Improving the prognosis for uveal melanoma

Results of a commercially available, genetic profiling test for uveal melanoma can be used to determine patients at high risk for metastasis. Understanding of the genetics of the tumor is also being applied to develop targeted therapies that may improve patient survival.

Take Home

Results of a commercially available, genetic profiling test for uveal melanoma can be used to determine patients at high risk for metastasis. Understanding of the genetics of the tumor is also being applied to develop targeted therapies that may improve patient survival.

Dr. Harbour

By Cheryl Guttman Krader; Reviewed by J. William Harbour, MD 

Miami, FL-Despite advances in the diagnosis of uveal melanoma and treatment for localized disease, the cancer-related mortality rate has remained largely unchanged. However, that situation will hopefully change thanks to recent developments in understanding the genetics of uveal melanoma, said J. William Harbour, MD.

“Gene expression profiling that divides uveal melanoma into 2 molecular subgroups (class 1 and 2) allows us to accurately predict which patients with uveal melanoma will develop metastatic disease, and that technology is now available as a routine clinical test (DecisionDX-UM, Castle Biosciences),” said Dr. Harbour, professor of ophthalmology and Mark J. Daily, MD, Endowed Chair, Bascom Palmer Eye Institute, University of Miami Miller School of Medicine, Miami, FL.

“The genetic information also holds the key to developing effective targeted therapies for patients with metastatic disease and those at high-risk,” he continued. “Currently, many such treatments are being evaluated in clinical trials, and I would expect the number of investigational agents will expand rapidly in the next several years.”


Construction of the mutational landscape of uveal melanoma began in 2010, and within the few ensuing years thereafter, 5 major driver mutations were identified.

“Now, after having sequenced the whole exome of many tumors, the number of new mutations being found is rapidly diminishing. Therefore, it seems we are coming into clearer focus about the mutational landscape of uveal melanoma that will allow us to develop an approach to targeted therapy,” Dr. Harbour said.

Mutations in two closely related genes-GNAQ and GNA11-were the first uveal melanoma mutations identified. One or the other of these genes, which encode two closely related G-coupled protein receptor subunits, are mutated in almost all uveal melanomas (up to 90%) as an early event, resulting in MAP kinase pathway activation that promotes tumor growth and proliferation. However, the GNAQ/GNA11 mutations are not associated with metastasis.

About 40% of uveal melanomas develop a mutation in BAP1 (BRCA associated protein 1) that confers a high risk of metastasis and is associated with monosomy 3 (mutation on one copy of chromosome 3 with loss of the second, tumor suppressor copy).  BAP1 is a deubiquitinase enzyme encoding a protein involved in epigenetic regulation, stem cell biology, and cell proliferation. Loss of BAP1 causes uveal melaonoma cells to behave like cancer stem cells and become metastatic.


As an aside, Dr. Harbour noted that the unexpected finding of a germline BAP1 mutation in an uveal melanoma patient led to further investigations identifying germline BAP1 mutations in several families with uveal melanoma and cutaneous melanoma, gastric cancer, mesothelioma, and renal cell carcinoma.

“A growing number of papers are describing a BAP1 familial cancer syndrome, and we now routinely ask patients about a personal or family history of these other cancers,” he said. “We predict that about 2% to 3% of uveal melanoma patients carry a germline BAP1 mutation, and we have gene testing available to offer those who are concerned.”

Tumors without the BAP1 mutation can continue to enlarge and develop some ability to metastasize. Exome sequencing has identified two mutations associated with a good prognosis for these tumors: one involving SF3B1, a splicing factor, and the other eukaryotic translation initiation factor 1A, x-linked (EIF1AX), a translational initiation factor. These mutations are mutually exclusive with BAP1 mutations, and their biological significance is under investigation, Dr. Harbour said.


Targeted treatments

Genetic profiling of uveal melanomas has provided a foundation for investigating targeted therapies. Although there are no treatments that directly target the GNAQ/GNA11 mutations, commercially available and investigational kinase inhibitors that affect the downstream proteins activated by these mutations are being studied, and in fact, selumetinib (AstraZeneca) was the first drug ever reported to have efficacy in treating advanced metastatic uveal melanoma, Dr. Harbour said.

As class 1 uveal melanomas (ie., tumors without the BAP1 mutation) retain expression of melanocytic antigens as they enlarge, the tumor cells may be recognized by the immune system if metastasis occurs. Therefore, metastatic class 1 uveal melanomas may respond to immunotherapy, and recently some major cancer centers have reported on the efficacy of ipilimumab (Yervoy, Bristol-Myers) for these tumors, Dr. Harbour said.   

Available epigenetic modulators-ie., agents that inhibit the activity of histone deacetylase or DNA methyl transferase-may reverse the effects of BAP1 mutations. There are a number of clinical trials underway investigating histone deacetylase inhibitors, including valproic acid and vorinostat (Zolinza, Merck), as treatment for patients with class 2 uveal melanoma who have developed metastases or as an adjuvant in those at high-risk.



J. William Harbour, MD


Dr. Harbour is the inventor of a commercially available gene expression profile for uveal melanoma (Castle Biosciences), receives royalties from its commercialization and is a paid consultant to Castle Biosciences, licensee of the intellectual property.