- Email: [email protected]
Iris melanocytoma: Clinical features and natural course in 47 cases
Iris Melanocytoma: Clinical Features and Natural Course in 47 Cases HAKAN DEMIRCI, MD, ARMAN MASHAYEKHI, MD, CAROL L. SHIELDS, MD, RALPH C. EAGLE, JR., MD, AND JERRY A. SHIELDS, MD
● PURPOSE: To describe the clinical features, natural course, management and histopathologic features of iris melanocytoma. ● DESIGN: Single-center retrospective case series. ● METHODS: PATIENT POPULATION: Forty-seven consecutive patients (47 eyes) with iris melanocytoma. INTERVENTION PROCEDURE: Data regarding patient and tumor features were analyzed for their impact on the main outcome measures using univariate and multivariate regression models. Kaplan-Meier estimates were used to analyze the main outcomes as a function of time. MAIN OUTCOME MEASURES: Increased intraocular pressure (IOP), tumor seeding, and tumor growth. ● RESULTS: Associated findings at initial presentation included iris stromal seeds in 20 patients (43%), and anterior chamber angle seeds in 12 (26%). Intrinsic vascularization and sector cataract were not seen in any eyes. The management at presentation included observation in 39 patients (83%), tumor removal by sector iridectomy/iridocyclectomy in 7 (15%), and enucleation for blind painful eye with secondary increased IOP in 1 (2%). The diagnosis was confirmed by histopathologic examination in 11 patients (23%). The mean follow-up was 58 months. Using Kaplan-Meier estimates, clinical evidence of growth was observed in 23% at 5 years, 48% at 10 years, and 74% at 15 years. New tumor seeds developed in 34% at 5 years, 63% at 10 years, and 75% at 15 years. Increased IOP was observed in 11% at 5 years, 11% at 10 years, and 55% at 15 years. Accepted for publication Oct 4, 2004. From the Oncology Service (H.D., C.L.S., A.M., J.A.S.) and Department of Pathology (R.C.E., Jr.), Wills Eye Hospital, Thomas Jefferson University, Philadelphia, Pennsylvania. Presented in part at the annual meeting of The Association for Research in Vision and Ophthalmology, Ft. Lauderdale, Florida, May 2003 (H.D.) and at the annual meeting of American Academy of Ophthalmology, New Orleans, Louisiana, October, 2004 (H.D.). Support provided by the International Award of Merit in Retina Research, Houston, Texas (J.A.S.), Rosenthal Award of the Macula Society (C.L.S.), Macula Foundation, New York, New York (C.L.S.), the Noel T. and Sara L. Simmonds Endowment for Ophthalmic Pathology, Wills Eye Hospital (R.C.E., Jr.) and the Eye Tumor Research Foundation, Philadelphia, Pennsylvania (C.L.S., J.A.S.). Inquiries to Carol L. Shields, MD Oncology Service, Wills Eye Hospital, 840 Walnut Street, Philadelphia, PA 19107; fax: 215-928-1140; e-mail: [email protected].
468
©
2005 BY
● CONCLUSIONS:
Iris melanocytoma represented only 3% of all iris nevi. Related iris stromal and anterior chamber angle seeds were common, and secondary glaucoma occurred in 11% at 5 years. Growth was observed in 23% at 5 years but no malignant transformation was found. (Am J Ophthalmol 2005;139:468 – 475. © 2005 by Elsevier Inc. All rights reserved.)
M
ELANOCYTOMA IS A VARIANT OF MELANOCYTIC
nevus with distinctive clinical and pathologic features.1 Clinically, it appears as a deeply pigmented tumor that is usually located on or adjacent to the optic disk.2,3 In rare instances, it can arise in iris, ciliary body, choroid, or conjunctiva.4 In 1965, Zimmerman suggested the term “melanocytoma” to describe this tumor and reported two patients with iris melanocytoma, a 60-year-old white man with a deeply pigmented iris mass and a 34-year-old white man with recurrent iritis and pigmented iris mass. Iris melanoma was suspected in both cases, but both lesions proved to be iris melanocytoma after enucleation.1 Since then, a few other reports of iris melanocytoma have been published.5–14 The natural course of iris melanocytoma is variable. Iris melanocytoma can undergo spontaneous necrosis with resultant pigment dispersion, causing secondary glaucoma and heterochromia.5,6,8 –13 In some cases, iris or ciliary body melanocytoma can show progressive growth or even extrascleral involvement making clinical differentiation from malignant melanoma difficult.7,15,16 In rare cases, it can transform into iris melanoma.8,14 Failure to recognize iris melanocytoma could lead to misdirected therapy, including enucleation. To further expand our understanding of this unique tumor, we report our experience with 47 patients with iris melanocytoma and describe its clinical and histopathologic features, management, and natural course.
PATIENTS AND METHODS THE MEDICAL RECORDS OF ALL PATIENTS DIAGNOSED WITH
iris nevus at the Oncology Service at Wills Eye Hospital between January 1974 and February 2003 were reviewed
ELSEVIER INC. ALL
RIGHTS RESERVED.
0002-9394/05/$30.00 doi:10.1016/j.ajo.2004.10.008
TABLE 1. Clinical Features at Presentation of 47 Consecutive Patients with Iris Melanocytoma
TABLE 2. Related Anterior Segment Findings at Presentation of 47 Patients with Iris Melanocytoma
No of patients (%)
Color Darkly pigmented Light brown Configuration Nodular Diffuse Involved quadrant Inferior Lateral Nasal Superior Anteroposterior location Iris root Midzone Pupil margin Diffuse
44 (94%) 3 (6%) 40 (85%) 7 (15%) 21 (45%) 18 (38%) 4 (9%) 4 (9%) 32 (68%) 7 (15%) 5 (11%) 3 (6%)
retrospectively, and the records of those patients who were specifically classified as having iris melanocytoma were retrieved for further analysis. Institutional Review Board approval was obtained for this retrospective study. The clinical features of iris melanocytoma have been described in the literature2 and include a darkly pigmented, dome shaped or sessile iris stromal tumor with a granular surface and no intrinsic vasculature. Our criteria for entry into this study included tumors with the above features confined to the iris or with extension of anterior chamber angle. Patients with ciliary body or choroidal melanocytoma were excluded. Throughout the manuscript, the terminology “anterior chamber seeds” and “iris stromal seeds” are used to imply the cells that have been shed from necrotic tumor or pigment containing macrophages. The patient data collected at initial evaluation at the Oncology Service included age (years), race (Caucasian, African-American, Asian, Hispanic), gender (male, female), and systemic disease. The ocular data included affected eye (right or left), laterality (unilateral or bilateral), visual symptoms, best-corrected Snellen visual acuity, intraocular pressure (mm Hg), and ocular melanocytosis (present or absent). All patients underwent detailed slit-lamp biomicroscopy, gonioscopy, fundus examination, and transillumination. Ultrasound biomicroscopy was performed in 11 patients. The tumor data included tumor color, configuration (nodular, plateau), size (mm), thickness (mm), quadrantic location of the tumor epicenter (superior, nasal, inferior, temporal), and anteroposterior location of the tumor epicenter (pupillary margin, midzone, root, or diffuse). The presence of secondary glaucoma, heterochromia, ectropion iridis, satellite tumors, iris stromal seeds, anterior chamber angle seeds, intrinsic VOL. 139, NO. 3
Clinical finding
No of patients (%)
Heterochromia Sentinel vessel Ectropion irides Satellite tumors Number of satellite tumors Mean Median Range Iris stromal seeds Extent of iris stromal seeds (clock hours) Mean Median Range AC angle seeds Extent of AC angle seeds (clock hours) Mean Median Range AC inflammation Hyphema Pigmented keratic precipitate Intrinsic vascularization Sector cataract
6 (13%) 7 (15%) 3 (6%) 12 (26%) 4 3 1–15 20 (43%) 5 3 1–12 12 (26%) 8 7 3–12 2 (4%) 0 (0%) 1 (2%) 0 (0%) 0 (0%)
AC: Anterior chamber
vessels, anterior chamber reaction, hyphema, pigmented keratic precipitates, and episcleral sentinel vessels were recorded. The presence or absence of intrinsic vessels is evaluated by slit lamp examination and when available, iris fluorescein angiography. Tumor management after referral (observation, fine-needle biopsy, sector iridectomy, or enucleation) was listed. Follow-up data were collected regarding change in visual acuity, development of increased intraocular pressure (IOP), development of anterior chamber angle seeds, development of iris stromal seeds, and tumor growth. The clinical data were analyzed with regard to three main outcome measures: development of tumor seeding, development of increased IOP, and tumor growth. The effect of each individual demographic datum and clinical variable on the final visual outcome was analyzed by a series of univariate logistic regression analyses. All variables were analyzed as discrete categories except patient age, intraocular pressure, tumor base, and tumor thickness, which were analyzed as continuous variables and subsequently grouped into discrete categories to derive cutoff values. The correlation among the predictor variables was determined by Pearson correlations. The variables that were significant on a univariate level (P ⱕ .05) were entered into multivariate logistic regression analysis using forward stepwise (likehood ratio)
IRIS MELANOCYTOMA
469
FIGURE 1. A 53-year-old male with iris melanocytoma at the left eye. (Top) At presentation. (Middle) Histopathologic examination displaying copious pigment obscuring nuclear details on routine hematoxylin-eosin sections (hematoxylin-eosin ⴛ 50). (Bottom) Bleached section disclosing bland nuclei and benign nuclear/cytoplasmic ratio (hematoxylin-eosin ⴛ 100).
FIGURE 2. A 15-year-old female who presented with iris melanocytoma at the right eye. (Top) At presentation. (Middle) Histopathologic examination showing densely pigmented tumor composed of plump, polyhedral nevus cells with large quantities of melanin pigment that tends to obscure nuclear details (Hematoxylin-eosin ⴛ 50). (Bottom) Bleached section disclosing bland nuclei and low nuclear/cytoplasmic ratio. (Hematoxylin-eosin ⴛ 100)
method. For variables that showed a high degree of correlation, only one variable from the set of associated variables was entered at a time in subsequent multivariate models. A final multivariable model fitted variables identified as significant predictors (P ⬍ .05)
in the stepwise model and variables considered clinically important for each main outcome. Kaplan-Meier survival estimates were used to analyze main outcome measures as a function of time from the initial examination at the Oncology Service.
470
AMERICAN JOURNAL
OF
OPHTHALMOLOGY
MARCH 2005
TABLE 3. Kaplan-Meier Estimates of Tumor Growth, Elevated IOP and Development of New Tumor Seeds in 47 Patients
Tumor growth Increased IOP Development of new tumor seeds
At 1 year
At 3 years
At 5 years
At 10 years
At 15 years
(0/19) 0% (0/19) 0% (1/19) 5%
(4/12) 23% (2/13) 11% (5/11) 27%
(4/10) 23% (2/10) 11% (6/9) 34%
(7/3) 48% (2/5) 11% (10/3) 63%
(8/1) 74% (3/1) 55% (11/1) 75%
IOP: Intraocular pressure
RESULTS OF APPROXIMATELY 1400 PATIENTS WITH IRIS NEVUS MAN-
aged at the Oncology Service over 29 years, 47 patients (3%) were classified as having iris melanocytoma. The mean age at presentation was 37 years (median 40 years, range 3 to 67 years). Of 47 patients, 41 (87%) were Caucasian, 4 (9%) African-American, 1 (2%) Asian, and 1 (2%) was Hispanic. There were 30 females (64%) and 17 males (36%). One patient (2%) had a history of skin melanoma and one patient (2%) had ocular melanocytosis. The initial patient symptoms and signs were the presence of a dark spot on the iris in 7 patients (15%), blurred vision in 5 patients (11%), and conjunctival hyperemia in 2 patients (4%). Thirty-three patients (70%) were asymptomatic. The right eye was affected in 18 patients (38%) and the left eye in 29 (64%). There was no patient with bilateral involvement. The best corrected visual acuity was 20/20 to 20/40 in 45 patients (96%), 20/50 to 20/400 in 1 patient (2%), and finger counting in 1 patient (2%). At presentation, the mean intraocular pressure was 17 mm Hg (median 16 mm Hg, range 9 to 45 mm Hg). The intraocular pressure was higher than 21 mm Hg in 2 patients (4%). The clinical features and associated clinical findings in 47 patients with iris melanocytoma at presentation are shown in Tables 1 and 2. The mean diameter of iris melanocytoma was 3 mm (median 3 mm, range 1 mm to 8 mm) and the mean thickness was 2 mm (median 2 mm, range 1 mm to 4 mm). Ultrasound biomicroscopy was performed on 11 patients (23%) and showed an echogenic nodular thickening of iris in all 11 cases. Management included observation in 39 patients (83%), tumor removal by sector iridectomy in 7 patients (15%), and enucleation in 1 patient (2%). The reasons for sector iridectomy were failure of clinical and noninvasive diagnostic techniques to differentiate from melanoma in 3 patients (6%), tumor growth in 3 patients (6%), and the presence of secondary glaucoma in 1 patient (2%). The reason for enucleation was blind painful eye from secondary glaucoma in 1 patient (2%). Three patients (6%) had fine needle aspiration biopsy for diagnostic confirmation and to exclude iris melanoma. Histopathologic examination of resected lesions showed characteristic features including an intensely pigmented VOL. 139, NO. 3
tumor composed of plump, polyhedral nevus cells with copious quantities of maximally pigmented cytoplasm that obscure nuclear details. Bleached sections disclose tumor cells with a low nuclear cytoplasmic ratio and bland nuclei with inconspicuous nucleoli (Figure 1, top, middle, and bottom; Figure 2, top, middle, and bottom). The patients were followed for a mean of 58 months (median 20 months, range 1 to 364 months). Using Kaplan-Meier estimates, the development of increased IOP was observed in 0% at 1 year, 11% at 5 years, 11% at 10 years, and 55% at 15 years (Table 3). All patients with increased IOP developed new tumor seeds and two showed tumor growth. New iris stromal or anterior chamber angle seeds developed in 5% at 1 year, 34% at 5 years, 63% at 10 years, and 75% at 15 years (Table 3) (Figure 3, top, middle, and bottom). Using Kaplan-Meier estimates, tumor growth was observed in 0% at 1 year, 23% at 5 years, 48% at 10 years, and 74% at 15 years (Table 3) (Figure 4, left and right). Overall, there was a mean growth of 30% in basal dimension (median 28%, range 20% to 50%) and a mean growth of 34% in thickness (median 33%, range 20% to 63%). The lesions that showed growth were managed by careful observation in 4 patients (50%) and sector iridectomy in 4 (50%). Tumor growth was documented in 50% of the eyes that developed new tumor seeds. Univariate predictive factors for development of increased IOP, development of tumor seeding and tumor growth are shown in Table 4. Multivariate analysis showed no significant factor predictive of any of the outcome measures.
DISCUSSION IRIS MELANOCYTOMA IS A RARE VARIANT OF IRIS NEVUS
with fairly typical clinical and histopathologic features. In a clinical study by Shields and associates, of 200 patients referred for suspected iris melanoma, 158 patients (76%) were found to have lesions other than melanoma and these are termed as pseudomelanomas.17 The most common pseudomelanomas included iris cyst (38%) and iris nevus (31%). In that series, only one pseudomelanoma (0.5%) was an iris melanocytoma.17 In a pathology study of 189 specimens clinically suspected
IRIS MELANOCYTOMA
471
series, the mean patient age was 37 years (median 40 years, range 3 to 67 years) and 4 patients (9%) were under the age of 10 years. Iris melanocytoma has subtle clinical features that distinguish it from other iris nevi or melanoma. Iris melanocytoma typically presents as a darkly-pigmented iris nodule with mossy, irregular surface. Similar to other melanocytic lesions of the iris, it is usually located in the inferior quadrant of the iris. However, the clinical findings associated with other melanocytic lesions of the iris, such as ectropion iridis, intrinsic vessels, or sector cataract, are less commonly seen with iris melanocytoma. In a review of 175 patients with iris nevus, Territo and associates reported that iris nevi were associated with ectropion iridis in 24% of patients, intrinsic vessels in 20%, and sector cataract in 3%.19 Similarly, in a review of 169 patients with iris melanoma, Shields and coworkers reported an irregular tumor surface in 93% of patients, ectropion iridis in 44%, intrinsic vascularization in 43%, and sector cataract in 14%.20 In the present study, iris melanocytoma was associated with ectropion iridis in only 6% of patients and intrinsic vascularization and sector cataract were not seen in any case. In this study, 11(23%) of 47 patients had histopathologic confirmation of iris melanocytoma. The other patients were diagnosed clinically based on their distinctive features. In the present study, using Kaplan-Meier estimates, increased IOP developed in 11% of patients at 5 years, 11% at 10 years, and 55% at 15 years. Electron microscopic studies of eyes with secondary glaucoma attributable to iris melanoma (melanomalytic glaucoma) showed that pigment phagocytosis by trabecular endothelial cells contributed to the mechanical obstruction of trabecular meshwork and subsequent increased IOP.21,22 The majority of intratrabecular pigment in melanocytomalytic glaucoma is contained within macrophages instead of trabecular endothelial cells.6 Tumor necrosis is an important factor in the pathogenesis of melanocytomalytic glaucoma. Melanocytoma can undergo spontaneous necrosis and such tumors can contain numerous macrophages. Two of the three patients with melanocytomalytic glaucoma reported by Fineman and coworkers developed crater-like cavities in the surface of their tumors that were clinical evidence of tumor necrosis.6 Of the 13 published cases of iris melanocytoma, 12 cases (92%) were found to have anterior chamber angle seeds and only 9 (69%) developed melanocytomalytic glaucoma (Table 5).1,5–14 Similarly, we did not find that anterior chamber angle seeds were a risk factor for development of increased IOP. In our study, pigmented keratic precipitates and anterior chamber inflammation were factors predictive of the development of increased IOP in the univariate analysis, emphasizing the role of macrophages in the anterior chamber angle. Multivariate analysis showed no factor predictive for the development of increased IOP.
FIGURE 3. A 23-year-old male who presented with iris melanocytoma at the left eye. (Top) At presentation. (Middle) Gonioscopy showing dense anterior chamber seeds at inferior anterior chamber angle. (Bottom) Histopathologic examination showing extensively necrotic tumor involving peripheral iris.
to be iris or ciliary body melanoma, Jakobiec and Silbert found 10 (5%) to be melanocytoma.18 Thus iris melanocytoma can closely simulate iris melanoma. Iris melanocytoma generally is diagnosed in adults and has been reported in children.1,5–14 Of the 13 welldocumented cases of iris melanocytoma in the literature, the mean patient age was 32 years and only 4 (9%) were under the age of 10 years (Table 5).1,5–14 Similarly, in our 472
AMERICAN JOURNAL
OF
OPHTHALMOLOGY
MARCH 2005
FIGURE 4. A 38-year-old male patient with iris melanocytoma at the right eye that showed growth. (Left) In 1972. (Right) In 2003, the melanocytoma had doubled in size compared with 1972.
TABLE 4. Univariate and Multivariate Analysis of Clinical Factors Correlated with Increased IOP, Development of New Iris Stromal and Anterior Chamber Angle Seeds and Growth in 47 Patients with Iris Melanocytoma*
Factor
Increased IOP Univariate analysis Race (Non-Caucasian** vs Caucasian) Pigmented keratic precipitates (Present** vs Absent) Anterior chamber inflammation (Present** vs Absent) New iris stromal and ACA seeds Univariate analysis Race (Non-Caucasian** vs Caucasian) Pigment keratic precipitates (Present** vs Absent) Growth Univariate analysis Anterior chamber angle seeds (Present** vs Absent)
P
Relative Risk (95% Confidence Interval)
0.008
27.5 (2.3–323.4)
0.04
11.5 (1.1–129.1)
0.04
18.4 (1.1–295.4)
0.04
18.5 (1.2–295.7)
0.04
18.5 (1.2–295.7)
0.04
5.5 (1.1–26.3)
*Multivariate analysis was performed for each main outcome, but there were no factor found to be a significant predictor for any of the outcome measures. **Reference variable
VOL. 139, NO. 3
Iris melanocytoma can undergo spontaneous necrosis with resultant pigment dispersion or shedding of melanocytoma cells and/or melanophages into the anterior chamber, causing secondary glaucoma or heterochromia. In the present study, iris melanocytoma was associated with anterior chamber angle seeds in 40% of patients and iris stromal seeds in 53% of patients. In comparison, Territo and associates found iris stromal seeds in only 9% of patients with standard iris nevi.19 In our study, new anterior chamber angle and/or iris stromal seeds developed in 34% of patients at 5 years, 63% at 10 years, and 75% at 15 years. Increased seeding could be related to partial tumor necrosis. The cause of spontaneous necrosis is unknown, although Teichmann and Karcioglu have speculated that it might be related to lack of circulation in a highly metabolic tumor such as melanocytoma.9 We found that pigmented keratic precipitates were a risk factor for development of anterior chamber and/or iris seeds in univariate analysis. Multivariate analysis showed no factor predictive for development of new tumor seeds. Iris melanocytoma usually has been considered to be a relatively stationary lesion. Of 13 published cases, only 3 cases (23%) showed clinical evidence of growth (Table 5).5–14 In the present study, growth occurred in 23% of the patients at 5 years and 48% at 10 years. The only factor predictive of growth was the presence of anterior chamber angle seeds in univariate analysis. Multivariate analysis showed no significant risk factor for tumor growth. In a review of 175 patients with iris nevus, Territo and associates found that only 5% of lesions showed growth at a mean follow-up of 5 years and the only predictive risk factor for growth was presence of iris stromal and anterior chamber angle seeds.19 Growth of iris melanocytoma does not necessarily imply malignant transfor-
IRIS MELANOCYTOMA
473
TABLE 5. Clinical Findings in 13 Well Documented Cases of Iris Melanocytoma in the Literature*
Eye
Symptoms or signs
Color
Location
ACA seeds
Author (Year)
Age/Race/Sex
Zimmerman (1965) Zimmerman (1965) Thomas (1969)
62 WM
OS Asymptomatic Black Inferonasal NA
34 WM
OS Recc. iritis
20 AAF
Shields (1977) Nakazawa (1984) Cialdini (1989)
Iris stromal seeds
Growth
Management
NA
NA
Absent
Enuc
Black Inferotemp Present NA
Absent
Absent Absent
Enuc
Black Superior
Present NA
Absent
Absent Present
Enuc
23 WM
OS Spot on iris, pain, redness OS Spot on iris
Black Inferior
Present Present
Present
Absent Absent
46 AF
OD Discomfort
Black Diffuse
Present Present
Present
Absent Absent
34 WF
OS Spot on iris, Black Temporal pain, redness OS Asymptomatic Black Inferior
Present Absent
Present
Present Present
Sector iridectomy Trabeculectomy and Inc Bx Sector iridectomy
Present NA
Present
Absent Absent
Fountain (1992) Teichmann (1995)
18 AAF
Green (1996) Fineman (1998) Fineman (1998) Kiratli (2001) Shields (2002)
28 ArF
NA
Glaucoma
Association with melanoma
Sector iridectomy Sector iridectomy
Brown Nasal
Present Absent
Present
Absent Absent
6F
OS Spot on iris, pain, redness OS Spot on iris
Black Superior
Present NA
Present
Absent Absent
Enuc
61 WM
OS Spot on iris
Black Inferonasal Present Present
Present
Absent Absent
Enuc
48 WF
OD Spot on iris
Brown Inferior
Present Present
Present
Present Absent
27 WF
OS Spot on iris
Black Inferior
Present Absent
Present
Absent Absent
9 WM
OS Spot on iris
Brown Inferotemp Present Absent
Absent
Present Absent
Sector iridectomy Sector iridectomy Sector iridectomy
ACA ⫽ Anterior chamber angle; W ⫽ White, AA ⫽ African-American; A ⫽ Asian; Ar ⫽ Arabic; M ⫽ Male; F ⫽ Female; OD ⫽ Right eye; OS ⫽ Left eye; Enuc ⫽ Enucleation; Recc ⫽ Recurrent; Inferotemp ⫽ Inferotemporal; Inc Bx ⫽ Incisional biopsy *. Well-documented cases published in English literature.
mation. Shields and associates reported a 9-year-old male who showed progressive growth of a pigmented iris lesion over 6 years and on histopathologic examination it proved to be an iris melanocytoma.7 In the present study, we did not observe malignant transformation on histopathologic examination of the lesions with documented growth that were resected. The management of suspected iris melanocytoma is usually cautious observation. When noninvasive diagnostic techniques fail to establish the diagnosis or tumor growth is documented, fine-needle aspiration biopsy or local resection using iridectomy, iridogoniectomy, or iridogoniocyclectomy is performed.23 Local resection can also be used to treat secondary glaucoma caused by necrotic melanocytoma.5 This surgical technique not only provides tissue for histopathologic diagnosis but may also assist in resolution of the secondary glaucoma.5,11 Secondary glaucoma can be treated medically 474
AMERICAN JOURNAL
by sector iridectomy or by glaucoma-filtering surgery.12 However, diagnostic confirmation of the lesion by histopathology or cytopathology is warranted before filtering surgery. Enucleation is usually reserved for blind painful eyes. In conclusion, iris melanocytoma is an uncommon variant of iris nevus with distinctive clinical and histopathologic features. It usually presents as a darkly-pigmented nodular mass with granular surface and no intrinsic vessels. It can undergo spontaneous necrosis resulting in iris stromal and anterior chamber angle seeding and dispersion of melanophages leading to secondary elevated IOP. Frequent observation of the lesion in every six months is usually warranted. Local resection is reserved for treatment of increased IOP or for diagnostic purposes when malignancy cannot be ruled out. Despite the benign nature of iris melanocytoma, clinical evidence of growth was observed in 23% of patients at 5 years and 48% at 10 years. OF
OPHTHALMOLOGY
MARCH 2005
REFERENCES 1. Zimmerman LE. Melanocytes, melanocytic nevi and melanocytomas. The Jonas Friedenwald Memorial Lecture. Invest Ophthalmol 1965;4:11– 41. 2. Shields JA, Shields CL. Posterior uveal melanocytoma. In: Shields JA, Shields CL, editors. Atlas of intraocular tumors. Philadelphia: Lippincott, Williams and Wilkins, 1999:61–72. 3. Shields JA, Demirci H, Mashayekhi A, Shields CL. Melanocytoma of optic disc in 115 cases: The 2004 Samuel Jackson Memorial Lecture, part I. Ophthalmology 2004;111:1739 – 1746. 4. Shields JA, Augsburger JJ, Bernardino V and Eller AW. Melanocytoma of the ciliary body and iris. Am J Ophthalmol 1980;89:632– 635. 5. Shields JA, Annesley WH Jr, Spaeth GL. Necrotic melanocytoma of iris with secondary glaucoma. Am J Ophthalmol 1977;84:826 – 829. 6. Fineman M, Eagle RC Jr, Shields JA, Shields CL, De Potter P. Melanocytomalytic glaucoma in eyes with necrotic iris melanocytoma. Ophthalmology 1998;105:492– 496. 7. Shields JA, Eagle RC Jr, Shields CL, Nelson LB. Progressive growth of an iris melanocytoma in a child. Am J Ophthalmol 2002;133:287–289. 8. Cialdini AP, Sahel JA, Jalkh AE, Weiter JJ, Zakka K, Albert DM. Malignant transformation of an iris melanocytoma. Graefe’s Arch Clin Exp Ophthalmol 1989;227:348 –354. 9. Teichmann KD, Karcioglu ZA. Melanocytoma of the iris with rapidly developing secondary glaucoma. Surv Ophthalmol 1995;40:136 –144. 10. Nakazawa M, Tamai M. Iris melanocytoma with secondary glaucoma. Am J Ophthalmol 1984;97:797–799. 11. Kiratli H, Bilgic S, Gedik S. Late normalization of melanocytomalytic intraocular pressure elevation following excision of iris melanpcytoma. Graefe’s Arch Clin Exp Ophthalmol 2001;239:712–715.
VOL. 139, NO. 3
12. Fountain TR, Goldberg MF, Green WR. Glaucoma and a melanocytic iris lesion in an-18-year-old. In: Schachat AP, editor. Current practice in ophthalmology. St. Louis: Mosby, 1992:371–380. 13. Green WR. Uveal tract. In: Spencer WH, editor. Ophthalmic pathology. An atlas and textbook. Philadelphia: WB Saunders, 1996:1608 –1609. 14. Thomas CI, Purnell EW. Ocular melanocytoma. Am J Ophthalmol 1969;67:79 – 86. 15. Scheie HG, Yanoff M. Pseudomelanoma of the ciliary body. Arch Ophthalmol 1967;77:81– 83. 16. Rummelt V, Naumann GOH, Foldberg R, Weingeist TA. Surgical management of melanocytoma of the ciliary body with extrascleral extension. Am J Ophthalmol 1994;117:169 –176. 17. Shields JA, Sanborn GE, Augsburger JJ. The differential diagnosis of malignant melanoma of the iris. A clinical study of 200 patients. Ophthalmology 1983;90:716 –720. 18. Jakobiec FA, Silbert A. Are most iris “melanomas” really nevi? A clinicopathological study of 189 lesions. Arch Ophthalmol 1981;99:2117–2132. 19. Territo C, Shields CL, Shields JA, Augsburger JJ, Schroeder RP. Natural course of melaonocytic tumors of the iris. Ophthalmology 1988;95:1251–1255. 20. Shields CL, Shields JA, Materin M, Gershenbaum E, Singh AD, Smith A. Iris melanoma. Risk factors for metastasis in 169 consecutive patients. Ophthalmology 2000;108:172–178. 21. Van Buskirk EM, Leure-DuPree AE. Pathophysiology and electron microscopy of melanomalytic glaucoma. Am J Ophthalmol 1978;85:160 –166. 22. McMenamin PG, Lee WR. Ultrastructural pathology melanomalytic glaucoma. Br J Ophthalmol 1986;70:895–906. 23. Shields JA, Shields CL, Ehya H, Eagle RC, De Potter P. Fine-needle aspiration biopsy of suspected intraocular tumors. The 1992 Urwick lecture. Ophthalmology 1993;100: 1677–1684.
IRIS MELANOCYTOMA
475
● PURPOSE: To describe the clinical features, natural course, management and histopathologic features of iris melanocytoma. ● DESIGN: Single-center retrospective case series. ● METHODS: PATIENT POPULATION: Forty-seven consecutive patients (47 eyes) with iris melanocytoma. INTERVENTION PROCEDURE: Data regarding patient and tumor features were analyzed for their impact on the main outcome measures using univariate and multivariate regression models. Kaplan-Meier estimates were used to analyze the main outcomes as a function of time. MAIN OUTCOME MEASURES: Increased intraocular pressure (IOP), tumor seeding, and tumor growth. ● RESULTS: Associated findings at initial presentation included iris stromal seeds in 20 patients (43%), and anterior chamber angle seeds in 12 (26%). Intrinsic vascularization and sector cataract were not seen in any eyes. The management at presentation included observation in 39 patients (83%), tumor removal by sector iridectomy/iridocyclectomy in 7 (15%), and enucleation for blind painful eye with secondary increased IOP in 1 (2%). The diagnosis was confirmed by histopathologic examination in 11 patients (23%). The mean follow-up was 58 months. Using Kaplan-Meier estimates, clinical evidence of growth was observed in 23% at 5 years, 48% at 10 years, and 74% at 15 years. New tumor seeds developed in 34% at 5 years, 63% at 10 years, and 75% at 15 years. Increased IOP was observed in 11% at 5 years, 11% at 10 years, and 55% at 15 years. Accepted for publication Oct 4, 2004. From the Oncology Service (H.D., C.L.S., A.M., J.A.S.) and Department of Pathology (R.C.E., Jr.), Wills Eye Hospital, Thomas Jefferson University, Philadelphia, Pennsylvania. Presented in part at the annual meeting of The Association for Research in Vision and Ophthalmology, Ft. Lauderdale, Florida, May 2003 (H.D.) and at the annual meeting of American Academy of Ophthalmology, New Orleans, Louisiana, October, 2004 (H.D.). Support provided by the International Award of Merit in Retina Research, Houston, Texas (J.A.S.), Rosenthal Award of the Macula Society (C.L.S.), Macula Foundation, New York, New York (C.L.S.), the Noel T. and Sara L. Simmonds Endowment for Ophthalmic Pathology, Wills Eye Hospital (R.C.E., Jr.) and the Eye Tumor Research Foundation, Philadelphia, Pennsylvania (C.L.S., J.A.S.). Inquiries to Carol L. Shields, MD Oncology Service, Wills Eye Hospital, 840 Walnut Street, Philadelphia, PA 19107; fax: 215-928-1140; e-mail: [email protected].
468
©
2005 BY
● CONCLUSIONS:
Iris melanocytoma represented only 3% of all iris nevi. Related iris stromal and anterior chamber angle seeds were common, and secondary glaucoma occurred in 11% at 5 years. Growth was observed in 23% at 5 years but no malignant transformation was found. (Am J Ophthalmol 2005;139:468 – 475. © 2005 by Elsevier Inc. All rights reserved.)
M
ELANOCYTOMA IS A VARIANT OF MELANOCYTIC
nevus with distinctive clinical and pathologic features.1 Clinically, it appears as a deeply pigmented tumor that is usually located on or adjacent to the optic disk.2,3 In rare instances, it can arise in iris, ciliary body, choroid, or conjunctiva.4 In 1965, Zimmerman suggested the term “melanocytoma” to describe this tumor and reported two patients with iris melanocytoma, a 60-year-old white man with a deeply pigmented iris mass and a 34-year-old white man with recurrent iritis and pigmented iris mass. Iris melanoma was suspected in both cases, but both lesions proved to be iris melanocytoma after enucleation.1 Since then, a few other reports of iris melanocytoma have been published.5–14 The natural course of iris melanocytoma is variable. Iris melanocytoma can undergo spontaneous necrosis with resultant pigment dispersion, causing secondary glaucoma and heterochromia.5,6,8 –13 In some cases, iris or ciliary body melanocytoma can show progressive growth or even extrascleral involvement making clinical differentiation from malignant melanoma difficult.7,15,16 In rare cases, it can transform into iris melanoma.8,14 Failure to recognize iris melanocytoma could lead to misdirected therapy, including enucleation. To further expand our understanding of this unique tumor, we report our experience with 47 patients with iris melanocytoma and describe its clinical and histopathologic features, management, and natural course.
PATIENTS AND METHODS THE MEDICAL RECORDS OF ALL PATIENTS DIAGNOSED WITH
iris nevus at the Oncology Service at Wills Eye Hospital between January 1974 and February 2003 were reviewed
ELSEVIER INC. ALL
RIGHTS RESERVED.
0002-9394/05/$30.00 doi:10.1016/j.ajo.2004.10.008
TABLE 1. Clinical Features at Presentation of 47 Consecutive Patients with Iris Melanocytoma
TABLE 2. Related Anterior Segment Findings at Presentation of 47 Patients with Iris Melanocytoma
No of patients (%)
Color Darkly pigmented Light brown Configuration Nodular Diffuse Involved quadrant Inferior Lateral Nasal Superior Anteroposterior location Iris root Midzone Pupil margin Diffuse
44 (94%) 3 (6%) 40 (85%) 7 (15%) 21 (45%) 18 (38%) 4 (9%) 4 (9%) 32 (68%) 7 (15%) 5 (11%) 3 (6%)
retrospectively, and the records of those patients who were specifically classified as having iris melanocytoma were retrieved for further analysis. Institutional Review Board approval was obtained for this retrospective study. The clinical features of iris melanocytoma have been described in the literature2 and include a darkly pigmented, dome shaped or sessile iris stromal tumor with a granular surface and no intrinsic vasculature. Our criteria for entry into this study included tumors with the above features confined to the iris or with extension of anterior chamber angle. Patients with ciliary body or choroidal melanocytoma were excluded. Throughout the manuscript, the terminology “anterior chamber seeds” and “iris stromal seeds” are used to imply the cells that have been shed from necrotic tumor or pigment containing macrophages. The patient data collected at initial evaluation at the Oncology Service included age (years), race (Caucasian, African-American, Asian, Hispanic), gender (male, female), and systemic disease. The ocular data included affected eye (right or left), laterality (unilateral or bilateral), visual symptoms, best-corrected Snellen visual acuity, intraocular pressure (mm Hg), and ocular melanocytosis (present or absent). All patients underwent detailed slit-lamp biomicroscopy, gonioscopy, fundus examination, and transillumination. Ultrasound biomicroscopy was performed in 11 patients. The tumor data included tumor color, configuration (nodular, plateau), size (mm), thickness (mm), quadrantic location of the tumor epicenter (superior, nasal, inferior, temporal), and anteroposterior location of the tumor epicenter (pupillary margin, midzone, root, or diffuse). The presence of secondary glaucoma, heterochromia, ectropion iridis, satellite tumors, iris stromal seeds, anterior chamber angle seeds, intrinsic VOL. 139, NO. 3
Clinical finding
No of patients (%)
Heterochromia Sentinel vessel Ectropion irides Satellite tumors Number of satellite tumors Mean Median Range Iris stromal seeds Extent of iris stromal seeds (clock hours) Mean Median Range AC angle seeds Extent of AC angle seeds (clock hours) Mean Median Range AC inflammation Hyphema Pigmented keratic precipitate Intrinsic vascularization Sector cataract
6 (13%) 7 (15%) 3 (6%) 12 (26%) 4 3 1–15 20 (43%) 5 3 1–12 12 (26%) 8 7 3–12 2 (4%) 0 (0%) 1 (2%) 0 (0%) 0 (0%)
AC: Anterior chamber
vessels, anterior chamber reaction, hyphema, pigmented keratic precipitates, and episcleral sentinel vessels were recorded. The presence or absence of intrinsic vessels is evaluated by slit lamp examination and when available, iris fluorescein angiography. Tumor management after referral (observation, fine-needle biopsy, sector iridectomy, or enucleation) was listed. Follow-up data were collected regarding change in visual acuity, development of increased intraocular pressure (IOP), development of anterior chamber angle seeds, development of iris stromal seeds, and tumor growth. The clinical data were analyzed with regard to three main outcome measures: development of tumor seeding, development of increased IOP, and tumor growth. The effect of each individual demographic datum and clinical variable on the final visual outcome was analyzed by a series of univariate logistic regression analyses. All variables were analyzed as discrete categories except patient age, intraocular pressure, tumor base, and tumor thickness, which were analyzed as continuous variables and subsequently grouped into discrete categories to derive cutoff values. The correlation among the predictor variables was determined by Pearson correlations. The variables that were significant on a univariate level (P ⱕ .05) were entered into multivariate logistic regression analysis using forward stepwise (likehood ratio)
IRIS MELANOCYTOMA
469
FIGURE 1. A 53-year-old male with iris melanocytoma at the left eye. (Top) At presentation. (Middle) Histopathologic examination displaying copious pigment obscuring nuclear details on routine hematoxylin-eosin sections (hematoxylin-eosin ⴛ 50). (Bottom) Bleached section disclosing bland nuclei and benign nuclear/cytoplasmic ratio (hematoxylin-eosin ⴛ 100).
FIGURE 2. A 15-year-old female who presented with iris melanocytoma at the right eye. (Top) At presentation. (Middle) Histopathologic examination showing densely pigmented tumor composed of plump, polyhedral nevus cells with large quantities of melanin pigment that tends to obscure nuclear details (Hematoxylin-eosin ⴛ 50). (Bottom) Bleached section disclosing bland nuclei and low nuclear/cytoplasmic ratio. (Hematoxylin-eosin ⴛ 100)
method. For variables that showed a high degree of correlation, only one variable from the set of associated variables was entered at a time in subsequent multivariate models. A final multivariable model fitted variables identified as significant predictors (P ⬍ .05)
in the stepwise model and variables considered clinically important for each main outcome. Kaplan-Meier survival estimates were used to analyze main outcome measures as a function of time from the initial examination at the Oncology Service.
470
AMERICAN JOURNAL
OF
OPHTHALMOLOGY
MARCH 2005
TABLE 3. Kaplan-Meier Estimates of Tumor Growth, Elevated IOP and Development of New Tumor Seeds in 47 Patients
Tumor growth Increased IOP Development of new tumor seeds
At 1 year
At 3 years
At 5 years
At 10 years
At 15 years
(0/19) 0% (0/19) 0% (1/19) 5%
(4/12) 23% (2/13) 11% (5/11) 27%
(4/10) 23% (2/10) 11% (6/9) 34%
(7/3) 48% (2/5) 11% (10/3) 63%
(8/1) 74% (3/1) 55% (11/1) 75%
IOP: Intraocular pressure
RESULTS OF APPROXIMATELY 1400 PATIENTS WITH IRIS NEVUS MAN-
aged at the Oncology Service over 29 years, 47 patients (3%) were classified as having iris melanocytoma. The mean age at presentation was 37 years (median 40 years, range 3 to 67 years). Of 47 patients, 41 (87%) were Caucasian, 4 (9%) African-American, 1 (2%) Asian, and 1 (2%) was Hispanic. There were 30 females (64%) and 17 males (36%). One patient (2%) had a history of skin melanoma and one patient (2%) had ocular melanocytosis. The initial patient symptoms and signs were the presence of a dark spot on the iris in 7 patients (15%), blurred vision in 5 patients (11%), and conjunctival hyperemia in 2 patients (4%). Thirty-three patients (70%) were asymptomatic. The right eye was affected in 18 patients (38%) and the left eye in 29 (64%). There was no patient with bilateral involvement. The best corrected visual acuity was 20/20 to 20/40 in 45 patients (96%), 20/50 to 20/400 in 1 patient (2%), and finger counting in 1 patient (2%). At presentation, the mean intraocular pressure was 17 mm Hg (median 16 mm Hg, range 9 to 45 mm Hg). The intraocular pressure was higher than 21 mm Hg in 2 patients (4%). The clinical features and associated clinical findings in 47 patients with iris melanocytoma at presentation are shown in Tables 1 and 2. The mean diameter of iris melanocytoma was 3 mm (median 3 mm, range 1 mm to 8 mm) and the mean thickness was 2 mm (median 2 mm, range 1 mm to 4 mm). Ultrasound biomicroscopy was performed on 11 patients (23%) and showed an echogenic nodular thickening of iris in all 11 cases. Management included observation in 39 patients (83%), tumor removal by sector iridectomy in 7 patients (15%), and enucleation in 1 patient (2%). The reasons for sector iridectomy were failure of clinical and noninvasive diagnostic techniques to differentiate from melanoma in 3 patients (6%), tumor growth in 3 patients (6%), and the presence of secondary glaucoma in 1 patient (2%). The reason for enucleation was blind painful eye from secondary glaucoma in 1 patient (2%). Three patients (6%) had fine needle aspiration biopsy for diagnostic confirmation and to exclude iris melanoma. Histopathologic examination of resected lesions showed characteristic features including an intensely pigmented VOL. 139, NO. 3
tumor composed of plump, polyhedral nevus cells with copious quantities of maximally pigmented cytoplasm that obscure nuclear details. Bleached sections disclose tumor cells with a low nuclear cytoplasmic ratio and bland nuclei with inconspicuous nucleoli (Figure 1, top, middle, and bottom; Figure 2, top, middle, and bottom). The patients were followed for a mean of 58 months (median 20 months, range 1 to 364 months). Using Kaplan-Meier estimates, the development of increased IOP was observed in 0% at 1 year, 11% at 5 years, 11% at 10 years, and 55% at 15 years (Table 3). All patients with increased IOP developed new tumor seeds and two showed tumor growth. New iris stromal or anterior chamber angle seeds developed in 5% at 1 year, 34% at 5 years, 63% at 10 years, and 75% at 15 years (Table 3) (Figure 3, top, middle, and bottom). Using Kaplan-Meier estimates, tumor growth was observed in 0% at 1 year, 23% at 5 years, 48% at 10 years, and 74% at 15 years (Table 3) (Figure 4, left and right). Overall, there was a mean growth of 30% in basal dimension (median 28%, range 20% to 50%) and a mean growth of 34% in thickness (median 33%, range 20% to 63%). The lesions that showed growth were managed by careful observation in 4 patients (50%) and sector iridectomy in 4 (50%). Tumor growth was documented in 50% of the eyes that developed new tumor seeds. Univariate predictive factors for development of increased IOP, development of tumor seeding and tumor growth are shown in Table 4. Multivariate analysis showed no significant factor predictive of any of the outcome measures.
DISCUSSION IRIS MELANOCYTOMA IS A RARE VARIANT OF IRIS NEVUS
with fairly typical clinical and histopathologic features. In a clinical study by Shields and associates, of 200 patients referred for suspected iris melanoma, 158 patients (76%) were found to have lesions other than melanoma and these are termed as pseudomelanomas.17 The most common pseudomelanomas included iris cyst (38%) and iris nevus (31%). In that series, only one pseudomelanoma (0.5%) was an iris melanocytoma.17 In a pathology study of 189 specimens clinically suspected
IRIS MELANOCYTOMA
471
series, the mean patient age was 37 years (median 40 years, range 3 to 67 years) and 4 patients (9%) were under the age of 10 years. Iris melanocytoma has subtle clinical features that distinguish it from other iris nevi or melanoma. Iris melanocytoma typically presents as a darkly-pigmented iris nodule with mossy, irregular surface. Similar to other melanocytic lesions of the iris, it is usually located in the inferior quadrant of the iris. However, the clinical findings associated with other melanocytic lesions of the iris, such as ectropion iridis, intrinsic vessels, or sector cataract, are less commonly seen with iris melanocytoma. In a review of 175 patients with iris nevus, Territo and associates reported that iris nevi were associated with ectropion iridis in 24% of patients, intrinsic vessels in 20%, and sector cataract in 3%.19 Similarly, in a review of 169 patients with iris melanoma, Shields and coworkers reported an irregular tumor surface in 93% of patients, ectropion iridis in 44%, intrinsic vascularization in 43%, and sector cataract in 14%.20 In the present study, iris melanocytoma was associated with ectropion iridis in only 6% of patients and intrinsic vascularization and sector cataract were not seen in any case. In this study, 11(23%) of 47 patients had histopathologic confirmation of iris melanocytoma. The other patients were diagnosed clinically based on their distinctive features. In the present study, using Kaplan-Meier estimates, increased IOP developed in 11% of patients at 5 years, 11% at 10 years, and 55% at 15 years. Electron microscopic studies of eyes with secondary glaucoma attributable to iris melanoma (melanomalytic glaucoma) showed that pigment phagocytosis by trabecular endothelial cells contributed to the mechanical obstruction of trabecular meshwork and subsequent increased IOP.21,22 The majority of intratrabecular pigment in melanocytomalytic glaucoma is contained within macrophages instead of trabecular endothelial cells.6 Tumor necrosis is an important factor in the pathogenesis of melanocytomalytic glaucoma. Melanocytoma can undergo spontaneous necrosis and such tumors can contain numerous macrophages. Two of the three patients with melanocytomalytic glaucoma reported by Fineman and coworkers developed crater-like cavities in the surface of their tumors that were clinical evidence of tumor necrosis.6 Of the 13 published cases of iris melanocytoma, 12 cases (92%) were found to have anterior chamber angle seeds and only 9 (69%) developed melanocytomalytic glaucoma (Table 5).1,5–14 Similarly, we did not find that anterior chamber angle seeds were a risk factor for development of increased IOP. In our study, pigmented keratic precipitates and anterior chamber inflammation were factors predictive of the development of increased IOP in the univariate analysis, emphasizing the role of macrophages in the anterior chamber angle. Multivariate analysis showed no factor predictive for the development of increased IOP.
FIGURE 3. A 23-year-old male who presented with iris melanocytoma at the left eye. (Top) At presentation. (Middle) Gonioscopy showing dense anterior chamber seeds at inferior anterior chamber angle. (Bottom) Histopathologic examination showing extensively necrotic tumor involving peripheral iris.
to be iris or ciliary body melanoma, Jakobiec and Silbert found 10 (5%) to be melanocytoma.18 Thus iris melanocytoma can closely simulate iris melanoma. Iris melanocytoma generally is diagnosed in adults and has been reported in children.1,5–14 Of the 13 welldocumented cases of iris melanocytoma in the literature, the mean patient age was 32 years and only 4 (9%) were under the age of 10 years (Table 5).1,5–14 Similarly, in our 472
AMERICAN JOURNAL
OF
OPHTHALMOLOGY
MARCH 2005
FIGURE 4. A 38-year-old male patient with iris melanocytoma at the right eye that showed growth. (Left) In 1972. (Right) In 2003, the melanocytoma had doubled in size compared with 1972.
TABLE 4. Univariate and Multivariate Analysis of Clinical Factors Correlated with Increased IOP, Development of New Iris Stromal and Anterior Chamber Angle Seeds and Growth in 47 Patients with Iris Melanocytoma*
Factor
Increased IOP Univariate analysis Race (Non-Caucasian** vs Caucasian) Pigmented keratic precipitates (Present** vs Absent) Anterior chamber inflammation (Present** vs Absent) New iris stromal and ACA seeds Univariate analysis Race (Non-Caucasian** vs Caucasian) Pigment keratic precipitates (Present** vs Absent) Growth Univariate analysis Anterior chamber angle seeds (Present** vs Absent)
P
Relative Risk (95% Confidence Interval)
0.008
27.5 (2.3–323.4)
0.04
11.5 (1.1–129.1)
0.04
18.4 (1.1–295.4)
0.04
18.5 (1.2–295.7)
0.04
18.5 (1.2–295.7)
0.04
5.5 (1.1–26.3)
*Multivariate analysis was performed for each main outcome, but there were no factor found to be a significant predictor for any of the outcome measures. **Reference variable
VOL. 139, NO. 3
Iris melanocytoma can undergo spontaneous necrosis with resultant pigment dispersion or shedding of melanocytoma cells and/or melanophages into the anterior chamber, causing secondary glaucoma or heterochromia. In the present study, iris melanocytoma was associated with anterior chamber angle seeds in 40% of patients and iris stromal seeds in 53% of patients. In comparison, Territo and associates found iris stromal seeds in only 9% of patients with standard iris nevi.19 In our study, new anterior chamber angle and/or iris stromal seeds developed in 34% of patients at 5 years, 63% at 10 years, and 75% at 15 years. Increased seeding could be related to partial tumor necrosis. The cause of spontaneous necrosis is unknown, although Teichmann and Karcioglu have speculated that it might be related to lack of circulation in a highly metabolic tumor such as melanocytoma.9 We found that pigmented keratic precipitates were a risk factor for development of anterior chamber and/or iris seeds in univariate analysis. Multivariate analysis showed no factor predictive for development of new tumor seeds. Iris melanocytoma usually has been considered to be a relatively stationary lesion. Of 13 published cases, only 3 cases (23%) showed clinical evidence of growth (Table 5).5–14 In the present study, growth occurred in 23% of the patients at 5 years and 48% at 10 years. The only factor predictive of growth was the presence of anterior chamber angle seeds in univariate analysis. Multivariate analysis showed no significant risk factor for tumor growth. In a review of 175 patients with iris nevus, Territo and associates found that only 5% of lesions showed growth at a mean follow-up of 5 years and the only predictive risk factor for growth was presence of iris stromal and anterior chamber angle seeds.19 Growth of iris melanocytoma does not necessarily imply malignant transfor-
IRIS MELANOCYTOMA
473
TABLE 5. Clinical Findings in 13 Well Documented Cases of Iris Melanocytoma in the Literature*
Eye
Symptoms or signs
Color
Location
ACA seeds
Author (Year)
Age/Race/Sex
Zimmerman (1965) Zimmerman (1965) Thomas (1969)
62 WM
OS Asymptomatic Black Inferonasal NA
34 WM
OS Recc. iritis
20 AAF
Shields (1977) Nakazawa (1984) Cialdini (1989)
Iris stromal seeds
Growth
Management
NA
NA
Absent
Enuc
Black Inferotemp Present NA
Absent
Absent Absent
Enuc
Black Superior
Present NA
Absent
Absent Present
Enuc
23 WM
OS Spot on iris, pain, redness OS Spot on iris
Black Inferior
Present Present
Present
Absent Absent
46 AF
OD Discomfort
Black Diffuse
Present Present
Present
Absent Absent
34 WF
OS Spot on iris, Black Temporal pain, redness OS Asymptomatic Black Inferior
Present Absent
Present
Present Present
Sector iridectomy Trabeculectomy and Inc Bx Sector iridectomy
Present NA
Present
Absent Absent
Fountain (1992) Teichmann (1995)
18 AAF
Green (1996) Fineman (1998) Fineman (1998) Kiratli (2001) Shields (2002)
28 ArF
NA
Glaucoma
Association with melanoma
Sector iridectomy Sector iridectomy
Brown Nasal
Present Absent
Present
Absent Absent
6F
OS Spot on iris, pain, redness OS Spot on iris
Black Superior
Present NA
Present
Absent Absent
Enuc
61 WM
OS Spot on iris
Black Inferonasal Present Present
Present
Absent Absent
Enuc
48 WF
OD Spot on iris
Brown Inferior
Present Present
Present
Present Absent
27 WF
OS Spot on iris
Black Inferior
Present Absent
Present
Absent Absent
9 WM
OS Spot on iris
Brown Inferotemp Present Absent
Absent
Present Absent
Sector iridectomy Sector iridectomy Sector iridectomy
ACA ⫽ Anterior chamber angle; W ⫽ White, AA ⫽ African-American; A ⫽ Asian; Ar ⫽ Arabic; M ⫽ Male; F ⫽ Female; OD ⫽ Right eye; OS ⫽ Left eye; Enuc ⫽ Enucleation; Recc ⫽ Recurrent; Inferotemp ⫽ Inferotemporal; Inc Bx ⫽ Incisional biopsy *. Well-documented cases published in English literature.
mation. Shields and associates reported a 9-year-old male who showed progressive growth of a pigmented iris lesion over 6 years and on histopathologic examination it proved to be an iris melanocytoma.7 In the present study, we did not observe malignant transformation on histopathologic examination of the lesions with documented growth that were resected. The management of suspected iris melanocytoma is usually cautious observation. When noninvasive diagnostic techniques fail to establish the diagnosis or tumor growth is documented, fine-needle aspiration biopsy or local resection using iridectomy, iridogoniectomy, or iridogoniocyclectomy is performed.23 Local resection can also be used to treat secondary glaucoma caused by necrotic melanocytoma.5 This surgical technique not only provides tissue for histopathologic diagnosis but may also assist in resolution of the secondary glaucoma.5,11 Secondary glaucoma can be treated medically 474
AMERICAN JOURNAL
by sector iridectomy or by glaucoma-filtering surgery.12 However, diagnostic confirmation of the lesion by histopathology or cytopathology is warranted before filtering surgery. Enucleation is usually reserved for blind painful eyes. In conclusion, iris melanocytoma is an uncommon variant of iris nevus with distinctive clinical and histopathologic features. It usually presents as a darkly-pigmented nodular mass with granular surface and no intrinsic vessels. It can undergo spontaneous necrosis resulting in iris stromal and anterior chamber angle seeding and dispersion of melanophages leading to secondary elevated IOP. Frequent observation of the lesion in every six months is usually warranted. Local resection is reserved for treatment of increased IOP or for diagnostic purposes when malignancy cannot be ruled out. Despite the benign nature of iris melanocytoma, clinical evidence of growth was observed in 23% of patients at 5 years and 48% at 10 years. OF
OPHTHALMOLOGY
MARCH 2005
REFERENCES 1. Zimmerman LE. Melanocytes, melanocytic nevi and melanocytomas. The Jonas Friedenwald Memorial Lecture. Invest Ophthalmol 1965;4:11– 41. 2. Shields JA, Shields CL. Posterior uveal melanocytoma. In: Shields JA, Shields CL, editors. Atlas of intraocular tumors. Philadelphia: Lippincott, Williams and Wilkins, 1999:61–72. 3. Shields JA, Demirci H, Mashayekhi A, Shields CL. Melanocytoma of optic disc in 115 cases: The 2004 Samuel Jackson Memorial Lecture, part I. Ophthalmology 2004;111:1739 – 1746. 4. Shields JA, Augsburger JJ, Bernardino V and Eller AW. Melanocytoma of the ciliary body and iris. Am J Ophthalmol 1980;89:632– 635. 5. Shields JA, Annesley WH Jr, Spaeth GL. Necrotic melanocytoma of iris with secondary glaucoma. Am J Ophthalmol 1977;84:826 – 829. 6. Fineman M, Eagle RC Jr, Shields JA, Shields CL, De Potter P. Melanocytomalytic glaucoma in eyes with necrotic iris melanocytoma. Ophthalmology 1998;105:492– 496. 7. Shields JA, Eagle RC Jr, Shields CL, Nelson LB. Progressive growth of an iris melanocytoma in a child. Am J Ophthalmol 2002;133:287–289. 8. Cialdini AP, Sahel JA, Jalkh AE, Weiter JJ, Zakka K, Albert DM. Malignant transformation of an iris melanocytoma. Graefe’s Arch Clin Exp Ophthalmol 1989;227:348 –354. 9. Teichmann KD, Karcioglu ZA. Melanocytoma of the iris with rapidly developing secondary glaucoma. Surv Ophthalmol 1995;40:136 –144. 10. Nakazawa M, Tamai M. Iris melanocytoma with secondary glaucoma. Am J Ophthalmol 1984;97:797–799. 11. Kiratli H, Bilgic S, Gedik S. Late normalization of melanocytomalytic intraocular pressure elevation following excision of iris melanpcytoma. Graefe’s Arch Clin Exp Ophthalmol 2001;239:712–715.
VOL. 139, NO. 3
12. Fountain TR, Goldberg MF, Green WR. Glaucoma and a melanocytic iris lesion in an-18-year-old. In: Schachat AP, editor. Current practice in ophthalmology. St. Louis: Mosby, 1992:371–380. 13. Green WR. Uveal tract. In: Spencer WH, editor. Ophthalmic pathology. An atlas and textbook. Philadelphia: WB Saunders, 1996:1608 –1609. 14. Thomas CI, Purnell EW. Ocular melanocytoma. Am J Ophthalmol 1969;67:79 – 86. 15. Scheie HG, Yanoff M. Pseudomelanoma of the ciliary body. Arch Ophthalmol 1967;77:81– 83. 16. Rummelt V, Naumann GOH, Foldberg R, Weingeist TA. Surgical management of melanocytoma of the ciliary body with extrascleral extension. Am J Ophthalmol 1994;117:169 –176. 17. Shields JA, Sanborn GE, Augsburger JJ. The differential diagnosis of malignant melanoma of the iris. A clinical study of 200 patients. Ophthalmology 1983;90:716 –720. 18. Jakobiec FA, Silbert A. Are most iris “melanomas” really nevi? A clinicopathological study of 189 lesions. Arch Ophthalmol 1981;99:2117–2132. 19. Territo C, Shields CL, Shields JA, Augsburger JJ, Schroeder RP. Natural course of melaonocytic tumors of the iris. Ophthalmology 1988;95:1251–1255. 20. Shields CL, Shields JA, Materin M, Gershenbaum E, Singh AD, Smith A. Iris melanoma. Risk factors for metastasis in 169 consecutive patients. Ophthalmology 2000;108:172–178. 21. Van Buskirk EM, Leure-DuPree AE. Pathophysiology and electron microscopy of melanomalytic glaucoma. Am J Ophthalmol 1978;85:160 –166. 22. McMenamin PG, Lee WR. Ultrastructural pathology melanomalytic glaucoma. Br J Ophthalmol 1986;70:895–906. 23. Shields JA, Shields CL, Ehya H, Eagle RC, De Potter P. Fine-needle aspiration biopsy of suspected intraocular tumors. The 1992 Urwick lecture. Ophthalmology 1993;100: 1677–1684.
IRIS MELANOCYTOMA
475