Research

The CATT’s Out of the Bag – Avastin versus Lucentis at ARVO 2011
Results Put More Pressure on Lucentis from Medicare Costs to Patient Co-Pays
Regeneron’s VEGF Trap Could Struggle in New AMD Landscape

Download Full Report with Important Disclosures: Morning Note 05-02-11 ROG NVS REGN BAYN

The Association for Research in Vision and Ophthalmology (ARVO) annual meeting began on Sunday which included a presentation of the NIH National Eye Institute’s “Comparison of AMD Treatments Trials” (CATT) results. The trial was designed to see a.) if Avastin is equivalent to Lucentis and b.) if “as needed” or pro re nata (PRN) dosing is equivalent to monthly dosing.  While the results were already published in the New England Journal of Medicine on Thursday (see below), the study chairman, Daniel F. Martin, M.D., of the Cleveland Clinic Cole Eye Institute as well as other members of the CATT group, provided significant color on the 1 year results.

The most commonly used anti-VEGF drugs for the treatment of macular degeneration are Avastin (bevacizumab) and Lucentis (ranibizumab), both of which were developed by Genentech, now fully-owned by Roche (SW:ROG) and partnered with Novartis (NYSE:NVS). Avastin is a full-length humanized monoclonal antibody that was FDA approved for the treatment of multiple solid tumor cancers in 2004. Lucentis is a smaller molecule (antigen binding fragment) derived from the same murine monoclonal antibody that is used to construct Avastin. Lucentis was FDA approved for the treatment of “wet” or neovascular Age-Related Macular Degeneration (AMD) in 2006. The two drugs mechanisms of action are the same but they differ in size, affinity for VEGF, speed of clearance from the eye, and cost. Lucentis, which is FDA-approved for AMD, costs approximately $2,000 per dose or $24,000 per year while Avastin, used off-label in AMD, costs approximately $50 per dose or $600 per year.

Equivalence was considered ±5 letters in visual acuity with the results at 1 year demonstrating that Avastin is equivalent to Lucentis at 99.2% confidence (higher than the usual 95% confidence). Avastin was also shown to be equivalent to Lucentis when both were dosed PRN.

Specifically, Dr. Martin’s presentation summary was as follows:

• Lucentis and Avastin were equivalent (virtually identical) for visual acuity at all time points when administered at the same dosing regimen.
• PRN dosing with monthly evaluation produced average gain that was 2 letters less than monthly dosing but overall results still excellent (equivalent for Lucentis, inconclusive for Avastin).
• PRN dosing resulted in 4-5 fewer injections over 1 year than monthly dosing. Avastin patients received mean 0.8 more injections than Lucentis.
• Both drugs produced an immediate and substantial decrease in fluids.
• Neither drug eliminated fluid in the majority of eyes although more eyes were completely dry with Lucentis monthly.
• There was no drug difference in leakage on FA but more leakage observed in both PRN groups.
• No lesion growth with monthly treatment of either drug; some growth with PRN over 1 year.
• No difference in death, stroke, MI or HTN between drugs at 1 year.
• Non-Specific SAE differences require additional study.

The issue of safety was discussed extensively as the proportion of patients with serious systemic adverse events (primarily hospitalizations) was higher with Avastin than with Lucentis (24.1% vs. 19.0%). Dr. Martin pointed out that the Avastin patients, especially the PRN arm, were slightly older (overall median age >80 years old) and had more co-morbidities than the Lucentis patients. In addition, there were more adverse events in the PRN arm where the patients were given less drug. Finally, the excess number of events were distributed over many different types of conditions, most of which were not seen in cancer trials where patients had received 500X the Avastin dose that was used in the CATT trial. Based on this, Dr. Martin believed this difference may not be drug-related and that the trial would have required 10,000 patients to be powered to detect the differences to statistical significance. Rather, the CATT group panelists were hopeful the 2 year data would provide more clarity. Roche/Genentech has funded a Johns Hopkins study of Medicare patients since 2005 showing that Avastin has 11% higher risk of mortality (see Johns Hopkins Medicare Claims Study). This paper will be presented at ARVO on Tuesday.

Despite current regulations prohibiting CMS from making Medicare reimbursement decisions using comparative effectiveness research studies, we believe Roche/Genentech and Novartis are now caught in a significant political and economic controversy as a result of the dramatic 40X difference in drug costs with equivalent efficacy. Philip J. Rosenfeld, M.D., Ph.D. from the Bascom Palmer Eye Institute stated “Health care providers and payers worldwide will now have to justify the cost of using ranibizumab (Lucentis). Regulators in certain countries will be forced to reconsider their policies that make it illegal to use drugs off-label, particularly when so many of their citizens cannot afford ranibizumab. The CATT data support the continued global use of intravitreal bevacizumab (Avastin) as an effective, low-cost alternative to ranibizumab.”

We also believe that Regeneron and Bayer face significant headwinds for their VEGF Trap-Eye drug. Data thus far shows dosing every other month as equivalent to Lucentis dosed monthly and would presumably yield a cost savings. However, the CATT data demonstrated that PRN dosing at 7-8 injections per year was effective which significantly diminishes the VEGF Trap-Eye dosing advantage. Furthermore, Avastin, even if given monthly, makes the economic case for VEGF Trap-Eye extremely difficult.

Download Full Report with Important Disclosures: Morning Note 05-02-11 ROG NVS REGN BAYN