 |
|
Histological Effects in the Iris After 3 Months of Latanoprost Therapy
The Mainz 1 Study
N. Pfeiffer, MD;
I. Grierson, PhD;
H. Goldsmith, BSc;
D. Hochgesand, MD;
A. Winkgen-Böhres, MD;
P. Appleton, PhD
Arch Ophthalmol. 2001;119:191-196.
ABSTRACT
| |
Objective To determine whether 3 months of topical latanoprost treatment caused
proliferative or degenerative effects on the peripheral iris of patients with
glaucoma.
Methods Seventeen patients requiring filtering surgery for primary open-angle
glaucoma or pseudoexfoliation glaucoma were randomized to receive topical
latanoprost for 3 months (n = 8) or alternative medication (n = 9) before
surgery. A trabeculectomy and a peripheral iridectomy specimen was obtained
from each patient during surgery. The tissue was subjected to histological
and immunohistochemical evaluation using 2 cell cycle markers: proliferating
cell nuclear antigen and nuclear-associated protein (Ki-67).
Results No degenerative or pathological changes were seen in the latanoprost-treated
irides, including the one specimen in this series in which there was an eye
color change. Proliferating cell nuclear antigen and nuclear-associated protein
markers were negative for changes in all the test specimens.
Conclusion Short-term treatment with latanoprost does not produce morphological
changes or cellular proliferation changes in the iris.
INTRODUCTION
LATANOPROST IS an analogue of the prostaglandin F2
and, because it is an effective intraocular pressure (IOP)–lowering
agent,1-3 it is
being used increasingly in the management of glaucoma. The one major ocular
adverse effect of latanoprost is that it can cause darkening of the iris.
During 1 year of treatment, about 20% of eyes treated with latanoprost darken
to a noticeable extent. Uniformly brown and blue irides darken rarely, but
hazel eyes darken frequently.2, 4
The mechanism by which the color change is brought about is obscure. There
is, to our knowledge, only one published article5
on the morphological features of a human iris treated with latanoprost, and
this is a case report of a patient with drug-induced iris darkening. Monkeys
undergoing long-term treatment with topical naturally occurring prostaglandins
also show iris color change, so it does seem to be the case that iris darkening
is a more general effect of some prostaglandins rather than being a specific
property of the analogue latanoprost.6
Among the possible mechanisms by which color change could be brought
about include hyperplasia of stromal melanocytes or stimulation of melanogenesis.
Overproduction of melanin granules could result in their release into the
stroma with consequent iritis7 and tissue degeneration
or, if they accumulated in the chamber angle, glaucoma might ensue.8 Proliferation of stromal melanocytes, if it occurred,
would be a particularly worrying adverse effect.
The present single-masked study was undertaken to compare peripheral
iridectomy specimens taken from patients who had vs those who had not been
treated with latanoprost for a 3-month period before glaucoma surgery. The
histological features of the iris were evaluated, looking for degenerative
changes in the tissue. However, the main thrust of the investigation was an
attempt to identify cells by immunohistochemistry, using antibodies against
proliferating cell nuclear antigen (PCNA) and nuclear-associated protein (Ki-67),
which were in or had been in the cell cycle. In this way, we wanted to find
evidence to support or refute the option that latanoprost might provoke cell
proliferation in iris tissue.
PATIENTS AND METHODS
We performed a randomized, single-masked study and recruited 20 patients
aged 40 years or older scheduled for routine trabeculectomy in one eye because
of uncontrolled IOP caused by primary open-angle glaucoma or pseudoexfoliation
glaucoma. Exclusion criteria included laser trabeculoplasty within 3 months
before the study, signs of ocular infection, inflammation, and use of contact
lenses. Eye color was classified as brown, blue, green, gray, or hazel. The
first 4 were uniform coloring, while the hazel was brown mixed with 1 of the
other 3.
Patients gave written informed consent. Then the patients were allocated
to receive, in addition to their medical regimen, either approximately 20
µL of topical latanoprost, 50 µg/mL, once each evening or an alternative
IOP-lowering agent (acetazolamide or 2-agonist clonidine).
The allocation of patients to treatment was at random, but an attempt was
made to proportion between the 2 groups based on the color classification
(Table 1). The treatment was restricted
to a 3-month period, which is the usual waiting period for filtering surgery
at Mainz University, Mainz, Germany. The patients and the consulting ophthalmologists
(N.P., D.H., and A.W.-B.) were not masked to their treatment during the waiting
period. All patients underwent measurements of visual acuity, biomicroscopy,
ophthalmoscopy, and Goldmann applanation tonometry at baseline, at 2 weeks
and 3 months before the day of surgery, and at 6 weeks after surgery. Iris
and gonioscopic photography was done at baseline and at 3 months. The study
was evaluated and approved by the Ethical Review Committee of Rheinland Pfalz.
A fornix-based trabeculectomy without antimetabolites was performed
by one of us (N.P.) at the 12-o'clock position, rendering a 1.5-mm square
trabeculectomy and a peripheral iridectomy specimen.9
The specimens were placed in a coded bottle on a support of filter paper and
fixed with 10% buffered formaldehyde solution.
LABORATORY PROCESSING AND HISTOLOGICAL FEATURES
Paraffin sections (3-4 µm thick) placed on 3-amino propyl triethoxysilane–coated
slides were stained with hematoxylin-eosin, Masson trichrome, and periodic
acid–Schiff. The histological sections were examined with the intention
of trying to identify any degenerative alterations in any of the coded specimens.
Thus, we evaluated under the light microscope each of the sections available
for the following features: (1) posterior epithelial degeneration, (2) stromal
inflammation, (3) vascular alterations, (4) clump cell frequency, (5) presence
of free stromal melanin, (6) stromal degeneration, (7) any atypical cellular
features in stromal melanocytes, and (8) thickness of the anterior border
layer. These were features highlighted as being of some interest based on
the findings of a previous case report.5 The
following grading system was used throughout the study: - indicates
absent; +/-, marginal; +, minimal; ++, moderate; and +++, marked. The
grading was found to be consistent from section to section for a given specimen,
reproducible on reanalysis, and comparable on interobserver checks. Throughout
the study, specimens were coded so that it was not known by the sectioner
or the evaluators (I.G., H.G., and P.A.) which were specimens from latanoprost-treated
eyes.
IMMUNOHISTOCHEMISTRY
Wax sections were deparaffinized in xylene and rehydrated in graded
ethanols. The sections available for each specimen were microwaved under citrate
buffer at pH 6.0. Positive controls (tissue with cells known to be in the
cell cycle) were human tonsil and developmental human eye tissue. The negative
control was adult retina. The ocular material consisted of eyes ranging from
10 weeks of development to 9 months postnatal (n = 11) and adult human eyes
of various eye color (n = 10). Following antigen retrieval, endogenous peroxidase
was blocked with a 0.3% solution of hydrogen peroxide for 20 minutes. Background
staining was eliminated by exposure to normal goat serum for a further 20
minutes. The primary antibodies used for this study were against antigens
connected with the cell cycle, PCNA and Ki-67. In the study, the antibody
against PCNA was an IgG2a mouse monoclonal antibody (DAKO A/S, Glostrup, Denmark)
and that against Ki-67 was MIB-1, an IgG1 mouse monoclonal antibody (DAKO
A/S). Both were applied at dilutions of 1:100 to sections for 16 hours at
4°C. After washing, the secondary biotinylated antibody was applied for
2 hours; then the sections were exposed to peroxidase-conjugated streptavadin
(both from DAKO A/S). Positive reactive sites were revealed using on occasion
3,3'-diaminobenzidine, which stains brown, or 3-amino-9-ethylcarbazole,
which produces a red color (Sigma-Aldrich Corp, St Louis, Mo).10-11
MICROSCOPY
At least 3 different levels of tissue of each specimen were examined
by conventional light microscopy and differential interference contrast optics
(Polyvar, Reichert, Vienna, Austria).
RESULTS
PATIENTS
We recruited 20 patients for the study. Twelve were women, and all were
white and had either primary open-angle glaucoma or pseudoexfoliation glaucoma
(Table 1). Seventeen patients
completed the study according to protocol. Of the 3 who withdrew from the
study, 1 (from the nonlatanoprost treatment group) underwent the surgery at
his home hospital after 2 months. Of the 2 other latanoprost-treated patients,
1 declined surgery and the other had well-controlled IOPs with latanoprost
treatment, so surgery was no longer considered necessary.
At baseline, the mean IOP in the latanoprost-treated eyes was 23.4 ±
5.2 mm Hg; it was 24.5 ± 4.4 mm Hg in the nonlatanoprost-treated eyes.
At 3 months, the mean IOPs were 18.2 ± 3.9 mm Hg in the latanoprost
group and 23.2 ± 5.2 mm Hg in the controls. (Data are given as mean
± SD.)
Only one 54-year-old patient (patient A6) with hazel eyes from the latanoprost
group underwent an eye color change, appreciable in Figure 1A-B. The iridectomy specimen from this eye was also taken
from the 12-o'clock position.
|
|
|
|
Figure 1. Pictures of one patient (patient
A6) taking latanoprost, with iris darkening at baseline (A) and at 3 months
(B).
|
|
|
HISTOLOGICAL FEATURES OF IRIDECTOMY SPECIMENS
The specimens were examined for the presence of any histological abnormalities,
and there was no evidence in any of our specimens of posterior epithelial
degeneration (Figure 2, Figure 3, and Figure 4) other than traumatic disruption from handling (Figure 4B). The stroma of all specimens exhibited
no signs of inflammation, vascular alterations, or any degeneration that could
be appreciated by light microscopy (Figure
2, Figure 3, and Figure 4). Free stromal melanin could not
be seen easily by conventional optics; however, by differential interference
contrast optics, free melanin could be appreciated more readily but was not
abundant in any sample examined (Figure 3B and Figure 4). Clump
cells also were identified more readily using differential interference contrast
optics. Thickness of the anterior border could be appreciated in this mode
(Figure 2B, Figure 3B, and Figure 4A).
The anterior border histological features of 1 specimen were difficult to
assess because of the presence of an ephelis (Figure 2), which was noted as an anterior border area of marginally
increased cellularity but clearly increased intracellular pigmentation.
|
|
|
|
Figure 2. Sections through an iridectomy
specimen (specimen A16). A, The specimen has rolled up with the pigmented
epithelium (E) in the center. The section shows an ephelis on the anterior
border (arrow) (hematoxylin-eosin, original magnification x180). B,
The section is immunostained for proliferating cell nuclear antigen, with
no counterstain, and visualized by differential interference contrast optics.
There are no positive-staining nuclei, and the ephelis is pigmented. The arrow
and E have the same meanings as in A (original magnification x350).
|
|
|
|
|
|
|
Figure 3. A latanoprost-treated patient
(patient A6). A, Normal morphological features of the stroma (S) (hematoxylin-eosin,
original magnification x500). B, Section of the same iris visualized
by differential interference contrast microscopy. There are no labeled nuclei.
S indicates stroma; v, blood vessels; and arrows, 2 clump cells at the anterior
border (proliferating cell nuclear antigen immunostaining, original magnification
x500).
|
|
|
|
|
|
|
Figure 4. Nuclear-associated protein–immunostained
sections (Ki-67) of iris visualized by differential interference contrast
optics. A, A nonlatanoprost-treated patient (patient A12) (original magnification
x500). Arrows indicate clump cells; arrowheads, free melanin. B, A latanoprost-treated
patient (patient A5). The posterior epithelium (E) is disrupted by contact
with filter paper (original magnification x500). There are no positive-labeled
nuclei in either A or B.
|
|
|
Table 2 shows the grading
of the specimens (performed by the masked observer [I.G.]) for the various
histological features. Posterior epithelial changes, inflammation, vascular
alteration, and stromal degeneration were not recorded in any of the 17 iridectomy
specimens in the series. Clump cells were just as common in the nonlatanoprost-treated
irides as they were in latanoprost-treated specimens; the 2 specimens with
the highest scores (specimens A7 and A9) were both untreated. Free melanin
granules were found sparingly in the stroma of most irides. Melanocytes with
nuclear features that were atypical (nuclear chromatin margination, prominent
nucleoli, and invagination of the nucleus) were rare, and when identified
they usually did not have all the atypical characteristics; these cells were
called borderline atypical. Melanocytes with atypical or borderline atypical
nuclei were present in 6 of the 9 untreated specimens compared with 5 of the
8 latanoprost-treated irides. The incidence would seem to be the same, but
the only 2 specimens that were graded + were latanoprost treated. Anterior
border thickness was above marginal in 5 of the 9 untreated specimens and
in 5 of the 8 latanoprost-treated specimens. Only one iris had a moderate
(grade ++) border; it was the one mentioned earlier that had an ephelis (Figure 2) and was identified as being in
the nonlatanoprost group (specimen A16) (Table 2).
|
|
|
|
Table 2. Masked Gradings for the Various Features in the Iridectomy
Specimens That Were Subjected to Particular Evaluation*
|
|
|
IMMUNOHISTOCHEMISTRY OF CONTROLS AND IRIDECTOMY SPECIMENS
Human tonsil as a positive control showed intense staining for PCNA
and Ki-67. Adult human eye showed labeling for both antigens in some conjunctival
epithelial nuclei and in some nuclei at the lens bow but nowhere else in the
eye. In the developing eye, nuclei that labeled positively with Ki-67 and
PCNA were identified in the immature stromal tissue. At 6 months following
birth, and in all the control adult eye specimens, no labeling was ever found.
In contrast, despite extensive examination of different levels in each specimen
with the 2 antibodies, neither the latanoprost-treated nor the untreated irides
of the test studies exhibited any positive staining (Figure 2B, Figure 3B,
and Figure 4). In these staining
runs, human tonsil and human fetal iris always exhibited numerous positively
staining nuclei.
COMMENT
Latanoprost is one of the latest additions to the treatment of glaucoma.
Since its introduction, it has received considerable attention and its use
is widespread; it is an attractive antiglaucoma medication principally because
of its relatively powerful IOP-lowering action, its ease of use (once-daily
treatment), and the apparent absence of any systemic adverse effects.12 There are, however, local adverse effects, of which
darkening of the iris is the most obvious and spectacular.1-4
The present study sought to (a) tackle the question
of whether topical application of latanoprost had a deleterious effect on
iris tissue and (b) gain some further insight into
the mechanism by which prostaglandins bring about darkening of eye color.
In this respect, we have valuable initial information on the former but provide
little to resolve the problem associated with the latter.
At the level of light microscopy, our findings show that a short period
of 3 months of exposure to latanoprost did not produce either observable degeneration
in the structure of the iris or inflammation. Melanin granules are thought
to be cytotoxic,13 and ocular melanin, including
that from the iris,7 is known to be uveitogenic.
As a consequence, some inflammatory pathological features might have been
expected if there had been an accumulation of melanin granules within the
iris stromal melanocytes leading to rupture and scattering of the granules
in the tissues, but there were no indications that melanin release was happening
in latanoprost-treated irides. Only one of the latanoprost-treated eyes actually
darkened in color. The morphological analysis did not show any remarkable
change in melanocyte pigmentation in the iris specimen from this patient nor
did it pick up any extensive release of free melanin into the stroma. To some
extent, it would have been surprising if such findings had been evident. Monkeys
that underwent iris color change after being treated with latanoprost for
1 year did show some accumulation of mature melanin granules within stromal
melanocytes, but this could be detected only after careful and exacting image
analysis.6 In addition, comparison of melanocytes
from the extremes of blue to brown in normal irides showed that not much of
a difference in melanin content was needed in the melanocytes to account for
the color difference.14-15
The clump cell frequency was assessed in this study because this local
macrophagic cell16 is associated with melanin
granule incorporation. If there was or had been free melanin in the stroma,
the clump cells might have been expected to be abundant. As it turned out,
there was no difference between latanoprost-treated and untreated eyes, and
the 2 specimens with the highest incidence of these cells turned out to be
from untreated patients.
The most important observation from the present study was the finding
that neither the latanoprost-treated (including the specimen from a patient
with an eye color change) nor the untreated iridectomy specimens showed any
positive staining nuclei by immunohistochemistry (melanocytes or other cells
in the iris) using antibodies against the cell cycle antigens PCNA and Ki-67.
These methods may also pick up a few cells that have recently left the cycle.10 However, for the present study, the advantage gained
by high sensitivity with PCNA and Ki-67 offsets their poor specificity.
Our finding in the present study of an absence of cell division in latanaprost-treated
human irides is in keeping with the findings in latanoprost-treated monkeys,
which showed no increase in the numbers of iris stromal melanocytes after
1 year.6 Nuclear activation in the form of
nuclear atypia was noted in a previously described patient who had darkening
of the iris while enrolled in the latanoprost phase 3 trial. In addition,
the patient had a thickened anterior border layer, although this was in association
with a nevus.5 A pronounced anterior border
layer was associated also with the untreated patient in this series who had
an ephelis or nevus in the iridectomy specimen. No relation with latanoprost
treatment and thickness of the anterior border layer was found in our histological
investigation.
Atypia was noted in the specimen with the ephelis or nevus but was marginal.
After close histological examination of many brown irides, including those
that served as controls in this experiment, a few melanocytes could be recognized
that have features, like margination of nuclear chromatin and nuclear invagination,
in common with nuclear atypia. Some incidence of atypia was found in most
specimens in the present series, but usually only at the marginal level, and
it was just as common in untreated as it was in latanoprost-treated tissues.
In primary acquired melanosis, atypia is associated with marked labeling for
the PCNA and Ki-67 antigens.17 In these circumstances,
it seems to be an entirely different entity than what we are seeing in the
present study and what was seen in a previous case report.5
On the other hand, the only 2 specimens that recorded a grade of + were both
latanoprost treated, and one of these was from the patient whose eye color
darkened during the study. We are, however, dealing with small differences
in minimal numbers, so we are a long way from demonstrating that atypia increases
with latanoprost treatment; the finding may well be spurious, especially since
the ratio of positives to negatives for the presence of atypia in the treated
and untreated groups was much the same.
In conclusion, the present study shows that a 3-month treatment with
latanoprost does not produce morphological changes in the iris and does not
push iris cells into the cell cycle. The reservations should be emphasised
that only 1 of our series of latanoprost-treated patients underwent a clinically
noticeable eye color change and our morphological investigations were restricted
to a small segment of peripheral iris.
AUTHOR INFORMATION
Accepted for publication July 6, 2000.
This study was supported by Pharmacia & Upjohn, Stockholm, Sweden.
Corresponding author and reprints: N. Pfeiffer, MD, Department of
Ophthalmology, Mainz University, Langenbeckstrasse 1, 55131 Mainz, Germany
(e-mail: pfeiffer{at}augen.klinik.uni-mainz.de).
From the Department of Ophthalmology, Mainz University, Mainz, Germany
(Drs Pfeiffer, Hochgesand, and Winkgen-Böhres); and the Unit of Ophthalmology,
University of Liverpool, Liverpool, England (Drs Grierson and Appleton and
Ms Goldsmith).
REFERENCES
| |
1. Alm A, Stjernschantz J for the Scandanavian Latanoprost Study Group. Effects on intraocular pressure and side effects of 0.005% latanoprost
applied once daily, evening or morning. Ophthalmology. 1995;102:1743-1752.
ISI
| PUBMED
2. Watson P, Stjernschantz J for the Latanoprost Study Group. A six-month, randomized, double-masked study comparing latanoprost
with timolol in open-angle glaucoma and ocular hypertension. Ophthalmology. 1996;103:126-137.
ISI
| PUBMED
3. Camras CB for the United States Latanoprost Study Group. Comparison of latanoprost and timolol in patients with ocular hypertension
and glaucoma: a six-month, masked, multicentre trial in the United States. Ophthalmology. 1996;103:138-147.
ISI
| PUBMED
4. Wistrand PJ, Stjernschantz J, Olsson K. The incidence and time-course of latanoprost-induced iridial pigmentation
as a function of eye color. Surv Ophthalmol. 1997;41(suppl 2):S129-S138.
5. Grierson I, Lee WR, Albert DM. The fine structure of an iridectomy specimen from a patient with latanoprost-induced
eye color change. Arch Ophthalmol. 1999;117:394-396.
FREE FULL TEXT
6. Selen G, Stjernschantz J, Resul B. Prostaglandin-induced irideal pigmentation in primates. Surv Ophthalmol. 1997;41(suppl 2):S125-S128.
7. Broekhuyse RM, Kuhlmann ED. Experimental autoimmune anterior uveitis: the preparation of uveitogenic
ocular melanin. Invest Ophthalmol Vis Sci. 1993;34:698-700.
FREE FULL TEXT
8. Yanoff M, Fine BS. Ocular Pathology: A Text and Atlas. New York, NY: Harper & Row; 1988.
9. Pfeiffer N, Grehn F. Improved suture for fornix-based conjunctival flap in filtering surgery. Int Ophthalmol. 1992;16:391-396.
FULL TEXT
|
ISI
| PUBMED
10. van Dierendonck JH, Wijsman JH, Keijzer R, van de Velde CJH, Cornelisse CJ. Cell-cycle-related staining of antiproliferating cell nuclear antigen
monoclonal antibodies. Am J Pathol. 1991;138:1165-1172.
ABSTRACT
11. Yu CCW, Filipe MI. Update on proliferation-associated antibodies applicable to formalin-fixed
paraffin-embedded tissue and their clinical applications. Histochem J. 1993;25:843-853.
ISI
| PUBMED
12. Camras CB, Alm A, Watson PG, Stjernschantz J for the Latanoprost Study Groups. Latanoprost, a prostaglandin analog, for glaucoma therapy: efficacy
and safety after a year of treatment in 198 patients. Ophthalmology. 1996;103:1916-1924.
ISI
| PUBMED
13. McGinness JE, Kono R, Moorhead WD. The melanosome: cytoprotective or cytotoxic? Pigment Cell. 1979;4:270-276.
14. Imesch PD, Bindly CD, Khademian Z, et al. Melanocytes and iris color: electron microscopic findings. Arch Ophthalmol. 1996;114:443-447.
FREE FULL TEXT
15. Imesch PD, Wallow IHL, Albert DM. The color of the human eye: a review of morphologic correlates and
of some conditions that affect iridial pigmentation. Surv Ophthalmol. 1997;41(suppl 2):S117-S123.
16. Freddo TF. Ultrastructure of the iris. Microsc Res Tech. 1996;33:369-389.
FULL TEXT
|
ISI
| PUBMED
17. Chowers I, Livini N, Solomon A, et al. MIB-1 and PC-10 immunostaining for the assessment of proliferative
activity in primary aquired melanosis with and without atypia. Br J Ophthalmol. 1998;82:1316-1319.
FREE FULL TEXT
 CiteULike  Connotea  Del.icio.us  Digg  Reddit  Technorati  Twitter
What's this?
RELATED ARTICLE
Archives of Ophthalmology Reader's Choice: Continuing Medical Education
Arch Ophthalmol. 2001;119(2):315-316.
FULL TEXT
THIS ARTICLE HAS BEEN CITED BY OTHER ARTICLES
|
A Study of Histopathological Features of Latanoprost-Treated Irides With or Without Darkening Compared With Non-Latanoprost-Treated Irides
Albert et al.
Arch Ophthalmol 2008;126:626-631.
ABSTRACT
| FULL TEXT
Bimatoprost-Induced Periocular Skin Hyperpigmentation: Histopathological Study
Kapur et al.
Arch Ophthalmol 2005;123:1541-1546.
ABSTRACT
| FULL TEXT
A Study of Iridectomy Histopathologic Features of Latanoprost- and Non-Latanoprost-Treated Patients
Albert et al.
Arch Ophthalmol 2004;122:1680-1685.
ABSTRACT
| FULL TEXT
A 5-Year, Multicenter, Open-Label, Safety Study of Adjunctive Latanoprost Therapy for Glaucoma
Alm et al.
Arch Ophthalmol 2004;122:957-965.
ABSTRACT
| FULL TEXT
New topical drugs for open-angle glaucoma
DTB 2003;41:12-14.
ABSTRACT
| FULL TEXT
Fine Structural Evaluation of the Iris After Unilateral Treatment With Latanoprost in Patients Undergoing Bilateral Trabeculectomy (The Mainz II Study)
Pfeiffer et al.
Arch Ophthalmol 2003;121:23-31.
ABSTRACT
| FULL TEXT
|
