Replies to post #108959 on Biotech Values

[jq1234]

Bayer, J&J's Xarelto Bests Warfarin Without Raising Bleeding Risk in Study

Johnson & Johnson and Bayer AG’s blood thinner Xarelto prevented strokes in patients with an erratic heartbeat better than standard therapy with warfarin in a study, without raising the risk of bleeding.

Patients taking Xarelto once a day were 21 percent less likely to suffer a stroke or embolism than those on warfarin, a 56-year-old medicine first used as rat poison, researchers said at the American Heart Association’s annual meeting in Chicago. A second analysis of the data using more stringent methods found Xarelto was equal to warfarin. Bleeding risk, a feared side effect of therapy, was similar.

The study positions Xarelto to take at least a third of the market for warfarin replacements, following the approval of a rival drug called Pradaxa in the U.S. last month, Savant Ahmed, a London-based analyst for the Royal Bank of Scotland, said in an e-mail. Bayer has estimated the market for new blood-thinners including Xarelto could surpass $14 billion in annual sales.

“If you look at the data, Pradaxa and Xarelto are just a lot better than warfarin,” Leslie Iltgen, a Frankfurt-based analyst at Bankhaus Lampe KG, said in a telephone interview. The market is big enough to support both drugs, said Lampe, who recommends buying Bayer shares.

Xarelto may generate combined peak sales for Bayer and Johnson & Johnson of $3.9 billion by 2020, according to estimates from Sanford C. Bernstein & Co. Their calculations assumed the Bayer drug would be about as effective as Boehringer Ingelheim GmbH’s Pradaxa.



http://www.bloomberg.com/news/2010-11-15/bayer-j-j-s-xarelto-beats-warfarin-without-elevating-danger-of-bleeding.html

[DewDiligence]

Re: Xarelto’s superiority claim or lack thereof

I watched the live webcast from AHA this morning when Forbes’ Langreth questioned the presenters at the session for late-breaking results, which formed the basis of the article you posted.

The argument by Dr. Hylek dismissing the per-protocol analysis (which showed Xarelto superiority to warfarin) because of Xarelto’s relatively short half-life is a red herring, IMO. I thought Dr. Califf refuted Dr. Hylek’s argument during the Q&A session by noting that the “in treatment” period used for the PP analysis included two days beyond a patient’s last dose and also included time when a patient was taken off drug for surgery.

The Xarelto claim of superiority in the PP analysis should IMO be dismissed for a much simpler reason: that ITT analysis is the preferred yardstick for results in a superiority test. (In a non-inferiority test, on the other hand, PP analysis is preferred for reasons that have previously been discussed on this board, e.g. in #msg-42490033.)

Whether the lack of a superiority claim for Xarelto relative to warfarin in the FDA label will hinder Xarelto’s commercial success is harder to say. My guess is that it will make some difference, but not a lot.

Comments?

[jq1234]

Controversy Builds Over JNJ Blood Thinner Trial Design

Nov. 15 2010 - 1:06 pm | 105 views | 0 recommendations | 0 comments
By ROBERT LANGRETH

http://blogs.forbes.com/robertlangreth/2010/11/15/controversy-builds-over-jnj-blood-thinner-trial-design/

J&J is out today with a giant new study it claims shows the blood thinner rivaroxaban is superior to standard drug warfarin in preventing strokes in patients with atrial fibrillation. But the only analysis that showed the drug is superior did not include the many patients who dropped out of the trial. This is raising questions over whether J&J is emphasizing an unconventional analysis, one that will not be taken at face value by the Food and Drug Administration. It is not a question of whether the drug will be approved–it almost certainly will–but whether the drug will be able to snare marketing claims that give it an advantage over the competing drug from Boehringer Ingelheim.

Here is what analyst Derrick Sung and his colleagues at Bernstein Research had to say in a research note today:

Having seen the results, we continue to think that Pradaxa [the rival drug from Boehreinger Ingelheim] will be the market leader. Although it is always hard to compare between very different trials, Xarelto does not look compelling vs. Pradaxa on either efficacy… or safety. The Bayer and JNJ press releases suggested that Xarelto is “superior” to warfarin in terms of efficacy on the primary endpoint of stroke and embolism prevention, and broadly equivalent to warfarin in terms of major bleeds. However, the AHA press release, using a more stringent and more conventional statistical approach (see footnote for an explanation[1]), notes that efficacy of Xarelto was not statistically superior to warfarin on an “intent-to-treat” analysis. Pradaxa, on the other hand, did show statistical superiority to warfarin on this “intent-to-treat” basis. We think the FDA is likely to be similarly conservative in its statistical approach.

In a footnote, he added this about J&J’s unconventional data analysis:

But if, on the other hand, one is looking for superiority, one generally picks the “intention to treat” population because you don’t want to flatter an effective medicine that patients don’t actually take, for reasons of intolerability perhaps. Therefore, it seems to us that Bayer and JNJ were being unconventional in emphasizing superiority results derived from a per-protocol population, when the per-protocol analysis was really intended to demonstrate the primary endpoint of non-inferiority.

[DewDiligence]

Bayer reports Xarelto subgroup data from ROCKET-AF study for patients who had a prior stroke/TIA (the “secondary-prevention” indication):

http://www.press.bayer.com/baynews/baynews.nsf/id/Major-Subgroup-Analysis-Shows-Bayers-Rivaroxaban-Highly-Effective-Prevention-Recurrent-Strokes?Open&ccm=001

The secondary prevention subgroup of the ROCKET AF study, consisting of patients with a prior history of stroke or TIA (representing 55 per cent of the total study population), received either 20mg once daily rivaroxaban (or 15 mg for patients with moderate renal impairment), or dose-adjusted warfarin. The ROCKET AF principal safety outcome was major and non-major clinically relevant bleeding. Rivaroxaban showed a numerically lower (13%) risk compared to warfarin in the prevention of recurrent strokes or non-CNS embolism in the safety population while on-treatment (2.26 events per 100 patient-years for rivaroxaban and 2.60 events per 100 patient-years for warfarin, HR 0.87, 95% CI [0.69, 1.10]).

In other words, Xarelto was better than warfarin in stroke (or non-CNS embolism) prevention, but the 13% reduction in risk was not statsig. The reduction in risk would have had to be in the high teens (HR in the low 80s) for this comparison to have been statisg.

What about bleeding?

Similar rates of overall bleeding to those seen in the primary ROCKET AF analysis were found, with no difference between the treatment groups (13.31 safety outcome-related events per 100 patient-years for rivaroxaban and 13.87 events per 100 patient-years for warfarin, HR 0.96, 95% CI [0.87, 1.07]).

In other words, Xarelto’s bleeding rate was lower than warfarin’s by 4%, which was not close to being statisg. Bleeding rates in both arms of the prior-stroke subgroup were similar to those in the overall trial.

What about intracranial hemorrhage (the scariest form of bleeding)?

Rates of ICH were numerically lower in the rivaroxaban arm (0.59 events per 100 patient-years for rivaroxaban and 0.8 events per 100 patient-years for warfarin, HR 0.74, 95% CI [0.47, 1.15]).

In other words, Xarelto’s ICH rate was 26% lower than warfarin’s, but the number of events was too small for this comparison to be statisg.

All told, these Xarelto subgroup results are impressive, IMO, despite the fact that they are PP rather than ITT data. The subgroup data are consistent with the outcome in the trial as a whole, which found Xarelto non-inferior to warfarin on an ITT basis and superior to warfarin on a PP basis in stroke prevention, without a significant increase in bleeding (#msg-56712414).