Laird Harrison

September 14, 2017

BARCELONA — Not long ago, a diagnosis of uveal melanoma meant losing an eye, if the patient was lucky. In the worst cases, it meant chemotherapy followed by a painful death as the cancer spread. But now, specialists report that everything is changing, and the innovations are poised to save the eyesight and lives of many.

Trials of genetic testing, laser ablation, a first-in-class targeted light-activation therapy, and immunotherapy have all delivered striking preliminary results. "There's a lot going on in oncology," said Carol Shields, MD, from Thomas Jefferson University in Philadelphia.

Some of these treatments could enable retina specialists to take on uveal melanoma in their own offices, rather than referring patients to oncologists as they do now.

The cancer has long bedeviled even the best ocular oncologists. About 7% of white adults have a choroidal nevus, and about 1 in 8000 of these progresses to melanoma. Often, by the time uveal melanoma is detected, it has already spread, with related cancer cells typically appearing in the liver.

"The difficulty is differentiating these suspicious lesions," said Rana'a Al-Jamal, MD, PhD, from Moorfields Eye Hospital in London. "And when we are in doubt, the difficult question is whether we should treat or wait."

If a nevus has three or more risk factors, it should be treated, or at least followed very closely, Dr Al-Jamal told Medscape Medical News. "If we don't treat it and it turns out to be melanoma, the risk of metastases is higher," he said here at the 17th EURETINA Congress, where he presented research on the treatment of small choroidal lesions.

In 1991, most doctors chose to err on the side of extreme caution when treating uveal melanoma. "The standard of care was enucleation," said Timothy Murray, MD, from Murray Ocular Oncology and Retina in Miami.

But patients who have undergone treatment are still at risk. "There are patients who have been enucleated and then, 20 years later, get metastatic disease," he reported. "We know it's not from the eye. We think you can have small cells elsewhere in the body, and something triggers them to become more active."

Preserving Eyes

Since 1991, brachytherapy has gradually become ascendant, and can be used to destroy lesions before they metastasize, preserving eyes. However, the radiation typically causes some vision loss, and vascular endothelial growth-factor (VEGF) inhibitors — originally developed as a cancer treatment but ineffective in the treatment of uveal cancer — are only somewhat helpful in ameliorating this adverse effect, Dr Murray explained.

VEGF inhibitors do not have to be administered by an oncologist. "When I started doing this, there was not even a radiation specialist who would do it," he told Medscape Medical News. "It was off label. But everything we do in our field is off label. Now, probably at least half of my patients can get treated by their local retinal specialist. They come back to me every 4 months instead of every 4 weeks."

In other research, genetic tests have been developed that use biopsy tissue to determine which tumors are most likely to metastasize.

One RNA gene-expression profiling test (DecisionDx, Castle Biosciences) identifies tumors as class 1A, which have a 2% chance for metastases, class 1B, which have a 21% chance for metastases, and class 2, which have a 72% chance for metastases.

And cytogenetics can be used to analyze abnormalities in chromosomes. "Chromosomes 3 and 8 are really the bad guys. If they are mutated, the patient is in trouble," Dr Shields pointed out.

This test quantifies risk well, said Dr Shields, who presented results of a study on the cytogenetic abnormalities of uveal melanoma at the recent American Society of Retina Specialists (ASRS) annual meeting (Ophthalmology. 2017;124:609-618).

She said she is able to tell patients, on the basis of the number of mutations, whether their risk for metastases is, for example, 11-fold or 120-fold.

Dr Murray asserted that cytogenetics requires a larger sample, and that the results are more often inconclusive, but Dr Shields said she disputes his figures.

In children, evidence is emerging that uveal melanoma has a different genetic profile, Dr Al-Jamal reported. Some of these tumors are associated with a mutation in the BAP1 gene, which is a tumor-suppressor gene, he explained.

However, although uveal melanoma is not as common in children as it is in adults, diagnosis and treatment are the same, Norbert Bornfeld, MD, from the University of Essen in Germany, said during a EURETINA session.

Genetic Insight

In most cases, the genetic information is only being used for research or to help patients with life planning. And some patients would rather not know the probability that their tumor is going to metastasize, said Dr Al-Jamal.

But Dr Murray said he sees another application. He described, at the ASRS annual meeting, how he used RNA classification in a series of 198 patients to offer a novel treatment.

At baseline, 192 of the patients had exudative focal retinal detachment and 112 had growing tumors. Mean visual acuity in the cohort was 20/100, and mean apical tumor height was 1.9 mm.

In all these patients, Dr Murray performed a pars plana vitrectomy, then removed the hyaloid membrane and the vitreoretinal traction overlying the tumor. He ablated the tumors with direct endolaser confluent treatment, passed a 25-gauge needle through the tumor multiple times to collect a sample for a biopsy, inspected for bleeding, and then injected intravitreal triamcinolone acetonide.

At a mean follow-up of 6 months, mean visual acuity in the cohort had improved to 20/40, and at 43 months, it was significantly better, at 20/30 (P < .02).

These are incredibly exciting early data.

Gene-expression profiling revealed that 166 of the tumors were class 1A, 19 were class 1B, and 10 were class 2; three tumors could not be classified. Dr Murray treated the 10 class 2 tumors with plaque brachytherapy.

At 43 months, mean apical height had shrunk to 1.4 mm, and only five of the patients had progressive retinal detachment. None of the tumors metastasized. Local tumor control was established in 195 of the patients. There were no infections and only five inflammatory events. "These are incredibly exciting early data," said Dr Murray.

Results in the 10 eyes with class 2 tumors were particularly striking, because 72% of class 2 tumors metastasize in 5 years, he pointed out.

This approach could put the care of most uveal melanomas in the hands of retina specialists. "The treatments I'm talking about could potentially change the field because every excellent retinal surgeon does what I do on a daily basis during other surgical procedures," he explained.

The approach is crying out for a randomized controlled trial, he acknowledged, but so far he has been unable to attract a funder for such an expensive undertaking.

In the meantime, one of the other therapies being examined is AU-011, from Aura Bioscience, in which chromophores attach to particles of virus capsules that bind selectively to heparin sulfate on melanoma cells. When activated with an ophthalmic laser, the particles heat up enough to destroy the cell membranes.

"In studies of rabbits and dogs, it led to complete necrosis of the tumor with minimal inflammation," said Dr Shields, who is helping to test the drug. In a safety trial this year, six humans suffered only mild posterior and anterior uveitis, which resolved with time, she reported.

At 3-month follow-up, all six patients were still faring well, although it is too early to judge any effects on their vision, she added. The researchers are planning to launch a larger trial this month.

"We're all very excited for this," said Dr Shields. "If the safety profile is good with this medication and if it can cause tumor regression, we may be able to treat small melanomas, or even some medium melanomas. I foresee a time in the future when we treat suspicious or borderline nevi rather than waiting for them to grow."

Other possible therapies for metastatic uveal melanoma are also emerging. One group of researchers has tried vaccinating patients with autologous dendritic cells loaded with melanoma antigens gp100 and tyrosinase. In a preliminary study, patients who showed tumor-specific immune responses lived longer than those who did not (J Immunother Cancer. 2015;3[Suppl 2]:P127).

Another type of immunotherapy is IMCgp100 from Immunocore. Dr Shields and other researchers have infused IMCgp100, which hooks onto the uveal melanoma human leukocyte antigen receptor CD3, in 20 patients. The protein resembles an antigen-presenting cell and attracts T-cells. "It basically marries the T-cell to the melanoma cell, and the T-cell gobbles it up," Dr Shields explained.

Two patients treated with IMCgp100 died within a year, but in some of the other patients, "the response is jaw-dropping," said Dr Shields. She described one patient whose multiple liver tumors have disappeared.

Dr Shields said she can envision a day when this treatment is used in melanoma that is not metastatic, although that will require years of safety data.

There have been speed bumps along the road. Checkpoint inhibitors, which strip cancer cells of some of their defenses against T-cells, have achieved a 50% response in skin melanoma, but they have only reached about 4% in uveal melanoma, she reported.

Still, the positive results seen with many therapies underscore the importance of identifying suspicious nevi early and referring patients to specialists for further investigation.

"The gold standard is a dilated fundus exam," said Dr Murray. "Looking at the eye is most important, and mainly the posterior pole. If you see any type of elevated lesion, if it's anything but flat, you're mandated to image it, and that typically means wide-field photography and ultrasound."

If the diagnosis comes back positive, at least the news won't have the ring of doom that it might have had a few years ago. With a little luck and persistence, the patient might even get into a trial of one of the new treatments, Dr Shield said.

Dr Shields reports serving on the scientific board of Aura Biosciences. Dr Al-Jamal and Dr Bornfeld have disclosed no relevant financial relationships. Dr Murray reports being a consultant for Alcon Surgical, Regeneron, and NCI.

Follow Medscape on Twitter @Medscape and Laird Harrison @LairdH


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.