American Academy of Ophthalmology Web Site:
Original URL:

January 2007

News in Review
A Look at Today's Ideas and Trends
By Linda Roach, Contributing Writer
Edited by Brian A. Francis, MD

• Predicting Speed of Eye Cancer’s Spread
• Soluble Protein Binds VEGF-A
• New Risk Calculator Ready for Prime Time
• New Device May Do Double Duty: Glaucoma and Retina

Predicting Speed of Eye Cancer’s Spread

Bit by bit over the last two decades, cancer research groups around the world have drawn a genetics-based picture of which uveal melanomas are likely to take the deadly turn of metastasizing to the liver.

This progress accelerated in the last several years as testing with DNA and RNA microarrays came on the scene. These laboratory tools aggregate large numbers of possibly relevant genetic probes into a single test, giving scientists a way to detect which genes are expressed at different stages of tumor development. By 2003 and 2004, testing of tumors from enucleated eyes had revealed that the most aggressive ocular melanomas had monosomy 3 (a missing copy of chromosome 3)1 and extra activity by genes on chromosome 8q.2

Then, last fall, came a big development in this ongoing story: The news that the more metastasis-prone uveal melanomas could be identified without enucleating the eye.

Writing in the Journal of Molecular Diagnostics, a research group at Washington University in St. Louis report using RNA samples taken by fine-needle biopsy to separate class 1, less aggressive tumors, from class 2, the type that is likely to spread to the liver and cause death.3 This would help ocular oncologists and patients make treatment decisions, even though their options remain limited.

“This is a really interesting paper. The beauty is that they were able to do this very detailed profiling just using a tiny biopsy sample taken with a 25-gauge needle,” said Carol L. Shields, MD, an ocular cancer specialist and professor of ophthalmology at Thomas Jefferson University, in Philadelphia. “In our practice, we do a similar test with a 27-gauge needle to look at chromosome 3, the most important chromosome in predicting melanoma metastasis. We have performed this testing in over 200 patients.”

J. William Harbour, MD, who leads the St. Louis researchers, said the group’s next paper will report that RNA-microarray screening appears to be more accurate than existing methods.

“We looked at over 60 tumors, all the samples we have, for their predictive value compared with monosomy 3 and other clinical parameters like tumor size. The microarray method had a class 1/class 2 sensitivity and specificity of about 90 percent, far and away better than any of the others,” said Dr. Harbour, who is associate professor of ophthalmology and visual sciences and of molecular oncology and cell biology at Washington University.

The researchers eventually hope to refine the method’s prognostic value further by analyzing gene-expression variations within the class 2 tumor type, he said. They also expect to propose a prospective, multicenter trial of the RNA-microarray’s prognostic value.

Dr. Harbour said this research should be more than a curiosity to general ophthalmologists, who frequently are the first clinicians to suspect uveal melanoma. First-line ophthalmologists will be crucial not only for referring patients to such trials, but also in helping patients gain access to the technology in the interim, Dr. Harbour said.

“When a general ophthalmologist sees a patient who might have an ocular melanoma, it is probably worth asking whoever you send those patients to whether they are collaborating with centers like ours to test the tumor’s genetics,” he said. “I get calls all the time from patients saying, ‘I just got irradiated. Can you test my tumor?’ And of course then it’s too late.

“Even if the patient is among the 10 percent in whom an eye has been removed, we can analyze those tumors after the fact, if the eye has been put in formalin,” Dr. Harbour added.

Unfortunately, Dr. Shields noted, ocular cancer specialists haven’t settled on optimal treatments for patients who are identified as high risk through genetic typing.

“We don’t have the treatment protocols sorted out for immune modulation or systemic chemotherapy in these patients,” she said. “But that is not a reason to avoid needle biopsy. We’re building our knowledge base with studies like this. In parallel, though, we need to work on better systemic treatments.”

1 Tschentscher F. et al. Cancer Res 2003;63(10):2578–2584.
2 Onken, M. D. et al. Cancer Res 2004;64(20):7205–7209.
3 Onken, M. D. et al. J Mol Diagn 2006;8(5):567–573.

Cornea Update

Soluble Protein Binds VEGF-A

VEGF in the cornea is immobilized not just by cell-bound receptors, but also by a free-floating, soluble protein that appears to be necessary and sufficient to prevent corneal angiogenesis, an international collaboration of researchers has concluded.

Labeled by the scientists as soluble VEGF receptor-1 (sVEGFR-1, or sflt-1), this molecule is unlike previously known VEGF receptors because it doesn’t need to be attached to a cellular membrane to lock up VEGF-A, the scientists report in Nature.1

And it is the solubility of sVEGFR-1 that is giving some angiogenesis scientists visions of harnessing a protein receptor (or its logical antagonists) that could be manipulated as easily as water can, said one of the team coleaders, Jayakrishna Ambati, MD, vice chairman of ophthalmology and visual sciences at the University of Kentucky.

The eye-research team, jointly led by Balamurali K. Ambati, MD, director of the cornea service at the Medical College of Georgia, stretched across 10 states and two oceans, to Australia, Japan and Europe.

It assembled diverse pieces of information from lab experiments and natural history studies to make the case that, under normal physiologic conditions in the human cornea, free-floating sVEGFR-1 binds up VEGF-A, the most potent member of this angiogenic family of molecules.

In the lab, mouse corneas would fill with blood vessels whenever the soluble receptor was absent or deactivated, the scientists report. Furthermore, they note, the only mammals with naturally vascularized corneas (Australian and Antillean manatees) lack this soluble VEGF receptor, but testing of humans and other large land and marine mammals with clear corneas (elephants, dolphins, whales and manatee-like dugongs) found that they do have the receptor.

“Before our report, researchers had identified a whole army of molecules thought to keep the cornea avascular. But when you create knockout mice that don’t have any of these molecules, the cornea still remains clear—meaning that those molecules, although they may be important in the context of trauma, are not keeping the cornea clear in the absence of injury,” Dr. J. Ambati said. “But this soluble receptor is completely different. This is the first molecule that when removed, blood vessels spontaneously invade the cornea.”

Earlier last year, a Massachusetts research team reported that one of these cell-surface molecules, VEGF receptor-3, binds VEGF-C and -D on the corneal epithelium.2 “I would look at these two novel and interesting mechanisms as being complementary,” said that study leader, Reza Dana, MD, MSc, MPH, senior scientist at the Schepens Eye Research Institute and director of the cornea service of Massachusetts Eye and Ear Infirmary. “What these two systems—soluble and cell-surface associated—do is they offer a complementary system of raising the threshold for inducing angiogenesis in the cornea.

“Normally, a tissue exposed to threats constantly, with material landing on it and causing inflammation, would get formation of new blood vessels as part of its response to these noxious stimuli,” Dr. Dana said. “In the cornea, evolution has led to the bar being raised really high, so that we can maintain corneal clarity and hence good vision.”

1 Ambati, B. K. et al. Nature 2006;443(7114):993–997.
2 Cursiefen, C. et al. Proc Natl Acad Sci USA 2006;103(30):11405–11410.

Glaucoma Update

New Risk Calculator Ready for Prime Time

The notion of risk stratification for predicting which ocular hypertensive patients are likely to develop primary open-angle glaucoma (POAG) has taken a step forward with the development of a quantitative calculator by investigators from the Ocular Hypertension Treatment Study (OHTS) and the European Glaucoma Prevention Study (EGPS).

“The model that we developed in OHTS has been validated in a totally separate European population,” said Michael A. Kass, MD, principal investigator for the OHTS.

The calculator, which assesses the five-year risk of progressing from ocular hypertension to glaucoma, is reported in this month’s Ophthalmology.

It is based on predictive factors identified in both the OHTS and the EPGS: baseline age, intraocular pressure, central corneal thickness, vertical cup/disc ratio and Humphrey visual field pattern standard deviation.

The need for risk assessment was underscored by the OHTS finding in 2002, that treating ocular hypertensive individuals reduced progression to POAG. Risk stratification is considered an alternative to treating all ocular hypertensives, which is regarded as neither medically nor economically justified.

While risk assessment narrows the playing field, it isn’t specific to each patient. That inability to individualize precisely who will benefit from close observation or treatment is a drawback shared by all existing calculators, explained Steven L. Mansberger, MD, MPH, creator of an early-model calculator. In other words, the less the individual patient resembles the average OHTS or EGPS patient, the less certain the prediction.

Still, Dr. Mansberger said the OHTS-EGPS calculator likely would become the gold standard, because it is based on a larger number of patients than earlier risk models—817 from the untreated OHTS group, plus 500 from the EGPS placebo group.

In time, more sensitive and specific models are expected to incorporate factors, such as better measures of the optic disc, nerve fiber layer and visual function, as well as genetics, diet and environmental exposures.

—Miriam Karmel

Technology Update

New Device May Do Double Duty: Glaucoma and Retina

It isn’t often that retina and glaucoma specialists hear about a technology that addresses clinical problems for both of them. But that happened at the Academy’s 2006 Subspecialty Day.

At the glaucoma symposium, ophthalmologists who have been working with iScience Interventional’s new iTrack presented papers about threading a 250-µm cannula through the entire circumference of Schlemm’s canal. At the retina gathering, they spoke of inserting a 370-µm version of the device through the suprachoroidal space to deliver medications to the posterior pole.

For the glaucoma specialist, this allows for viscodilation of the canal and passage of a 10-0 Prolene suture to enhance circumferential flow, keeping the drainage channel and trabecular meshwork open, said Richard A. Lewis, MD, a Sacramento ophthalmologist and clinical investigator for iScience Interventional.

“The operative procedure is called canaloplasty,” Dr. Lewis said. “It is a physiological procedure for enhancing outflow of aqueous and controlling intraocular pressure.”

In an observational study of more than 100 patients, he and other co-investigators found that those followed past 12 months had an average IOP of 14.6 mmHg. This procedure does not require a bleb and when compared to a trabeculectomy has fewer complications and easier postoperative care.

For the retina specialist, the iTrack offers the potential to treat retinal vascularization and other problems by targeting drug delivery into the suprachoroidal space, said Timothy Olsen, MD, professor of ophthalmology at the University of Minnesota, who published a different study about this in November.1

Indeed, the device might dose the retina more efficiently, the early results suggest. Triamcinolone acetonide delivered through the microcannula persisted in the choroid for 120 days without evidence of an impending decline, yet was very low in serum, Dr. Olsen said.


1 Am J Ophthalmol 2006;142(5):777–787.