Updated from 12:29 pm ET with slide showing baseline patient characteristics.

BOSTON (TheStreet) -- Biogen Idec (BIIB - Get Report) shares are up an astounding 8.4% at $470.00 shortly after noon Friday on the back of the strong (but expected) clinical trial results from the company's Alzheimer's drug BIIB037.

I wrote about the BIIB037 study results earlier Friday, but here are some follow-up observations, additional analysis and read-worthy links:

Evercore ISI analyst Mark Schoenebaum held a client conference call earlier Friday featuring a medical expert in Alzheimer's disease. Overall, the doctor was very positive about the cognitive and amyloid plaque-busting data presented on BIIB037, but he also raised concerns about the high rate of ARIA (brain swelling) seen with the highest and most effective (10 mg) dose of the drug. That's just one doctor's opinion, of course, so it's not necessarily a consensus view.

At that high dose of BIIB037, 55% of patients that carry the ApoE4 gene (denoting a higher risk for Alzheimer's) reported signs of ARIA, and 35% of those patients were discontinued from the study.

The doctor on Schoenebaum's conference call labeled this rate of ARIA "unacceptable," disagreeing with the study presenters who said the brain-swelling side effect was moderate and manageable. He also said an acceptable rate of ARIA for Alzheimer's patients with the ApoE4 gene would be less than 20%.

The safety concerns about BIIB037 must be viewed in context with the strong efficacy signal on cognition and amyloid plaque reduction. If the 10 mg dose carries an unacceptable safety risk, the lower doses become more important. Biogen showed strong results with a 3 mg dose of BIIB037 during Friday's presentation, although not as robust as the 10 mg dose.

There is a group of patients in the study on a 6 mg dose of BIIB037, but we didn't see that data Friday because patients have yet to complete the 54 weeks of treatment. Everyone wants to see how the 6 mg dose performs through the full study.

Which brings me to the two main efficacy slides on BIIB037 presented this morning (photos taken by Schoenebaum):

Notice the mixed picture on the 6 mg dose (the red line), which only has results through 26 weeks. On MMSE, 6 mg looks very much like placebo (the yellow line) at 26 weeks. The 6 mg dose looks more in line when CDR is measured.

The steep decline in the performance of placebo patients in the study is also something to notice. It helped BIIB037's performance. In Biogen's defense, this study was also more stringent in making sure patients who enrolled had a confirmed diagnosis of mild Alzheimer's. In previous studies of other drugs, patients thought to have mild disease later turned out not to have Alzheimer's at all.

Not everyone believes BIIB037 is destined to become a blockbuster Alzheimer's drug. Earlier today, I heard from a buyside source who is always skeptical. He writes:

"How many times must provocative phase 1/2 Alzheimer's data be seen through rose colored glasses instead of with the proper scrutiny it deserves? The sell side can enter the orgasmatron if they wish. I think this is a study that benefitted greatly from baseline imbalances and a poorly performing placebo arm. There have been so many phase 3 failures using the very same mechanism. In today's market, that is worth $35 billion."

Therein lies the dilemma: In the current manic biotech bull market, the bias is to believe every drug works, or at least to downplay the risks. Let's not forget that similar Alzheimer's drugs developed by Pfizer (PFE - Get Report), Johnson & Johnson (JNJ - Get Report) and Eli Lilly (LLY - Get Report) have failed despite promising early data. Biogen's drug might be better than all proceeding Alzheimer's drug but since Biogen first discussed the positive BIIB037 data in December, the stock has also added around $35 billion in market value.

BIIB Chart

Don't count on sell-side analysts to caution investors about the risk still inherent in BIIB037, which faces another three to four years of large phase III studies before we get definitive answers about efficacy and safety. Analysts notes published this morning are euphoric on Biogen for the most part. I'm waiting for the first analyst to assign a probability of success for BIIB037 at greater than 100%.

Here are a couple of good stories on BIIB037 to read. Forbes' Matt Herper runs the BIIB037 data in front of some Alzheimer's doctors. Also read chemist and blogger Derek Lowe's objective take on Friday's results.

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