—  SHORT COURSE  —

OPHTHALMIC PATHOLOGY FOR THE NON-SPECIALIST


CASE 6 – MALIGNANT MELANOMA OF CHOROID AND CILIARY BODY

J. Godfrey Heathcote, M.B.,Ph.D.  —  Janice R. Safneck, M.D.




History
An 84-year-old man presented with decreased visual acuity in his right eye of 8 days duration.

Diagnosis
Malignant melanoma, spindle cell type, of choroid and ciliary body

Histopathology
The eye is of normal size and appears unremarkable externally. On transillumination, a mass is visible inferiorly. Opening the eye reveals a pale tan nodular lesion with focal hemorrhage, basal diameter 12 mm and elevation 9 mm, arising from the choroid. Tumor has penetrated both choroid and retina and abuts on the posterior lens surface.


Case 6, Slide 14 - Malignant Melanoma of Choroid and Ciliary Body: Cut surface of eye showing a pale tan nodular lesion with focal hemorrhage, basal diameter of 12 mm and elevation of 9 mm, arising from the choroid. Tumor has penetrated both the choroid and retina and abuts the posterior surface of the lens.
Case 6, Slide 15 - Malignant Melanoma of Choroid and Ciliary Body: The tumor is a spindle cell melanoma which is essentially amelanotic. Note the mitotic figure.

Microscopically, a spindle cell melanoma which is essentially amelanotic is seen. It involves both choroid and ciliary body but does not extend into anterior chamber or iris. Tumor breaks through Bruch's membrane, penetrates the retina, and is present on adjacent retina's inner surface. Neoplastic cells are S100 positive and HMB-45 negative. Melanoma is noted in a vortex vein, extending extraocularly within that vein. Mitoses average 7/40 high power fields.

Discussion
Primary intraocular melanomas arise in the uveal tract (iris, ciliary body and choroid) and are the most common primary intraocular malignancy with an annual incidence of 5-8 cases per 1,000,000 people.1  Patients with uveal melanoma are typically between 50 and 70 years of age with less than 2% of cases occurring in individuals less than 20. Almost all uveal melanomas are unilateral, bilateral tumors accounting for less than 0.2%. The etiology of uveal melanoma is unknown. Risk factors include race, age, ocular/oculodermal melanocytosis, neurofibromatosis, uveal nevi, and heredity. Uveal melanomas are rare in African Americans compared to Caucasians, with an incidence ratio of 1:98.2 

Uveal melanomas, usually unilateral3  but occasionally bilateral,4  develop in 1-2% of patients with ocular/oculodermal melanocytosis. Ocular melanocytosis is defined as increased melanocytic pigmentation involving uvea, sclera and episclera and is typically congenital, unilateral and diffuse, although bilateral and sectoral types occur. Microscopically, episclera, sclera and choroid contain hyperpigmented melanocytes; as well these cells can be identified in iris and anterior chamber angle structures in three quarters of cases, optic nerve in one third and conjunctiva in approximately 14%.3  Oculodermal melanocytosis (nevus of Ota) is ocular melanocytosis associated with increased skin pigmentation in the distribution of the fifth cranial nerve ophthalmic, maxillary and sometimes mandibular divisions. Choroidal nevi are found in approximately 5-10% of healthy individuals.1  They are a risk factor for melanoma since occasional lesions thought to be nevi clinically suddenly grow and on enucleation are malignant melanomas. Neurofibromatosis is associated with an increased risk for uveal melanoma. The roles of the dysplastic nevus syndrome, ultraviolet irradiation and pregnancy in uveal melanoma development are controversial. Clinically patients with uveal melanoma typically complain of visual disturbances or have a mass on examination. Most tumors involve the choroid; iris tumors are the least frequent (3-10%).1  In many instances, clinical diagnosis results in enucleation or local resection from which a histologic diagnosis is made; however, fine needle aspiration biopsy has successfully diagnosed anterior and posterior uveal melanomas5  and occasionally incisional biopsies are employed.

Pathological findings
Pathological findings classically regarded as prognostically important include tumor size, tumor location including location of the anterior margin, cell type, extraocular extension and mitotic rate.

Tumor size is an important prognostic indicator, specifically the largest tumor dimension (either maximum height or greatest length of scleral contact).6  Measurements taken from the gross specimen can be larger than those obtained from glass slides6  but both correlate with outcome. Small tumors have a diameter <10mm or height <3 mm; medium tumors a diameter >10mm but <15 mm or a height <5 mm; large tumors a diameter >15 mm or height <5 mm; the largest measurement is used to determine the category. Grossly, uveal melanomas are typically circumscribed, vary from white to dark brown, at times even within the same tumor, and crescentic, oval, nodular, irregular or mushroom-shaped, the latter in choroidal tumors breaking through Bruch's membrane. Diffuse flat melanomas also can occur but are rare (3-5%).7  Other rare forms include multifocal melanomas and anterior ring melanomas.

Tumor location has been assessed with regard to prognosis. Combined ciliary body and choroidal tumors have a poorer prognosis likely the result of their large size. Iris melanomas, on the other hand, have a much lower metastatic rate, in the order of 5%, but this too may be due to their generally small size.8  In a recent series of more than 1500 uveal melanomas, over 96% of medium and large melanomas were located entirely or partially posterior to the equator, with 34.5% partially in the ciliary body, 7.9% in the trabecular meshwork and 4.7% in the iris.9  The vast majority of large tumors (82.4%) have an anterior margin at or anterior to the equator.9 

Callender's classification recognized 6 cell types and growth patterns: spindle A, spindle B, epithelioid, mixed, fascicular and necrotic. Spindle A cells are characterized by fine nuclear chromatin, a nuclear groove, indistinct nucleoli and indistinct cell borders. Spindle B cells have coarser chromatin and definite nucleoli with indistinct cell borders. Epithelioid cells are polygonal with well defined cell borders, less cohesion and large prominent nucleoli. Mixed cell tumors contain both epithelioid and spindle cells. Fascicular pattern involves palisading of spindle cells. Necrotic tumors have no well-preserved areas which would allow them to be classified into the other types. Although useful, problems with this classification arise from lack of interobserver reproducibility, overlap of spindle A cells with nevus cells, no category for tumors composed of spindle A and B cells which is a frequent occurrence, lack of prognostic value of the fascicular pattern and no criteria to specify what numbers of a second cell type would have to be present to classify a tumor as mixed.

The modified Callender classification categorizes tumors as spindle cell, epithelioid cell, mixed cell or necrotic; mixed and epithelioid cell melanomas often are combined as their outcome is similar. Epithelioid cells may be small or large. Mixed cell tumors need only contain a small quantity of second cell type and the presence of even a few epithelioid cells in an otherwise spindle cell tumor alters the prognosis.10  However, the classification still suffers from difficulties in interobserver reproducibility, problems in distinguishing spindle B from epithelioid cells since there is a morphological continuum between such cells, and lack of a universally accepted definition of how many epithelioid cells are needed for a mixed cell tumor (one easily recognizable epithelioid cell per 5 x 400 fields has been suggested)11  or how many spindle cells are permissible in an epithelioid one. Mixed cell tumors constitute half of posterior uveal melanomas followed in frequency by spindle cell tumors; pure epithelioid cell tumors are uncommon.12  Mortality for spindle cell tumors is in the order of 27-29%, mixed cell 59-62%, epithelioid 58-81% and necrotic 72%.1 

Extraocular extension from tumor invading and penetrating the sclera typically around vessels and nerves (extension through emissary canals) can be seen grossly and/or microscopically, posteriorly or anteriorly in approximately 17% of uveal melanomas.13  Such tumors often are large with a prominent component of epithelioid cells and it is uncertain how much additional prognostic significance the extrascleral extension has in comparison to tumor size and cell type.13  However, tumors with microscopic to moderate gross extraocular extension often are given post- enucleation radiotherapy so it is important to note this finding. Intravascular involvement of vortex veins also may be seen and carries a poor prognosis so these vessels should be submitted for histology. Tumor can invade optic nerve in approximately 2-7% of cases9  although rarely extensively; a cross section of optic nerve from the resection line should be examined microscopically.

Mitotic rate has prognostic significance. Uveal melanomas typically have few mitoses and therefore they are counted over 40 high power fields (x400).12 

In the search for better prognostic factors,lymphocytic infiltration, mean nucleolar diameter, and tumor PAS positive patterns have been examined. 12% of melanomas contain 100 or more lymphocytes, predominantly T cells, per 20 high power fields (x400); such tumors appear to have a poorer prognosis.14  The Collaborative Ocular Melanoma Study found epithelioid cell tumors contained more inflammatory cells than other cell types.9  The mean of the longest nucleoli measured in tumor sections stained with colloidal silver nitrate stain for nucleolar organizing regions using an ocular micrometer and x1000 magnification also correlates with poor outcome, although technical modifications have made evaluation easier, this prognostic predictor is not routinely employed at the present time. 15  Study of PAS positive patterns in uveal melanomas has shown one closed loop (PAS positive tissue surrounds at least three quarters of a tumor lobule), networks (3 back to back loops) and parallel arrays with cross linking, correlate with decreased survival.16, 17  Recently, arcs or arcs with branching and parallel channels have been noted to correlate with the presence of loops and networks.18  There is some controversy as to whether the PAS positive structures represent blood vessels or connective tissue19  but general agreement as to the usefulness of these patterns in separating low from high-risk categories for metastasis.

Cytogenetic analysis of uveal melanoma frequently has shown monosomy 3, gain of 8q material, and 6p alterations with approximately 70% of individuals with monosomy 3 dying of metastases; the possibility of two tumor suppressor genes on this chromosome has been raised.20  An ever increasing number of genes and proteins are being studied in uveal melanoma but none have yet proved useful as routine diagnostic and/or prognostic measures.

Complications of uveal melanoma include glaucoma, retinal detachment, intraocular hemorrhage and cystoid macular edema. Large tumors and iris or iris and ciliary body neoplasms are more likely to be associated with glaucoma than small or choroidal ones.21,22  For ciliary body tumors, intraocular pressure may depend on the balance between the tumors' effects on aqueous outflow and production by ciliary body epithelium.22 

Treatment of uveal melanoma includes photocoagulation, local resection, plaque or particle beam radiotherapy and enucleation.1  Exenteration is reserved for cases with marked orbital extension. Enucleation usually is used for larger tumors or ones that have failed other forms of therapy. Irradiated melanomas, regardless of the method employed, show histologically increased inflammation, necrosis, fibrosis, balloon cells, and blood vessel damage with hemorrhage and the same or fewer mitoses in comparison to non-irradiated tumors.23 

Overall 5, 10 and 15 year survival rates for uveal melanoma are 65%, 52% and 46% respectively.24  Intraocular lymphatics do not exist; exceedingly rare instances of regional lymph node metastases occurred only after extraocular spread and invasion of conjunctival lymphatics. Thus uveal melanomas essentially metastasize hematogenously. Approximately 2-3% of patients with ciliary body and choroidal melanoma, typically with large tumors, have metastases at the time of diagnosis.1  Common sites of metastatic disease include the liver (93%), lung (24%) and bone (16%) with multiple sites of involvement in 87% of patients.25  On histology, metastatic melanomas consist predominantly of epithelioid or unclassifiable cells regardless of the primary cell type. HMB-45 remains the best marker immunohistochemically with S100 staining approximately 66% of metastases.24  Uveal melanoma may metastasize decades after the primary tumor was diagnosed.

Differential Diagnosis
Although the diagnosis of uveal melanoma usually is straightforward, occasionally both clinically and microscopically, problems may arise. Differential diagnosis of uveal malignant melanoma includes metastatic tumors and benign and malignant primary intraocular neoplasms. Metastatic tumors are the most common malignant intraocular neoplasms and primarily involve the choroid. Patients may be asymptomatic or have decreased vision.26  Breast (40-47%) and lung (21-30%) carcinomas are the most common tumors to metastasize intraocularly;26,27  most patients with the former have widespread metastatic disease but frequently lung carcinoma intraocular spread is detected before the primary is found.26  It should be noted that occasional uveal melanomas may contain substantial numbers of cells with clear cytoplasm, a potential source of confusion with metastatic renal cell carcinoma.28  Also, pigmented neuroendocrine tumors metastatic to uvea may be misdiagnosed as melanoma,29  illustrating the importance of performing immunohistochemical and other ancillary studies in neoplasms appearing atypical for melanoma. Cutaneous melanomas may metastasize to uvea but such metastases are often multiple, involve other intraocular tissues and are typically part of widespread systemic disease.30  Rarely, lymphoma may present as a discrete uveal mass simulating uveal melanoma.31 

Primary ocular tumors which may be confused with malignant melanoma include uveal tract nevi, melanocytoma (magnocellular nevus),32  adenoma/adenocarcinoma of pigmented ciliary body epithelium,33  leiomyoma/leiomyosarcoma34,35  and melanotic schwannoma,36  as well as reactive and neoplastic retinal pigment epithelial proliferations.

Occasionally eyes enucleated for uveal malignant melanoma do not have this tumor.9  Simulating lesions include choroidal nevi, central and peripheral exudative chorioretinopathy, congenital and reactive hypertrophy of retinal pigment epithelium, hemangiomas, hemorrhage, retinal detachment and choroiditis. On the other hand, eyes enucleated because of trauma, glaucoma, phthisis, endophthalmitis or painful blindness, sometimes contain malignant melanoma. Unsuspected uveal malignant melanoma also may be discovered at autopsy either as an incidental finding37  or in patients with metastases but no apparent primary, illustrating the importance of examining eyes post mortem.

References

  1. Seregard S: Posterior uveal melanoma. The Swedish perspective. Acta Ophthalmol Scand 1996;74:315-329.
  2. Margo CE, McLean IW: Malignant melanoma of the choroid and ciliary body in black patients. Arch Ophthalmol 1984;102:77-79.
  3. Gonder JR, et al: Uveal malignant melanoma with ocular and oculodermal melanocytosis. Ophthalmology 1982;89:953-960.
  4. Singh AD, et al: Bilateral primary uveal melanoma. Bad luck or bad genes? Ophthalmology 1996;103:256-262.
  5. Shields JA, et al: Fine-needle aspiration biopsy of suspected intraocular tumors. The 1992 Urwick lecture. Ophthalmology 1993;100:1677-1684.
  6. Folberg R, et al: Comparison of direct and microslide pathology measurements of uveal melanomas. Invest Ophthalmol Vis Sci 1985;86:1788-1791.
  7. Shields CL, et al: Diffuse choroidal melanoma. Clinical features predictive of metastasis. Arch Ophthalmology 1996;114:956-963.
  8. Jakobiec FA, Silbert G: Are most iris "melanomas" really nevi? A clinicopathologic study of 189 lesions. Arch Ophthalmol 1981;99:2117-2132.
  9. Histopathologic characteristics of uveal melanomas in eye enucleated from the collaborative ocular melanoma study group. COMS report no. 6. Am J Ophthalmol 1998;125:745-766.
  10. McLean IW, et al: Modifications of Callender's classification of uveal melanoma at the Armed Forces Institute of Pathology. Am J Ophthalmol 1983;96:502-509.
  11. Gamel JW, McLean IW: Quantitative analysis of the Callender classification of uveal melanoma cells. Arch Ophthalmol 1977;95:686-691.
  12. McLean IW: Prognostic features of uveal malignant melanoma. Ophthalmol Clin N A 1995:8:143-153.
  13. Seregard S, Kock E: Prognostic indicators following enucleation for posterior uveal melanoma. A multivariate analysis of long-tern survival with minimized loss to follow-up. Acta Ophthalmol Scand 1995;73:340-344.
  14. de la Cruz PO Jr, Specht CS, McLean IW: Lymphocytic infiltration in uveal malignant melanoma. Cancer 1990;65:112-115.
  15. Moshari A, McLean IW: Uveal melanoma: mean of the longest nucleoli measured on silver-stained sections. Invest Ophthalmol Vis Sci 2001;42:1160-1163
  16. Folberg R, et al: The morphologic characteristics of tumor blood vessels as a marker of tumor progression in primary uveal melanoma: a matched case-control study. Hum Pathol 1992;23:1298-1305.
  17. Folberg R, et al: The prognostic value of tumor blood vessel morphology in primary uveal melanoma. Ophthalmology 1993;100:1389-1398.
  18. Folberg R, et al: Microcirculation patterns other than loops and networks in choroidal and ciliary body melanomas. Ophthalmology 2001;108:996-1001.
  19. McDonald DM, Munn L, Jain RK: Vasculogenic mimicry: how convincing, how novel, and how significant? Am J Pathol 2000;156:383-388.
  20. Tschentscher F, et al: Partial deletions of the long and short arm of chromosome 3 point to two tumor suppressor genes in uveal melanoma. Cancer Res 2001;61:3439-3442.
  21. Yanoff M: Mechanisms of glaucoma in eyes with uveal malignant melanomas. Int Ophthalmol Clin 1972;12:51-62.
  22. Shields MB, Klintworth GK: Anterior uveal melanomas and intraocular pressure. Ophthalmology 1980;87:503-517.
  23. Saornil MA, et al: Histopathologic study of eyes after Iodine I 125 episcleral plaque irradiation for uveal melanoma. Arch Ophthalmol 1997;115:1395-1400.
  24. Luyten GPM, et al: Metastatic uveal melanoma. A morphologic and immunohistochemical analysis. Cancer 1996;78:1967-71.
  25. Assessment of metastatic disease status at death in 435 patients with large choroidal melanoma in the collaborative ocular melanoma study (COMS). COMS report no.15. Arch Ophthalmol 2001;119:670-676.
  26. Shields CL, et al: Survey of 520 eye with uveal metastases. Ophthalmology 1997;104:1265-1276.
  27. Ferry AP, Font RL: Carcinoma metastatic to the eye and orbit. I. A clinicopathologic study of 227 cases. Arch Ophthalmol 1974;93:276-286.
  28. Clear cell differentiation in choroidal melanoma. COMS report no. 8. Arch Ophthalmol 1997;115:894-898.
  29. Eagle RC Jr., et al: Choroidal metastasis as the initial manifestation of a pigmented neuroendocrine tumor. Arch Ophthalmol 2000;118:841-845.
  30. Fishman ML, Tomaszewski MM, Kuwabara T: Malignant melanoma of the skin metastatic to the eye: frequency in autopsy series. Arch Ophthalmol 1976;94:1309-1311.
  31. Fredrick DR, et al: Solitary intraocular lymphoma as an initial presentation of widespread disease. Arch Ophthalmol 1989;107:395-39.
  32. Reidy JJ, et al: Melanocytoma: nomenclature, pathogenesis, natural history and treatment. Surv Ophthalmol 1985;29:319-327.
  33. Rennie IG, Faulkner MK, Parsons MA: Adenoma of the pigmented ciliary epithelium. Br J Ophthalmol 1994;78:484-485.
  34. Foss AJE, et al: Are most intraocular "leiomyomas" really melanocytic lesions? Ophthalmology 1994;101:919-924.
  35. Shields JA, et al: Observations on seven cases of intraocular leiomyoma. The 1993 Byron Demorest Lecture. Arch Ophthalmol 1994;112:521-528.
  36. Shields JA, et al: Melanotic schwannoma of the choroid. Immunohistochemistry and electron microscopic observations. Ophthalmology 1994;101:843-849.
  37. Butnoer KJ, Proia AD: Unexpected autopsy findings arising from postmortem ocular examination. Arch Pathol Lab Med 2001;125:1193-1196.