Lou Gehrig's Disease - Motor Neuron Disease - Amyotrophic Lateral Sclerosis
Thought it had been cured by now? Still no known cause. Still no cure. Still quickly fatal. Still outrageous.

Wednesday, November 18, 2020

Home Run. Left Field Fence.

Yesterday news broke.  Hearts broke.  

There were statistics.  There was spin.  There were excuses.  There were reasons. 

But there wasn't a home run over the left field fence. 

The news:

There is no doubt to those of us with friends in the Brainstorm clinical trials that the therapy was helpful to some, yet the trial did not deliver statistics to back that up.  We've seen this before in ALS trials.  Unless we stop this madness, we'll see it again. 

In clinical trials, you have to call your home run shot before you take a swing.  Left field fence?  Ok. You have to stand in the batter's box and point to your home run.  Left field fence.  That's your endpoint.  If you swing and it goes into right field (or even clears the right field fence), that's nice but it's not the home run you called.  You said left field fence.  Otherwise it doesn't count.

You also get to call the pitch you want. That should make it easier, right?  Those are your inclusion criteria.  Low in the strike zone with a fast ball.  You get the pitch you want and you have to hit your home run where you say it will clear the fence.  If you call the wrong pitch, you miss your shot.  Unfortunately with ALS, we're not very good at having a pitching machine that puts the ball precisely where you want it. 

In baseball where lives aren't at stake, we analyze swings and pitches and statistics.  Coaches and players work tirelessly to refine their games and improve.  In ALS research where lives are indeed at stake, we make a lot of mistakes over and over. We settle for excuses.  We accept reasons.  It is madness.

The NurOwn trials from Brainstorm have been painfully long and cruel to watch.  The last thing we expected was a right-field double when they said the trial had to be designed with 50:50 placebo and exclusion criteria to give them their statistical left-field-fence home run.  

There were also humane features that were missing.  They couldn't afford them, Brainstorm said.  Perhaps we'll figure out that sponsors can't afford not to include features that give trial participants continuing access to the real therapy if the participants feel it was helpful.  Think of the compelling data (albeit non-trial data) that they would have generated by this time.

A trial participant made an interesting and informed observation yesterday.  He felt that the neuroprotective properties of the therapy helped his still-viable motor neurons but not motor neurons that were too far gone.  That makes a lot of sense in light of the data we saw.  That also makes it all the more cruel to let people who had been in the trial and felt improvement to be abandoned after trial participation.  More motor neurons died as they lost access to their neuroprotection.  And many of them died.

Again in this trial, we have the challenge of the overachieving placebo group.  The disease is heterogeneous.  The random placebo group can help you or hurt you.  Get a bunch of fast progressors on placebo and even a mediocre therapy will look good.  Get a bunch of slow progressors in the placebo group and your therapy is statistical toast.  Hey, it's random. right?  We have this problem over and over and over.  It is madness.

We have somebody stranded on second base as a result of the Brainstorm trial.  Do we call the game over?  

I hope not.

As much as I don't want to reward inhumane trial designers, can we use the trial data and the tools that the FDA will have available to keep advancing the runner?  We can be creative and still deliver good science.  We need to do the same things that good coaches do.  Analyze and react quickly and make the most of what we have.

And we need to make sure that the next batter up does better -- better called shot, better swing, better pitch, better result.