Melanoma, Uveal


A number sign (#) is used with this entry because mutations in several genes may underlie susceptibility to uveal melanoma. Susceptibility loci have been mapped to chromosome 3q (UVM1; 606660) and chromosome 3p (UVM2; 606661).

See also 614327 for a tumor predisposition syndrome that may include the development of uveal melanoma and is caused by germline mutation in the BAP1 gene (603089) on chromosome 3p21.

Clinical Features

Uveal melanoma is the most common primary intraocular malignancy. Canning and Hungerford (1988) found reports of 12 families with adequate documentation of the occurrence of uveal melanomas in 2 or more members. They described 2 more families, each with 2 affected members: a brother and sister who presented at ages 69 and 63 years, respectively, and a mother and son who presented at ages 33 and 20, respectively. Young et al. (1994) reported 11 families in which 2 or more members had uveal melanoma. The median age at diagnosis was 56.5 years, similar to the median age at diagnosis of the other patients treated at their institution. Of the 24 affected persons, 13 were male and 11 were female.

In a review of medical charts of 4,500 patients with uveal melanoma, Singh et al. (1996) found 56 patients in 27 families with a family history. The uveal melanoma was unilateral in all 56 familial cases. In 17 cases (63%), the second affected relative was a first-degree relative; second-degree relatives accounted for 22% and third-degree relatives for 15%. In 25 families (93%), only 2 members were affected, and in 2 families (7%), 3 members had uveal melanoma. Patients with familial uveal melanoma were 4 times as likely to have a second primary malignant neoplasm than persons in the general population; however, unaffected relatives of these patients were apparently not at higher risk of having another primary malignant neoplasm. The authors stated that the occurrence of familial uveal melanoma is rare, comprising only 0.6% of all uveal melanoma patients.

Singh and Topham (2003) determined the incidence of primary uveal melanoma in the United States over a 25-year period (1973-1997). They found 2,493 cases of uveal melanoma, representing 2.9% of all recorded cases of melanoma. The mean age-adjusted incidence was 4.3 per million. Most cases (97.8%) occurred in the white population. There was a significant variation of incidence between genders (males, 4.9; females, 3.7). There was no significant variation of incidence by the geographic location of the registry. The mean age-adjusted incidence of uveal melanoma in the United States (4.3 per million) was similar to that reported from European countries. The age-adjusted incidence rate of uveal melanoma had remained stable for over the 25 years.

Singh et al. (2000) reported a man who presented with choroidal melanoma at the age of 41 and his 68-year-old father who presented with the same condition 6 months after his son. Using principles of probability, they stated that the likelihood of familial occurrence of uveal melanoma was approximately 1 in 10 million.

Li et al. (2000) reviewed 1,848 consecutive primary choroidal and/or ciliary body melanoma patients treated with proton beam irradiation. They concluded that patients with melanomas of presumed ciliary body origin are subject to a higher risk of death resulting from melanoma metastasis.

Al-Jamal et al. (2003) reported that the mean of the 10 largest nucleoli (MLN) was an independent predictor of survival in patients with uveal melanoma, when adjusting in turn for the presence of epithelioid cells, loops and networks, and microvascular density (MVD). The authors concluded that MLN and microvascular loops and networks were unrelated, independent predictors of survival. MLN and MVD were found to be partially interrelated. Multivariate models that included MVD in addition to MLN fitted better with observed melanoma-specific survival than models that excluded MVD.

Subretinal fluid is a strong risk factor for growth of choroidal melanocytic tumors. Espinoza et al. (2004) found that optical coherence tomography was useful in distinguishing active subretinal fluid from chronic retinal changes overlying a choroidal melanocytic tumor and might be of predictive value in identifying tumors that are likely to grow and require treatment.

Toivonen et al. (2004) investigated the relationship between progression to hepatic metastasis and tumor-infiltrating macrophages and microcirculation attributes in uveal melanoma, a cancer that almost invariably disseminates hematogenously to the liver. The authors found that the hepatic metastases had a significantly lower grade of pigmentation, more frequent epithelioid cells, more intermediate and dendritic types of CD68 (153634)-immunopositive macrophages than round ones, and a higher MVD than the primary uveal melanomas that spawned the metastases. Patient survival after diagnosis of disseminated disease tended to be shorter if hepatic metastases had a high MVD.

Shields et al. (2004) reported the clinical variation and natural course of optic disc melanocytoma in 115 patients. Although optic disc melanocytoma is generally considered to be a benign, stationary lesion, they found that it can produce several local complications, can cause visual loss, can grow slowly, and can, rarely, undergo malignant transformation into melanoma. Shields et al. (2004) concluded that patients with optic disc melanocytoma should undergo periodic ocular examination.

Zografos et al. (2004) reviewed 37 cases of optic disc melanocytoma in European patients. Tumor growth was demonstrated in 6 of 9 patients followed for at least 6 years. In 2 cases, presumed malignant transformation required treatment. Zografos et al. (2004) concluded that the risk of melanocytoma progression, although sometimes occurring as much as several years after the initial diagnosis, justifies a cautious approach with long-term regular surveillance.

Elner et al. (2004) reviewed the clinical and echographic characteristics of a chronic, inactive choroidal nevus and a malignant choroidal melanoma. They concluded that documented growth is not an unequivocal indicator of malignant transformation for small melanocytic tumors. Elner et al. (2004) stated that the Collaborative Ocular Melanoma Study Group (1990) reported histopathologic examination of 413 eyes with a clinical diagnosis of choroidal melanoma and a misdiagnosis rate of 0.48%. All of these melanomas had a height greater than 2.5 mm, and Elner et al. (2004) noted that with lesions less than 2 mm in height, it is more difficult to distinguish a nevus from a melanoma.

Hadden and Damato (2003) reported the occurrence of 2 entirely separate choroidal melanomas in the same eye of 1 patient, a 30-year-old woman. The first developed in her left eye. It was successfully treated with proton beam radiotherapy. One year later, she developed a pigmented lesion on her right calf. Excision biopsy revealed melanoma-in-situ. Four years later, a small choroidal tumor was noted in her left eye and was presumed to be a nevus. It was observed for 1 year. Because of rapid growth, choroidal melanoma was diagnosed. This second choroidal melanoma was treated with a ruthenium plaque. Six years later, both tumors had continued to regress and the patient remained in good health. Only 19 cases of multifocal uveal melanoma had been reported previously. Hadden and Damato (2003) concluded that a second uveal melanoma in the same eye suggested a genetic predisposition.

Smith et al. (2007) evaluated a kindred in which several members had both uveal and cutaneous (see 155600) melanomas. The proband and his mother had uveal melanoma, 3 cutaneous melanomas occurred between 2 sibs, and 2 other sibs had basal cell carcinomas. No germline mutations were detected in the melanoma-associated tumor suppressor genes p16(INK4A) (600160) and p14(ARF) (see 600160). Seven of 10 sibs had a history of cutaneous and/or ocular nevi. Of the 3 subjects without nevi, 2 had histories of eye or skin malignancies (1 uveal melanoma and 1 basal cell carcinoma). Three of the 10 sibs had relevant ocular findings (2 choroidal nevi and 1 uveal melanoma). Six were also found to be in the 'high-risk' classification for cutaneous malignancies. Smith et al. (2007) concluded that their results strengthened the association between uveal melanoma, atypical nevi, and cutaneous melanoma.


Goto et al. (2001) reported that single photon emission computed tomography (SPECT) images using a radiopharmaceutical was both sensitive and specific for the diagnosis of uveal melanoma. Eight of 20 eyes showed high accumulation of the radiopharmaceutical in the late phase in the area corresponding to the uveal tumor, and the diagnosis of uveal melanoma was confirmed histologically in 7 enucleated eyes. The 12 patients with no accumulation were found to have iris nevus, choroidal nevus, or other intraocular or adnexal conditions, clinically or histologically.

Clinical Management

Singh et al. (2000) reported the occurrence of retinoblastoma and uveal melanoma in the same patient. The patient had had enucleation of her right eye at age 3.5 years for unilateral sporadic retinoblastoma. Fifty years later, in 1990, she developed a visual field defect in her remaining left eye and was diagnosed with choroidal nevus. Barrier photocoagulation was performed in 1995 to limit subretinal fluid leakage. Because of growth of the lesion, choroidal melanoma was diagnosed in 1997. Transpupillary thermotherapy enhanced with indocyanine green was performed in an attempt to treat the tumor without risk of visual loss in her 20/20 left eye. Because of continued tumor growth, iodine-125 radioactive plaque was applied. At the 1-year follow-up, the patient had a visual acuity of 20/40 and the tumor had regressed. The authors calculated the probability of occurrence of unilateral retinoblastoma followed by uveal melanoma in the remaining eye at 1:165 billion.

The Collaborative Ocular Melanoma Study Group (2001) described metastatic disease status at death in patients with large choroidal melanoma. Of the 1,003 patients enrolled in the trial, 457 had died. Thus, estimated median survival from time of enrollment was 7.4 years. The authors determined disease status at time of death for 435 of 457 patients (95%). Sixty-two percent had histopathologically confirmed melanoma metastasis at the time of death. Metastasis was suspected in a further 21% on the basis of imaging or other tests but was not histopathologically confirmed. Common metastatic sites were liver (93%), lung (24%), and bone (16%). Multiple sites were identified in 87% of patients with metastasis. The likelihood of 3 or more sites increased more than 4-fold when autopsy results were available (6%). The authors also suggested guidelines for the evaluation of patients in clinical studies of choroidal melanoma.

Gombos et al. (2004) found significant differences among ocular oncologists in North America and Europe in the use of techniques to screen for metastatic uveal melanoma. North American members (from the United States and Canada) of the Collaborative Ocular Melanoma Study (COMS) centers relied primarily upon liver function tests and chest x-rays. The majority of European centers used liver ultrasonography. The authors urged ocular oncologists to design a cost-effective screening protocol for patients with uveal melanoma and to implement it uniformly.

Shields et al. (2002) evaluated tumor control and treatment complications following plaque radiotherapy combined with transpupillary thermotherapy for 270 consecutive choroidal melanomas. Prior to treatment, the median base of the tumor was 11 mm (range 4-21 mm) and the median thickness was 4 mm (range 2-9 mm). All patients received plaque radiotherapy (median dose was 9000 rad to the tumor apex). Transpupillary thermotherapy (median dose 700 mW) was applied in 3 sessions beginning at plaque removal and continuing at 4-month intervals. At 5 years, treatment-related complications included maculopathy in 18% of patients, papillopathy in 38%, macular retinal vascular obstruction in 18%, vitreous hemorrhage in 18%, rhegmatogenous retinal detachment in 2%, cataract in 6%, and neovascular glaucoma in 7%. Enucleation for radiation complications was necessary in 3 cases (1%). The authors concluded that their protocol provided excellent tumor control with only 3% recurrence at 5 years' follow-up.

Li et al. (2003) compared various methods of modeling the relationship between tumor dimension and metastatic death. They analyzed 1,204 consecutive patients with primary choroidal melanoma treated with proton beam irradiation. They found that tumor basal area was a better prognostic indicator than largest tumor diameter or tumor volume in the prediction of metastatic death after proton beam irradiation for uveal melanoma.

Shields et al. (2002) presented the results of primary transpupillary thermotherapy in 256 patients with newly diagnosed small choroidal melanomas. After a mean of 3 treatment sessions, complete tumor control without recurrence was found in 91% of patients, with recurrence in 9%. Mean time to recurrence was 22 months. Tumor-related mortality occurred in 1% of patients. The authors concluded that transpupillary thermotherapy is an effective treatment for certain small choroidal melanomas and that appropriate tumor selection is critical to successful treatment. They cautioned that patients with tumors abutting or overhanging the optic disc, or those requiring more than 3 sessions for tumor control are more likely to develop recurrence. They also observed that transpupillary thermotherapy can damage the retina, leading to visual loss shortly after treatment.

Krause et al. (2003) evaluated an infrared laser in the destruction of pigmented choroidal melanomas implanted in the subchoroidal space of rabbits. Twenty animals received a single treatment of focused 1047-nm light. The rate of complete tumor eradication was 91% in 10 of 11 animals receiving a dosage of 125 J/cm2. Continuous tumor growth was observed in all animals treated with collimated radiation and in untreated controls. Their data suggested that a single treatment with a focused, raster-scanned beam of 1047 nm might play a role in the destruction of pigmented choroidal melanomas.

Singh et al. (2003) reported clinicopathologic correlations in 10 eyes that required enucleation after transpupillary thermotherapy for choroidal melanoma. Indications for enucleation included recurrence in 5 of 10, progression in 3 of 10, retinal detachment in 1 of 10, and neovascular glaucoma in 1 of 10. Interval between transpupillary thermotherapy and enucleation ranged from 6 to 70 months (mean 19.4 months). Despite satisfactory ophthalmoscopic evidence of regression, intrascleral extension occurred in some patients. The authors recommended routine B-scan ultrasonography to detect extrascleral extension in patients with choroidal melanoma.

Dithmar et al. (2000) showed the usefulness of recombinant alpha-interferon (see 147562) for decreasing hepatic metastases from intraocular melanoma in a murine model. The authors concluded that adjuvant recombinant IFN alfa-2b treatment given before or at the time of enucleation might be a treatment option for patients with uveal melanoma with high-risk factors for developing metastatic disease.

Kiratli and Bilgic (2001) reported that, although clinically detectable calcification in choroidal melanomas is extremely rare, progressive calcification as well as tumor regression in a posterior uveal melanoma may occur after treatment with infrared diode laser transpupillary thermotherapy.

Singh et al. (2001) reported a 49-year-old patient who, 15 years after brachytherapy for left choroidal melanoma with excellent regression, developed a metastatic melanoma to the contralateral choroid as the first sign of metastasis. Systemic evaluation disclosed multiple metastases confined to the liver. The authors concluded that metastasis to the contralateral choroid may be the first sign of metastasis from choroidal melanoma.

Hadden et al. (2004) reviewed the histopathology of eyes enucleated after endoresection of choroidal melanoma. As with other treatments conserving the eye, these authors stressed that enucleation should be performed if adequate ocular examination is no longer possible, and that follow-up should be lifelong.

Kaiserman et al. (2004) evaluated trends in liver function tests (LFTs) before detection of liver metastases from uveal melanoma. At the time of diagnosis of liver metastases by imaging, 50% of patients had at least 1 abnormal LFT, compared with only 5% of the control group. Based on likelihood ratios, alkaline phosphatase and lactate dehydrogenase levels were the most predictive tests. The authors concluded that monitoring the changes in selected LFTs, even within normal limits, could help predict metastatic uveal melanoma.

Primary malignant melanoma of the choroid and ciliary body has traditionally been treated without histologic staging, using purely clinical indicators. The presence of extravascular matrix patterns (EMPs) in histologic sections of uveal melanoma is an independent indicator of metastatic risk (Folberg et al., 1993). Coleman et al. (2004) presented evidence suggesting that ultrasound can be used to detect noninvasively the presence of clinically significant prognostic features of uveal melanoma related to EMP patterns, and for patient stratification to different treatment modalities based on assignment to high- or low-risk groups.

The Collaborative Ocular Melanoma Study Group (2004) reported that at 10-year follow-up, no survival advantage was attributable to pre-enucleation radiation for large choroidal melanomas. They concluded that mortality rates by baseline characteristics should facilitate counseling of patients who have large choroidal melanoma and no evidence of metastasis or another malignancy at diagnosis.

Sisley et al. (2011) reviewed the background of GNAQ mutations in uveal melanoma and discussed them as a therapeutic target.

Biochemical Features

Endocrine influence on cutaneous and uveal melanomas has been a topic of interest since Pack and Scharnagel (1951) reported that the prognosis for cutaneous melanoma is worse in pregnant women. Grostern et al. (2001) evaluated choroid melanomas in enucleated eyes for the presence of estrogen receptor-1 (ESR1; 133430). No tumors showed immunohistochemical evidence of ESR1.

All-Ericsson et al. (2002) investigated the expression of insulin-like growth factor I receptor (IGF1R; 147370), focusing on its role in cell growth in uveal melanoma. Their data suggested a significant association between high IGF1R expression and death due to metastatic disease. Because IGF1R is produced mainly in the liver, the preferential site for uveal melanoma metastases, these results pointed to the possibility of interfering therapeutically with IGF1R in uveal melanoma that appears to follow an aggressive clinical course.

Apte et al. (2001) assayed 7 uveal melanoma cell lines for their ability to produce angiostatin (173350) in vitro. Three cell lines generated 38-Kd angiostatin molecules that significantly inhibited bovine endothelial cell proliferation in vitro. Enucleation of eyes containing angiostatin-producing cell lines exacerbated the metastatic potential of the uveal melanomas. In contrast, enucleation of eyes containing a nonangiostatin-producing cell line had no significant effect on liver metastasis. The authors concluded that, in some circumstances, enucleation of a melanoma-containing eye might unwittingly exacerbate the metastatic potential of uveal melanoma.

Ardjomand et al. (2003) investigated the distribution of somatostatin receptor subtypes 2 (SSTR2; 182452), 3 (SSTR3; 182453), and 5 (SSTR5; 182455) in uveal melanomas and their diagnostic and possible therapeutic value. All 25 uveal melanomas studied were positive for SSTR2: SSTR2A was expressed in 15 of 25; SSTR2B in 23 of 25; SSTR3 in 7 of 25; and SSTR5 in 13 of 25. A Kaplan-Meier survival curve showed a significantly better ad vitam prognosis for patients with tumors expressing high levels of SSTR2. Because a melanoma cell proliferation assay showed an inhibitory effect of up to 36% +/- 6% using octreotide or vapreotide, somatostatin analogs might be beneficial in the treatment of patients with ocular melanomas.

Clarijs et al. (2003) stated that, in both primary uveal and cutaneous melanoma, 9 different patterns of extracellular matrix deposition have been identified by conventional periodic acid-Schiff (PAS) staining, and these patterns appear to be of prognostic significance. In particular, the presence of PAS-positive arcs, loops, and network patterns has been associated with poor survival. Along with PAS-positive patterns, numerous macrophages are present, and their recruitment into tumor tissue is mediated by chemotactic cytokines. Clarijs et al. (2003) showed that, in uveal melanoma, macrophages accumulate at sites of endothelial monocyte-activating polypeptide-2 (EMAP2; 603605) expression. They hypothesized that the chemotaxis process is facilitated by EMAP2-dependent expression of ICAM1 (147840) on vascular endothelial cells and concomitantly leads to localized vascular damage, as indicated by release of von Willebrand factor (VWF; 613160).

Klisovic et al. (2003) reported that histone deacetylase inhibitors (HDACIs) such as depsipeptide inhibited primary and metastatic uveal melanoma cell growth in vitro. The apoptosis they observed was probably mediated through the Fas/FasL (134637, 134638) signaling pathway rather than via Bcl2 (151430).

Reiniger et al. (2007) evaluated osteopontin (OPN; 166490) as a potential marker for screening and detection of metastatic uveal melanoma. Mean plasma concentration of OPN in 8 patients with uveal melanoma without metastasis was 46.8 ng/ml. Increased median levels of OPN of 170.7 ng/ml were seen in 8 patients with proven metastatic disease. In healthy control subjects, median plasma OPN concentration was 54.6 ng/ml. Reiniger et al. (2007) suggested that OPN is a promising tumor marker.

Because of the potential immunogenicity of the melanocortin-1 receptor (MC1R; 155555), Lopez et al. (2007) evaluated its expression in uveal melanoma. Their results demonstrated that MC1R was expressed by uveal melanoma to a significantly greater extent than other melanoma markers. MC1R was found in 95% of melanoma tissues tested, including 1 liver metastasis. Even though MC1R was mainly located intracellularly, its cell surface expression could be promoted by cytokines, such as interferon-gamma (147570) and tumor necrosis factor-alpha (191160). The data supported MC1R as a new marker for the diagnosis of uveal melanoma and as a putative therapeutic target.

Other Features

Hurks et al. (2001) assayed 11 human uveal melanoma cell lines and 17 primary uveal melanomas for expression of HLA-G (142871), a nonclassical HLA class I molecule that is a critical mediator in the inhibition of natural killer (NK) cell-mediated cytolysis. Because none of the cell lines and none of the primary melanomas expressed HLA-G, the authors concluded that it is unlikely that HLA-G plays a role, direct or indirect, in the modulation of cellular immunity against uveal melanoma tumors.


In 19 patients with uveal melanoma, Mukai and Dryja (1985, 1986) sought evidence of homozygosity or hemizygosity for DNA polymorphisms that were heterozygous in the host. The rationale follows that which has been successful in the study of retinoblastoma and Wilms tumor. They found that 2 of the 15 informative patients had, in their tumors, lost alleles at loci on chromosome 2.

Using DNA markers, Prescher et al. (1992) found loss of chromosome 3 alleles and multiplication of 8q alleles in uveal melanoma. Cytogenetic analysis had previously shown monosomy of chromosome 3 and multiplication of chromosome 8q material as nonrandom chromosomal aberrations. Horsthemke et al. (1992) interpreted the findings as indicating that a tumor suppressor gene on chromosome 3 and an oncogene on chromosome 8 are involved in the formation or progression of uveal melanoma. In 4 of 10 posterior uveal melanomas, all derived from the ciliary body, Sisley et al. (1992) demonstrated monosomy 3 and i(8q). See also Prescher et al. (1994).

Tschentscher et al. (2001) stated that monosomy 3, an unusual finding in most tumors, is present in approximately 50% of uveal melanomas and is significantly correlated with metastatic disease. To obtain positional information on putative tumor suppressor genes on chromosome 3, they investigated tumors from 333 patients by comparative genomic hybridization, microsatellite analysis, or conventional karyotype analysis. A partial deletion of the long arm was found in 8 tumors, and the smallest region of deletion overlap (SRO) spanned 3q24-q26. They found 6 tumors with a partial deletion of the short arm and defined a second SRO of about 2.5 Mb in 3p25. This SRO did not overlap with the von Hippel-Lindau disease gene (608537). These findings suggested a role for 2 tumor suppressor genes (see 606660 and 606661) in metastasizing uveal melanoma and may explain the loss of an entire chromosome 3 in these tumors.

Molecular Genetics

Hearle et al. (2003) screened 385 patients with uveal melanomas for germline mutations in the BRCA2 (600185), p16(INK4A)/p14(ARF) (600160), and p15 (600431) genes. Their findings suggested that fewer than 2% of uveal melanoma cases could be ascribed to germline mutations in BRCA2, p16/p14, or p15. The authors concluded that it is likely that mutations in other genes contribute to an inherited predisposition to uveal melanoma.

Somatic Mutations

Maat et al. (2008) noted that mutations in the genes that control call proliferation in cutaneous melanoma are generally uncommon in uveal melanoma. However, using the very sensitive pyrophosphorolysis-activated polymerization (PAP) assay to screen for mutations in exon 15 of the BRAF gene (164757) in 11 uveal melanoma cell lines and 45 primary uveal melanomas, Maat et al. (2008) identified mutations in 2 cell lines (V600E; 164757.0001) and 6 primary tumors. Direct sequencing of the exon 15 PCR product did not reveal the mutations found with the PAP assay, indicating a low frequency of the mutant allele in primary samples. Maat et al. (2008) concluded that the relative scarcity of the BRAF mutations excluded an elemental role for them in uveal melanoma.

Van Raamsdonk et al. (2009) reported frequent somatic mutations in the heterotrimeric G protein alpha-subunit (GNAQ; 600998) in blue nevi (603670) (83%) and ocular melanoma of the uvea (46%). The mutations occurred exclusively in codon 209 in the Ras-like domain and resulted in constitutive activation, turning GNAQ into a dominant-acting oncogene. Van Raamsdonk et al. (2009) concluded that their results demonstrated an alternative route to MAP kinase activation in melanocytic neoplasia, providing new opportunities for therapeutic intervention.

Populo et al. (2011) identified the GNAQ Q209 mutation in 36% of 22 enucleated uveal melanomas. No associations were found between the presence of the GNAQ mutation and prognostic parameters, the expression of ERK1/2 (MAPK3, 601795/MAPK1, 176948), phosphorylated ERK1/2, and cell cycle markers. Populo et al. (2011) suggested that GNAQ-mutated uveal melanomas do not exhibit a higher deregulation of proliferation or higher activation of the MAP kinase signaling pathway than uveal melanomas without GNAQ activation.

Van Raamsdonk et al. (2010) identified somatic mutations affecting residue Q209 of the GNA11 gene (139313) in 7% of blue nevi, 32% of primary uveal melanomas, and 57% of uveal melanoma metastases. Mutations in the same codon (Q209) of the paralogous GNAQ gene were found in 55% of blue nevi, 45% of primary uveal melanomas, and 22% of uveal melanoma metastases. The sample group included a total of 713 melanocytic neoplasms. Sequencing of exon 4 of both these genes, affecting residue R183, in 453 melanocytic neoplasms showed a lower prevalence of mutations, in 2.1% of blue nevi and 4.9% of primary uveal melanomas. The mutations were mutually exclusive, except for a single tumor that carried mutations at both Q209 and R183 in GNA11. In total, 83% of all uveal melanomas examined had oncogenic mutations in either GNAQ or GNA11. Mice injected with cells transduced with the GNA11 Q209L mutation developed rapidly growing tumors and metastases, whereas injection with GNA11 R183C-transduced cells showed lesser potency. Western blot analysis of melanocytes transduced with GNA11 Q209L showed constitutive activation of the MAPK pathway. Although GNA11 mutations appeared to have a more potent effect on melanocytes than did GNAQ mutations, there was no difference in patient survival among those with GNA11 mutations compared to those with GNAQ mutations.

Harbour et al. (2010) used exome capture coupled with massively parallel sequencing to search for metastasis-related mutations in highly metastatic uveal melanomas of the eye. Inactivating somatic mutations were identified in the gene encoding BRCA1-associated protein-1 (BAP1; 603089) on chromosome 3p21.1 in 26 of 31 (84%) metastasizing tumors, including 15 mutations causing premature protein termination and 5 affecting its ubiquitin carboxyl-terminal hydrolase domain. One tumor harbored a frameshift mutation that was germline in origin, thus representing a susceptibility allele. Harbour et al. (2010) concluded that their findings implicated loss of BAP1 in uveal melanoma metastasis.

Wiesner et al. (2011) found somatic mutations in the BAP1 gene in 13 (40%) of 33 uveal melanomas. All the uveal melanomas with BAP1 mutations also carried mutations at codon 209 in either GNAQ or GNA11.

Harbour et al. (2013) described mutations occurring exclusively at codon 625 of the SF3B1 gene (605590) in low-grade uveal melanoma with good prognosis.

Using exome sequencing, Martin et al. (2013) identified recurrent somatic mutations in EIF1AX (300186) and SF3B1, specifically occurring in uveal melanomas with disomy 3, which rarely metastasize. Targeted resequencing showed that 24 of 31 tumors with disomy 3 (77%) had mutations in either EIF1AX (15; 48%) or SF3B1 (9; 29%). Mutations were infrequent (2 of 35; 5.7%) in uveal melanomas with monosomy 3, which are associated with poor prognosis. Resequencing of 13 uveal melanomas with partial monosomy 3 identified 8 tumors with a mutation in either SF3B1 (7; 54%) or EIF1AX (1; 8%). All EIF1AX mutations caused in-frame changes affecting the N terminus of the protein, whereas 17 of 19 SF3B1 mutations encoded an alteration of arg625. Resequencing of 10 uveal melanomas with disomy 3 that developed metastases identified SF3B1 mutations in 3 tumors, none of which targeted arg625.

Animal Model

Murine uveal melanoma cells express have been found to secrete Vegf both locally and at distant sites of metastases, and local VEGF secretion has been observed in human uveal melanoma. VEGF secretion stimulates angiogenesis and can enhance metastatic potential. Crosby et al. (2011) evaluated Vegf expression in sera from mice inoculated with intraocular melanoma and correlated this with the number and location of hepatic metastases. Serum Vegf levels rose after inoculation of C56BL/6 mice eyes with B16LS9 cutaneous melanoma cells which, similar to human uveal melanoma, express high levels of c-met (164860) and metastasize to the liver. Beginning on day 14, there was a statistically significant increase in Vegf levels. Peak serum Vegf levels correlated with the total number of hepatic micrometastases (R = 0.444), and there was moderate correlation of peak Vegf serum levels with micrometastases in more hypoxic locations (R = 0.572). Vegf mRNA expression by micrometastases was highest in the most hypoxic regions of the hepatic lobule. Crosby et al. (2011) concluded that serum VEGF may be clinically useful as an early marker of micrometastasis or progression.