area free of superficial vessels (C, G) was measured with ImageJ software using the green channel of the native bright light slit‐lamp photograph (A, E). The green channel (D, H) of the native blue cobalt slit‐lamp photograph (B, F) was used to assess fluorescein staining. LSC transplantation was successful in this eye. Visual acuity improved from 20/3,991 preoperatively to 20/50 36 months after transplantation (+19 lines), fluorescein staining decreased from 97.7 to 22.3, and the corneal area free of superficial vessels increased from 21% to 74% of the total corneal area.

Figure 2

Survival of the transplanted epithelium after LSC transplantation. The primary outcome measure was defined as absence of recurrence of the clinical signs of LSC deficiency (opacification of the corneal epithelium, irregularity of the corneal epithelium with late fluorescein staining, superficial corneal vascularization) in the central cornea. Survival was significantly better for autografts than allografts, whereas no significant differences were found between limbal tissue grafts and cultured LSC grafts. Abbreviations: alloLSC, allogeneic cultured LSC transplantation; alloLT, allogeneic limbal tissue transplantation; autoLSC, autologous cultured LSC transplantation; autoLT, autologous limbal tissue transplantation; LSC, limbal stem cell.

Figure 3

Change in visual acuity as a function of postoperative time point in 30 eyes with stage III limbal deficiency treated with transplantation of limbal stem cells. Overall vision improved after transplantation (A). From baseline to last visit, visual acuity improved by an average of 9.2 lines in the autoLSC group and 3.3 lines in the autoLT group, whereas it decreased by 0.7 lines in the alloLSC group and 1.9 lines in the alloLT group (B). Abbreviations: LogMAR, logarithm of the minimum angle of resolution; LSC, limbal stem cell.

Figure 4

Subjective score (A), Schirmer test (B), fluorescein staining (C), and superficial vascularization‐free corneal area (D) as function of postoperative time point after cultured LSC transplantation. The subjective score was calculated as the mean of quantitative evaluations of redness, pain, itching, foreign body sensation, and blurred vision. Fluorescein staining was assessed with blue cobalt slit‐lamp images as the mean grayscale level of the green channel. We show the change in fluorescein staining from baseline. This decreased with postoperative time in the autoLSC group but not in the alloLSC group. Superficial vascularization‐free corneal area was assessed with the green channel of white slit‐lamp image. We show the percentage of the corneal area that was free of superficial vascularization. Abbreviation: LSC, limbal stem cell.