Arguably the most exciting news from this year’s Association for Vision in Research and Ophthalmology (ARVO) annual meeting was the one-year outcomes of the CATT (Comparison of Age-Related Macular Degeneration Treatments Trial) study. This trial addressed the relative safety and efficacy of bevacizumab (Avastin) and ranibizumab (Lucentis), both manufactured by Genentech, in the treatment of age-related macular degeneration (AMD).
While past clinical trials have established the efficacy of ranibizumab in the treatment of AMD, the off-label use of Avastin was commonly practiced, despite the absence of clinical trials establishing an equivalent efficacy to Lucentis. This was primarily because bevacizumab was available at lower cost.
These clinical pearls will be devoted to discussing the results of the trial and what they mean to us as clinicians. The full article describing the results of the CATT study was published in the New England Journal of Medicine on April 28.
The CATT study had four arms. Two of the arms compared bevacizumab to ranibizumab on a monthly treatment regimen. The other two arms compared the same drugs on an as-needed treatment schedule, in which re-injection after the initial mandatory dose was based on specific optical coherence tomography criteria and treatment response.
A total of 1,105 subjects completed the one-year follow up. One of the primary outcomes measured was mean change in visual acuity from baseline to one year. Change in foveal thickness, total number of injections over one year in the as needed treatment arms, adverse events and total drug cost per arm were among some of the secondary outcomes measured.
In short, the change in visual acuity at one year was equivalent in the monthly Avastin and Lucentis groups. Additionally, Avastin given as needed was equivalent to Lucentis given as needed. Moreover, Lucentis given monthly was equivalent to Lucentis given as needed, while the comparison of monthly Avastin with as-needed Avastin was inconclusive.
When looking at the secondary outcome of foveal thickness, the Lucentis arms did better than their corresponding Avastin arms. So, while visual outcomes were statistically the same, anatomically there was a significant difference in foveal thicknesses. This incongruity between anatomic and visual outcomes appears to be a common, but as yet unexplained, phenomenon in treatment of AMD and other macular disorders.
No significant differences in ophthalmic adverse events were noted between the two drugs. With regard to serious, systemic adverse events, the statistical power did not detect a difference in the rates of death, arteriothrombotic events or venous thrombotic events, all of which are adverse events commonly identified with systemic, intravenous Avastin therapy for cancer treatment.
However, in the CATT study, the rate of serious systemic adverse events, particularly hospitalizations, was higher in the Avastin groups. Critics and skeptics of the study have already asserted that this finding makes Avastin a riskier drug in the treatment of AMD.
However, this assertion requires further study, because the systemic events noted in the CATT study were not associated with organ systems that are typically affected by systemic, intravenous Avastin therapy when administered for cancer treatment at 500 times those used in intravitreal injections. Additionally, rates of adverse events were higher in the as-needed arms than in the monthly arms, indicating a non-dose related effect.
The cost of a single dose of Lucentis is approximately $2,000. The cost of a single dose of Avastin is $50. In the CATT study, average yearly cost of Lucentis (in the monthly group) was $23,400; Avastin (in the monthly group) cost $595; Lucentis as needed $13,800; and Avastin as needed $385.
The response from Roche Holding AG, which markets both drugs through its Genentech unit in the United States and through Novartis AG elsewhere, stated that the study showed that Lucentis eliminated fluid from the eye better than Avastin. Roche also noted that Lucentis is specifically designed for treatment of macular degeneration, while Avastin is specially designed for treatment of cancer.
The Wall Street Journal quoted Novartis executive Trevor Mundel as saying the study's visual-acuity measurement was a "pretty crude measure" of efficacy, and a weakness of the study.
Additionally, a Genentech-funded study by Johns Hopkins researchers (lead author Emily Gower, PhD), part of which was also presented at ARVO this year, entailed a retrospective review of 77,886 Medicare claims between 2005 and 2009. The abstract for the study stated an 11 percent higher risk in overall mortality and a 57 percent higher risk of hemorrhagic cerebrovascular accident in those patients receiving Avastin. The abstract said the study was “limited by incomplete information on some important confounding factors, e.g., smoking, lipid and blood pressure levels…”
The response from Roche/Genentech/Novartis is not unexpected, as the company has taken prior actions to limit physician access to Avastin. On October 2007, Genentech announced it would no longer allow compounding pharmacies to purchase Avastin directly from wholesale distributors. After much outcry from vitreoretinal specialists, many of whom obtain Avastin through compounding pharmacies, the Academy and the American Society of Retina Specialists worked with Genetech to loosen the restriction, as announced in December 2007.
The current extent to which clinicians use Avastin in the United States is not known. Brechner et al note that, in 2008, of all Medicare beneficiaries who underwent treatment for AMD, approximately 65 percent of them received Avastin, while 35 percent received Lucentis. Those numbers do not include the use of Avastin for other vitreoretinal disorders in younger patients, such as for diabetic retinopathy and less-common causes of choroidal neovascularization.
Prior to the CATT study, clinicians were in a position where they often felt obligated to defend their decision to use Avastin over Lucentis. The difference now, as Philip Rosenfeld, MD, PhD, wrote in his recent editorial in the The New England Journal of Medicine, is that clinicians and insurance companies will have to justify and defend the cost of using Lucentis, instead of Avastin, in the treatment of AMD.
The standard has been set. Not only do we have a treatment available to stabilize and improve vision in patients with exudative AMD, but even more amazingly, it costs only $50 per dose. As future drugs or treatments for wet AMD come into the market, they will have to be significantly better than Avastin, either in terms of efficacy, cost or both, in order for clinicians to consider using them.
And if we assume that such a future treatment exists, how much better will it have to be for clinicians to use it instead of Avastin? What is the efficacy/cost threshold for clinicians? Will the safety concerns for Avastin, which have not yet been elucidated, impact clinical decision-making? Will the treatment burden for patients, in terms of frequency of treatments, play a role? These are thought-provoking questions that every vitreoretinal specialist will ask themselves as newer treatments come into the market.
Dr. Rosenfeld deserves credit for pioneering the use of Avastin in the eye. Additionally, Daniel Martin, MD, and all the other investigators in the CATT Research Group should be thanked for spearheading and participating in this important study.
Issue Index | Related Articles | YO Info Archive
* * *
About the author: Manju Subramanian, MD, is a vitreoretinal specialist at Boston Medical Center, the VA Boston Healthcare System and Dedham Ophthalmic Consultants and an assistant professor in ophthalmology at Boston University School of Medicine. She serves as vice-chairman of clinical operations for six clinical sites affiliated with Boston University Eye Associates and Boston Medical Center.
As principal investigator and lead author in a study conducted at the VA Boston Healthcare System, Dr. Subramanian published one-year outcomes of a small, head-to-head, double-masked trial comparing Avastin and Lucentis, finding that both drugs failed to show a difference in efficacy when administered on an as-needed basis (Eye (Lond). 2010 Nov;24(11):1708-15).
Brechner RJ, et al. Pharmacotherapy for neovascular age-related macular degeneration: an analysis of the 100% 2008 medicare fee-for-service part B claims file.
Gower, EW et al. Adverse Event Rates Following Intravitreal Injection of Avastin or Lucentis for Treating Age-Related Macular Degeneration. Abstract, The Association for Research in Vision and Ophthalmology, Inc. May 3, 2011