CATT: Beyond Visual Acuity
Year 2 results did much more than just confirm that the two rival drugs are equivalent.
Frank Celia, Contributing Editor
Hanging over every injection of Avastin (bevacizumab) administered in the last 12 months was a gnawing question: would the statistically significant anatomical benefits of Lucentis (ranibizumab) in AMD found in the first year of the CATT study translate into worse vision among Avastin patients at year 2?
To the collective relief of practitioners and millions of Avastin patients worldwide, that didn't happen. Despite clear evidence that Lucentis produced thinner, drier maculas, visual acuity benefits among the two study groups held firm when the drugs were delivered monthly. Both anti-VEGF therapies appear to preserve and enhance vision among AMD patients with more or less equal power.
Figure 1. A photo of a patient with exudative AMD, taken with an Optos camera (courtesy of Justis Ehlers, MD).
That resolved, the Comparison of Age-Related Macular Degeneration Treatment Trials (CATT) year 2 data unveiled last month at the annual meeting of the Association for Research in Vision and Ophthalmology still presented serious questions for retina subspecialists to ponder. This article addresses three big ones.
Safety: A concern from the beginning, bevacizumab's potential systemic health risks are now more worrying. As in year 1, a significantly greater number of Avastin patients suffered serious systemic adverse events. Though a strong argument can still be made for Avastin's safety, many believe the CATT data weakened it.
Dose regimen resilience: When patients injected monthly for a year were switched to as-needed dosing, a surprising decline in vision ensued, almost as if they had been on PRN dosage from the outset. This raises concerns about anti-VEGF therapy's sustainability and the treatment burden patients must shoulder.
Geographic atrophy: Are drier retinas always healthier? Though patients receiving monthly Lucentis demonstrated less subretinal fluid, they also showed a greater incidence of geographic atrophy.
A STRONG MICROSCOPE
CATT assigned 1,185 wet AMD patients into one of four treatment groups: ranibizumab monthly or PRN, and bevacizumab monthly or PRN. After year 1, some of the monthly treatment patients were switched to as-needed treatment. Patients underwent examinations every four weeks and were treated when OCT scans detected fluid or, in the presence of hemorrhage, loss of visual acuity or dye leakage during fluorescein angiography.
Neither drug proved more efficacious when the data for all regimens of the drugs were combined. Unlike the year 1 primary outcome data, there was not enough statistical power in year 2 to evaluate the six groups, so we will not know how the Avastin PRN group fared in the second year. However, we do know that visual outcomes were slightly better among monthly patients than as-needed ones, by about 2.4 letters of visual acuity, when both drugs were combined. Overall, after two years, the mean increase from baseline was 8.8 letters in the ranibizumab-monthly group, 7.8 letters in the bevacizumab-monthly group, 6.7 letters in the ranibizumab-PRN group, and 5.0 letters in the bevacizumab-PRN group.
Patients switched from monthly to as-needed treatment in the second year experienced a small but statistically significant drop in vision of about 2.2 letters. They also showed a 19% decrease in retinas free of subretinal fluid, the study found.
Though some visual acuity differences proved significant, CATT's lead author emphasized that, broadly speaking, on a 60-letter vision scale, they were relatively small. “I want to stress that we were looking at this under a very strong microscope,” Daniel F. Martin, MD, chairman of the Cole Eye Institute at the Cleveland Clinic in Ohio, told the ARVO audience at the CATT year 2 presentation. “The magnitude of these differences is very small.”
HARD TO EXPLAIN
Rates of death and arteriothrombotic events were similar for both drugs. However, like the results of the first year, the proportion of patients with one or more systemic serious adverse event (SAE) was higher in the Avastin group than the Lucentis group. In the Avastin group 39.9% of patients experienced such events, while in the Lucentis group 31.7% did. The year 1 data found 24.1% of Avastin patients experienced SAEs compared to 19.0% in the Lucentis group. In other words, not only did both years see a significant increase in SAEs among Avastin patients, but that trend grew more pronounced during the second year.
In a statement released last month, Genentech, which manufactures both drugs, reaffirmed its position that Avastin may egress into the patient's system and cause serious health risks, some perhaps unknown at this time. “We specifically designed Lucentis for use in the eye and to clear quickly from the bloodstream to minimize side effects,” the company wrote, unlike Avastin, which was designed as a systemic agent “and intended to be retained in the bloodstream for weeks at a time.”
Such has been Genentech's stance since off-label Avastin prescribing became popular several years ago. Heretofore, the general consensus among retinal physicians was that the company had more interest in its bottom line than safety. However, that consensus now appears harder to maintain.
“There was a little bump in SAEs in the first year,” notes Pravin Dugel, MD, a Phoenix-based retinal specialist. “Many people thought maybe that was statistical noise. Avastin patients were slightly older, after all, and statistical anomalies do happen. However, when that finding was consistent in year 2, now it becomes a little harder to explain. Avastin lasts longer in the systemic circulation. That's a fact. You cannot ignore that.”
Nevertheless, Dr. Dugel and others point out that a compelling argument can still be made for Avastin's safety. For starters, more of the SAEs occurred in the PRN arms of the study than in the monthly arms, which would seem to argue against drug causation. Additionally, as the CATT study team noted, most of the excess events reported in the trial have not been associated previously with systemic anti-VEGF therapy for cancer.
However, the issue of safety is concerning and warrants further study. Indeed, in his opinion, the safety data were among CATT's most interesting findings, says Richard Kaiser, MD, an associate professor of ophthalmology at Wills Eye Institute in Philadelphia. “While the safety results may not make sense based on our preconceived notions of the accepted risk of these drugs, you can't ignore a statistically significant endpoint. When you perform a cost-benefit analysis, patients have benefited more from Avastin than any other anti-VEGF drug, but the safety data need to be further investigated. We need more information.”
Switching to PRN dosing after one year of monthly treatment with either drug resulted in a mean 2.2-letter decrease in visual acuity. In other words, the modest gains produced by monthly dosing all but vanished soon after the injection rate declined. Furthermore, the mean total retinal thickness increased significantly in the switched patients (ranibizumab, +31 µm; bevacizumab, +19 µm), as did the number of patients with evidence of any fluid on OCT.
Many had hoped those therapeutic advantages would have had a little more longevity. The reason behind switching dosing regimens was to determine how much histologic benefit anti-VEGF therapy imparted, according to Dr. Martin. “To me the take-home message is this: if we are going to treat monthly for that small advantage in vision — 2.4 letters — we can't stop; it has to be continuous,” he said at ARVO.
The switch data can be viewed as casting a pall over anti-VEGF monotherapy in general, observes Dr. Dugel: “What this tells me is that even if you give anti-VEGF injections monthly for year, it confers no structural, permanent advantage whatsoever. So I should be having a conversation with my patients before we begin, asking them which track they want to be on. Another way to put it is, ‘What are 2.4 letters of vision worth to you? Are they worth twice the number of injections and the increased risk of endophthalmitis? Do 2.4 letters of vision have an impact on the quality of life?'
“This also says something incredibly interesting about the physiology and mechanism of action of anti-VEGF monotherapy,” Dr. Dugel continues. Without a structural advantage, there is less hope for a therapy endpoint for these patients, raising the specter of an immense treatment burden. “We simply cannot continue to treat like this and pretend that what we are doing is sustainable,” he concludes. He emphasizes that the most dire current problem, treatment burden, was not designed to be addressed by the CATT study. “All patients, even in the as-needed arm, were seen on a monthly basis. The results of the CATT study may not be duplicated if patients are not seen monthly. Many recent studies and clinical experience suggest that very few patients are seen monthly in most practices.”
Combination therapy may offer a solution, Dr. Dugel adds. An experimental therapy such as the anti-PDGF aptamer E10030 (being developed by Ophthotech) could potentially better attack the roots of neovascularization, for instance. E10030 is targets PDGF-B, a key molecule involved in the recruitment and maturation of pericytes, which play a major role in anti-VEGF treatment resistance. E10030 strips the pericytes from the neovascular tissue, rendering them highly sensitive to an anti-VEGF attack, according to company literature.
Dr. Kaiser also looks forward to potential new therapies. “If you analyze the angiograms from the ANCHOR and MARINA studies, which paved the way for anti-VEGF drugs, they showed no regression of the choroidal vascular lesions. In fact, there was actually growth of the lesions, despite the impressive visual results,” he explains. “Although anti-VEGF therapy decreases vascular permeability, which decreases leakage and edema and the kind of changes that obscure vision, it does not cause regression of the choroidal vascular complex. What we need is a drug to induce regression. But no therapy to date has shown that.”
The highest proportion of dry retinas was found in the ranibizumab monthly group. However, these patients also showed the highest incidence of geographic atrophy among the six treatment groups. “This may be important to consider when weighing the risks and benefits of monthly versus as-needed treatment and in the selection of drug,” the CATT paper states. “Although most geographic atrophy in the CATT was nonfoveal through the two-year period of the study, adverse effects on visual function, such as lower reading speed and extension to the foveal center with time, are common in the natural history of geographic atrophy.”
This was another unexpected result, because for years ophthalmologists have sought to give their patients the driest, thinnest retinas possible. Add the finding that the wetter, thicker Avastin retinas of year 1 did not experience acuity loss at year 2, and some experts have begun to question whether a totally dry, very thin macula ought to remain the gold standard for every patient.
“The observation that eyes in the Lucentis monthly arm of the CATT study showed an increased incidence of geographic atrophy was surprising,” notes David F. Williams, MD, MBA, former president of the American Society of Retinal Specialists. “The pathophysiological mechanism of the GA in these eyes is unknown, but we can speculate that VEGF may play a role in maintenance of RPE function and morphology, and that constant and robust suppression of VEGF may impair RPE physiology and promote GA in some eyes.”
How this might impact anti-VEGF treatment remains open for debate, says Dr. Williams, adding he believes multiple other parameters should be considered, such as visual acuity, lesion activity, intraretinal vs subretinal fluid, and lesion growth over time. “However, it suggests to me that we should use anti-VEGF agents only enough to suppress active CNV and its associated exudation, and that something less than monthly treatment, as is done in both PRN and treat-and-extend protocols, may be a safer option for some eyes.”
Treat-and-extend dosing regimens were not studied by CATT, but they pertain to any discussion of PRN vs monthly schedules, because in real-world clinical practice, treat-and-extend is far more likely to be employed than any other schedule. A Norwegian trial, LUCAS (Lucentis Compared to Avastin Study), currently under way, is examining treat-and-extend therapy, and its results are eagerly anticipated.
BEYOND SIBLING RIVALRY
Ophthalmology's utmost concern, perforce, is visual acuity. But other, seemingly secondary, factors often have a long-term impact on standards of care, practice management, and research goals. CATT began as a study of the efficacy of two rival drugs, but could very well end up having profound and enduring effects on the profession that five years ago no one could have foreseen. RP