« More on C&E News |
Main
| Two Bad Ideas »
April 20, 2010
Bits And Pieces
Posted by Derek
Still traveling, so not much time to update things. Here are a couple of interesting links, though:
Adam Feuerstein's take on the Rexahn clinical data (which I spoke about here). He's not all that impressed, either, to put it delicately.
Big Pharma bonds getting downgraded - orrg. Smaller companies tend to raise more money with equity, but debt financing gets more and more important as time goes on, so this isn't a good sign.
Comments (4)
+ TrackBacks (0) | Category: Current Events
POST A COMMENT
- RELATED ENTRIES
- › Why Not Share More Bioactivity Data?
- › An Anticoagulant Antidote
- › Merck's Liptruzet: A Cause For Shame?
- › Your Brain Shifts Gears
- › Total Synthesis in Print
- › Things I Won't Work With: Dimethylcadmium
- › Another Germ Theory Victory - Back Pain?
- › An Update on Deuterium Drugs
1. JOhn on April 20, 2010 9:47 PM writes...
I think the concerns about p values are misguided, and I'm not really too concerned about retrospective subgroup analysis in an exploratory trial.
But what I think would have wrapped this up for me even before seeing the results of this trial was the realization that they were testing doses that are on the order of 4 - 10% that contained in a single Augmentin tablet. Maybe if they drop down to 0.01% the Augmentin dose they can cure cancer too?
I don't think Adam picked up on this or he wouldn't have been asking about the side effect profile.
Permalink to Comment2. Hap on April 21, 2010 11:40 AM writes...
No, but I think even before he got to that, he would have hit the showstoppers of why the effects were seen in low doses but not mentioned in high doses, why the trial results were segmented as they were (presumably after the fact), and why no overall data has been presented.
Feuerstein's takedown seemed pretty thorough and rational, but I don't know enough relevant to know if he's missing or eliding anything.
Permalink to Comment3. Skeptic on April 21, 2010 10:57 PM writes...
"Big Pharma bonds getting downgraded"
Go see your doctor.
Permalink to Comment4. TFox on April 22, 2010 6:21 PM writes...
Feuerstein points out two things I missed on reading the press release, despite my attempts to read it carefully. 1) The dose group for which they report positive results is the smallest dose, suggesting a lack of dose response. 2) They measured multiple scales of depression, but reported on only one. (The one they reported, MADRS, was their primary outcome measure, but still.) While we don't know what the data is for those other groups, it's not unreasonable to assume that it looks worse than the data they did report. The missing dose-response relationship is particularly troubling, and suggests that the positive result they did find is just a fluke.
I still think you can use a nonsignificant result to estimate an effect size in order to help plan a larger trial, but you're still not allowed to cherry-pick data.
Permalink to Comment