2018 NPB-14

This case was originally published in 2018. The information provided in this case was accurate and correct at the time of initial program release. Any changes in terminology since the time of initial publication may not be reflected in this case.

A 47-year-old man presented with vision loss in the left eye. Funduscopic examination revealed a subretinal mass adjacent to the optic nerve head. An enucleation was performed.

Tissue Site
Left eye

The whole slide image provided is an H&E stained slide of the left eye from an enucleation.

The uvea of the eye is a vascular tunic comprised of the iris, ciliary body, and choroid. Located between the sclera and the retina, the uvea contains dendritic pigmented melanocytes which have the potential to give rise to malignant melanoma. Patients with choroidal melanoma typically present as adults with painless monocular vision loss, while a cataract or glaucoma may be the presenting feature of an anterior segment melanoma. Approximately half of patients with choroidal and ciliochoroidal melanomas eventually die from their tumors. Prognosis is better in cases localized to the iris, presumably because they are recognized earlier.

Most uveal melanomas produce at least some melanin pigment. For amelanotic cases, confirmatory IHC stains for melanoma, such as SOX10 and HMB45 (Image A), are essential. Many uveal melanomas perforate Bruch membrane, enter the subretinal space, and invade the retina (Image B), resulting in cystic atrophy of the overlying retina.

Although a small percentage of melanomas diffusely infiltrate the uvea, the vast majority are well-circumscribed. The presence of distinct margins allows for accurate assessment of tumor size. Measurement of largest tumor diameter at the base has prognostic significance. The current case demonstrated basal largest tumor diameter of 9 mm (Image C). The 5-year survival of small (<10 mm), medium (10 to 15 mm), and large (>15 mm) melanomas are 86%, 66%, and 56%, respectively.

Histologic assessment of uveal melanoma requires subtyping based on the predominance of particular morphologic cell types: spindle type A, spindle type B, and epitheloid. This cytomorphologic subtyping schema is known as the modified Callender classification, after a system originated in 1931 by Dr. George Russel Callender and then modified in 1978. Spindle type A cells have elongated nuclei, inapparent nucleoli, scant cytoplasm, and indistinct cell borders. Spindle type B (Image D) cells have oval nuclei with identifiable nucleoli and moderate cytoplasmic volume. Epithelioid cells (Image E) are larger, with rounded nuclei, prominent nucleoli, and distinct cell borders. Patients with pure spindle tumors have the best prognosis. Indeed, tumors composed entirely of spindle A cells are better classified as benign spindle cell nevi. On the other end of the spectrum, tumors composed purely of epithelioid cells have the worst prognosis. Those with a mix of spindle and epithelioid cell types, which represent the majority of cases, have an intermediate prognosis.

In addition to basal largest tumor diameter and cytologic phenotype, other histologic features shown to have some negative prognostic implication include: lymphocytic infiltration (Image F and Image G), macronucleoli, elevated mitotic count, looping vascular networks, and extrascleral extension.

Although the histomorphology of cutaneous and uveal melanomas can be similar, genetic tumorigenesis differs. Early mutations in BRAF characterize cutaneous melanoma while mutations in GNAQ and GNA11 are typical of uveal melanomas. Chromosomal and genetic features also impact prognosis. Monosomy for chromosome 3 indicates a worse prognosis and is present in approximately half of cases. Gene expression profiling using RT-PCR has emerged as a method of stratifying patients according to risk for metastatic disease and has been reported to be prognostically superior to clinical and histopathologic features. Uveal melanomas naturally cluster into two groups according to expression of thirteen different genes. Class 1 melanomas tend not to metastasize, while class 2 melanomas have a high risk of metastasis. Not unexpectedly, class 2 tumors tend to exhibit epithelioid cytomorphology and monosomy for chromosome 3. Elevated expression of the PRAME (preferentially expressed antigen in melanoma) gene on chromosome 22 is an additional indicator of increased risk for metastasis. Increased PRAME expression has been shown to identify the small percentage of class 1 uveal melanomas that metastasize as well as indicate a shorter time to metastasis for class 2 tumors.

Therapeutic options such as brachytherapy and focused radiation may obviate the need for enucleation in smaller tumors. For those who undergo enucleation, long-term surveillance includes monitoring liver enzymes as well as hepatic imaging, as the liver is the metastatic site first discovered in 80% of cases. Approximately half of patients with uveal melanoma develop metastatic disease, and more than half of patients who have metastatic uveal melanoma die within one year.

• Comprised of the iris, ciliary body, and choroid, the uvea contains dendritic pigmented melanocytes which have the potential to give rise to malignant melanoma.
• The modified Callender classification is a system used to stratify metastatic risk for uveal melanoma based on the preponderance of morphologic cell type (spindle type A, spindle type B, and epithelioid).
• Largest tumor diameter, the presence of macronucleoli, elevated mitotic count, looping vascular networks, lymphocytic infiltration, and extrascleral extension are additional histologic features in uveal melanoma that have been associated with worse prognosis.
• Early mutations in BRAF characterize cutaneous melanoma while mutations in GNAQ and GNA11 are typical of uveal melanomas.
• Gene expression profiling has emerged as a strong indicator of prognosis, with uveal melanomas naturally clustering into two groups. Class 1 melanomas tend not to metastasize, while class 2 melanomas have a high risk of metastasis.

1. Bronkhorst IH, Vu TH, Jordanova ES, et al. Different subsets of tumor-infiltrating lymphocytes correlate with macrophage influx and monosomy 3 in uveal melanoma. Invest Ophthalmol Vis Sci. 2012;53:5370-78.
2. Eagle RC Jr. Eye Pathology: An Atlas and Text. 2nd ed. Lippincott Williams & Wilkins; 2011:178-88.
3. Field MG, Decatur CL, Kurtenbach S, et al. PRAME as an independent biomarker for metastasis in uveal melanoma. Clin Cancer Res. 2016;22:1234-42.
4. Shoushtari AN, Carvaial RD. GNAQ and GNA11 mutations in uveal melanoma. Melanoma Res. 2014 Dec;24(6):525-34.
5. Singh AD, Turell ME, Topham AK. Uveal melanoma: trends in incidence, treatment, and survival. Ophthalmology. 2011;118:1881-85.