The

n e w e ng l a n d j o u r na l

of

m e dic i n e

original article

Limbal Stem-Cell Therapy and Long-Term

Corneal Regeneration
Paolo Rama, M.D., Stanislav Matuska, M.D., Giorgio Paganoni, M.D.,
Alessandra Spinelli, M.D., Michele De Luca, M.D., and Graziella Pellegrini, Ph.D.

A BS T R AC T
Background

Corneal renewal and repair are mediated by stem cells of the limbus, the narrow
zone between the cornea and the bulbar conjunctiva. Ocular burns may destroy the
limbus, causing limbal stem-cell deficiency. We investigated the long-term clinical
results of cell therapy in patients with burn-related corneal destruction associated
with limbal stem-cell deficiency, a highly disabling ocular disease.
Methods

We used autologous limbal stem cells cultivated on fibrin to treat 112 patients with
corneal damage, most of whom had burn-dependent limbal stem-cell deficiency.
Clinical results were assessed by means of KaplanMeier, KruskalWallis, and univariate and multivariate logistic-regression analyses. We also assessed the clinical
outcome according to the percentage of holoclone-forming stem cells, detected as
cells that stain intensely (p63-bright cells) in the cultures.

From San Raffaele Scientific Institute,

Ophthalmology Unit, Milan (P.R., S.M.,
G.P., A.S.); and the Center for Regenerative Medicine Stefano Ferrari, University
of Modena and Reggio Emilia, Modena,
Italy (M.D.L., G.P.). Address reprint requests to Professor Pellegrini at the Center for Regenerative Medicine Stefano Ferrari, University of Modena and Reggio
Emilia, Via Glauco Gottardi 100, Modena,
Italy, or at graziella.pellegrini@unimore.it.
This article (10.1056/NEJMoa0905955) was
published on June 23, 2010, at NEJM.org.
N Engl J Med 2010;363:147-55.
Copyright 2010 Massachusetts Medical Society.

Results

Permanent restoration of a transparent, renewing corneal epithelium was attained

in 76.6% of eyes. The failures occurred within the first year. Restored eyes remained
stable over time, with up to 10 years of follow-up (mean, 2.911.99; median, 1.93).
In post hoc analyses, success that is, the generation of normal epithelium on donor stroma was associated with the percentage of p63-bright holoclone-forming
stem cells in culture. Cultures in which p63-bright cells constituted more than 3%
of the total number of clonogenic cells were associated with successful transplantation in 78% of patients. In contrast, cultures in which such cells made up 3% or less
of the total number of cells were associated with successful transplantation in only
11% of patients. Graft failure was also associated with the type of initial ocular
damage and postoperative complications.
Conclusions

Cultures of limbal stem cells represent a source of cells for transplantation in the
treatment of destruction of the human cornea due to burns.

n engl j med 363;2 nejm.org july 8, 2010

The New England Journal of Medicine

Downloaded from nejm.org on December 1, 2016. For personal use only. No other uses without permission.
Copyright 2010 Massachusetts Medical Society. All rights reserved.

147

The

n e w e ng l a n d j o u r na l

clear cornea is essential to visual

acuity and depends on stromal avascularity and epithelial integrity.1 Corneal renewal and repair are mediated by stem cells of the
limbus, the narrow zone between the cornea and
the bulbar conjunctiva.2 Ocular burns may destroy
the limbus, causing limbal stem-cell deficiency.
In such cases, the cornea acquires an epithelium
through the invasion of bulbar conjunctival cells.
This process leads to neovascularization, chronic
inflammation, and stromal scarring, with corneal
opacity and loss of vision.3 Allogeneic corneal
transplantation (keratoplasty) restores transparency temporarily, but eventually, the conjunctival
cells begin to invade and resurface the cornea. The
only way to prevent this invasion is to restore the
limbus. Such restoration has been attained in cases
of unilateral limbal stem-cell deficiency through
the grafting of limbal fragments obtained from
the uninjured eye.4
Under appropriate culture conditions, human
keratinocytes generate holoclones, meroclones,
and paraclones, all of which are capable of proliferation.5 Holoclone-forming cells are the stem
cells of virtually all human squamous epithelia2,6,7
and are required to permanently restore massive
epithelial defects.6-8 In humans, holoclone-forming cells are located in the limbus but not in the
central cornea.9,10 Paraclones, which are derived
from holoclones, have the properties of transient
amplifying progenitor cells.5,9 Meroclones have
intermediate proliferative and clonogenic potential; they give rise to and are a reservoir for paraclones.5,9
Human limbal stem cells can be identified,
both in vivo and in vitro, through their expression of the p63 transcription factor.11-13 The TP63
gene generates six isoforms.14 Transcription from
different promoters generates two different pre
messenger RNAs: TAp63 and Np63. Alternative
splicing of each transcript produces , , and
isoforms.14 Ocular keratinocytes may contain all
the N isoforms, but Np63 (hereafter referred
to as p63) is by far the most abundant11,12; it is
present in the limbus but not in the uninjured
central cornea12,13 and is expressed in holoclones
but not in paraclones.12 The protein p63 sustains
the proliferative potential of limbal stem cells,
and the C/EBP transcription factor maintains the
self-renewal of these cells and regulates their mitotic rate.13 During corneal repair, limbal stem
cells that stain intensely (hereafter referred to
148

of

m e dic i n e

as p63-bright cells) are released from C/EBPdependent mitotic constraints, multiply, and migrate to regenerate corneal epithelium.13
The finding that human limbal cell cultures15
contain holoclones9 led to the first therapeutic use
of such cultures in the regeneration of corneal
epithelium.16 Autologous limbal cell cultures also
offer an opportunity to treat patients who have
severe bilateral loss of corneal epithelium, provided that a tiny part of the limbus is spared in
one of the two eyes; thus, it is not applicable in
cases of total bilateral limbal stem-cell deficiency.
Related studies have been reported,6,17 but they
have been limited because of the comparatively
small numbers of cases, the relatively short follow-up, and the heterogeneity among causes of
the disorders and among the sources of transplants and cell cultures. We report the long-term
results of a study in which we used this approach
to treat patients with corneal damage from common causes.

Me thods
Patients

Patients with severe or total, unilateral or partial

burn-related lateral limbal stem-cell deficiency were
included in the study. (For details, see Table 1 in
the Supplementary Appendix, available with the
full text of this article at NEJM.org.)
Cell Culture and Treatment

Autologous limbal stem cells were obtained from

a biopsy specimen (1 to 2 mm2) taken from the
patients contralateral eye, and the cells were cultured on fibrin, a natural substrate that preserves
holoclone-forming cells.18-20 Clinical-gradecertified, lethally irradiated 3T3-J2 cells from a master cell bank were used as feeder cells. Details of
cell culture, graft preparations, immunohisto
chemical tests, and surgical and postoperative procedures are described in the Supplementary Appendix.
Follow-up and Assessment of Outcome

Follow-up evaluations were performed according

to standard procedures; for details, see the Supplementary Appendix. The outcome of treatment
was assessed at 1 year. The treatment was considered to be successful if all symptoms had disappeared and a transparent, avascular, and stable
corneal surface had been restored and was con-

n engl j med 363;2 nejm.org july 8, 2010

The New England Journal of Medicine

Downloaded from nejm.org on December 1, 2016. For personal use only. No other uses without permission.
Copyright 2010 Massachusetts Medical Society. All rights reserved.

Stem-Cell Ther apy and Corneal Regener ation

sidered to be partially successful if most symptoms had disappeared but superficial neovascularization had recurred, even if it was not as
extensive as at the time of admission. Treatment
failure was defined as the presence of symptoms,
recurrent epithelial defects, pannus, and inflammation at 1 year. Three observers assessed the
clinical results; two were involved in the patients
care, and the third did not participate in any biologic or clinical procedure.
Statistical Analysis

We analyzed results with the use of the Kaplan

Meier, chi-square, MannWhitney, McNemar, and
KruskalWallis tests as well as univariate and
multivariate stepwise logistic-regression analyses
(as described in the Supplementary Appendix).

R e sult s
Characteristics of the Patients

We studied 113 eyes from 112 patients enrolled

between 1998 and 2006. The mean (SD) age of
the patients was 46.514.4 years (range, 14 to 80),
and 78.6% of them were men. The limbal stemcell deficiency ranged from severe (in 35.4% of
the patients) to total (in 64.6%) (Table 1 in the
Supplementary Appendix). The most common
causes of limbal stem-cell deficiency (in 97.3% of
cases) were chemical burns (83 alkali, 16 acid,
and 4 other) and thermal burns (7). Two patients
had a bacterial infection and one patient had undergone orbit irradiation. Patients had unilateral
(87.5%) or bilateral (12.5%) burns; only two patients had bilateral severe limbal stem-cell deficiency (Patients 24 and 83 in Table 1 in the Supplementary Appendix). Most of the eyes (84%) had
been treated previously with surgery (in some cases multiple surgeries), but all surgical treatments
had failed, probably because of an insufficient

number of residual limbal stem cells to regenerate corneal epithelium. Twenty-eight patients had
undergone a single unsuccessful keratoplasty,
and 20 had undergone multiple unsuccessful keratoplasties. Before this study, optimally corrected
visual acuity was less than 0.1 (i.e., light perception, hand movement, and counting fingers) in
88.5% of the patients and was 0.1 to 0.5 in 11.5%
of the patients. The baseline features of the patients, including diagnoses and grading of limbal stem-cell deficiency, are provided in the Supplementary Appendix.
Grafts

A total of 125 cultures were grafted in 113 eyes

(112 patients) between July 1998 and December
2007. These included a bilateral graft in Patient
24 plus 12 regrafts. Both eyes of Patient 24 and the
left eye of Patient 83 were grafted with cultures
prepared from a single biopsy specimen taken from
a spared limbal area.
The clinical outcome of limbal stem-cell transplantation was judged to be successful in 76.6%
of the eyes (Table 1).
The mean time between injury and the first
graft was 18.316.7 years (median, 12.9; range,
0.8 to 56.4). No intraoperative complications were
reported. Postoperative adverse events included
hemorrhage (in 12 patients), residual fibrin on the
third postoperative day (in 11 patients), inflammation (in 59 patients), blepharitis with epithelial
involvement (in 35 patients), and herpetic keratitis (in 3 patients) (Table 1 in the Supplementary
Appendix), all of which were treated with the use
of standard therapy.
Regeneration of Corneal Epithelium

The human corneal epithelium is renewed approximately every 9 to 12 months.1 Since we therefore
surmised that the regenerated epithelium would

Table 1. Clinical Outcomes of Limbal Stem-Cell Grafts in 112 Patients.*

Variable

Number
of Eyes

Success

107

73 (68.2)

Partial
Success

Failure

number of eyes (percent)

Outcome of first graft
Outcome of subsequent graft
Final outcome

18 (16.8)

16 (15.0)

12

9 (75.0)

2 (16.7)

1 (8.3)

107

82 (76.6)

14 (13.1)

11 (10.3)

* Six of the 112 patients were excluded because they did not complete the study. One patient had bilateral limbal stemcell deficiency and received two grafts.

n engl j med 363;2 nejm.org july 8, 2010

The New England Journal of Medicine

Downloaded from nejm.org on December 1, 2016. For personal use only. No other uses without permission.
Copyright 2010 Massachusetts Medical Society. All rights reserved.

149

The

n e w e ng l a n d j o u r na l

A Grafted Limbal Stem-Cell Survival after One Graft

Proportion Free of Failure (%)

100
80
60
40
20
0

10

10

Years since Randomization

No. at Risk

107

90

41

29

13

10

B Grafted Limbal Stem-Cell Survival after More Than One Graft

Proportion Free of Failure (%)

100
80
60
40
20
0

Years since Randomization

No. at Risk

107

95

50

36

18

14

Figure 1. KaplanMeier Estimates of Grafted Limbal Stem-Cell Survival.

Panel A shows the survival estimates for the cultures after one graft was
placed, with partial or total success attained in 68.2% of the eyes treated.
Panel B shows survival estimates after a second graft was placed in 11 eyes
(a total of 12 additional grafts, since 1 eye was regrafted twice), indicating
either partial success or failure. After regrafting, 9 of these eyes regenerated normal epithelium. Thus, the final clinical outcome was deemed a success in 76.6% of the eyes treated. All failures occurred within the first year
after grafting, whereas successful cultures remained stable for up to 10
years of follow-up.

persist for 1 year after transplantation only if the

stem cells had successfully engrafted,8 we carried
out the first conclusive clinical evaluation after at
least 1 year of follow-up. Six patients did not
complete the study and were excluded from the
analysis. We report the results obtained during a
maximum of 10 years of follow-up for the remaining 106 patients, with a mean follow-up period of 2.911.99 years and a median of 1.93 years
(interquartile range, 1.02 to 3.67).
One year after surgery, the 107 transplants were
scored as success, partial success, or failure in 73
150

of

m e dic i n e

eyes (68.2%), 18 eyes (16.8%), and 16 eyes (15%),

respectively (Table 1). Of 11 eyes (representing 12
grafts, since 1 eye was regrafted twice), 6 were
partially successful and 5 were considered to have
failed; these 11 were all regrafted with cultures
prepared from frozen cells taken from the original biopsy specimen or from cells cultured from
a new biopsy specimen. In 9 of these eyes, normal
epithelium was regenerated after regrafting.
Thus, the final clinical outcome was classified
as a success in 82 eyes (76.6%), a partial success
in 14 eyes (13.1%), and a failure in 11 eyes (10.3%)
(Table 1). Grafts that were considered successful
at 1 year after surgery invariably remained stable
thereafter, with a transparent avascular cornea
covered by normal corneal epithelium (Fig. 1).
Grafts that were considered failures at 1 year did
not worsen in subsequent years as compared with
baseline.
Univariate logistic-regression analysis showed
that failures were associated with a number of
factors, including age, the cause of limbal stemcell deficiency, previous surgical procedures, the
severity of damage, culturing and graft-transport
complications, postoperative complications, and
inflammation (Table 2). On multivariate logisticregression analysis, failures were still significantly
associated with the severity of damage and with
culturing, graft-transport, and postoperative complications (Table 2). Post hoc analyses showed no
significant difference in the likelihood of success
between eyes that were also treated with keratoplasty (71%) and eyes that were not (79%).
Criteria for Graftable Limbal Cultures

The average percentage of clonogenic cells detected in cultures was similar for successful, partially
successful, and failed transplants (37.2%, 31.1%,
and 35.5%, respectively). But the percentage of
colony-forming cells is inadequate as an indication of the percentage of stem cells, because it is
not possible to visually identify holoclones.5,9
Determining the number of holoclones by
means of clonal analysis is both cumbersome as
a routine procedure5,9 and unsuitable for preoperative quality control because the results become
available only after grafting. In contrast, quantitative immunodetection of p63, a marker of holoclones, is straightforward and can be performed
before grafting.6,21
We were able to examine the relationship between the percentage of p63-bright cells in each
culture and the clinical results in 82 patients. (We

n engl j med 363;2 nejm.org july 8, 2010

The New England Journal of Medicine

Downloaded from nejm.org on December 1, 2016. For personal use only. No other uses without permission.
Copyright 2010 Massachusetts Medical Society. All rights reserved.

Stem-Cell Ther apy and Corneal Regener ation

Table 2. Association between Characteristics of the Patients and 1-Year Outcomes.*

Partially
Successful
P Value on
or Failed Grafts Univariate
(N=34)
Analysis

Variable

Successful
Grafts
(N=73)

Age yr

49.114.0

41.414.3

0.02

54 (74)

29 (85)

0.23

7 (10)

9 (26)

0.04

Male sex no. of patients (%)

Odds Ratio
(95% CI) on
Multivariate
Analysis

P Value

Cause of burn no. of eyes (%)

Acid

55 (75)

23 (68)

Previous intervention no. of eyes (%)

Alkali

58 (79)

33 (97)

0.017

Severe damage no. of eyes (%)

41 (56)

28 (82)

0.008

4.05 (1.3412.2)

0.013

8 (11)

8 (24)

0.089

3.86 (1.0813.75)

0.037

Postoperative complications no. of

eyes (%)

43 (59)

31 (91)

<0.001

7.64 (1.9929.33)

0.003

Inflammation no. of eyes (%)

33 (45)

26 (76)

<0.003

22.5

28

0.46

1641

1743

12.9

7.7

Culture or graft-transport complications

no. of eyes (%)

Follow-up period
Biopsy to transplantation days
Median
Interquartile range
Burn to transplantation yr
Median
Interquartile range

5.633

0.15

4.827

* Plusminus values are means SD. CI denotes confidence interval.

P values were calculated with the use of the chi-square or MannWhitney test. Variables significantly associated with
risk of failure on univariate analysis (P<0.1) were included in the multivariate stepwise logistic-regression model.
Odds ratios are shown only for variables with significant associations. Odds ratios for continuous variables are for an
increase of 1 unit.

were unable to analyze this relationship in all patients because p63 was not found to be a marker
of limbal stem cells until 2001.11) A post hoc
KruskalWallis analysis showed that cultures giving rise to successful grafts contained a significantly greater percentage of p63-bright cells than
those giving rise to unsuccessful grafts (P<0.001)
(Fig. 2A). Grafts associated with partial success
were obtained from cultures with an intermediate percentage of p63-bright cells.
In additional post hoc analyses, we evaluated
the clinical results according to culture status.
Cultures that contained more than 3% p63-bright
cells led to successful corneal epithelial regeneration in 78% of the eyes (Fig. 2B), and only 8% and
14% of such cultures were associated with failed
or partial regeneration, respectively. In contrast,
cultures in which 3% or less of the cells were
p63-bright were successful in only 11% of eyes and
were unsuccessful or partially successful in 67%

and 22% of eyes, respectively. Thus, the percentage of p63-bright cells in the culture was positively associated with the clinical outcome of limbal stem-cell grafting. These data are consistent
with the finding that the human limbus contains
at least 5% stem cells.9 On the basis of these data,
we now use only cultures that contain more than
3% p63-bright cells to prepare grafts.
That said, a defined percentage of stem cells is
necessary but not sufficient to ensure a good
clinical result. Approximately 20% of cultures that
contain more than 3% p63-bright cells were associated with failed or only partially successful
treatment (Fig. 2), suggesting that other factors,
such as the severity of injury and the presence or
absence of complications, also influence outcome.
Restoration of Visual Acuity

The regeneration of normal avascular corneal epithelium was associated with amelioration of symp-

n engl j med 363;2 nejm.org july 8, 2010

The New England Journal of Medicine

Downloaded from nejm.org on December 1, 2016. For personal use only. No other uses without permission.
Copyright 2010 Massachusetts Medical Society. All rights reserved.

151

The

n e w e ng l a n d j o u r na l

p63-Bright Cells (%)

10

P<0.001

8
6
4
2
0

Success

Partial Success

Failure

B
100
90

Success

Partial success

Failure

P<0.001

Clinical Result (%)

80
70
60
50
40
30
20
10
0

>3% p63-Bright Cells

3% p63-Bright Cells

Figure 2. Relation between Clinical Results and

Percentage of p63-Bright Cells in Limbal Stem-Cell
Cultures.
The percentage of p63-bright cells was calculated on
subconfluent primary cultures used to seed graftable
secondary cultures. These analyses were performed on
only 82 cultures, since the relation between p63 cells
and limbal stem cells was not discovered until 2001.11
The box plots in Panel A, which are based on a
KruskalWallis analysis, show the relationship between
the percentage of p63-bright cells and the three outcomes of success, partial success, and failure. The horizontal line within each box represents the median value; the bottom and top lines of the box represent the
25th and 75th percentiles, respectively; and the horizontal lines below and above the box represent the
lowest and highest values, respectively, with outliers
excluded. In Panel B, also based on a KruskalWallis
test, grafts scored as successes, partial successes, and
failures were evaluated in relation to the percentage of
p63-bright holoclone-forming cells (>3% vs. 3%). The
cultures that contained more than 3% p63-bright cells
were successful in 78.1% of eyes and failed in 8.2% of
eyes, whereas cultures that contained smaller percentages of such cells were successful in 11.1% of eyes and
failed in 66.7% of eyes. Thus, the percentage of p63bright cells can be used to predict the clinical outcome
of limbal stem-cell cultures.

152

of

m e dic i n e

toms (burning, pain, and photophobia) in all patients whose grafts were considered successful
(see Table 1 in the Supplementary Appendix). Normal vision was restored in only those patients with
undamaged corneal stroma, however (Fig. 3, and
Table 1 in the Supplementary Appendix). To improve the visual acuity of eyes with stromal scarring, we performed corrective surgical procedures
12 to 24 months after grafting the limbal stemcell cultures: 46 patients underwent penetrating
keratoplasty (89%), lamellar keratoplasty (9%), or
phototherapeutic keratectomy (2%) to replace the
damaged stroma. We observed no intraoperative
complications and used standard therapeutic approaches for postoperative adverse events (Table 1
in the Supplementary Appendix). Histologic examination of the central cornea removed at the
time of keratoplasty showed that the regenerated
epithelium expressed keratin 12 (a marker of corneal epithelium) but not keratin 19 (a marker of
conjunctival epithelium) (Fig. 2 in the Supplementary Appendix). We observed regeneration of the
corneal epithelium that was sufficient to resurface the donor stroma in all the patients treated
with keratoplasty. Permanent recovery of at least
0.6 visual acuity (range, 0.6 to 1.0) was attained
in 21 patients (Table 1 in the Supplementary Appendix). The remaining 25 patients had partial
recovery of vision (up to 0.5 visual acuity). Figure
3B shows the results of limbal stem-cell grafting
followed by penetrating keratoplasty in the eyes
of three patients who had total limbal stem-cell
deficiency, complete corneal opacification, stromal
scarring, and reduced visual acuity (<0.1, counting fingers and perceiving hand movements) before treatment. After keratoplasty, which was performed between 4 and 6.5 years after grafting, the
cornea was transparent in each of the patients and
their visual acuity ranged from 0.3 to 0.9.
Figure 3 in the Supplementary Appendix shows
the eyes of Patient 24 who had severe bilateral
limbal stem-cell deficiency that resulted from an
alkali burn in 1948. He was treated with grafts of
limbal stem-cell cultures prepared from a single
biopsy specimen (taken from the left eye), followed
by penetrating keratoplasty. Both corneal surfaces
were restored. Follow-up at 2 years (right eye) and
5 years (left eye) showed that both eyes were stable. At follow-up his best-corrected visual acuity
was 0.6 and 0.7 in the left and right eyes, respectively.

n engl j med 363;2 nejm.org july 8, 2010

The New England Journal of Medicine

Downloaded from nejm.org on December 1, 2016. For personal use only. No other uses without permission.
Copyright 2010 Massachusetts Medical Society. All rights reserved.

Stem-Cell Ther apy and Corneal Regener ation

Discussion
Autologous cultures of limbal cells provide an adequate long-term source of transplant tissue for
the treatment of corneal damage due to burns.
We observed that a minimum of approximately
3000 stem cells, detected as p63-bright holocloneforming cells, was required to achieve clinical
success. (A limbal culture contains a minimum
of 3105 cells,20 about 30% of which are clonogenic; thus, about 3% of these clonogenic cells
should be holoclones.) Analysis of our findings
suggested that outcomes differ significantly depending on whether the transplanted cultures contain more than 3% p63-bright holoclone-forming
stem cells or 3% or less the success rates were
78% with the larger number of stem cells and 11%
with the smaller number. On the basis of this observation, we speculate that corneal regeneration
cannot be ascribed to a nonspecific stimulatory
effect of epithelial cultures, fibrin, or surgical manipulation on spared residual limbal cells (if such
cells are even present).
Preservation of holoclones requires culture with
selected 3T3 feeder cells and fetal-calf serum,6,22,23
and this culture method has been used worldwide
since the 1980s6,7 to treat patients with massive
full-thickness burns.18,19,24 During the past 30
years, no adverse effects have been reported, and
this method has been approved for use in the
United States, Japan, Italy, and South Korea.6,7
Retention of holoclones also requires appropriate substrates for the cultivation of cells; both
fibrin and plastic have been shown to preserve
holoclone-forming cells.18-20 Alternative methods
involving other reagents have been proposed that
obviate the use of feeder cells, serum, or both,6,17
since some investigators consider these reagents
to be potentially harmful.25 The retention of stem
cells when these alternative methods are used has
not been investigated.
Allogeneic and buccal keratinocytes have been
used as sources of grafts in previous clinical experimental studies to treat limbal stem-cell deficiency.6,17 Allogeneic keratinocytes even when
frozen or lyophilized improve healing of partial-thickness or small skin wounds by stimulating resident cells.26,27 However, engraftment of
these allogeneic cells is not permanent,26 and it
seems unlikely that the presence of stem cells is
required for their clinical effect. Donor epithelial
cells have not been recovered from the ocular sur-

A Cultures Only
Before

After

1.3 yr

B Cultures plus Keratoplasty

Before Treatment

After Treatment

Patient
22
6 yr

Patient
26
6.5 yr

Patient
46
4 yr

Figure 3. Regeneration of a Functional Corneal Epithelium and Restoration

of Visual Acuity.
Panel A shows the left eye of Patient 93 (see Table 1 in the Supplementary
Appendix, available with the full text of this article at NEJM.org), who had total limbal stem-cell deficiency due to an acid burn (image at left). His visual
acuity was reduced to counting fingers. A graft of autologous limbal cultures
was sufficient to regenerate functional corneal epithelium (image at right)
and to restore normal vision (visual acuity, 0.7), since the eye had no stromal
scarring. Panel B shows the eyes of Patients 22, 26, and 46 (see Table 1 in the
Supplementary Appendix), which were damaged by alkali burns and were
treated with unsuccessful surgery 13, 30, and 3 years before admission, respectively. All three eyes had total limbal stem-cell deficiency, complete corneal opacification, and stromal scarring (images at left). Vision was reduced
to counting fingers (in Patient 22) or perceiving hand movements (in Patients 26 and 46). In all three patients, autologous limbal stem-cell cultures
successfully regenerated functional corneal epithelium. To improve their visual acuity after grafting, the patients underwent penetrating keratoplasty. In all
three eyes, the engrafted limbal stem cells resurfaced the donor stroma. At
the last follow-up visits (at 6, 6.5, and 4 years, respectively), all eyes were covered by stable corneal epithelium (images at right). The keratoplasty resulted
in complete restoration of visual acuity in Patients 22 and 46 (0.9 and 0.8, respectively). The visual acuity of Patient 26 increased to only 0.3 because of a
concomitant amblyopia (the alkali burn had occurred 30 years before admission). In Patient 46, the follow-up image shows that the conjunctival vessels
stop at the conjunctivalcorneal boundary (arrowheads); they do not invade
the restored corneal surface.

n engl j med 363;2 nejm.org july 8, 2010

The New England Journal of Medicine

Downloaded from nejm.org on December 1, 2016. For personal use only. No other uses without permission.
Copyright 2010 Massachusetts Medical Society. All rights reserved.

153

The

n e w e ng l a n d j o u r na l

of

m e dic i n e

face in studies that looked at long-term out- stem cells thus represent a source of cells for
comes,28,29 and it seems unlikely that these cells transplantation in the treatment of burn-induced
would regenerate functional epithelium in patients destruction of the human cornea.
with total limbal stem-cell deficiency, nor is there
Supported by grants from the Italian Ministry for Education,
evidence that buccal keratinocytes can generate University and Research and Ministry of Health; lAssociation
bona fide corneal epithelium.30 If limbal stem-cell Franaise contre les Myopathies et les Maladies Neuromuscu
laires (AFM) Telethon; the European Commission Sixth Framedeficiency is only partial, allogeneic or oral kera- work Program for Research and Technological Development
tinocytes may be sufficient to stimulate resident (corneal engineering) and Seventh Framework Program; Optilimbal cells to regenerate corneal epithelium. Ad- mization of Stem Cell Therapy for Degenerative Epithelial and
Muscle Diseases (OptiStem, HEALTH-F5-2009-223098); Regione
ditional studies to test this hypothesis are war- EmiliaRomagna (area 1b, medicina rigenerativa); and the Italranted. The same argument might hold true for ian Ministry of Health.
Disclosure forms provided by the authors are available with
the effect of autologous cultures on corneal regenthe full text of this article at NEJM.org.
eration. Nevertheless, the association that we obThis article is dedicated to the memory of Professor Stefano
served between the percentage of p63-bright Ferrari, whose inspiring passion and commitment to excellence
cells and clinical success was never related to in science and education brought about the Center for Regenerative Medicine, and also to Professor Giovanni Rama, who was
a nonspecific stimulatory effect of the transplant- involved in the first phase of the project.
ed culture on resident stem cells.
We thank Dr. Howard Green (Harvard Medical School, BosIn conclusion, our study shows that in patients ton) for providing the original 3T3-J2 feeder cells; Patrizia Pater
na and Sergio Bondanza (Istituto Dermopatico dellImmacolata
with limbal stem-cell deficiency who received cul- [IDI]), and Marina Bertolin, Luca Canton, Alessandro Chiarello,
tured limbal stem-cell grafts for corneal trans- Barbara Ferrari, Daniela Ferrari, Valeria Leuci, and Erika Nizza
plantation, the clinical results were successful at (Veneto Eye Bank Foundation [FBOV]) for their assistance with
cell culture; Osvaldo Golisano (IDI) and Enzo Di Iorio and Vanup to 10 years (at a median of 2 years) in more than essa Barbaro (FBOV) for immunohistochemical testing; and
75% of the patients treated. Cultures of limbal Giliola Calori and Laura Angelici for statistical analyses.
References
1. Wagoner MD. Chemical injuries of

the eye: current concepts in pathophysiology and therapy. Surv Ophthalmol 1997;
41:275-313.
2. Pellegrini G, Rama P, Mavilio F, De
Luca M. Epithelial stem cells in corneal
regeneration and epidermal gene therapy.
J Pathol 2009;217:217-28.
3. Dua HS, Azuara-Blanco A. Limbal
stem cells of the corneal epithelium. Surv
Ophthalmol 2000;44:415-25.
4. Kenyon KR, Tseng SC. Limbal autograft transplantation for ocular surface
disorders. Ophthalmology 1989;96:709-23.
5. Barrandon Y, Green H. Three clonal
types of keratinocyte with different capacities for multiplication. Proc Natl Acad
Sci U S A 1987;84:2302-6.
6. De Luca M, Pellegrini G, Green H. Regeneration of squamous epithelia from
stem cells of cultured grafts. Regen Med
2006;1:45-57.
7. Green H. The birth of therapy with
cultured cells. Bioessays 2008;30:897-903.
8. Mavilio F, Pellegrini G, Ferrari S, et al.
Correction of junctional epidermolysis
bullosa by transplantation of genetically
modified epidermal stem cells. Nat Med
2006;12:1397-402.
9. Pellegrini G, Golisano O, Paterna P, et
al. Location and clonal analysis of stem
cells and their differentiated progeny in
the human ocular surface. J Cell Biol
1999;145:769-82.
10. Majo F, Rochat A, Nicolas M, Jaoud

154

GA, Barrandon Y. Oligopotent stem cells

are distributed throughout the mammalian ocular surface. Nature 2008;456:250-4.
11. Pellegrini G, Dellambra E, Golisano
O, et al. p63 Identifies keratinocyte stem
cells. Proc Natl Acad Sci U S A 2001;
98:3156-61.
12. Di Iorio E, Barbaro V, Ruzza A, Ponzin
D, Pellegrini G, De Luca M. Isoforms of
DeltaNp63 and the migration of ocular
limbal cells in human corneal regeneration. Proc Natl Acad Sci U S A 2005;
102:9523-8.
13. Barbaro V, Testa A, Di Iorio E, Mavilio
F, Pellegrini G, De Luca M. C/EBPdelta
regulates cell cycle and self-renewal of
human limbal stem cells. J Cell Biol 2007;
177:1037-49.
14. Yang A, Kaghad M, Wang Y, et al. p63,
A p53 homolog at 3q27-29, encodes multiple products with transactivating, deathinducing, and dominant-negative activities. Mol Cell 1998;2:305-16.
15. Lindberg K, Brown ME, Chaves HV,
Kenyon KR, Rheinwald JG. In vitro propagation of human ocular surface epithelial
cells for transplantation. Invest Ophthalmol Vis Sci 1993;34:2672-9.
16. Pellegrini G, Traverso CE, Franzi AT,
Zingirian M, Cancedda R, De Luca M.
Long-term restoration of damaged corneal
surfaces with autologous cultivated corneal epithelium. Lancet 1997;349:990-3.
17. Shortt AJ, Secker GA, Notara MD, et
al. Transplantation of ex vivo cultured

limbal epithelial stem cells: a review of

techniques and clinical results. Surv Ophthalmol 2007;52:483-502.
18. Pellegrini G, Ranno R, Stracuzzi G, et
al. The control of epidermal stem cells
(holoclones) in the treatment of massive
full-thickness burns with autologous keratinocytes cultured on fibrin. Transplantation 1999;68:868-79.
19. Ronfard V, Rives JM, Neveux Y, Carsin
H, Barrandon Y. Long-term regeneration
of human epidermis on third degree
burns transplanted with autologous cultured epithelium grown on a fibrin matrix. Transplantation 2000;70:1588-98.
20. Rama P, Bonini S, Lambiase A, et al.
Autologous fibrin-cultured limbal stem
cells permanently restore the corneal surface of patients with total limbal stem cell
deficiency. Transplantation 2001;72:147885.
21. Di Iorio E, Barbaro V, Ferrari S, Ortolani C, De Luca M, Pellegrini G. Q-FIHC:
quantification of fluorescence immunohistochemistry to analyse p63 isoforms and
cell cycle phases in human limbal stem
cells. Microsc Res Tech 2006;69:983-91.
22. Rheinwald JG, Green H. Serial cultivation of strains of human epidermal keratinocytes: the formation of keratinizing
colonies from single cells. Cell 1975;6:33143.
23. Green H, Kehinde O, Thomas J.
Growth of cultured human epidermal cells
into multiple epithelia suitable for graft-

n engl j med 363;2 nejm.org july 8, 2010

The New England Journal of Medicine

Downloaded from nejm.org on December 1, 2016. For personal use only. No other uses without permission.
Copyright 2010 Massachusetts Medical Society. All rights reserved.

Stem-Cell Ther apy and Corneal Regener ation

ing. Proc Natl Acad Sci U S A 1979;76:
5665-8.
24. Gallico GG III, OConnor NE, Compton CC, Kehinde O, Green H. Permanent
coverage of large burn wounds with autologous cultured human epithelium. N Engl
J Med 1984;311:448-51.
25. Schwab IR, Johnson NT, Harkin DG.
Inherent risks associated with manufacture of bioengineered ocular surface tissue. Arch Ophthalmol 2006;124:1734-40.
26. De Luca M, Albanese E, Bondanza S,

et al. Multicentre experience in the treatment of burns with autologous and allogenic cultured epithelium, fresh or preserved in a frozen state. Burns 1989;15:
303-9.
27. De Luca M, Albanese E, Cancedda R,
et al. Treatment of leg ulcers with cryopreserved allogeneic cultured epithelium:
a multicenter study. Arch Dermatol 1992;
128:633-8.
28. Henderson TR, Coster DJ, Williams
KA. The long term outcome of limbal al-

lografts: the search for surviving cells. Br

J Ophthalmol 2001;85:604-9.
29. Daya SM, Watson A, Sharpe JR, et al.
Outcomes and DNA analysis of ex vivo expanded stem cell allograft for ocular surface reconstruction. Ophthalmology 2005;
112:470-7.
30. Pellegrini G. Changing the cell source
in cell therapy? N Engl J Med 2004;351:
1170-2.
Copyright 2010 Massachusetts Medical Society.

receive immediate notification when

a journal article is released early

To be notified when an article is released early

on the Web and to receive the table of contents
of the Journal by e-mail every Wednesday evening,
sign up through our Web site at
NEJM.org.

n engl j med 363;2 nejm.org july 8, 2010

The New England Journal of Medicine

Downloaded from nejm.org on December 1, 2016. For personal use only. No other uses without permission.
Copyright 2010 Massachusetts Medical Society. All rights reserved.

155