After years of debate within the ophthalmology community, the results of the first clinical trial comparing two Genentech drugs head-to-head, Avastin (bevacizumab) and Lucentis (ranibizumab), for the treatment of wet-age related macular degeneration show there is little reason to spend thousands of dollars on Lucentis when Avastin works as well for a fraction of the cost.

The results put Genentech, owned by Roche, and Novartis AG, which markets Lucentis in Europe, on the defensive. To maintain market share of their blockbuster brand Lucentis, the two firms have highlighted unanswered questions associated with Avastin, including potential safety issues and slight discrepancies in dosing.

The results of the Comparison of Age-Related Macular Degeneration Treatments Trial (CATT), a 1,200-patient trial sponsored by NIH's National Eye Institute, were released in The New England Journal of Medicine April 28, days before the data are scheduled to be presented at the Association for Research in Vision and Ophthalmology meeting in Fort Lauderdale, Fla. May 1.

"The CATT data support the continued global use of intravitreal bevacizumab as an effective, low-cost alternative to ranibizumab," wrote Phillip Rosenfeld, MD, PhD, a professor of ophthalmology at the University of Miami Miller School of Medicine, in an accompanying editorial. "Healthcare providers and payers worldwide will now have to justify the cost of using ranibizumab," he added.

Lucentis, which is FDA approved for wet AMD, costs $2,000 per dose, while Avastin, used in micro-doses compounded by pharmacists, costs just $50.

The results of CATT will pave the way for even more use of Avastin in wet AMD, despite the fact that it is not approved for the indication. Currently 250,000 patients are treated annually in the U.S. for wet AMD, the leading cause of blindness, and a majority receive Avastin. According to the American Academy of Ophthalmology, Avastin's market share in the U.S. is greater than 60%. Nonetheless, Medicare pays significantly more for Lucentis; in 2008 it spent $557 million for the antibody fragment versus $20 million for Avastin. That was for 480,000 Avastin injections versus 337,000 Lucentis injections. Deutsche Bank analyst Robyn Karnauskas included the data, which came from a widely-quoted study by Rosenfeld, in an April 19 presentation she made to investors on the potential implications of CATT.

Two central but related questions stem directly from the publication of the head-to-head results: Will the Centers for Medicare & Medicaid Services initiate a coverage review that would put limitations on the use of Lucentis, and will private payers take steps to limit Lucentis as a first-line treatment option? Medicare currently reimburses both products as do many private insurers.

Reimbursement Changes Ahead?

Director of the National Eye Institute Paul Sieving, MD, PhD, said the institute is "ready and eager" to present the data to Congress if requested. "I think the study stands as a measure of the value of taxpayer dollars being spent on vision research," he said during a same-day conference call announcing the data. The researchers declined, however, to comment on whether and how the data should be used to change reimbursement policies.

In an interview, Director of Health Policy at the American Academy of Ophthalmology William Rich, MD, said it would object to CMS changing its coverage policy for the drugs, despite the fact the study shows very strong evidence of similar efficacy and safety. "We think it would be a gross mistake," he said. "There is enough individual variability with response to different drugs."

"This is a case of the federal government getting it right," said Premera Blue Cross Pharmacy Manager John Watkins of the decision to initiate the comparison trial in an interview just ahead of the data release. Nonetheless, he said Premara has no immediate plans to change its reimbursement policy for the drugs based on the outcome. Premara currently reimburses both drugs. "At this point I don't have any immediate plans to recommend [limiting Lucentis] and we certainly would not do that without consulting with our opinion leaders," Watkins said. The data will be reviewed by Premera's pharmacy & therapeutics committee however.

Researchers Confident On Efficacy And Safety

During the press briefing, researchers expressed their confidence in the results of the trial. "I would say that Lucentis and Avastin are equivalent for visual acuity, that PRN dosing is a viable treatment option and that we saw no major safety differences - death, myocardial infarction and stroke being the three primary ones we were concerned about," said Daniel Martin, MD, the study chair for CATT and chairman of the Cole Eye Institute at the Cleveland Clinic.

Safety and PRN dosing - which refers to the practice of treating patients on an "as needed" basis as the disease progresses - were two of the data points industry observers were most highly anticipating. PRN dosing, which generally pushes back the time between doses to six or eight weeks, is used frequently with both Lucentis and Avastin, despite the fact that monthly dosing was used in the pivotal trial to support approval of Lucentis. CATT enrolled patients in four different arms: half the patients were given either Lucentis or Avastin every four weeks for one year, after which they were re-randomized to variable dosing with the same drug; the remaining patients were prescribed Lucentis or Avastin on a PRN basis, for two years.

CATT effectively appears to answer the PRN dosing question in the minds of the researchers. At one year Avastin was equivalent to Lucentis on the change in visual-acuity score within five letters on an eye chart from baseline, both when the drugs were given on a monthly basis and as needed, the researchers concluded. Nonetheless, some differences did not favor Avastin.

Lucentis given as needed was equivalent to Lucentis monthly with a mean difference of 1.7 letters. Avastin given as needed was equivalent to Avastin monthly through 36 weeks, with a difference of 1.6 letters. However, at 52 weeks, Avastin given PRN relative to monthly yielded slightly less benefit, with patients in the as needed cohort showing a gain in visual acuity of 2.1 letters less. That made the results inconclusive.

"CATT clearly shows that this is a very viable treatment option. The mean gain in visual acuity is the best that has ever been reported in less than monthly dosing," Martin said of the PRN dosing regimen during the conference call. "It was on average two letters less than monthly dosing, but that difference is small."

The NIH investigators also monitored the incidence of adverse events in the Lucentis and Avastin treatment arms. Of the major adverse events -- death, MI and stroke -- no difference was observed in patients taking the two medicines. However, patients prescribed Avastin were slightly more likely to experience non-specific serious adverse events, primarily hospitalizations, than those given Lucentis (24.1% versus 19%). These events were distributed over many different types of conditions, making it difficult to draw conclusions linking adverse events to treatment. The results of the second year of CATT, expected next year, could provide more clarity; moreover, the number of study patients was too few to detect adverse events, leaving the results subject to interpretation. Already Genentech and Novartis execs have seized upon the possible safety questions as a means of protecting their lucrative brand.

Genentech, Novartis Fired Up Over Safety

Indeed, Johns Hopkins University researchers will present retrospective data from a Medicare claims database analysis at AVRO May 3, showing an 11% higher risk in mortality and 57% higher risk of hemorrhagic cerebrovascular accident in patients treated with intravitreal injections of Avastin compared to Lucentis. The study was conducted with funding from Genentech and Novartis. [Editor's note: this paragraph was updated April 29 to clarify that Johns Hopkins carried out the study.]

"[CATT] is too small to show the differences on some of the side effects in the areas of stroke and heart attack, which we have seen in some of the big public database examinations," said Novartis' head of global pharmaceutical development, Trevor Mundel, MD, PhD, in an interview shortly after the data was released.

"From my perspective, I could not see this forming the basis for any submission that we would do for a product," he said. "It is one small study, and even within that one study it raises a number of issues that you would need further studies to sort out. I can't image that any of the regulators in Europe would have any basis to make any regulatory decisions based on this paper."

In a statement, Genentech said it continues to believe that Lucentis is the most appropriate treatment for wet-AMD. The safety findings from CATT "add to the emerging body of evidence from much larger analyses that suggest the risk of systemic adverse events may be higher when injecting Avastin into a person's eye compared to Lucentis," the company said.

From the day Lucentis launched in 2006, its use as an AMD therapy has been contentious. That's because the medicine is closely related to Avastin; indeed, Lucentis was derived from the same anti-VEGF mouse monoclonal antibody used to create the approved cancer medicine. However, Genentech never studied Avastin in AMD or filed for FDA approval for the indication for fear of under-cutting its Lucentis brand.

Despite the controversy, Lucentis, thus far, has been an unqualified commercial success. In 2010, it generated sales of CHF 1.46 billion ($1.65 billion) for Roche and $1.53 billion for its marketing partner Novartis. Some physicians are more comfortable prescribing Lucentis because it is approved for wet AMD, as well as other eye diseases. But another reason for the utilization may be that under Medicare, which pays physicians a drug's average sales price plus 6%, an economic incentive exists to prescribe Lucentis over Avastin.

The CATT data could also have implications for the pricing and commercial uptake of newer AMD therapies, most notably Regeneron Pharmaceuticals Inc./Bayer AG's aflibercept. Regeneron's VEGF Trap-Eye is being assessed by U.S. regulators, with an Aug. 20 PDUFA date after being granted a priority review. Regeneron owns rights to the drug in the U.S. and shares rights outside the U.S. with Bayer.

In the two Phase III clinical trials for the VEGF Trap-Eye, the drug was compared to Lucentis and shown to be non-inferior with fewer doses, every two months versus every month for Lucentis (Also see "Regeneron/Bayer Bullish On Wet-AMD Drug But Commercial Success Is No Guarantee" - Pink Sheet, 22 November, 2010.). While important, given the elderly patient population that most frequently uses the drugs and the requirement for regular lengthy in-office procedures, a dosing advantage alone may not be enough for Regeneron to ride on, however.

-Jessica Merrill (j.merrill@elsevier.com)

Lucentis And Avastin Stack Up In CATT, Forcing Genentech/Novartis To Play Defense

More from Archive

More from Pink Sheet