What to Learn From the Sonus Blowup

There's nothing I'd like to do more than forget that Sonus Pharmaceuticals(SNUS Quote) exists, let alone remind readers that I looked favorably upon the company's cancer drug Tocosol paclitaxel in an August column.

But the disastrous results from the phase III study released Monday, and the absolute thrashing of the stock -- it plummeted 84% to 70 cents -- requires a review of what went wrong.

There's a lesson or two in this debacle, I hope.

I guess I deserve this to some extent, so I'll start by letting reader Larry K. take a swing at me:

"My goodness, how wrong can an analyst be," he asks (rhetorically, I assume.) "You're off my read list. Terrible analysis up and down."

Well, I actually think my analysis of Tocosol paclitaxel was OK, based on publicly available clinical data from previous studies. The problem was that the phase II data on Tocosol paclitaxel (which I'll call TocP from now on) turned out to be totally unreliable. It's not yet clear why.

So, my big mistake was placing too much faith in those past studies and not factoring in enough risk that the phase III trial could end up in a doomsday scenario of lower efficacy and higher toxicity -- which is what happened precisely.

When investment firms or stock-pickers warn you that past results are not a good predictor of future results, they mean it. The same goes with drug development. It's an old saw in biotech circles: Phase II results are almost always better than any data from a phase III study.