Efforts to improve treatments for melanoma of the eye

Although primary uveal melanoma can be effectively treated with radiation or surgery, patients with metastatic disease — or those at high risk for metastases — have few proven options. But specialists in medical oncology, radiation oncology, ophthalmology, surgery and translational research are working to improve those options.

Uveal melanoma is rare and accounts for only about 5% of melanoma cases; however, metastatic disease from uveal melanoma is often fatal because the most common metastatic site is the liver. Even among patients who have no evidence of disease after treatment of the primary tumor, metastatic disease will occur in about 25% within five years and 50% at some point in their lives.

In an attempt to minimize the risk of death from metastatic disease in patients with uveal melanoma, physicians and scientists at MD Anderson Cancer Center are pursuing a multidisciplinary strategy that includes treating the primary tumor, identifying patients at high risk of metastatic disease and providing them with adjuvant therapy or increased surveillance, and exploring new treatments for metastatic disease.

Treating the primary tumor

Treatment of the primary tumor requires surgery, i.e., the removal of the eye, in about 30% of patients with uveal melanoma. Most patients can instead be treated with brachytherapy, in which a radioactive plaque is implanted in the patient and remains in place for two to four days, depending on the size of the tumor.

According to Beth Beadle, M.D., an assistant professor of Radiation Oncology, brachytherapy achieves local control in approximately 90% of patients with uveal melanoma in whom it is used. Iodine 125 is the most commonly used isotope in brachytherapy for uveal melanoma.

Another isotope used to treat the disease is ruthenium 106, which was available in the United States from 2003 until 2007, when it became unavailable on the U.S. market for commercial reasons. However, ruthenium 106 has remained available in other countries and is widely used in Europe.

“Ruthenium gives a very high radiation dose to a very small area — even more so than iodine — so the toxicity to normal tissues tends to be less,” Beadle said.

Ruthenium 106 recently became available again in the U.S., and MD Anderson was among the first institutions in the nation to resume its use. The decision to reinstate its use at MD Anderson was based on a review of 40 patients treated with ruthenium 106 plaque brachytherapy for uveal melanoma at the institution between 2003 and 2007. In the review, Beadle and her colleagues found that ruthenium 106 achieved local control equivalent to that of iodine 125, but with fewer toxic effects.

“Because of the narrow depth of penetration of the radiation, we can only use ruthenium for very small lesions, those less than 5 mm in depth; but that describes the majority of lesions we see,” Beadle said. “Iodine is still a very good treatment, but for patients with very small tumors, ruthenium seems to be even better.”

Tumors diagnosed early are likely to be small enough to be treated with ruthenium 106. “If we catch these tumors early, we can salvage the globe and also offer the patients a treatment that causes less toxicity to the eye,” said Dan Gombos, M.D., a professor of Head and Neck Surgery and chief of Ophthalmology. While early diagnosis increases treatment options for the primary tumor, whether early treatment decreases a patient’s chance of developing metastatic disease is unknown.

Assessing the risk of metastases

“There are two schools of thought about metastases from uveal melanoma,” Beadle said. “One is that the longer the patient has uncontrolled local disease, the greater the opportunity for it to spread; and the other is that metastatic disease is determined by the tumor’s biology and isn’t affected by what is done to control local disease.”

The effect of tumor biology on the development of uveal melanoma metastases has been elucidated in recent years. A commercially available test (DecisionDx-UM, Castle Biosciences) analyzes tumor RNA for a group of gene mutations associated with a high risk of uveal melanoma metastasis. This gene expression profile is used to classify a patent’s risk of developing metastases after successful treatment of the primary tumor.

“About 50% of the patients identified by the test as having high-risk disease go on to develop metastases by three years after treatment, and more than 70% develop metastases by five years,” said Sapna Patel, M.D., assistant professor of Melanoma Medical Oncology.

One limitation of the test is that it requires a tumor sample, which in most patients must be obtained by a needle biopsy. Because uveal melanoma is diagnosed on the basis of clinical features, a biopsy historically was considered unnecessary.

“Twenty years ago, we never biopsied these tumors,” Gombos said, adding that this philosophy is changing and that many ocular oncologists and ophthalmologists perform a needle biopsy before sewing on the radiation plaque. “There was a concern that a needle biopsy might increase the risk of disease spread, but the consensus now is that this risk is exceptionally low.”

Another limitation of the test is that physicians have few options for patients whose tumors are identified as high risk. “This test is highly prognostic in terms of telling people they’re going to have a bad outcome,” Patel said. “The problem is that in uveal melanoma, not only is there no standard of care for metastatic disease, there are no proven effective adjuvant therapies.”

Toward effective adjuvant therapy

After local control of the primary tumor has been achieved, all uveal melanoma patients at MD Anderson — especially those whose gene expression profile indicates a high risk of metastasis — are referred to Patel or one of her colleagues in the Department of Melanoma Medical Oncology for a consultation. The oncologists offer the options of surveillance (the standard of care) or experimental or off-protocol adjuvant therapy aimed at destroying micrometastases before clinically detectable metastatic disease can develop in the liver or elsewhere.

Patel was the principal investigator for a clinical trial of adjuvant therapy for patients who had been treated for primary uveal melanoma and had a high risk of metastases. Patients in the study were given ipilimumab to boost the immune system’s surveillance for cancer. The trial is no longer enrolling patients because of funding issues.

For patients with a high risk of metastasis after primary treatment of uveal melanoma, the off-label use of ipilimumab for adjuvant therapy is generally unfeasible.

“Right now, ipilimumab is one of the most expensive agents in cancer medicine, and it is not approved for adjuvant therapy,” Patel said. “Insurance companies would pay for treatment if a patient has metastatic disease, but these patients don’t have metastatic disease; they are at high risk of developing metastasis.”

However, other off-protocol adjuvant approaches often are used to treat such patients.

Because metastatic uveal melanoma most commonly emerges in the liver, one off-protocol approach is liver-directed therapy. “The idea is that there might be micrometastases in the liver that we can’t detect on a computed tomography scan. The tumor burden we suspect would be extremely small, so we bathe the liver in chemotherapy drugs to wipe out the tumors,” Patel said.

In liver-directed therapy, an interventional radiologist inserts a catheter through the groin into the hepatic circulation to deliver chemotherapeutic agents. Treatments alternate so that one round of chemotherapy infuses the right side of the liver and the next round infuses the left side.

Another approach to adjuvant therapy is to restore BAP1 protein function with histone deacetylase inhibitors such as vorinostat; inactivating mutations in the BAP1 tumor suppressor gene are common in uveal melanoma and indicate a high risk of metastasis. Patel said that valproic acid, which is commonly used to treat seizures, also inhibits histone deacetylase and is inexpensive. However, she added, “It’s not clear what dose of valproic acid would be needed to prevent uveal melanoma metastases via histone acetylation.”

Treating metastatic disease

As with adjuvant treatment for uveal melanoma, much is yet to be learned about treatment for metastatic disease. “Once a patient has metastases in the liver, it’s very challenging to cure the disease,” Gombos said. “There are some classes of drugs that we are excited about, but we don’t have a proven treatment for metastatic disease.”

In a phase II trial under way at MD Anderson and other centers, patients with metastatic uveal melanoma are randomly assigned to receive the MEK inhibitor trametinib with or without the AKT inhibitor GSK2141795. “The MEK and AKT pathways seem complementary in driving metastatic disease, so blockade of both should be important,” Patel said.

A clinical trial of an anti–PD-L1 antibody for patients with metastatic uveal melanoma recently completed enrollment, but more trials dedicated to metastatic uveal melanoma will open soon at MD Anderson. Also, patients with metastases from uveal melanoma may be eligible for several open studies enrolling patients with various cancer types.

Ongoing research

In addition to clinical studies, basic science and translational research play important roles in the search for ways to prevent or cure metastasis from uveal melanoma.

Although anti-CTLA-4, anti-PD-1 and anti-PD-L1 antibodies have extended survival durations for patients with metastatic cutaneous melanoma (see New Approaches Revolutionize the Treatment of Advanced Melanoma), the effects of these drugs in uveal melanoma patients are not as well known.

Elizabeth Grimm, Ph.D., professor of Melanoma Medical Oncology, explained that uveal melanoma does not always respond to the agents that are effective against cutaneous melanoma because the two diseases are biologically distinct. “To date, there are no known common mutations between uveal and cutaneous melanoma,” Grimm said. “Uveal melanoma has its own subset of unique genomic characteristics, notably chromosome 3 monosomy.”

Understanding uveal melanoma’s genomic characteristics and finding agents that can exploit them is a major goal for Grimm and other investigators in the Melanoma Medical Oncology. For example, Scott Woodman, M.D.,Ph.D., and Chandrani Chattopadhyay, Ph.D., are planning a major drug screening — a template of 6,000. Chattopadhyay is investigating the effects of hepatocyte growth factor and insulin-like growth factor receptor on liver metastases from uveal melanoma in vitro.

“These specific growth factors and receptors cause things to grow in the liver, so we’re studying primary uveal melanomas and liver metastases for their dependence on these factors,” Grimm said.

For the projects described above and others, Grimm and her colleagues, including Bita Esmaeli, M.D., professor of Plastic Surgery, have been collecting blood and tumor samples from uveal melanoma patients for 15 years. Many of these tumors have been submitted to The Cancer Genome Atlas for analysis.

Grimm said the rarity of uveal melanoma and the small size of biopsy samples limit researchers’ resources, but she and her colleagues overcome these challenges through multi-institutional collaboration.

“We have a worldwide network of laboratories,” she said. “It’s a very active field. We’re one of the leading centers for uveal melanoma, but we collaborate with centers around the world.”

Anticipating a new era

Although currently physicians have limited options to offer patients with metastatic uveal melanoma or a high risk of developing metastases, Gombos is optimistic. He said, “I think we’re at the crossroads of a new era in uveal melanoma where we can begin to offer directed therapy with better options than in the past. MD Anderson has assembled a truly unique multidisciplinary team of clinicians and researchers at the forefront of this rare ocular malignancy.”

This article originally appeared in the November/December 2014 issue of OncoLog.