Abstract
To treat unilateral limbal stem cell (LSC) deficiency, we developed cultivated autologous limbal epithelial cells (CALEC) using an innovative xenobiotic-free, serum-free, antibiotic-free, two-step manufacturing process for LSC isolation and expansion onto human amniotic membrane with rigorous quality control in a good manufacturing practices facility. Limbal biopsies were used to generate CALEC constructs, and final grafts were evaluated by noninvasive scanning microscopy and tested for viability and sterility. Cultivated cells maintained epithelial cell phenotype with colony-forming and proliferative capacities. Analysis of LSC biomarkers showed preservation of “stemness.” After preclinical development, a phase 1 clinical trial enrolled five patients with unilateral LSC deficiency. Four of these patients received CALEC transplants, establishing preliminary feasibility. Clinical case histories are reported, with no primary safety events. On the basis of these results, a second recruitment phase of the trial was opened to provide longer term safety and efficacy data on more patients.
CALEC is a frontier in cultivated autologous cell therapy to treat corneal limbal stem cell deficiency.
INTRODUCTION
Corneal epithelial stem cells are adult somatic stem cells located at the limbus, a niche area between the cornea and conjunctiva/sclera representing the source of transparent and intact corneal epithelium. When limbal stem cells (LSCs) become dysfunctional or deficient, the cornea is unable to maintain its surface epithelial integrity and LSC deficiency (LSCD) develops (1, 2). The clinical hallmarks of LSCD include conjunctivalization, neovascularization, recurrent or persistent epithelial defects, inflammation, and corneal scarring. These changes lead to loss of vision, pain, and impaired quality of life (3).
Conventional corneal transplantation replaces the central cornea but cannot treat LSCD. Therapeutic strategies have been developed to replace limbal epithelium, including limbal autografts for unilateral LSCD, where tissue is obtained from the fellow eye, and limbal allografts for bilateral LSCD, where tissue is obtained from cadaveric donor (4, 5). Limbal allografts, which carry a high risk of rejection, require long-term systemic immunosuppression (4, 5). Autologous limbal transplantation (conjunctival limbal autograft) involves removal of a substantial amount of limbal tissue from the donor eye and risks inducing LSCD to the donor eye (6–8). More recently, simple limbal epithelial transplantation has also been used in some cases of LSCD (9–11).
To avoid these risks, methods to expand LSC ex vivo were developed. A small limbal biopsy of the patient’s healthy eye was used for cultivated limbal epithelial transplantation (CLET) onto the diseased eye. In 1997, Pellegrini et al. (12) reported the successful use of ex vivo cultivated autologous limbal epithelial stem cells for two subjects with unilateral LSCD from chemical burns (12) with favorable long-term outcomes (13), and several follow up studies have shown variable success (14). CLET is not available in the United States, as the manufacturing procedure uses xenogeneic murine feeder cells and fetal bovine serum (FBS) and is not compatible with current good manufacturing practices (cGMPs). The use of murine fibroblasts carry a risk of viral contaminants and inflammatory response against murine antigens (15, 16). Although the use of FBS in cultivating LSC has not been associated with immunogenicity, its use in cultivating mesenchymal stromal cells has been shown to elicit immune responses in the host (17, 18). These components are also sources of nonhuman sialic acid Neu5Gc, which can induce immune responses and jeopardize engraftment (19, 20).
We developed a manufacturing technique to address these shortcomings: cultivated autologous limbal epithelial cells (CALECs). We removed antibiotics and murine feeder cells, used highly reproducible methods for LSC isolation and expansion onto human amniotic membrane, and developed rigorous quality criteria for the manufactured tissue graft. Here, we report the development of the CALEC manufacturing process and clinical outcomes of the first four patients who received CALEC grafts.
RESULTS
Preclinical development of CALEC manufacture
Ex vivo cultivation of LSCs
Cadaveric limbal biopsies of 1 clock hour were used to generate a single-cell suspension of epithelial cells, which became adherent in six-well plates. Colonies gradually grew in size, forming a confluent monolayer of P0 cells in 9.1 ± 2.5 days (Fig. 1A). P0 cells (3.0 × 104 to 5.0 × 104) were harvested and seeded onto de-epithelialized AmnioGraft (Bio-Tissue Inc.) where they generated confluent monolayers in 6.7 ± 1.4 days (Fig. 1, B and C). After the initial P0 culture, the average cell viability of P0 cells was 83.9 ± 11.6% on the basis of trypan blue dye exclusion. The optimal cell seeding dose for a single amniotic membrane was determined to be 4.0 ± 1.0 × 104 cells. To avoid disruption of the final cellular construct (P1), the cell viability at this stage was determined by measuring the lactate dehydrogenase (LDH) in the culture supernatant (Fig. 1C). CALEC constructs have a final cell count of 7.5 ± 2.1 × 105 determined in situ by scanning the surface of CALEC construct with the EVOS cell imaging system. The constructs contained a uniform layer of cobblestone-shaped cells, as visualized by conventional and confocal microscopy (Fig. 1, D and E).
Fig. 1. Ex vivo cultivation of limbal epithelial cells and characterization of cultured cells.
Limbal biopsies were obtained from cadaveric donors and single-cell suspensions were obtained after enzymatic digestion. (A) P0 cells became adherent on plastic, formed small colonies, and grew into a confluent monolayer. (B) P1 cells were plated on a decellularized amniotic membrane and grew to confluence. (C) Average culture duration, cell count, and viability of P0 and P1 cells (N = 16). P0 cell viability was determined using trypan blue staining and expressed as %. P1 cell viability was determined by measuring LDH concentration in the supernatant (U/liter). (D) In situ cobblestone morphology of the cultured cells at the end of P1. (E) Confocal microscopy showing cell morphology at low, mid, and high density using vinculin (green) and actin F (red) staining during P1 expansion. The nuclei were stained with 4′,6-diamidino-2-phenylindole (DAPI) (blue). (F) Flow cytometric analysis of cell makers demonstrated the predominantly epithelial phenotype of the cultured cells (N = 13). (G) P0 cells cultured for 10.0 to 14.0 days grew into colonies and yielded an average colony-forming efficiency (CFE) of 7.6 ± 2.7% (N = 7). (H) Intracellular adenosine 5′-triphosphate (iATP) assay demonstrated that cells have varying proliferative capacities, and an average reference line was created (N = 8). (I) In situ hybridization proliferation and viability of cultured cells using EdU incorporation and LIVE/DEAD assay, respectively. CALEC, cultivated autologous limbal epithelial cell.
Phenotype and proliferative capacity of cultivated LSCs
Flow cytometry was used to characterize the cells from the limbal biopsy after they had expanded to confluence at stage P0. The five-marker flow cytometry panel showed that P0 cells expressed high levels (~95%) of epithelial markers (CD49F and CD340), low levels (3.9%) of leukocyte/endothelial markers (CD45/CD31), and low levels (6.4%) of a mesenchymal stem cell marker (CD13), confirming the predominantly epithelial phenotype of cultivated LSC (Fig. 1F).
Colony-forming efficiency (CFE) assay (Fig. 1G) was used to evaluate cell “stemness.” P0 cells were cultured at low concentration, and colonies were counted after 10 to 14 days. The average CFE was 7.6 ± 2.7%. A functional, quantitative, and standardized assay measuring intracellular adenosine 5′-triphosphate (iATP) was used to determine the cell proliferation potential of cells after P0 expansion (Fig. 1H) (21). P0 cells were cultured at three different cell concentrations for 7 days. iATP measurements demonstrated that the proliferative capacity of the cultured LSC increased in a dose-dependent manner. On the basis of results obtained from eight individual donor corneas, an average curve (Fig. 1H, solid line labeled “average”) was created as a reference for future patient sample comparison.
To evaluate cells in the final CALEC construct (P1), in situ proliferation and viability measurements were carried out using 5-ethynyl-2′-dexyuridine (EdU) incorporation and a commercially available LIVE/DEAD assay, respectively. As shown in Fig. 1I, 25.0 ± 2.0% of cells were EdU-positive, hence synthesizing new DNA. Similarly, the in situ viability (LIVE/DEAD) assay data showed 83.0 ± 1.0% viable cells (Fig. 1I). These preclinical results with constructs prepared with limbal biopsies from cadaveric donors demonstrate that cells in the final CALEC constructs have high viability and proliferative capacity.
We detected putative stem cell markers p63, p63α, and C/EBPδ in human limbus (Fig. 2A), whereas cytokeratin 12 (Krt12), a marker for differentiated corneal epithelial cells, was noted in the cornea and more superficial cells in the limbus. P0 cells maintained the expression of these stem cell markers (Fig. 2B), suggesting that stemness was preserved in cell culture. CALEC construct, in both whole mount (Fig. 2C) and cryosection (Fig. 2D), showed staining of these stem cell markers. The quantitative analysis detected that the bright positive rate was 5.5 ± 1.9% for p63, 5.4 ± 1.1% for p63α, 1.2 ± 0.4 for C/EBPδ, and 2.3 ± 1.0 for Krt12 (Fig. 2E).
Fig. 2. Expression of stem cell markers in CALEC constructs.
(A) Detection of putative stem cell markers p63, p63α, C/EBPδ, and marker for differentiated corneal epithelial cells Krt12 in human limbal biopsy. Scale bar, 100 μm. (B) Immunostaining of these markers in P0 cell culture. Whole mounts (C) and cryosection (D) of P1 construct showing representative images of p63, p63α, C/EBPδ, and Krt12. Scale bars, 50 μm (B to D). (E) Rate of p63-, p63α-, C/EBPδ-, and Krt12-positive cells in whole mount was counted and expressed as means ± SD (n = 4).
Final formulation and packaging of final construct
To determine the optimal method for transporting the manufactured CALEC from the cGMP laboratory to the operating room, CALECs were first cultured and stored in hypothermic conditions (2° to 8°C for 6 days). Cell death measured by LIVE/DEAD assay was correlated with LDH levels in the supernatant. The high correlation (r = 0.99) between the cell death rate and LDH concentration indicated that LDH release provides a nondestructive surrogate for determining the cell viability of CALEC constructs. Cell viability remained high (cell death rate less than 10%) when constructs were maintained in hypothermic conditions for up to 72 hours (Fig. 3, A and B).
Fig. 3. Optimization of storage conditions to maintain viability and metabolic activity of CALEC constructs.
(A) Cultivated limbal epithelial cells at P0 were seeded into seven dishes and maintained in hypothermic conditions (2° to 8°C) for up 6 days. Cell viability was determined using LIVE/DEAD assay, and the LDH concentration in the culture supernatant was measured every 24 hours. (B) There was a strong (r = 0.99) correlation between the % dead cells and LDH concentration. (C) Cells were maintained in three different transportation media, and HypoThermosol yielded the lowest LDH level and cell death rate. (D) After 24 hours of cold storage in HypoThermosol FRS, CALEC constructs were recovered and put back in corneal growth media at 37°C for 24 hours. LDH levels in the culture supernatant were comparable for fresh constructs maintained at 37°C, after 24 hours of cold storage, and after 24 hours of recovery at 37°C. CSN, cell supernatant. (E) Metabolic activity of CALEC constructs was determined by measuring glucose consumption and lactate production in the same supernatant. Fresh CALEC constructs are metabolically active, while constructs stored in HypoThermosol FRS had higher glucose and lower lactate levels in supernatant, indicating metabolic inactivity. After the recovery from hypothermic storage for 24 hours, constructs resume metabolic activity to levels similar to prior to cold storage. (F) Final construct is transported in a reusable EVO transporting device with temperature and tilt monitoring by GPS device. HT, hypothermosol.
Comparison of different storage media showed that HypoThermosol free radical scavenger (FRS) yielded the lowest cell death rate (Fig. 3C). The effect of hypothermic storage on CALEC viability in HypoThermosol was evaluated by comparing LDH levels in supernatants of fresh constructs at 2° to 8°C storage and after recovery at 37°C (Fig. 3D). CALEC constructs stored in HypoThermosol FRS had higher glucose and lower lactate levels in supernatant, indicative of metabolic inactivity but metabolic activity recovered from hypothermic storage after constructs were again placed at 37°C (Fig. 3E). On the basis of these results, final CALEC constructs are transferred to HypoThermosol FRS media in a six-well plate and transported from the cGMP facility to the operating room in a reusable EVO device with temperature and tilt monitoring by GPS (Fig. 3F). The EVO container is set to maintain the CALEC graft at 2° to 8°C during transport.
CALEC transplantation in four subjects
Following the development, standardization, and preclinical testing of CALEC, a phase 1 clinical trial was implemented to assess the safety and feasibility along with preliminary efficacy estimates of CALEC grafts performed for unilateral LSCD (IND #16102; NCT02592330). The biopsy of the donor eye of 1 clock hour (averaging ~3.1 mm by 2.5 mm) was performed at Massachusetts Eye and Ear (MEE) and transferred to a cGMP laboratory for manufacturing. The overall process for manufacture and transplant of the CALEC grafts is summarized in Fig. 4. Of the five patients enrolled, four received CALEC transplantation (fig. S2). Cells from one biopsy grew normally and were harvested at P0 but did not expand on the de-epithelialized AmnioGraft. A CALEC graft was successfully manufactured from a second biopsy. Cells from a fifth subject did not expand at P0. For all five subjects undergoing biopsy, the donor eye healed without complications and the visual acuity returned to baseline within 4 weeks. No primary safety events occurred.
Fig. 4. Overall schema for manufacturing CALEC grafts for clinical transplantation.
The limbal biopsy is surgically removed and transported to the cGMP laboratory where the epithelial cells are harvested, isolated, and plated in the primary culture (P0). At the end of P0 culture, cell number, viability (trypan blue staining), phenotype, colony formation, and proliferation are assessed. Cells are then isolated and seeded onto denuded human amniotic membrane, AmnioGraft, in the secondary (P1) culture. At the end of P1 culture, cell number, viability (LDH level in supernatant), and endotoxin and mycoplasma levels are determined. Once the construct meets the release criteria, it is preserved and transported for transplantation within 24 hours. The recipient eye undergoes corneal scar removal and reconstruction to prepare it for CALEC transplantation. The CALEC construct is trephinated and sutured onto the recipient eye. FACS, fluorescence-activated cell sorting.
Clinical course of subject 1
A 46-year-old male with a history of unilateral sodium hydroxide ocular burn underwent extensive symblepharon lysis before study enrollment (Fig. 5A). After 11 months, he presented with stage III LSC, 12 clock hours of pannus, corneal opacification (Fantes grade 4), and an epithelial defect measuring 5.0 by 3.2 mm (NEI 15). CALEC transplantation occurred 16 days after the limbal biopsy. A central intrastromal hemorrhage occurred during surgery but resolved before 9 months. The opacity cleared peripherally (Fantes 3) at 3, 9, and 12 months. At 12 months, his vision remained hand motion due to intrastromal opacity, but he experienced reduction in neovascularization and resolution of the epithelial defect (NEI 0), preparing him to undergo corneal transplantation for visual rehabilitation.
Fig. 5. Clinical slit lamp photography of the first 4 patients receiving CALEC transplantation.
Preoperative baseline and postoperative months 3, 9, and 12 slit lamp images taken with diffuse white light (top row) and cobalt blue light with fluorescein staining (bottom row). (A) Subject 1 is a 46-year-old with a history of sodium hydroxide ocular burn in the left eye. (B) Subject 2 is a 31-year-old male with a history of a plaster chemical burn in the right eye. (C) Subject 3 is a 36-year-old male with a history of a fireproofing chemical in the left eye. (D) Subject 4 is a 52-year-old male with an acid burn from a motor vehicle accident the left eye.
Clinical course of subject 2
A 31-year-old male with a history of unilateral plaster chemical burn underwent ocular surface reconstruction 8 years before the study (Fig. 5B). At baseline, he had corneal opacification obscuring iris details (Fantes 3) and 8 clock hours of pannus with epithelial staining (NEI 11) (stage IIB LSC). CALEC transplantation occurred 22 days after the limbal biopsy. Corneal opacity (Fantes 3) persisted through the 3- and 9-month visits but reduced (Fantes 2) at the 12-month visit. Corneal staining improved at 3-, 9-, and 12-month visit (NEI 0). His symptoms resolved, and the best-corrected visual acuity (BCVA) vision improved from 20/40 (baseline) to 20/30 (12 months).
Clinical course of subject 3
A 36-year-old male presented with unilateral LSC due a fireproofing chemical injury 5 months before study enrollment (Fig. 5C). At baseline, he had complete corneal opacification (Fantes 4), 12 clock hours of neovascularization, and an epithelial defect measuring 4.6 by 4.7 mm (NEI 15) (stage III LSC). CALEC transplant occurred 13 days after the limbal biopsy. Opacification was significantly reduced (Fantes 1) at 3, 9, and 12 months. Pseudopterygium (<1 mm) was present at 3 and 9 months but ghosted at 12 months. The epithelial defect resolved (NEI 0) at the 12-month visit. BCVA vision improved from hand motion (baseline) to 20/30 (12 months).
Clinical course of subject 4
A 52-year-old male sustained a unilateral acid burn (stage III LSCD) 30 years previously (Fig. 5D). He had complete opacification (Fantes 4), 12 clock hours of neovascularization, dense late corneal staining indicating conjunctivalization (NEI 15), and lagophthalmos. The first biopsy attempt did not result in a viable construct; he subsequently underwent a successful biopsy 3 years later. At 3 months, the bandage contact lens (BCL) was lost and, due to lagophthalmos, he developed an epithelial defect, which resolved without recurrence after resuming a BCL and tape tarsorrhaphy. BCVA improved from hand motion (baseline) to counting fingers (12 months). Corneal opacity was reduced, albeit iris details remained obscured (Fantes 3) due to intrastromal scarring. Conjuntivalization resolved, and neovascularization and corneal staining (NEI 5/15) improved at 12 months, preparing him to undergo corneal transplantation for visual rehabilitation.
Manufacture, quality control, and biomarker analysis of CALEC constructs of clinical subjects
Figure 6A summarizes manufacturing data for the four transplanted CALEC constructs. The mean duration from biopsy to P0 harvest was 9.0 ± 1.8 days and 9.0 ± 4.2 additional days to final construct, resulting in a total culture time of 13.0 to 22.0 days. The mean viability of the P0 cultures was 78.7 ± 4.2%. Flow cytometry results confirmed the predominant epithelial cell phenotype of cells at the end of P0. P0 cells from subject 2 did not form colonies in vitro (CFE assay), but these cells proliferated well when transferred to the amniograft. Assays for initial and final sterility, mycoplasma and endotoxin showed no evidence of bacterial contamination. The mean final cell count was 0.8 ± 0.2 × 106 cells per construct. All final products met release criteria established for the LDH viability assay (≤150 U/liter). P0 cells from two subjects proliferated above and two below the reference line for iATP assay (Fig. 6B). Two product failures occurred within the first cohort of five patients. See table S1 for quality control reports for these product failures.
Fig. 6. Manufacturing characteristics and biomarker analysis of patient CALEC constructs.
(A) In-process analysis and final product release assays for CALEC constructs manufactured for the enrolled four patients (subject 4 had CALEC construct from second biopsy). NOS, no organisms seen. (B) iATP assay demonstrates varying degree of cell proliferative capacity from patient cells obtained at P0 (dashed lines). Solid line represents mean reference values obtained from P0 cells obtained from eight cadaveric donor biopsies. (C) Backup CALEC constructs of subjects 1 to 3 were stained for p63/Krt12 and (D) C/EBPδ/p63. (E) Positive rate of each biomarker in each subject. IHC, immunohistohemistry.
Three subjects had duplicate backup constructs manufactured at the same time as the primary CALEC grafts. These constructs were used to examine LSC biomarker expression by immunohistochemistry. Figure 6 (C and D) demonstrates the expression of p63, Krt12, p63a, and C/EBPδ in whole mount sections. A range of biomarker expression was observed, with average positive rates of 2.9, 2.8, 4.3, and 2.7% for Krt12, p63, p63α, and C/EBPδ, respectively (Fig. 6E).
DISCUSSION
Before CALEC, techniques for culturing autologous and allogeneic limbal epithelial cells for treatment of LSCD used a variety of manufacturing procedures and substrates on which the cells are grown, ranging from petrolatum gauze (12), contact lenses (12, 21–23), fibrin (13, 15, 24–26), keratin (27), silk fibroin (28), and collagen (29). However, despite landmark publications (13, 30, 31) describing this approach decades ago, no previous clinical studies have evaluated cultivated LSC grafting in the United States. To address this limitation, our study presents the development of a critical technologic innovation that allowed the first-in-human trial of CALEC using xenobiotic-free, serum-free, and antibiotic-free manufacturing method that led to a successful transplantation of resulting autologous grafts in a highly vetted clinical trial. The rigorous development process led to an innovative two-stage cell manufacturing method to ensure safety, consistency, and viability of the final CALEC formulation. In the first stage, limbal epithelial cells were grown on plastic in defined media that optimized LSC growth until they became confluent (P0). In the second stage, P0 cells were transferred to de-epithelialized AmnioGraft, which provided a Food and Drug Administration (FDA)–approved biocompatible substrate for continued growth until cells again become confluent and released for autologous transplantation (P1). Novel quality control assays were developed to characterize the cellular product at each stage of manufacturing.
A major challenge was the development of methods to determine cell counts, viability, and phenotype without damaging the final cellular construct. To address this critical issue, a cell counting method using live cell imaging was developed and measurements of factors in media from the final product formulation were used to assess viability. Release of LDH in culture media is a well-established surrogate for cell death (32, 33), and levels ≤150 U/liter reflected excellent viability and were therefore required for product release. EVOS cell imaging was used to estimate the number of cells growing on the amniograft membrane at the end of processing. The phenotype of cells at P0 was determined by flow cytometry, and phenotype of cells in the final product was evaluated in a backup product manufactured at the same time that was used for immunohistochemistry after the primary graft was successfully transplanted.
The distinguishing characteristics of LSC include slow cycling properties, high proliferative potential, and clonogenicity. Several LSC biomarkers have been identified, including p63 (34), ΔNp63α (35), C/EBPδ (36), Bcrp1/ABCG2 (37), ABCB5 (38), and cytokeratin 14 (39, 40). The relationship between biomarker levels and cellular function during epithelial regeneration or in an ex vivo cultivated construct is not known. Overall, levels of LSC marker expression in the backup CALEC constructs for three patients were consistent with published literature (34–40), with some variability between patients. The expression of p63 nuclear transcription factor and its isoform p63α, as well as C/EBPδ, have been identified in LSC holoclones and used to differentiate cultivated limbal graft success (13). All cases had positive p63, p63α, and C/EBPδ, at varying amounts: p63 (1.9 to 4.3%) and p63α (1.9 to 7.1%). Some had p63 levels less than those indicative of CLET success (13), but these products nevertheless led to successful CALEC transplants in our study. In addition, biomarker analysis was performed using backup grafts where cells have undergone two passages, while other studies performed this analysis on cells after only one passage (13, 15).
Development of a potency assay that reliably measures and predicts a fully functional cell therapy product is essential (41). Because transplantable CALEC constructs offer limited options for analysis of the final product, we implemented additional assays of cells obtained after completion of the first manufacturing stage (P0). Similar to previous reports (42, 43), CFE measurements alone are inadequate because it is not possible to visually identify holoclones or to identify LSC populations in a timely manner as the process of seeding P0 cells to form colonies in 10 to 14 days took longer than the growth and release of final products (13). In preclinical studies, changes in iATP concentration due to cellular and mitochondrial function correlated with cellular number and proliferation as determined by 5-bromo-2′-deoxyuridine and in situ viability and was indicative of a LSC population. Possibly, the reference iATP values obtained from cadaveric limbal biopsies does not adequately represent the variability in LSC taken from actual patients and will need to be recalibrated after compiling the clinical trial data. The final criteria for acceptance, which will be determined with a larger set of patients, will ideally include a numerical range representative of biological activity and clinical effectiveness.
The details of cellular therapy product development and level of regulatory compliance of prior products are unclear (13–15, 24, 44–47). Three products used murine 3T3 fibroblasts as feeder cells, five used serum and animal-derived growth factors, and all used antibiotics in their process. One detected 3T3 cells in 5% of transplanted grafts, potentially stimulating an inflammatory response against murine antigens (15). We developed a manufacturing procedure without feeder cells or serum using completely defined reagents with additional testing (i.e., bovine adventitious virus testing). Modifications and standardizations also included the development of reproducible methods for cell isolation, expansion, and quality control of the resultant CALEC sheets.
During the initial recruitment phase of the clinical trial, two biopsies failed to generate viable CALEC constructs (fig. S1). One failure was attributed to incomplete de-epithelialization of the amniotic membrane, which prevented LSC attachment and growth. There appeared to be no issues with LSC function as P0 cells proliferated faster than control iATP values; CFE was 7.2%. A second biopsy was obtained, which was able to generate a successful CALEC product (subject 4). In contrast, the second failure had inadequate growth of LSC with no proliferative activity on iATP and CFE assay. Because these results suggested poor LSC function in the “normal” eye due to contact lens wear, a second biopsy was not pursued, consistent with previous reports (21). Four CALEC participants with variable LSCD severity received CALEC transplant and achieved stable ocular surface at 12 months, supporting the validity of this approach.
In summary, the initial phase of this trial has established feasibility of the product manufacturing method with no immediate safety concerns, enabling a second recruitment phase to provide longer term safety and efficacy data on more patients. Use of the small biopsy carries low risk of inducing LSCD in the donor eye, as seen in one example showing the repeatability of the CALEC procedure. Following this preliminary 12-month safety report, the ongoing trial will provide more data on a larger cohort of patients including 18-month follow-up for rigorously defined efficacy end points and correlation of these outcomes with marker and potency assay data from autologous products. If successful, this first use of CALEC in the United States will serve as a steppingstone for establishing cellular therapy products as viable options for patients with LSCD.
METHODS
Ex vivo cultivation of limbal epithelial cells
In the preclinical phase, cadaveric donor corneas were obtained from the National Disease Research Interchange. Later in the clinical phase, patient limbal biopsies were obtained (surgical details below). For primary (P0) cell culture, a 3 mm–by–3 mm biopsy was harvested from the limbus, and dissected limbal epithelial sheets were digested with Liberase MNP-S (Roche, Basel, Switzerland) and then dissociated into a single-cell suspension. After 30 min of enzymatic digestion, the cells were centrifuged (450g) for 5 min at room temperature. The supernatant was collected for sterility testing (BacT, bioMerieux). The cells were resuspended and plated into one well of a six-well culture plate in 2 ml of serum-free complete corneal epithelial cell basal media [American Type Culture Collection (ATCC), Manassas, VA, USA] and placed in a humidified incubator at 37°C and 5% CO2. Media changes were done every 2 days during the primary culture. Corneal epithelial cell basal medium contains essential and non-essential amino acids, vitamins, other organic compounds, trace minerals, and inorganic salts and was supplemented with a corneal epithelial cell specific growth kit (ATCC, Manassas, VA, USA), which contains apo-transferrin, epinephrine, extract P, and hydrocortisone hemisuccinate. When using this complete media system, the growth of corneal epithelial cells is supported without the use of feeder layers, extracellular matrix proteins, or other substrates.
When adherent cells achieved 90 to 100% confluence, cells were harvested and plated onto an amniotic membrane (P1 culture). Cells were detached with 2 ml of CTS TrypLE (Thermo Fisher Scientific, Waltham, MA, USA) for 7 min at 37°C. The cell suspension was collected in a 15-ml conical tube containing 1 ml of media. The well is rinsed with 1 ml of media and added to the 15-ml conical tube. The tube was centrifuged (450g for 5 min at room temperature) and resuspended in 1 ml of complete corneal epithelial cell media. The AmnioGraft (Biotissue, Miami, FL, USA) was thawed and rinsed with phosphate-buffered saline (PBS). The graft was transferred to a petri dish containing prewarmed CTS TrypLE for de-epithelialization. After incubation at 37°C for 30 min, amniotic epithelial cells were scraped off the membrane using a cell scraper. The denuded membrane was then transferred to a new petri dish and rinsed with PBS two times. Complete amniotic epithelial cell removal was determined to be essential for limbal epithelial cell growth, and validation of amniotic membrane processing was performed by staining the de-epithelialized membranes with 4′,6-diamidino-2-phenylindole (DAPI), a blue dye that stains nuclei. The absence of blue dye staining indicating the complete removal of endogenous amniotic epithelial cells was determined using a fluorescent microscope.
Before seeding on the membrane, limbal cells were counted using trypan blue to assess viability; 5 × 104 viable cells were seeded on the basement side of the de-epithelialized AmnioGraft plated in a transwell to generate CALEC construct. Remaining cells were taken for quality control (phenotype, CFE, proliferation potential assay, endotoxin, mycoplasma, and sterility).
Flow cytometry to characterize cell phenotype
Approximately 100,000 harvested P0 cells were washed twice by adding 2 ml of PBS and centrifuged at 450g for 5 min at room temperature. Washed supernatant was carefully decanted, and the following antibodies (BD Biosciences, San Jose, CA, USA) were added to the cell suspension: CD45–fluorescein isothiocyanate (FITC) (20 μl; catalog no. 347463), CD31-FITC (20 μl; catalog no. 555445), CD13-phycoerythrin (20 μl; catalog no. 555394), CD49F–PerCP-Cy5.5 (5 μl; catalog no. 562495), and CD340-allophycocyanin (5 μl; catalog no. 340554). The tube was mixed and incubated in the dark at room temperature for 15 min. After incubation, the cells were washed twice by adding 2 ml of PBS and centrifuged at 450g for 5 min at room temperature. As control, 100,000 unstained cells were washed a total of four times with PBS and centrifuged at 450g for 5 min at room temperature. Both unstained and stained cells were resuspended in 450 μl of PBS and analyzed using a Beckman Coulter Navios with CXP software (gating strategy shown in fig. S2). Nonspecific staining from the control cells was subtracted from the results of the test cells before reporting the results.
iATP assay
We developed a quantitative assay to evaluate the proliferative potential of limbal epithelial cells by measuring iATP. Changes in iATP concentration as a result of mitochondrial activity correlate directly with cell number and proliferation. Cells were harvested at the end of the primary culture (P0). Three different cell doses (125, 250, and 500 cells per well) were plated in six replicates in a 96-well plate in corneal growth media and placed at 37°C, 5% CO2, and 90% relative humidity. iATP concentration was measured after 7 days by bioluminescence using a SpectraMax i3 Multi-Mode Microplate Reader (Molecular Devices, San Jose, CA). ATP standards, controls, and enumeration reagents are all from Hemogenix (Colorado Springs, CO, USA). The cell dose per well is plotted against the mean iATP concentration (mM) per well, and the slope of the linear regression provided the proliferation potential of each sample.
CFE assay
Clonal potential of limbal epithelial cells was also assessed using CFE assay, which is commonly used to assess epithelial cell cultures. Cells at the end of primary culture (P0) were seeded at low density (1 to 200 cells/cm2), cultured for 10 to 14 days, fixed and stained with rhodamine B. Photomicrographs of colonies were scored using an inverted microscope. Colony scores were then reported as a % CFE as determined by the ratio of the number of colonies counted to the number of inoculated cells × 100%.
LDH assay
LDH activity is a well-established marker/indicator of cellular toxicity and lysis (32). LDH is a cytosolic enzyme that is released in the media only when the plasma membrane is damaged and is a surrogate marker for cell death. LDH was measured by a commercial colorimetric assay (Roche Diagnostics) on the CEDEX Bio Analyzer (Roche Diagnostics) and reported in a bar graph as the average LDH value ± SD.
EdU assay
To assess the quality of the final product (during process development), we evaluated in situ proliferation and viability at the end of the process (before formulation for transport). CALEC constructs were pulsed with the modified thymidine analog, EdU, for 4 hours and then processed for analysis. EdU is efficiently incorporated into newly synthesized DNA and fluorescently labeled with a bright, photostable Alexa Fluor dye. Nuclei were counterstained with DAPI, and images were acquired on an Olympus FluoView FV1000 confocal microscope. Cells that are synthesizing new DNA and incorporating EdU are green/light blue, while EdU-negative cells are dark blue. The EdU+ and EdU− cells were counted, and the percentage of proliferating cells were estimated manually or using automated image analysis software (Imaris, Bitplane).
LIVE/DEAD assay
To assess cell viability and qualify the LDH assay during process development, CALEC constructs were stained with LIVE/DEAD assay (Thermo Fisher Scientific, Waltham, MA, USA). It is a two-color assay to determine viability of cells in a population based on plasma membrane integrity and esterase activity. This method discriminates live from dead cells by simultaneously staining with green fluorescent calcein-AM to indicate intracellular esterase activity (live cells) and red fluorescent ethidium homodimer-1 to indicate loss of plasma membrane integrity (dead cells). Cells were analyzed under a fluorescence microscope (Floid cell imaging station). For each sample, five different fields were captured, and for each field, the % of dead cells was calculated as follow: % Dead Cells = (#dead cells)/(#dead cells + #live cells) * 100.
Glucose consumption and lactate production
CALEC construct metabolic activity was measured by glucose consumption and lactate production pre- (0 hours) and post-hypothermic (24 hours) storage. CALEC construct media supernatant samples were analyzed for LDH release (LDH assay) and metabolic activity (glucose consumption and lactate production) using the CEDEX Bio Analyzer (Roche Diagnostics) as per the manufacturer’s instructions. The corneal media and HypoThermosol FRS contain about 5 and 4 nM glucose (and no lactate), respectively.
Immunofluorescence staining for biomarker analysis
To test the stemness of CALEC constructs, the central construct was separated from the transwell insert by using the 17-mm corneal trephine. The construct was processed for whole mount immunostaining analysis (p63 was coimmunostained with Krt12 in one quadrant, and p63α was coimmunostained with C/EBPδ in the other quadrant) and embedded in OCT solution for cryosection. Human limbal biopsy, cultured P0 cells, P1 CALEC construct whole mount, and cryosections (12 μm in thickness) were fixed with 4% paraformaldehyde for 20 min at room temperature. The specimens were rehydrated in PBS and then incubated in 0.05% Triton X-100 for 20 min. After rinsing the samples with PBS and pre-incubating with 2% bovine serum albumin for 1 hour, the specimens were incubated with pan-p63 antibody (1:3 dilution; 790-4509, Roche), Keratin 12 antibody (1:500 dilution; ab185627, Abcam), p63α (1:100 dilution; 13109, Cell Signaling Technology), or C/EBPδ antibody (1:50 dilution; sc-365546, Santa Cruz Biotechnology) at 4°C overnight. After washing with PBS, the specimens were then incubated with the Alexa Fluor 488– and/or 594–conjugated immunoglobulin G secondary antibody (1:300 dilution; A21202 or A21207, Life Technologies) for 1 hour. The specimens were counterstained with DAPI mounting medium (H-1200, Vector Laboratories) and photographed under confocal laser scanning (SP8, Leica) or immunofluorescence microscopy (DMi8, Leica).
Clinical trial design
Following the development, standardization, and preclinical testing of CALEC manufacturing techniques designed to enhance safety and predictability of clinical outcomes, a phase 1 clinical trial was developed. The FDA approved an Investigational New Drug Application (IND #16102) to perform the CALEC graft for unilateral LSCD under a single center clinical trial at the MEE. The study adhered to the tenets of the Declaration of Helsinki and was approved by the MEE Institutional Review Board. Informed consent was obtained from all participants. The CALEC protocol is listed on www.clinicaltrials.gov (NCT02592330).
The primary aim of the clinical trial is to establish the safety and feasibility of CALEC transplantation in the treatment of unilateral LSCD. The inclusion and exclusion criteria are listed in table S2. Participants are 18 to <90 years old with unilateral LSCD as determined by conjunctivalization of the cornea defined by fibrovascular pannus more than 2 mm from the limbus into the cornea for ≥6 clock hours. LSCD was classified according to the global consensus on LSCD criteria (48). The trial visit schedule, testing procedures, and primary outcome definitions are detailed in the CALEC protocol (posted on https://public.jaeb.org/calec). The primary safety events of interest were defined as (i) ocular infection [defined as endophthalmitis or microbial keratitis (bacterial, fungal, and parasitic)], (ii) corneal perforation, and (iii) graft detachment of ≥50%.
During study visits, BCVA was assessed with Snellen acuity. Slit lamp examination was performed by a certified independent study investigator who was not the investigator performing the biopsy or the transplant. Corneal opacification was graded according to the Fantes (49) scale ranging from 0 (transparent) to grade 4 that is a totally dense opacity completely obscuring intraocular features. Corneal fluorescein staining was performed using a sodium fluorescein strip with sterile saline. The entire corneal staining was graded according to the National Eye Institute (NEI) grading scale where five corneal zones (superior, nasal, central, inferior, and temporal) were graded ranging from 0 (no staining) to 3 (severe) with the total score ranging from 0 to 15 points (50). Epithelial defects, if present, were measured at the greatest horizontal and vertical dimensions. Slit lamp photography was performed using diffuse white light and cobalt blue light at ×10 magnification. Corneal neovascularization was recorded in vessel length, clock hours of involvement, depth, and activity.
There were two recruitment phases of the trial. In the initial recruitment phase, enrollment was staggered to occur after completion of a 2-week posttransplant safety data review by an independent data and safety monitoring committee for each preceding participant. After preliminary feasibility for the first three successful CALEC transplants was confirmed with no immediate safety concerns, a second recruitment phase of open enrollment began. The current paper reports on the feasibility and a descriptive clinical case history through the first 12 months after transplant of the initial recruitment phase participants. A future paper will report the data for participants enrolled in the entire study and will include safety and efficacy outcome data through 18 months of follow-up.
CALEC biopsy and surgery
The CALEC biopsy and transplant procedures are described in the CALEC protocol (posted on https://public.jaeb.org/calec). All participants started on preoperative topical fluoroquinolone in the operated eye. The biopsy of the donor eye (approximately 3.0 mm × 3.0 mm) was performed for each patient under topical anesthesia and placed in HypoThermosol FRS animal component free solution (BioLife Solutions, Bothell, WA) for transfer to the Connell and O’Reilly Families Cell Manipulation Core Facility, Dana-Farber Cancer Institute, a cGMP facility, in a reusable EVO transporting device (BioLife Solutions, Bothell, WA) with temperature and tilt monitoring by GPS. After manufacturing was completed, the CALEC graft was released on the basis of the criteria detailed in Fig. 4. In most cases, sufficient number of cells were obtained after P0 culture to seed two Amniograft membranes. Two CALEC grafts constructs were delivered to the operating room in the six-well plate using the EVO transport device. In the operating room, the construct with higher cell count was used as a primary graft. The pannus tissue was carefully resected from the recipient eye, and mitomycin C (0.4 mg/ml) was applied to the ocular surface sparing the cornea and limbus to prevent fibrosis. The CALEC construct with the insert was removed from the six-well plate with sterile forceps and placed on a sterile silicone platform. The HypoThermosol was irrigated with balanced salt solution. Under the operating microscope, the CALEC construct was trephined with a 17-mm trephine, lifted from the culture insert, and placed on the recipient eye by holding the trephined insert bottom with tying forceps, which were further used to pull the construct from the insert onto the cornea. The CALEC grafts were sutured with interrupted 10-0 nylon sutures at the limbus, and fluorescein light was used to confirm cellular integrity. A BCL was placed on the eye. The post-operative regimen consisted of preservative-free moxifloxacin, autologous serum tears, and prednisolone (1%) eye drops. If epithelial defect was noted on the week 1 or 2 after the surgery, then preservative-free methylprednisolone was used. The pathology of the excised pannus revealed findings consistent with LSCD.
Acknowledgments
We thank the operations committee members (A.R.A., U.V.J., R. Lindblad, M. Maguire, M. Redford, and J.R.); data safety and monitoring committee members [J. Wittes (Chair), D. Barnbaum, D. McKenna, D. Roop, M. Rosenblatt, E. Sugar, and J. Weiss]; clinical center staff from the MEE, Harvard Medical School (U.V.J., J.Y., L.K.J., S.E., L. Langone, E. Fitzgerald, X. Chen, M. Stenerson, B. Barakati, M. E. Gutierrez Escalante, J. Hyojung Ha, M. Cheung, D. Raoof, L. Foster, R. Gupta, T. Hall, C. Prescott, C. Hamil, L. Sullivan, R. Dana, A. Gauthier, S.L., and S.O.T.); cell manufacturing staff from the Connell and O’Reilly Families Cell Manipulation Core Facility, Dana-Farber Cancer Institute (J.R., K.L.S., R. Khelladi, D.E.H.R., H.D., A. Cunningham, S. Richard, H.N., and H. Anderson); coordinating center staff from Jaeb Center for Health Research (A.R.A., D. Rojas, J. Shah, M. Maguire, L.S., W. Woodall, R. Parsons, and N. Cohen); and medical monitors staff from EMMES (O. J. Daniel, J. A. Sliman, A. E. Fiky, and R. Lindblad).
Funding: Research reported in this publication was supported by the NEI of the National Institutes of Health under award numbers UG1EY026508 (MEE Infirmary), UG1EY027726 (Cell Manipulation Core Facility), and UG1EY027725 (Coordinating Center). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. Preclinical development of the CALEC manufacturing process was supported by the Production Assistance for Cellular Therapy (PACT) Program of the NHLBI (contract HHSN268201000009C).
Author contributions: Writing—original draft: U.V.J., A.R.A., and J.R. Writing—review and editing: U.V.J., J.Y., L.K.J., S.L., H.N., K.L.S., L.S., A.R.A., A.K., K.K., A.G., S.O.T., A.R.K., S.E., D.E.H.R., H.D., R.D., M.A., and J.R.
Competing interests: All authors have completed and submitted the ICMJE form for disclosure of potential competing interest. Patent pending (U.V.J, R.D., J.R., and M.A.). U.V.J. and R.D. have a financial interest in OcuCell Inc. This company is developing living ophthalmic cell-based therapies for treating eye disease. U.V.J. and R.D. interests were reviewed and are managed by MEE Infirmary and Mass General Brigham in accordance with their competing interest policies. The authors declare that they have no other competing interests.
Data and materials availability: All data needed to evaluate the conclusions in the paper are present in the paper and/or the Supplementary Materials (table S3, A and B).
Supplementary Materials
This PDF file includes:
Tables S1 to S3
Figs. S1 to S3
REFERENCES AND NOTES
- 1.G. Cotsarelis, S. Z. Cheng, G. Dong, T. T. Sun, R. M. Lavker, Existence of slow-cycling limbal epithelial basal cells that can be preferentially stimulated to proliferate: Implications on epithelial stem cells. Cell 57, 201–209 (1989). [DOI] [PubMed] [Google Scholar]
- 2.H. S. Dua, A. Azuara-Blanco, Limbal stem cells of the corneal epithelium. Surv. Ophthalmol. 44, 415–425 (2000). [DOI] [PubMed] [Google Scholar]
- 3.E. J. Holland, Management of limbal stem cell deficiency: A historical perspective, past, present, and future. Cornea 10, S9–S15 (2015). [DOI] [PubMed] [Google Scholar]
- 4.M. Haagdorens, S. I. Van Acker, V. Van Gerwen, S. N. Dhubhghaill, C. Koppen, M. J. Tassignon, N. Zakaria, Limbal stem cell deficiency: Current treatment options and emerging therapies. Stem Cells Int. 2016, 9798374 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.J. Yin, U. Jurkunas, Limbal stem cell transplantation and complications. Seminars Ophthalmol. 33, 134–141 (2018). [DOI] [PubMed] [Google Scholar]
- 6.P. Haamann, O. M. Jensen, P. Schmidt, Limbal autograft transplantation. Acta Ophthalmol. Scandinavica 76, 117–118 (1998). [DOI] [PubMed] [Google Scholar]
- 7.K. R. Kenyon, S. C. Tseng, Limbal autograft transplantation for ocular surface disorders. Ophthalmology 96, 709–723 (1989). [DOI] [PubMed] [Google Scholar]
- 8.D. T. Tan, L. A. Ficker, R. J. Buckley, Limbal transplantation. Ophthalmology 103, 29–36 (1996). [DOI] [PubMed] [Google Scholar]
- 9.S. Basu, S. P. Sureka, S. S. Shanbhag, A. R. Kethiri, V. Singh, V. S. Sangwan, Simple limbal epithelial transplantation: Long-term clinical outcomes in 125 cases of unilateral chronic ocular surface burns. Ophthalmology 123, 1000–1010 (2016). [DOI] [PubMed] [Google Scholar]
- 10.V. M. Borderie, D. Ghoubay, C. Georgeon, M. Borderie, C. de Sousa, A. Legendre, H. Rouard, Long-term results of cultured limbal stem cell versus limbal tissue transplantation in stage III limbal deficiency. Stem Cells Transl. Med. 8, 1230–1241 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.S. X. Deng, F. Kruse, J. A. P. Gomes, C. C. Chan, S. Daya, R. Dana, F. C. Figueiredo, S. Kinoshita, P. Rama, V. Sangwan, A. R. Slomovic, D. Tan, Global consensus on the management of limbal stem cell deficiency. Cornea 39, 1291–1302 (2020). [DOI] [PubMed] [Google Scholar]
- 12.G. Pellegrini, C. E. Traverso, A. T. Franzi, M. Zingirian, R. Cancedda, M. de Luca, Long-term restoration of damaged corneal surfaces with autologous cultivated corneal epithelium. Lancet 349, 990–993 (1997). [DOI] [PubMed] [Google Scholar]
- 13.P. Rama, S. Matuska, G. Paganoni, A. Spinelli, M. De Luca, G. Pellegrini, Limbal stem-cell therapy and long-term corneal regeneration. N. Engl. J. Med. 363, 147–155 (2010). [DOI] [PubMed] [Google Scholar]
- 14.K. Sejpal, M. H. Ali, S. Maddileti, S. Basu, M. Ramappa, R. Kekunnaya, G. K. Vemuganti, V. S. Sangwan, Cultivated limbal epithelial transplantation in children with ocular surface burns. JAMA Ophthalmol. 131, 731–736 (2013). [DOI] [PubMed] [Google Scholar]
- 15.E. Di Iorio, S. Ferrari, A. Fasolo, E. Böhm, D. Ponzin, V. Barbaro, Techniques for culture and assessment of limbal stem cell grafts. Ocular Surf. 8, 146–153 (2010). [DOI] [PubMed] [Google Scholar]
- 16.R. Mendoza, A. E. Vaughan, A. D. Miller, The left half of the XMRV retrovirus is present in an endogenous retrovirus of NIH/3T3 Swiss mouse cells. J. Virol. 85, 9247–9248 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.C. A. Gregory, E. Reyes, M. J. Whitney, J. L. Spees, Enhanced engraftment of mesenchymal stem cells in a cutaneous wound model by culture in allogenic species-specific serum and administration in fibrin constructs. Stem Cells 24, 2232–2243 (2006). [DOI] [PubMed] [Google Scholar]
- 18.M. Sundin, O. Ringdén, B. Sundberg, S. Nava, C. Götherström, K. Le Blanc, No alloantibodies against mesenchymal stromal cells, but presence of anti-fetal calf serum antibodies, after transplantation in allogeneic hematopoietic stem cell recipients. Haematologica 92, 1208–1215 (2007). [DOI] [PubMed] [Google Scholar]
- 19.A. Heiskanen, T. Satomaa, S. Tiitinen, A. Laitinen, S. Mannelin, U. Impola, M. Mikkola, C. Olsson, H. Miller-Podraza, M. Blomqvist, A. Olonen, H. Salo, P. Lehenkari, T. Tuuri, T. Otonkoski, J. Natunen, J. Saarinen, J. Laine, N-glycolylneuraminic acid xenoantigen contamination of human embryonic and mesenchymal stem cells is substantially reversible. Stem Cells 25, 197–202 (2007). [DOI] [PubMed] [Google Scholar]
- 20.P. Tangvoranuntakul, P. Gagneux, S. Diaz, M. Bardor, N. Varki, A. Varki, E. Muchmore, Human uptake and incorporation of an immunogenic nonhuman dietary sialic acid. Proc. Natl. Acad. Sci. U.S.A. 100, 12045–12050 (2003). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.S. Bobba, S. Chow, S. Watson, N. Di Girolamo, Clinical outcomes of xeno-free expansion and transplantation of autologous ocular surface epithelial stem cells via contact lens delivery: A prospective case series. Stem Cell Res. Ther. 6, 23 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.N. Di Girolamo, M. Bosch, K. Zamora, M. T. Coroneo, D. Wakefield, S. L. Watson, A contact lens-based technique for expansion and transplantation of autologous epithelial progenitors for ocular surface reconstruction. Transplantation 87, 1571–1578 (2009). [DOI] [PubMed] [Google Scholar]
- 23.I. R. Schwab, Cultured corneal epithelia for ocular surface disease. Trans. Am. Ophthalmol. Soc. 97, 891–986 (1999). [PMC free article] [PubMed] [Google Scholar]
- 24.R. A. Colabelli Gisoldi, A. Pocobelli, C. M. Villani, D. Amato, G. Pellegrini, Evaluation of molecular markers in corneal regeneration by means of autologous cultures of limbal cells and keratoplasty. Cornea 29, 715–722 (2010). [DOI] [PubMed] [Google Scholar]
- 25.A. Fasolo, E. Pedrotti, M. Passilongo, G. Marchini, C. Monterosso, R. Zampini, E. Bohm, F. Birattari, A. Franch, V. Barbaro, M. Bertolin, C. Breda, E. Di Iorio, B. Ferrari, S. Ferrari, M. Meneguzzi, D. Ponzin, Safety outcomes and long-term effectiveness of ex vivo autologous cultured limbal epithelial transplantation for limbal stem cell deficiency. Br. J. Ophthalmol. 101, 640–649 (2017). [DOI] [PubMed] [Google Scholar]
- 26.P. Rama, S. Bonini, A. Lambiase, O. Golisano, P. Paterna, M. De Luca, G. Pellegrini, Autologous fibrin-cultured limbal stem cells permanently restore the corneal surface of patients with total limbal stem cell deficiency. Transplantation 72, 1478–1485 (2001). [DOI] [PubMed] [Google Scholar]
- 27.Y. Feng, M. Borrelli, T. Meyer-Ter-Vehn, S. Reichl, S. Schrader, G. Geerling, Epithelial wound healing on keratin film, amniotic membrane and polystyrene in vitro. Curr. Eye Res. 39, 561–570 (2014). [DOI] [PubMed] [Google Scholar]
- 28.J. Liu, B. D. Lawrence, A. Liu, I. R. Schwab, L. A. Oliveira, M. I. Rosenblatt, Silk fibroin as a biomaterial substrate for corneal epithelial cell sheet generation. Investig. Ophthalmol. Vis. Sci. 53, 4130–4138 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.C. Petsch, U. Schlötzer-Schrehardt, E. Meyer-Blazejewska, M. Frey, F. E. Kruse, B. O. Bachmann, Novel collagen membranes for the reconstruction of the corneal surface. Tissue Eng. Part A 20, 2378–2389 (2014). [DOI] [PubMed] [Google Scholar]
- 30.T. Nakamura, T. Inatomi, C. Sotozono, N. Koizumi, S. Kinoshita, Successful primary culture and autologous transplantation of corneal limbal epithelial cells from minimal biopsy for unilateral severe ocular surface disease. Acta Ophthalmol. Scandinavica 82, 468–471 (2004). [DOI] [PubMed] [Google Scholar]
- 31.R. J. Tsai, L. M. Li, J. K. Chen, Reconstruction of damaged corneas by transplantation of autologous limbal epithelial cells. N. Engl. J. Med. 343, 86–93 (2000). [DOI] [PubMed] [Google Scholar]
- 32.T. Decker, M. L. Lohmann-Matthes, A quick and simple method for the quantitation of lactate dehydrogenase release in measurements of cellular cytotoxicity and tumor necrosis factor (TNF) activity. J. Immunol. Methods 115, 61–69 (1988). [DOI] [PubMed] [Google Scholar]
- 33.C. Legrand, J. M. Bour, C. Jacob, J. Capiaumont, A. Martial, A. Marc, M. Wudtke, G. Kretzmer, C. Demangel, D. Duval, J. Hache, Lactate dehydrogenase (LDH) activity of the cultured eukaryotic cells as marker of the number of dead cells in the medium [corrected]. J. Biotechnol. 25, 231–243 (1992). [DOI] [PubMed] [Google Scholar]
- 34.G. Pellegrini, E. Dellambra, O. Golisano, E. Martinelli, I. Fantozzi, S. Bondanza, D. Ponzin, F. McKeon, M. De Luca, p63 Identifies keratinocyte stem cells. Proc. Natl. Acad. Sci. U.S.A. 98, 3156–3161 (2001). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.E. Di Iorio, V. Barbaro, A. Ruzza, D. Ponzin, G. Pellegrini, M. De Luca, Isoforms of ∆Np63 and the migration of ocular limbal cells in human corneal regeneration. Proc. Natl. Acad. Sci. U.S.A. 102, 9523–9528 (2005). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.V. Barbaro, A. Testa, E. Di Iorio, F. Mavilio, G. Pellegrini, M. De Luca, C/EBPδ regulates cell cycle and self-renewal of human limbal stem cells. J. Cell Biol. 177, 1037–1049 (2007). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.C. S. de Paiva, Z. Chen, R. M. Corrales, S. C. Pflugfelder, D. Q. Li, ABCG2 transporter identifies a population of clonogenic human limbal epithelial cells. Stem Cells 23, 63–73 (2005). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.B. R. Ksander, P. E. Kolovou, B. J. Wilson, K. R. Saab, Q. Guo, J. Ma, S. P. McGuire, M. S. Gregory, W. J. Vincent, V. L. Perez, F. Cruz-Guilloty, W. W. Kao, M. K. Call, B. A. Tucker, Q. Zhan, G. F. Murphy, K. L. Lathrop, C. Alt, L. J. Mortensen, C. P. Lin, J. D. Zieske, M. H. Frank, N. Y. Frank, ABCB5 is a limbal stem cell gene required for corneal development and repair. Nature 511, 353–357 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.B. Chen, S. Mi, B. Wright, C. J. Connon, Investigation of K14/K5 as a stem cell marker in the limbal region of the bovine cornea. PLOS ONE 5, e13192 (2010). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.J. Collin, R. Queen, D. Zerti, S. Bojic, B. Dorgau, N. Moyse, M. M. Molina, C. Yang, S. Dey, G. Reynolds, R. Hussain, J. M. Coxhead, S. Lisgo, D. Henderson, A. Joseph, P. Rooney, S. Ghosh, L. Clarke, C. Connon, M. Haniffa, F. Figueiredo, L. Armstrong, M. Lako, A single cell atlas of human cornea that defines its development, limbal progenitor cells and their interactions with the immune cells. Ocul. Surface 21, 279–298 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.L. Gómez-Cid, L. Grigorian-Shamagian, R. Sanz-Ruiz, A. S. de la Nava, A. I. Fernández, M. E. Fernández-Santos, F. Fernández-Avilés, The essential need for a validated potency assay for cell-based therapies in cardiac regenerative and reparative medicine. A practical approach to test development. Stem Cell Rev. Rep. 17, 2235–2244 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Y. Barrandon, H. Green, Three clonal types of keratinocyte with different capacities for multiplication. Proc. Natl. Acad. Sci. U.S.A. 84, 2302–2306 (1987). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.G. Pellegrini, O. Golisano, P. Paterna, A. Lambiase, S. Bonini, P. Rama, M. De Luca, Location and clonal analysis of stem cells and their differentiated progeny in the human ocular surface. J. Cell Biol. 145, 769–782 (1999). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.S. Kolli, S. Ahmad, M. Lako, F. Figueiredo, Successful clinical implementation of corneal epithelial stem cell therapy for treatment of unilateral limbal stem cell deficiency. Stem Cells 28, 597–610 (2010). [DOI] [PubMed] [Google Scholar]
- 45.B. E. Ramírez, A. Sánchez, J. M. Herreras, I. Fernández, J. García-Sancho, T. Nieto-Miguel, M. Calonge, Stem cell therapy for corneal epithelium regeneration following good manufacturing and clinical procedures. Biomed. Res. Int. 2015, 408495 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46.A. J. Shortt, G. A. Secker, M. S. Rajan, G. Meligonis, J. K. Dart, S. J. Tuft, J. T. Daniels, Ex vivo expansion and transplantation of limbal epithelial stem cells. Ophthalmology 115, 1989–1997 (2008). [DOI] [PubMed] [Google Scholar]
- 47.V. S. Sangwan, S. Basu, G. K. Vemuganti, K. Sejpal, S. V. Subramaniam, S. Bandyopadhyay, S. Krishnaiah, S. Gaddipati, S. Tiwari, D. Balasubramanian, Clinical outcomes of xeno-free autologous cultivated limbal epithelial transplantation: A 10-year study. Br. J. Ophthalmol. 95, 1525–1529 (2011). [DOI] [PubMed] [Google Scholar]
- 48.S. X. Deng, V. Borderie, C. C. Chan, R. Dana, F. C. Figueiredo, J. A. P. Gomes, G. Pellegrini, S. Shimmura, F. E. Kruse, Global consensus on definition, classification, diagnosis, and staging of limbal stem cell deficiency. Cornea 38, 364–375 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 49.F. E. Fantes, K. D. Hanna, G. O. Waring 3rd, Y. Pouliquen, K. P. Thompson, M. Savoldelli, Wound healing after excimer laser keratomileusis (photorefractive keratectomy) in monkeys. Arch. Ophthalmol 108, 665–675 (1990). [DOI] [PubMed] [Google Scholar]
- 50.M. A. Lemp, Report of the National Eye Institute/Industry workshop on Clinical Trials in Dry Eyes. CLAO J. 21, 221–232 (1995). [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Tables S1 to S3
Figs. S1 to S3