21 April 2007

EPO and Aranesp

Few drugs have had the sustained attention from the drug safety community given to EPO and its derivatives. Oral contraceptives, anti-psychotics, perhaps NSAIDS are the only rivals in this regard. EPO's safety concerns began with its launch, when the issue of what would happen if the drug were overused in patients with marginal endogenous production of the substance was raised. Then, there was the pure red cell aplasia episode in Europe. That EPO was involved in the outbreak everyone accepts. Exactly how it occurred is still a bone of contention among some. Now we have safety concerns occurring with Aranesp.

EPO has been a phenomenally successful franchise for Amgen and J&J. For the former, it has been at the core of the company's ascent into the rarified air of multi-billion dollar in capitalization pharmaceutical companies. So successful has Amgen become that the city fathers in Thousand Oaks have supposedly told the company to expand elsewhere. Which it is doing, in South San Francisco, about 400 miles up the road in the international hub of the biotech industry and in the shadows of its rival, Genentech.

This all brings us to Aranesp, a second-generation EPO. Trying to develop its EPO franchise further with expanded indications for Aranesp, Amgen set up a trial with Aranesp to show that increasing hematocrit would lead to more oxygen deliver to tumor cells in head and neck cancer cases. The increased amount of oxygen would lead to the production of more free radicals during the radiation therapy at the core of treating such cancers. The unexpected result of the trial, though, was just the opposite of what was expected, and it resulted in some pundits suggesting Aranesp helped the tumor thrive and survive. No one questioned the safety of Aranesp in its labeled indication. Just this new area. However, since the drug is used in oncology, the risk of a safety problem in a key therapeutic area was no doubt troubling to Amgen. It was certainly troubling to investors. Everyone became concerned with every trial result for this drug.

Recently, one such trial came to conclusion without any indication of a safety problem. Investors (and analysts) sighed in relief. Except the trial really doesn't address anything but the crudest of questions, since the trial isn't really large enough (and wasn't designed to be large enough) to determine if there is a safety problem with the drug. What we do know is no registry data have yet been released indicating whether there's a problem. Nor have any trials an order of magnitude larger than those recently reported on been undertaken to demonstrate the product's safety. (It seems unlikely such a trial will be mounted any time soon.) What we also know is that there's lots of questions, but little clear evidence of a problem. Which isn't to say a problem doesn't exist.

Absence of evidence isn't the same thing as evidence of absence. Never has been, never will be.

16 April 2007

Ranolazine: Safety First

I'm sure you've heard of a company in Silicon Valley by the name CV Therapeutics. CVT has a product on the market by the name of Ranexa (generic name of ranolazine). This is a drug developed for people with angina. Just one hitch: it's associated with a prolonged QTc interval. This means the heart in a patient using the drug has some risk of developing an arrhythmias. Not a small problem in such a patient population. The concerns about this adverse event (nice understatement) have restrained the medical community in the United States from using this product--less than $25 million was sold last year in the U.S. The company is burning about $300 million per year, and it has about a year's worth of cash on hand, so unless the drug catches fire in the marketplace, it seems likely there will be some layoffs for the company. After a recent trial result missed its efficacy endpoint, Wall Street practically wrote the company off for the corporate dead. (Rumor has the trial costing nearly $100 million.) It's true that the same trial didn't confirm a safety problem, but such trials are never good for those purposes. If they were, then the safety problem is so onerous, the drug wouldn't stand a chance in the market.

One might think that the management of CVT would have done everything possible to position the drug to be accepted by the medical community. And one would wrong in that thinking. As a recent paper in Clinical Therapeutics (2006;28:1996-2007) noted, there hasn't been one cost-benefit analysis of the product published. It appear there is no registry of ranolazine users, either. Such registries are commonplace in the pharma industry today, and with a large registry of users followed for considerable periods of time, one can get enough statistical power to examine adverse events like QTc prolongation or arrhythmias. Certainly better than in a randomized trial! But CVT's management doesn't appear think such basic tools are part of risk management--and there's no clear reason. It also doesn't provide for much outside, independent evaluation of its risk management program to assess the effectiveness of the program. Let's hope the FDA is on the ball with this one, since it's the only one able to force any changes in the program.

Even more intriguing, the grapevine has the company just getting around to hiring a VP for drug safety. Kind of like closing the barn door after the horses are long gone. And yet the CEO has the time to serve on the Board of Directors of two other companies--Metabolex and Maxygen. Maybe he's pretty confident that the lack of sales of this product are a minor problem associated with 2006 and that the safety issues are now going to disappear?

Stay tuned--we're keeping track of this situation. Very curious indeed.

14 April 2007

Pfizer, torcetrapib, and HDLs

It's time to spill some ink on torcetrapib. Everyone knows this drug by now--"son of Lipitor" which couldn't make it to an NDA. In a late Phase 3 trial, persons taking torcetrapib had higher mortality than those not taking it. Definitely not a good thing. (Some years back, Bristol-Myers Squibb had an antibiotic, desquin, which had higher mortality associated with its use among persons with abdominal abscesses. After the company finally realized it would never get the drug to market (it also had a slight hypoglycemia problem), it gave the drug back to its Japanese discoverer, which promptly licensed it to Schering-Plough. Lord only knows what Fred Hassan was thinking when he made the deal. But I digress...) The idea with torcetrapib was to raise HDLs. Data from Framingham and lots of other epidemiologic data suggested that HDLs were important to reduce cardiovascular risk. Some suggested it was even more important than LDLs, which are the target of the omnipresent statins like Lipitor. Niacin, which raises HDLs (if you can tolerate the hot flashes), doesn't appear to raise mortality; on the contrary, it lowers it, enough for Abbott Labs to buy Kos Pharmaceuticals just to get into the niacin market. So what gives with torcetrapib?

It turns out it might not be a matter of increasing levels of HDL per se. It turns out HDL molecules carry lots of stuff inside beside lipoproteins. For example, protease inhibitors, which might help stabilize atherosclerotic lesions, aka plaque. That's just an example. There are lots of others. I doubt torcetrapib does anything about this other stuff. It works to increase HDL levels, period. While Pfizer shareholders will likely not be heartened by this fact, for the rest of us, this realization carries two messages: first, raising HDL alone may not be the best path to reducing cardiovascular disease (and identifying a potential blockbuster), and second, pharmacological approaches are often not as compelling as we first thought.

On the other hand, one wonders why the good folks at Pfizer didn't try pairing Lipitor with niacin (which is generic, I think). Maybe there's a patent issue, but I doubt it. It certainly would have been safer than torcetrapib. In any case, one of the lessons from the torcetrapib is that not all drug safety issues are post-marketing (after marketing approval has been obtained), and not all of drug safety is sturm and drang. While Pfizer shareholders didn't benefit much from torcetrapib, the rest of us did.

To CET: Many thanks.

11 April 2007

Merck and Arcoxia

Merck is still pursuing getting Arcoxia to market. Does anyone really think there will be much interest in the medical community in using this drug? I don't.

Why is Merck pursuing a drug likely to be a commercial failue? Good question you've asked. The simple reason is that Merck wants a defense in Vioxx cases. If Arcoxia, which is even more selective as a COX-2 blocking agent, can win approval from FDA, implying the drug's safety, then how can one suggest Vioxx was not safe for use by the general public?

The line of reasoning hold so long as Arcoxia is approved by FDA.

And there's the rub: I don't think it will be approved.

Safety: 1 Industry: 0

Update (14 April): Reason prevailed. The Advisory Committee overwhelmingly sided against Arcoxia.

09 April 2007

The Brave New World

There's finally lots of talk about changes in the US drug safety system. The notion of active surveillance is rearing its head. Don't be surprised if it supplants the existing passive system all the medical officers at FDA are used to. The passive system produces data no one can interpret. The active system doesn't leave a lot of room for interpretation--the rates are elevated or they're not. That's not to say there isn't room for differences of opinion. There's lots of that kind of room, and as usual, it's the opinion of the FDA which will govern what makes it onto the label and what doesn't. However, the possibility exists for some rational labels, ones based on data the essense of which everyone can agree. Sounds like too much sanity for the pharmaceutical world? Perhaps. But it beats having to license technology from the likes of Lincoln Technology just because everyone else (read: FDA) has it already. Drug Safety: 1 Data Mining: 0. That's a score which looks pretty from where I sit.

07 April 2007


This is a blog about drug safety. Because of its subject and includes lots of stuff from within the industry, it is written anonymously.

Will everything written about here be absolutely true? I wish I could say that it will be, but the truth is, I'm not everywhere and this isn't my day job. It will include, however, clear indications that something's been whispered and may not be established fact.

How often will I post? Don't know as yet. Probably not every day. But who knows, maybe I'll get to that point.