About that 17% bump in Athersys'  (ATHX - Get Report)  stock price Thursday: Don't be a sucker.

The Athersys stem-cell therapy MultiStem doesn't help patients recover from stroke any better than a placebo. Negative results from the randomized, double-blind, placebo-controlled study were disclosed last April, after which Athersys' stock price was cut in half.

Athersys used a "global test analysis" encompassing three different stroke-recovery measures -- the modified Rankin Scale (mRS), the NIH Stroke Scale and the Barthel Index -- to assess MultiStem's ability to improve recovery from a stroke after 90 days. On this measure, the study's primary efficacy endpoint, MultiStem was no different than placebo.

Secondarily, Athersys looked at each of the three stroke-recovery measures individually to detect a difference between MultiStem and placebo. The company found no difference.

The data from the Athersys study do not lie, yet Gil Van Bokkelen, the CEO of Athersys, hasn't conceded the failure of his company's only product in clinical trials. He and his team have spent the last 10 months making excuses and spinning statements about "positive" MultiStem stroke results. These statements would never hold up to regulatory scrutiny, but repeated often enough in press releases they take on a veneer of truthiness sufficient to lure retail investors who don't know better.

Last April, Athersys claimed patients treated with MultiStem 24-36 hours after having a stroke recovered better than those treated with a placebo. A promising result? Not really. The analysis -- conducted post-hoc, of course -- required Athersys to erase and ignore the detrimental outcomes reported from 60% of the patients treated with MultiStem.

If you still have trouble understanding why Athersys' MultiStem spin job doesn't work, think about this way:

"Captain Smith, how was the maiden voyage of the Titanic?"

"The trip was great, except for hitting the iceberg."

On Wednesday, Athersys claimed stroke patients treated with MultiStem demonstrated a statistically significant "higher rate of complete or nearly full recovery" compared to placebo patients after one year of follow-up.

Let's unpack this: For starters, stroke is an acute illness requiring immediate medical intervention. A clinically meaningful outcome for stroke victims is improvement or recovery measured days or weeks after the stroke, not one year later.

Athersys defines "complete or nearly full recovery" as patients achieving an "excellent outcome," which is just a dumbed-down revision of the more stringent "global test analysis" measure used as the primary endpoint in the study. [The primary endpoint which MultiStem couldn't achieve last April].

After 90 days of follow-up -- the only time point that matters -- MultiStem also failed to demonstrate an improvement over placebo using the "excellent outcome" definition of stroke recovery. This just underscores its weakness: The stem cell therapy couldn't even muster a win over placebo using a remedial definition of stroke recovery.

If MultiStem efficacy defined as an "excellent outcome" failed at 90 days, the same outcome measured at one year cannot become statistically significant. There are no magical bio-statistical mulligans allowed in clinical trials.

MultiStem was injected into patients once within 48 hours of their stroke. For Athersys to claim its stem-cell elixir is finally helping stroke patients one year later would be like reading this headline in the New York Times on April 15, 1913:

One Year After Sinking, Titanic Survivors Outlive Those Who Perished at Sea.


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.