Myriad Genetics Doesn't Make the Grade

Myriad Genetics(MYGN - Cramer's Take - Stockpickr) is one of several companies trying to hit the Alzheimer's disease jackpot by developing a drug that blocks the formation of toxic beta-amyloid protein deposits in the brain.

At the risk of sounding too much like a nattering nabob of negativism, I believe the odds for success here are low. (I know, first I take a bearish stance on Neurochem(NRMX - Cramer's Take - Stockpickr) and its Alzheimer's drug, and now I do the same with Myriad.)

What's my beef with experimental Alzheimer's drugs? There's no beef, I promise -- I simply believe these first two Alzheimer's drugs with disease-modifying potential are lacking.

Flurizan is Myriad's beta-amyloid-lowering drug now being evaluated in two phase III trials. On Tuesday, Myriad announced that it was punting on a planned 12-month interim analysis this fall of the first of these studies. Instead, the company will let the trial run through to completion in the first half of 2008.

So, with data from the Flurizan study not coming until next year, isn't it a bit early to worry about this as a stock-moving event? Agreed, but at the same time, I view the canceled interim analysis as a crack in Myriad's confidence in its Alzheimer's drug candidate.

Myriad's stock has been on a tear because the company's base genetics-testing business is profitable and growing. But my general feeling is that Myriad's valuation is overinflated because of Flurizan.

As a stand-alone business, genetic testing (or diagnostics) tends to have lower profit margins than selling drugs. So, subtract Flurizan from Myriad's plate and the company wouldn't (shouldn't) deserve the premium multiple it gets today.

Myriad shares closed Wednesday up 3.5% to $37.11.

All right, let's focus on Flurizan. Like with my analysis on Neurochem, my skepticism is grounded in poor phase II data, which suggests the phase III trials are high-risk.

Last July, Myriad announced results of a 24-month, follow-up analysis of its phase II study that found patients with mild Alzheimer's disease receiving high-dose Flurizan (800 milligrams dosed twice a day) "demonstrated a substantial benefit over other dose groups in the study and that this benefit continued to increase over 24 months, for each of the evaluated areas of cognition, memory loss, global function and activities of daily living."

These data were presented at the 10th International Conference on Alzheimer's Disease and Related Disorders, in Madrid, Spain, by Dr. Gordon Wilcock of the University of Oxford. Wilcock was the lead investigator for the phase II trial.

Here's one of the primary slides from Dr. Wilcock's presentation:

Click here for larger image.

Note that patients in both Flurizan dose groups, as well as the placebo patients, all seem to do equally well for the first nine months of the study. (The error bars for each three-month time point either overlap or almost touch each other, suggesting that the point differences shown on the graph are a bit suspect.)

From months 9 to 12, however, the 800-mg dose group and placebo patients decline dramatically. The steep slope declines of the graphs represent a loss of cognition, as measured by a mean change in the ADAS-cog test.

The primary endpoint of the phase II study was measured at 12 months, and it's at this point that the study failed to show a statistically significant difference between Flurizan and placebo using ADAS-cog and other tests.

But Myriad wants you to forget about the results at 12 months and instead focus on what happens to patients who continue taking Flurizan through 24 months. Here, in this voluntary continuation of the phase II study, Myriad claims that the cognitive ability of high-dose Flurizan patients stabilizes -- and even improves at some time points -- from 12 to 24 months, as illustrated by the flattening of the slope of the curve.

What might explain this dramatic improvement?