This article is from October 2006 and may contain outdated material.
Ocular neoplasms are relatively uncommon. And yet many remain potentially sight- and life-threatening despite ongoing progress in early detection and management.
The most common intraocular primary malignancy is melanoma, with an estimated 2,300 new cases in 2006.1 “We’ve become more focused on the far periphery of the eye routinely during our exams, and as the technology to examine the eye has improved, we’re now picking up tumors at an earlier point,” said Timothy G. Murray, MD, professor of ophthalmology and radiation oncology at Bascom Palmer Eye Institute in Miami. And thanks to this earlier diagnosis, many patients are being monitored and/or treated sooner in the hopes of minimizing the risk of metastasis.
“There is really no good systemic control for melanoma, whether it’s on the skin or in the eye,” said Carol L. Shields, MD, professor of ophthalmology at Wills Eye Hospital and Thomas Jefferson University in Philadelphia. “Hence, we’re focusing our efforts on detecting small melanomas so we don’t have to deal with metastatic disease in the long run. Our goal is to find the thinnest, earliest melanomas we can.”
Even as tumors are being diagnosed earlier, there is plenty of discussion and debate among specialists as to when small-sized tumors should be treated, whether it’s safe to only watch them until growth occurs, and what the optimal treatment should be.
Catch That Neoplasm Now
Early detection of very small tumors “is where all the action is right now in the management of melanoma—specifically, how to detect these tumors not just when they’re small, but when they’re very tiny,” said Dr. Shields. Ocular melanomas are now initially identified at an average thickness of 3.5 mm, compared with 5.5 mm 15 to 20 years ago.
Not only is diagnosis occurring earlier in the natural course of choroidal melanoma, but the diagnostic process has become more accurate. “The good news in our field is that in the last 30 years, we’ve gone from removing as many as 20 percent of eyes thought to have melanoma—but turned out not to have it—to an accuracy diagnostic rate of more than 99 percent in most centers,” said David H. Abramson, MD, chief of the ophthalmic oncology service at Memorial Sloan-Kettering Cancer Center in New York. “In the last five years of the Collaborative Melanoma Study, which involved 45 centers in North America, there were no incorrect diagnoses.”
However, small tumors under 2.5 mm in thickness still need to be better understood, added Dr. Abramson. “We neither know their biologic behavior nor their histologic or molecular details,” he said. “Yes, we’re better at seeing them today, but we’re still not sure what we should do about them.”
You See It, Now Identify It
Ultrasound. The gold standard in evaluating ocular tumors remains ocular ultrasound, including both A- and B-scan ultrasound. “It is clearly better than the best MRI scan, the best CT scan or the best PET scan that we have,” said Dr. Murray. “The only cases in which we’ll obtain adjunctive imaging tests such as MRI or CT scanning is if we think the patient’s tumor extends through the wall of the eye and that there is orbital involvement.”
FA. Although fluorescein angiography is widely used, it has recently been described as having less diagnostic utility than was once believed.2 Today, it appears most helpful in excluding other conditions that simulate melanoma. Nevertheless, Dr. Shields pointed out, “If fluorescein angiography shows multifocal spots of hyperfluorescence over the tumor, this could be a sign of activity and hence a melanoma.”
OCT. In patients with vision loss that appears related to a tumor, optical coherence tomography may be used, although it appears to have limitations. “Newer techniques like OCT don’t show details in the choroid, and they certainly don’t tell you whether a nevus is present,” said Dr. Abramson. “So although ophthalmologists are using OCT, it’s really not the answer in this area.”
When to Treat? Good Question
When opinions among ocular neoplasm specialists diverge, it is often over the optimal management strategies for very small lesions (less than 2 mm in height, and less than 10 mm in base diameter).
Watch and wait? These lesions rarely if ever need to be treated immediately, noted Dr. Murray. “They can be photographed and ultrasound can be obtained, and the patient can be seen again every three to four months,” he said. “The data suggest that in these patients, 70 percent will have no change in the lesion five years after you first see them. But if the lesion changes, I advocate that they get treated immediately upon the change.”
Dr. Murray pointed to data in press that examined survival outcomes in patients who were watched closely and treated only when their tumor changed. Survival in these subjects was similar to patients who were treated when the lesion was first seen.
Or is sooner better? However, some ophthalmologists have argued that even precancerous lesions should be treated, particularly if the lesion is considered at high risk for progressing to cancer. Wills Eye Hospital researchers have developed an algorithm that quantifies specific risk factors that can help the clinician differentiate small choroidal melanoma from choroidal nevus, and decide whether to treat small lesions under 3 mm in thickness. These risk factors—thickness >1 mm, associated subretinal fluid, visual symptoms, orange pigment and posterior tumor margin touching the optic disc—were determined through a multivariate analysis of nearly 1,300 patients with small choroidal melanocytic tumors.
“It used to be a guessing game,” observed Dr. Shields. “A patient would come in with a small pigmented choroidal lesion, and we would estimate whether it was a melanoma or not. But now we have risk factors to assist in finding small ocular melanomas.”
If a patient has three or more risk factors, there’s a 50 percent chance that the lesion will be larger within five years, and hence, it’s probably a melanoma. “So in these patients with at least three risk factors, we tend to treat immediately —we don’t wait for growth to occur,” said Dr. Shields. In patients with fewer than three risk factors, other criteria are weighed in the decision-making process, including how long the lesion has been present, patient vision and the status of the opposite eye.
According to Dr. Shields, serial observation turns out to be a viable option for many of the smallest tumors. “Especially if the patient has a small tumor in the macula or near the nerve, and there is no fluid and no orange pigment, we’d definitely observe it,” she said. “But once we start seeing orange pigment, and the tumor begins to gain a little thickness, we’re probably dealing with a melanoma.”
Deciding to Treat
Wills Eye researchers have developed an algorithm that can help distinguish a small choroidal melanoma from a choroidal nevus. Risk factors for lesions under 3 mm in thickness are:
- thickness >1 mm
- associated subretinal fluid
- visual symptoms
- orange pigment
- posterior tumor margin touching the optic disc
If a patient has three or more such risk factors, there’s a 50 percent chance that the lesion will be larger within five years, and so is probably a melanoma. For patients with fewer than three factors, other considerations for deciding whether to treat include vision, lesion history and the status of the fellow eye.
How to Treat
Radiation therapy, using either plaque (brachytherapy) or charged particles, has become the cornerstone of treatment of ocular melanoma. Of his series of small tumors managed with radiotherapy, Dr. Murray said, “We had a local tumor control rate of greater than 98 percent after five years. That was comparable whether patients were treated with brachytherapy or with charged-particle radiation.”
Brachytherapy delivers radiation via a radioactive episcleral plaque, and is by far the most common form of radiotherapy for ocular neoplasms. Dr. Shields explained, “We can irradiate a patient’s melanoma, and target the radiation to the tumor and allow a 2-mm safety rim outside the tumor margin, thus sparing the rest of the eye of substantial radiation.”
However, the risk of side effects remains troubling. Particularly in tumors within 3 mm of the optic nerve or 3 mm of the fovea, radiotherapy may lead to the loss of significant visual function within 36 months of treatment. “With plaque radiation therapy, patients tend to develop radiation vasculopathy or optic neuropathy in the back of the eye adjacent to the tumor,” said Dr. Murray. “They rarely develop complications in the front of the eye, such as secondary neovascular glaucoma, which is the most common complication from charged-particle therapy.”
The one-two punch. Although many centers use plaque radiotherapy alone, others are now combining treatments—most often, plaque radiation along with transpupillary laser thermotherapy (TTT), which delivers heat to tumor cells, using a low-energy diode laser. This dual therapy may cause tumors to shrink more quickly in some patients, but it is still uncertain whether it significantly alters tumor control rates or survival. Proponents of the combination approach tend to treat the majority of patients with small melanomas (3 mm or less in thickness) with plaque combined with thermotherapy.
The addition of TTT, said Dr. Shields, may be particularly useful in treating tumors near the nerve. “Juxtapapillary melanoma is the most difficult location to irradiate because you have to put a plaque right next to the nerve with a notch on the plaque, and make sure it’s perfectly set and secure, and designed so that the radiation isodose curve covers the entire tumor,” she said. “That’s why we add TTT in these cases, because this site poses the greatest risk for recurrence, and we want to minimize that.”
Remove that eye? Only rarely. Historically, the treatment of posterior uveal melanomas has been enucleation. But in recent years, ophthalmologists have moved away from routine enucleation, with studies showing that in all but large melanomas, more conservative alternatives may have comparable efficacy. In the Collaborative Ocular Melanoma Study, when iodine 125 brachytherapy was compared with enucleation in tumors 2.5 to 10 mm in apical height, there was no statistically significant difference in survival between the two treatments after five to 10 years.3
Although localized resection of melanomas is an option for some patients, it is complex surgery. “There’s some concern that patients undergoing this procedure are more likely to get recurrent disease in the orbit,” said Dr. Abramson. “Plaques work so well, it’s hard to justify doing this localized operation.”
Progress, or Running in Place?
Despite improvements in areas like early diagnosis and successes with radiation therapy, there’s also a sense of frustration among clinicians. “The survival rates of patients with melanoma have not changed in 100 years,” said Dr. Abramson. “That’s been very disappointing.”
But newer “customizable brachytherapy” may be available soon, in which the shape of plaques will match the shape of the tumor. This new generation of plaques will be capable of delivering uniform radiation to the surface of the tumor, rather than leaving some cold spots where the tumor is not treated as effectively, said Dr. Abramson.
Treat the tumor, but don’t forget the patient. Also, according to Dr. Abramson, there needs to be a shift in thinking about ocular melanoma. “For years, ophthalmologists have thought about choroidal melanoma the way they think about retinoblastoma—the smaller the tumor, the better the success rate,” he said. “But I believe we need a paradigm shift in thinking because choroidal melanoma is more like breast cancer. The advances in breast cancer were made only when it was recognized that no matter how small the tumor, the problem is not in the breast—it is a systemic disease. Improved survival in breast cancer has been accomplished by treating undetected systemic disease at first diagnosis. That’s what we need in ocular oncology.
“It’s my belief that in a modest period of time, 100 percent of patients with melanoma will begin lifelong systemic treatment at the same time that we treat their ocular melanoma—not using chemotherapy but rather using agents like molecular modifiers that shut down melanoma cells wherever they are, killing them within hours.”
At Memorial Sloan-Kettering, thousands of such agents are being screened. “In the future, some of these druglike chemicals that modify molecular signals may become available, and melanoma patients will take them for the rest of their lives,” said Dr. Abramson. “It’s very exciting, and it will happen soon.”
1 American Cancer Society Cancer Facts & Figures 2006, p. 4.
2 Damato, B. Clin Experiment Ophthalmol 2004;32:639–647.
3 Jampol, L. M. et al. Ophthalmology 2002;109:2197–2206.