Pharmaceutical sponsors across therapeutic categories may have breathed a sigh of relief when FDA’s Cardiovascular and Renal Drugs Advisory Committee said that Bayer/Johnson & Johnson’s Xarelto (rivaroxaban) need not show it is “as effective” as Boehringer Ingelheim GmbH’s anticoagulant Pradaxa (dabigatran) in atrial fibrillation patients.

At a Sept. 8 meeting, the majority of panel members said Xarelto’s sponsors should not be required to directly show comparable efficacy to dabigatran, a drug approved a mere 11 months earlier.

FDA sought the panel’s input on whether warfarin or dabigatran was the appropriate comparator for rivaroxaban in light of a 1995 policy that requires new drugs be “as effective” as approved treatments for serious diseases.

Requiring drugs for serious or life-threatening diseases to be compared head-to-head against the newest treatment on the block could throw a wrench in long-planned and ongoing clinical development programs that use a well-established standard of care as a comparator agent, some panelists said.

However, the relief at knowing that Bayer/J&J probably would not have to go back to the drawing board to conduct a head-to-head comparison of the two novel anticoagulants before approval carries a lesson of which all sponsors should take note. When crossing the regulatory finish line on the heels of a drug that has demonstrated superiority over a long-established treatment (in this case, warfarin), FDA is going to want to see efficacy data showing robust non-inferiority to the standard of care.

Even if FDA does not require a direct, head-to head trial of the newest treatments, the payer community might not be as accepting of data that merely show non-inferiority against an older therapy. FDA approval aside, sponsors may need evidence demonstrating comparable or better efficacy of their drug over a first-mover among new treatments.

For Xarelto, although its sponsors succeeded in obtaining a favorable panel recommendation, with a 9-2-1 vote for approval, challenges lay ahead, including the prospect of an unfavorable label when compared to that of Pradaxa (Also see "Xarelto Faces Approval, Labeling Obstacles Even After Cmte. Endorsement" - Pink Sheet, 12 September, 2011.).

Comparing ROCKET AF And RE-LY

In convening the cardio-renal advisory committee to consider rivaroxaban’s use in atrial fibrillation, FDA sought the panel’s views on efficacy data from the pivotal ROCKET AF study and whether the Factor Xa inhibitor was studied against an appropriate comparator (Also see "Xarelto's Comparative Efficacy To Warfarin, Pradaxa Will Be Focus At FDA Panel Review" - Pink Sheet, 6 September, 2011.).

The 14,000-patient ROCKET AF trial compared rivaroxaban against the standard of care, warfarin, for stroke and systemic embolism prevention in atrial fibrillation patients. The study was initiated in 2006, and the last patient contact occurred in September 2010.

FDA approved dabigatran in October 2010, one month after the last patient contact occurred in rivaroxaban’s pivotal trial against warfarin.

However, one month later, in October 2010, FDA approved dabigatran, making the direct thrombin inhibitor the first in a wave of novel oral anticoagulants to gain approval for use in atrial fibrillation. Approval followed the cardio-renal panel’s unanimous endorsement (see related story, Also see "Pradaxa Vs. Xarelto: How The Advisory Committees Voted" - Pink Sheet, 19 September, 2011. ).

The dabigatran 150 mg twice-daily dose was approved based upon data from the 18,000-patient RE-LY trial demonstrating a significant 35% reduced risk of stroke or systemic embolism compared to warfarin in the intention-to-treat analysis, with a similar risk of major bleeding (“FDA Pradaxa Approval Balances Efficacy And Bleeding Risk,” “The Pink Sheet,” Oct. 25, 2010).

Dabigatran’s labeling does not bear an overt superiority claim in the Indications section because FDA determined the drug’s better performance was driven by patients in the warfarin arm whose International Normalized Ratio (a measure of anticoagulation) was not well controlled. However, the label’s Clinical Studies section contains a “back door” claim reflecting the efficacy results and dabigatran's demonstration of superiority (Also see "Pradaxa Review Shows FDA Doubts About The Power Of REMS To Change Prescribing Practices" - Pink Sheet, 1 April, 2011.).

In ROCKET AF, rivaroxaban 20 mg once a day demonstrated non-inferiority against warfarin in the per protocol “on treatment” analysis and superiority in the safety “on treatment” population, but superiority was not shown in the ITT analysis preferred by FDA.

In briefing documents, FDA reviewers recommended a “complete response” action, saying the ROCKET AF data were not adequate to determine whether rivaroxaban is as effective as warfarin when the latter is used skillfully.

They maintained that interpretation of the ROCKET AF data was complicated by suboptimal INR control in the study. The mean INR time in therapeutic range was 55%, compared with 64% in RE-LY. Furthermore, FDA’s analysis showed rivaroxaban appeared to perform worse than warfarin at those sites where the mean TTR was 67.8% or higher.

Testing FDA’s “As Effective” Policy

FDA sought the advisory committee’s input on whether warfarin or dabigatran was the appropriate comparator for rivaroxaban in light of a 1995 agency policy that requires new therapies be “as effective” as approved treatments when the disease at issue is life-threatening or capable of causing irreversible morbidity.

Although the policy is broadly written, “stroke is a classic situation of a drug covered by this policy,” Division of Cardiovascular and Renal Products Clinical Reviewer Martin Rose told the committee, adding that the agency has previously applied the policy in this context.

For instance, FDA refused to approve the dabigatran 110 mg dose because it was inferior to the 150 mg dose. The agency also has stated that an NDA for a stroke prevention drug in atrial fibrillation cannot be based solely upon a trial that uses aspirin as an active comparator, Rose said.

The current landscape of anticoagulant therapy in A-fib is an important factor in answering rivaroxaban’s efficacy questions, FDA reviewer Rose said.

“I think that the current landscape of anticoagulant therapy for atrial fibrillation patients is an important factor in answering these questions,” Rose said. Although there was no good alternative to warfarin when ROCKET AF was designed in 2006, dabigatran’s approval changed all that.

“RE-LY demonstrated that dabigatran was robustly not inferior to warfarin in patients at sites with TTR above the study median of 67%,” Rose said. “I think it’s reasonable that rivaroxaban should have to meet the same standard, i.e., robust non-inferiority to well- managed warfarin.”

Since many patients are well managed with warfarin in clinical practice, “if they switch to another drug you should be comfortable that it is as good as warfarin when warfarin is used well,” Rose said. “The alternative could be that the patient will be less well anticoagulated and might have a stroke. Rivaroxaban does not meet this test.”

The sponsors sought to focus the panel’s attention on rivaroxaban’s efficacy as it compared to warfarin in ROCKET AF, not as it measured up against dabigatran’s efficacy in RE-LY.

Speaking on behalf of the sponsors, Robert Califf, Duke University, said it would be inappropriate to compare the treatment effects of rivaroxaban and dabigatran because of differences between ROCKET AF and RE-LY in study drug, trial design, population and geographic regions.

Comparing across clinical trials to rank drugs for safety and effectiveness “will be doing a disservice to the public health,” Duke’s Califf said.

“If we fall prey to a situation where indirect comparisons across clinical trials are used to rank drugs in terms of their safety and effectiveness, we will be doing a disservice to the public health,” Califf said.

The Oracle Of FDA: Bob Temple Weighs In

During the course of the meeting, Office of Drug Evaluation I Director Robert Temple indicated that although he believes warfarin is the appropriate comparator against which to judge rivaroxaban’s efficacy, the existence of dabigatran nevertheless changes the evidentiary threshold.

“You can’t avoid the fact that dabi exists,” Temple said. “I think it increases the insistence we have that the findings here seem robustly persuasive as to non-inferiority.”

Temple expanded on these views further when he was asked to clarify one of the questions posed to the committee. That question asked whether the ROCKET AF study population represented a sufficiently different group of patients from those in RE-LY as to render the “as effective” policy inapplicable to rivaroxaban.

ROCKET enrolled a greater proportion of subjects with multiple risk factors or a prior stroke than did RE-LY. FDA did not agree with the rivaroxaban sponsors’ plan to enroll higher-risk patients because they represented a sicker population than had been studied in historical stroke prevention trials that compared warfarin’s efficacy to placebo (see related story, Also see "ROCKET AF’s Design Causes Headaches For Xarelto Sponsors" - Pink Sheet, 19 September, 2011. ).

“This was not my favorite question,” said Temple, who proceeded to point out the bigger issue inherent in applying FDA’s “as effective” policy in a dynamic development and regulatory environment.

“There is an interesting and provocative issue here,” Temple said. “Suppose in the course of the study, something new and really hot comes along. Do we ever say, ‘I’m sorry, you compared it to a dog, we have something better now?’ We don’t usually do that, but I wouldn’t rule out the possibility. And this is a little bit about that possibility, but I don’t think we really meant to get too much into that discussion.”

FDA’s Temple said dabigatran’s existence warrants making sure that rivaroxaban’s comparison with warfarin “is really solid.”

Temple reiterated that the comparison the agency focused on when it discussed ROCKET AF’s design with the sponsors was rivaroxaban versus warfarin, and that remains the agency’s focus today.

However, “what dabi does is makes me at least more interested in being sure that the comparison with warfarin is really solid,” Temple said. “Whereas before there was any alternative, just having an alternative to the difficulty [involved in using warfarin] would have seemed like a very good deal, and any evidence of effectiveness might have been OK. I don’t think that’s true anymore because you have alternatives that are really solid, but I don’t think it makes dabi the comparator.”

Committee members generally agreed that the ROCKET AF population, though higher risk than that in RE-LY, was not sufficiently different as to warrant exempting rivaroxaban from demonstrating it was “as effective” as other treatments. The committee then was asked to opine on what therapy or therapies should be used as the comparator for the “as effective” test.

The committee’s non-voting industry representative, Jonathan Fox, AstraZeneca, suggested that interpreting the “as effective” policy to require comparisons of investigational drugs to newly approved therapies would wreak havoc on clinical trial programs, potentially requiring sponsors to stop ongoing studies that use established standard-of-care agents in favor of a newer comparator, even if the latter has not yet completed the FDA approval process.

When ROCKET AF was designed and conducted, dabigatran was still in development and had not been approved, “so dabigatran is irrelevant to the question as it’s driven by this policy,” Fox said.

Committee member Darren McGuire, University of Texas Southwestern Medical Center, said he interpreted the 1995 policy as requiring a demonstration of comparable efficacy against at least one approved therapy, and warfarin fits that bill. “I don’t think we should consider the dabigatran comparison,” McGuire said.

Like Fox, McGuire cited important practical considerations for drug development programs.

“It would be impossible to continue to execute these large-scale, randomized clinical trials without having to stop and start, stop and start as new agents come along,” he said, noting that other novel anticoagulants (such as Bristol-Myers Squibb/Pfizer’s apixaban) will soon be making their way down the approval pathway. “I think it just raises a bar that is impractical to achieve.”

Although the majority of panel members did not believe dabigatran should be the required comparator in assessing rivaroxaban’s efficacy, this view was not unanimously held.

Temporary voting member Thomas Fleming, University of Washington, said if there is sufficient data demonstrating dabigatran to be superior to warfarin, “then I am uncomfortable stating that the policy is addressed simply by showing you are non-inferior to something that is inferior to an existing therapy.”

By Sue Sutter

Xarelto Review Puts FDA’s Comparative Effectiveness Stance To The Test

More from Archive

More from Pink Sheet