Response -- both supportive and critical -- has poured in over my story Wednesday on the Alzheimer's drug bapineuzumab that is being developed by Elan ( ELN) and Wyeth ( WYE).

To summarize, the story detailed how the two drugmakers are using their own newly created endpoint in a phase III trial for bapineuzumab, rather than the industry standard that has been the basis for all Alzheimer drug approvals to date.

The story is important enough (in my humble opinion) and the feedback strong enough to justify a special Friday edition of the Biotech Mailbag. I'll devote this space entirely to the Elan-Wyeth story so that I can leave room for lots of other good stuff in my regular Saturday Mailbag.

I'll start with a research note issued by Corey Davis, the specialty pharmaceuticals analyst at Natixis Bleichroeder. He has a buy rating on Elan and a $31 price target, so naturally, he took issue with my cautious, if not bearish, thesis. (Natixis has no banking relationship with either Elan or Wyeth.)

Davis says my story wasn't new news because "Elan has been very public for years that both the NTB Neuropsychological test battery and the more traditional ADAS-Cog will be examined as a measure of cognition in their Alzheimer's studies."

He goes on to write that the Food and Drug Administration has vetted Elan's approach to the phase III trials, including approving the company's NTB endpoint. He also wrote that the company has an agreement from the FDA that it can adjust the endpoints of the phase III studies depending on how the phase II bapineuzumab study turns out. We get to see those data mid year.

"Officially in the Investigator's brochure, NTB is listed as the primary endpoint of the phase III trials, but again, they can change this at some point well before the data are unblinded," Davis wrote. "There can always be improvements in scales used for clinical trials -- especially ones that are as subjective as Alzheimer's. Elan is not trying to pull a bait and switch but simply trying to advance the science and pioneer improvements."

Davis fairly encapsulates much of the criticism of my story, which boils down to a belief that Elan and Wyeth wouldn't invest tens, or even hundreds, of millions of dollars in an Alzheimer's drug program without making sure that everything they're doing is by the book and green-lighted by the FDA.

I do appreciate Davis' further confirmation of a central point in my story -- that the NTB is listed as the primary endpoint of the phase III studies. I find that interesting for a simple reason: If bapineuzumab is truly the effective drug against Alzheimer's that Elan bulls believe it to be, then why the need for a new and unproven cognitive endpoint to prove it?

Let me put that another way: Given that Elan took a look at interim data from the bapineuzumab phase II study before signing off on a decision to move to phase III, why not just use ADAS-cog -- the gold standard test for cognition -- as the primary endpoint of the phase III studies if the drug was, indeed, demonstrating a great improvement on that scale in the phase II study?

I don't know what the phase II data on bapineuzumab are going to look like. I readily concede that. But it doesn't make sense to me for Elan to be fiddling with a new endpoint if the old and well understood endpoint is working just fine.

We will know for sure when the phase II data are released by Elan in a few months. If bapineuzumab causes a dramatically positive improvement on ADAS-cog, then I'll gladly acknowledge the error in my thesis.

There is one thing that Davis and I do agree on, and he says it later in his research note: "We doubt very much the FDA would ever approve a drug like bapineuzumab that showed zero effect on the ADAS-Cog and did show an effect on NTB."

The data will speak for itself, soon.

Moving on, I received this note from Robb K.: "Take care of an AD Alzheimer's patient for a week and then write the article. Cognition is great if you are working and need to think, just being able to get around the house and find your shoes and wallet is the big part. Until you personally live it with someone that you love, you have no expletive deleted clue. I pray that you and all of the others of your generation never have to experience the soul-wrenching experience of watching someone you love have the mind slowly dissolve before your eyes."

I sympathize with Robb greatly. I do have some personal familiarity with age-related senility and dementia, but nowhere close to the experience or heartbreak that's suggested in his letter. Alzheimer's is a truly devastating disease for the person afflicted and his families and caregivers.

With that said, I'll make two points. First, don't get emotionally involved in your stock portfolio. These are investments, period. That requires a cool, objective detachment. Emotions cloud judgment, no matter how good the intentions.

Second, I focused my story on the requirements to show cognitive improvement; what Robb talks about are functional measures -- stuff like getting dressed by yourself, or being able to brush your teeth or find your keys. The FDA wants to see improvements in both cognitive and functional measures before granting approval to any new Alzheimer's drug. It's not an either/or option for bapineuzumab.

Harry Tracy of NI Research and the publisher of the NeuroInvestment newsletter chides me for being somewhat unfair to Elan and Wyeth over trying to come up with a better cognitive test.

"The ADAS-cog is reviled by most dementia researchers that I know as a blunt instrument which misses potentially important aspects of neuropsychological function -- delayed verbal recall and executive functions being some of the most important omissions," he wrote.

Tracy goes on to say, "The NTB's components are far more validated, where Elan and Wyeth are pioneering is in trying to develop a composite score from those pieces. Somebody had to eventually challenge the FDA's reliance on a substandard measure -- Elan and Wyeth may be self-serving, given failed Alzheimer drug AN-1792's history -- but there are a lot of companies rooting for the NTB... to become an accepted cognitive measure. I am too." (Tracy's newsletter has no position in Elan or Wyeth.)

I respect Tracy's opinion a great deal, but I'd be more comfortable if Elan and Wyeth were not both trying to develop bapineuzumab and the cognitive test by which the drug will be judged. Isn't that a conflict of interest?

Likewise, the FDA may very well agree with Elan and Wyeth that the NTB is a better test than ADAS-cog, but, as I detailed in my story, I wonder why the companies didn't secure a Special Protocol Assessment with the regulatory agency to lock that agreement into place. The lack of an SPA suggests, to me at least, that the FDA remains unconvinced about the validity or objectivity of the NTB.

Let me end this special Mailbag with a reader who likes my story. Dan S. writes, "Awesome article. Incredibly well written."


Adam Feuerstein writes regularly for In keeping with TSC's editorial policy, he doesn't own or short individual stocks, although he owns stock in He also doesn't invest in hedge funds or other private investment partnerships. Feuerstein appreciates your feedback; click here to send him an email.