CATT, Hopkins analysis add fervor to ranibizumab vs. bevacizumab debate

Emerging data regarding the efficacy and safety of ranibizumab and bevacizumab for the treatment of neovascular age-related macular degeneration is prompting heavy debate among ophthalmologists, who must weigh and apply this evidence to the clinical management of an array of patients.

The recent publication of the National Eye Institute’s Comparison of Age-Related Macular Degeneration Treatment Trials (CATT) 1-year results, as well as the presentation of the Johns Hopkins Medicare claims data at the Association for Research in Vision and Ophthalmology meeting, has spurred substantial discussion regarding the use of Lucentis (ranibizumab, Genentech) and Avastin (bevacizumab, Genentech).

While the debate is not new, questions raised by CATT and the Hopkins analysis are numerous. Researchers continue to investigate whether bevacizumab, which is not approved by the U.S. Food and Drug Administration for the treatment of AMD, is as safe as ranibizumab, which is FDA approved. Moreover, clinicians confront the persistent issue of cost, in that a roughly $2,000 injection of ranibizumab is 40 times as expensive as bevacizumab and Medicare beneficiaries receive only 80% coverage.

While the CATT 1-year results suggest that the two drugs may be equivalent in terms of visual acuity outcomes, and while no pre-specified adverse events have been demonstrated for bevacizumab thus far, the Hopkins analysis suggests a higher incidence of mortality and hemorrhagic stroke for patients using the less expensive drug.

However, researchers said that the differing designs of the two studies make their results virtually incomparable.

According to Daniel F. Martin, MD, differences in study designs make comparisons of CATT and Hopkins study difficult. Image: Cleveland Clinic

“You are comparing apples and oranges. … Multicenter, randomized clinical trial data will always provide a higher level of evidence than any retrospective analysis can, but retrospective analyses can be important, too,” Daniel F. Martin, MD, study chair for CATT, said. “You just have to be careful in understanding the caveats that any retrospective analysis of an administrative database shows.”

Conversely, while study design may make the Hopkins analysis inconclusive, the size of CATT may preclude it from detecting minor differences in adverse events.

“Although CATT is a well-designed clinical trial, it is not powered to detect statistically significant differences in rare events,” Emily W. Gower, PhD, lead investigator of the Hopkins review, said. “We view the Hopkins analysis as a hypothesis-generating study that hopefully will increase discussion between physician and patient in making a decision on what drug to choose. However, it should not be considered a definitive analysis.”

The American Academy of Ophthalmology has publicly applauded CATT for its design and implementation; however, representatives from the organization are opposed to discontinued financial support for ranibizumab.

“The position of the Academy is that we continue to make decisions based on what we think is right, safe and the best thing to do for patients. CATT and others have shown Lucentis and Avastin to be equal in terms of efficacy and safety, but the AAO would still strongly oppose the Centers for Medicare and Medicaid Services only funding Avastin,” William L. Rich III, MD, director of health policy for AAO, said. “Why? Because people’s physiology is different. There are variations in response to the drugs, and to preclude the use of either on a financial basis is wrong and not good public policy.”

CATT

The prospective, multicenter, randomized CATT included 1,185 patients with neovascular AMD from 44 clinical centers in the United States. The patients were assigned to intravitreal injections of ranibizumab or bevacizumab on either a monthly or an as-needed basis and were monitored every 28 days.

After a year, treatment with monthly ranibizumab and bevacizumab was equivalent in terms of visual acuity outcomes, with patients gaining 8.0 and 8.5 letters, respectively. For those on an as-needed schedule, 5.9 and 6.8 letters were achieved, respectively. While outcomes with monthly and as-needed ranibizumab were shown to be equivalent, a comparison between the two dosing regimens for bevacizumab was inconclusive.

Peter K. Kaiser

“If you were a Lucentis user, you now have a trial that validates using as-needed therapy. You get the same results as monthly treatment as long as the patients are followed closely. … If you were an Avastin user, you now have a clinical trial that validates what you were doing, namely that Avastin is similar to Lucentis,” Peter K. Kaiser, MD, OSN Retina/Vitreous Board Member, said. “The caveat is that, until we have the second-year results, if you use Avastin, you need to be treating patients monthly, not as needed.”

Ranibizumab and bevacizumab demonstrated similar results for virtually all visual acuity metrics, including mean visual gain at 1 year, gain of three lines or loss of three lines, and achievement of 20/40 vision or better, Dr. Martin said.

In terms of safety, rates of pre-specified systemic adverse events, including death, myocardial infarction and stroke, were similar for patients who received either drug. However, the occurrence of serious systemic adverse events, in particular hospitalizations, was more frequent in the bevacizumab group.

According to Dr. Martin, these events were of a broad variety and were not of previous concern. Traditional intravenous bevacizumab dosages are 500 times the potency of those used in CATT, meaning complications observed during CATT should have been evident previously, Dr. Rich said. Moreover, increased rates of complications were not linked to increased exposure to either drug. Chance, imbalances in baseline health favoring ranibizumab patients and legitimate risk were all suggested as possible explanations by the study authors.

“These are elderly patients, median age of 79. They have a lot of adverse events. Most of them are probably unrelated to Lucentis or Avastin, so there is a lot of noise happening here. The question is: Is there a signal that emerges through that noise? The answer is: We still do not know,” Dr. Martin said.

The FDA requires a safety warning for ranibizumab acknowledging the increased risk of stroke, which Dr. Martin categorized as small and as being outweighed by the drug’s potential benefits. He believes safety issues with bevacizumab to be on the same level.

“CATT is the first to begin to prospectively examine this question [of safety]. … There are safety issues that could arise in 5 to 10 years for Avastin; there are safety issues that could arise for Lucentis. We are equally unknowing for either of those,” he said, adding that CATT’s 2-year duration was determined by funding from the National Institutes of Health.

William L. Rich III

However, Dr. Rich said that drug complications should present themselves in the short term, because anti-VEGF therapy is traditionally associated with vascular effects, which are not cumulative.

Some physicians believe that CATT is not capable of detecting complication rates.

“I think the [CATT] was one of the best studies ever performed in ophthalmology, both in the study design and the way it was reported. … [However,] it is too small of a sample size. You probably need close to 10,000 patients to find some of these differences in stroke rates and thromboembolic effects, etc., between the drugs,” Dr. Kaiser said.

Dosing, retinal fluid

According to Dr. Martin, patients on as-needed regimens did better than any reported large-scale group of patients receiving non-monthly therapy, gaining an average of two letters less than their monthly counterparts. There are clinical situations in which that two-letter gain is significant, for example, for monocular patients, and clinical situations in which that difference is insignificant, he said.

Sophie J. Bakri

Sophie J. Bakri, MD, emphasized the value of these findings, saying that fewer injections may decrease the risk of endophthalmitis. However, she added that patients on an as-needed regimen must be monitored monthly.

“What the CATT study showed is what we know instinctively, that more shots are better and that [as-needed] reduced injection burden by three to four injections a year but did not reduce patient visit burden at all. The risk is that doctors are going to misinterpret [as-needed] to mean something different than shown in this trial,” Scott W. Cousins, MD, said.

Many physicians, including Dr. Kaiser, are enthusiastic about CATT’s second-year results, which may provide a stronger understanding of dosing. Some patients who were on a monthly regimen during the first year will be switched to as-needed to determine if they maintain visual gains.

As a secondary outcome, CATT also analyzed decreases in central retinal thickness, which, at 1 year, were greater in the ranibizumab monthly group.

“The important thing to emphasize is that the study confirmed that the two drugs were equivalent in terms of top-line result, vision, but there was no evidence that Avastin was superior to Lucentis in terms of durability. Also, Lucentis seemed better at drying the retina and controlling of fluorescein leak,” Dr. Cousins said.

Whether the slight superiority of ranibizumab in decreasing retinal fluid will have a long-term effect on visual acuity remains uncertain.

“The average amount of fluid remaining in the retina with either drug was tiny. It is conceivable that this will cause no difference in visual acuity in the long term if it represents tiny amounts of fluid overlying areas of scarring or atrophy or a small non-foveal pigment epithelial detachment. On the other hand, this could represent fluid over a reasonable-looking fovea, in which case it is possible that it could have an adverse effect on visual acuity,” Dr. Bakri said.

Johns Hopkins Medicare analysis

The retrospective Medicare claims database analysis, which was funded by a research contract from Genentech, reviewed beneficiaries with 1+ claim for neovascular AMD between 2005 and 2009, seeking to assess the relative safety of ranibizumab and bevacizumab.

According to Dr. Gower, two sets of analyses were performed. The first focused on patients for whom specific medical billing codes were available, enabling direct identification of their injected drug. The second included patients whose injections were hypothesized based on a combination of unclassified drug codes and charge data.

The direct identification analysis assessed data for 77,886 beneficiaries and showed an 11% higher risk in mortality and a 57% higher risk in hemorrhagic cerebrovascular accident for bevacizumab patients, risks that Dr. Gower said were not evident in the second analysis. She recently met with Senate and Medicare representatives to discuss the study methodology.

“Previous media articles focused specifically on our first analysis, which showed statistically significant differences in stroke and death. However, little mention has been made of our second analysis, which showed no statistically significant differences in some of these outcomes. I wanted to make sure that [lawmakers and Medicare officials] understood both analyses and their differing results,” Dr. Gower said.

The study attempted to control for certain baseline factors, including comorbidities, demographics and socioeconomic status, but additional information, such as smoking status and blood pressure levels, was not available.

“I do not think there is any statistical significance to the claims made in that paper once you take into account the health status of the beneficiaries. … People who are sicker and poorer are not going to get the $2,000 drug, Lucentis,” Dr. Rich said, emphasizing that underinsured bevacizumab patients are likely to have higher baseline comorbidities.

Dr. Cousins said that the retrospective nature of the study makes it incapable of providing definitive results, but he emphasized that it raises concerns regarding a potential safety issue with bevacizumab.

“These results should not change clinical practice, other than to make clinicians become aware that there may be a potential safety signal with Avastin. But it is up to the scholarly community to do more research on the topic, to figure out who is at risk and how big that risk is,” Dr. Cousins said, expressing the hope that future studies be funded by a neutral third party.

History of bevacizumab

Ranibizumab is a derivative of bevacizumab; it is one of two Fab fragments on the bevacizumab monoclonal antibody that has been tweaked to increase its binding affinity, Dr. Martin said. Initial experiments by Genentech suggested that the bevacizumab molecule, which is roughly 150 kD, was too large to cross the internal limiting membrane and reach the subretinal space, he said, so researchers engineered the smaller 50 kD fragment.

However, according to Dr. Martin, during the period when physicians were waiting for ranibizumab to be approved by the FDA, it was discovered that early experiments by Genentech were faulty and that bevacizumab could reach the subretinal space. As the drug was already approved for certain types of cancer, ophthalmologists jumped on the opportunity to treat patients with it.

“We knew we had this great drug that was being evaluated by the FDA, but we had nothing to offer our patients at that point that would have any chance of stabilizing or improving vision. Because of that, when it was first announced by Dr. Rosenfeld that Avastin worked, it spread like wildfire,” Dr. Kaiser said.

Since ranibizumab’s approval, physicians have shown continued support for bevacizumab despite the higher profit gained through use of ranibizumab. The two are Part B drugs, meaning they are ordered by the physician and delivered under his or her supervision, and more importantly, their administration results in federal reimbursement of an additional 6% of the drug’s cost, Dr. Rich said. That translates to roughly $120 per injection for ranibizumab.

Dr. Rich believes continued support of bevacizumab is a testament to the ophthalmic community’s belief in the drug’s safety and desire to lessen seniors’ financial burden.

In fact, he said that surgeons have been lobbying in its defense for many years, as Genentech tried to halt sales of bevacizumab to compounding pharmacies that were repackaging the drug for ocular use in 2007, and surgeons subsequently put pressure on Capitol Hill to enable continued access.

“[Genentech’s] explicit rationale was that they were concerned about safety issues with repackaging a drug that was designed for intravenous use that was being repackaged and compounded for intraocular use,” Dr. Cousins said.

Research going forward

In addition to generating substantial debate, CATT has also led to a revamping of the government’s approach to clinical research.

“Before we started CATT, it was not possible for the federal government to compare two different drugs, two different operations, two different modes of care or two different devices. This trial resulted in a complete rewrite of the Center for Medicare and Medicaid Services clinical trial policy,” Dr. Rich said.

The 2010 Patient Protection and Affordable Care Act established the Patient-Centered Outcomes Research Institute, an entity aimed at comparative effectiveness research that will help physicians and patients make more informed decisions. Dr. Rich believes that CATT’s design will serve as a model for future studies.

“It is imperative that the government continues to fund this kind of research. In the case of comparative studies, government-funded studies are really the only ones who can be believed,” Dr. Kaiser said.

Dr. Rich further highlighted CATT’s effects on the global community, as its validation of bevacizumab use may prompt other nations to move forward with treatment.

“There are parts of the world where people were forbidden to use Avastin off label, because, in this country, it was not FDA approved. Our hope is that now that we have this trial in the United States, other countries will be able to prevent the leading cause of blindness in Caucasians in Europe and other parts of the world,” he said.

Dr. Martin hopes that, in its second year, CATT will identify subgroups of patients who respond differently to anti-VEGF therapy.

“It has been of interest to many that there are some patients who only need two or three injections over the course of a year. … And then there are other patients who need six, seven or eight injections and others who need 12. The question is, why? CATT provides an opportunity for the first time to prospectively study these groups and attempt to identity subgroups and tailor their treatment,” he said.

While the ranibizumab vs. bevacizumab debate will continue beyond CATT and the Hopkins analysis, these trials have already enhanced dialogue between physicians and between physicians and patients.

“Retinal surgeons have been injecting intravitreal antibiotics, steroids and gases into the eye for decades. These agents have not been approved for intraocular use. Without these, countless patients would have gone blind from endophthalmitis or retinal detachment,” Dr. Bakri said. – by Michelle Pagnani

Which drug do you prefer for the treatment of wet AMD, ranibizumab or bevacizumab?

Lindstrom's Perspective
AMD trials provide valuable data for physicians, patients

References:

• CATT Research Group, Martin DF, Maguire MG, Ying GS, Grunwald JE, Fine SL, Jaffe GJ. Ranibizumab and bevacizumab for neovascular age-related macular degeneration. N Engl J Med. 2011;364(20):1897-1908.
• Gower EW, Cassard SD, Chu L, et al. Risk of adverse event following intravitreal injection of bevacizumab or ranibizumab for treating age-related macular degeneration. Presented at: The Association for Research in Vision and Ophthalmology; May 1-5, 2011; Fort Lauderdale, Fla.

• Sophie J. Bakri, MD, can be reached at Mayo Clinic, 200 First St. SW, Rochester, MN 55905; 507-284-3701; email: bakri.sophie@mayo.edu.
• Scott W. Cousins, MD, can be reached at Duke University Eye Center, DUMC Box 3802, Durham, NC 27710; 919-684-3090; fax: 919-681-6474; email: scott.cousins@duke.edu.
• Emily W. Gower, PhD, can be reached at Wilmer Eye Institute, 116 Wilmer Building, 600 N. Wolfe St., Baltimore, MD 21287; 410-614-2874; email: egower1@jhmi.edu.
• Peter K. Kaiser, MD, can be reached at Cole Eye Institute, Division of Ophthalmology, A31, 9500 Euclid Ave., Cleveland, OH 44195; 216-444-6702; email: pkkaiser@aol.com.
• Daniel F. Martin, MD, can be reached at Cleveland Clinic Main Campus, 9500 Euclid Ave., Cleveland, OH 44195; 216-444-0430; email: martind5@ccf.org.
• William L. Rich III, MD, can be reached at American Academy of Ophthalmology, Governmental Affairs Division, 20 F St. NW, Washington, DC 20001-6701; 202-737-6662; fax: 212-737-7061; email: hyasxa@aol.com.
• Disclosures: Dr. Bakri has served as a consultant for Allergan. Dr. Cousins is a consultant for Genentech. Dr. Kaiser is a consultant for Genentech, Regeneron, Bayer and Novartis. Dr. Gower, Dr. Martin and Dr. Rich have no financial interests in any of the products discussed in this article nor are they paid consultants for any companies mentioned.