Sunday, September 6, 2009

Saphris®: It's Different Without Actually Being. . . "Erh, Different": The MacGuffin Blog


First, a sincere "Hat-Tip" to Marilyn Mann, for pointing me to this blogger's content!

Now, if you don't mind occasionally-blue language, this snarky review of Kenilworth's efforts to market Schering-Plough's recently-approved asenapine product -- under the trade name Saphris® -- is both provocative, and illuminating. You'll need to visit The MacGuffin's blog, to follow and understand his footnote links, so do go read it all (graphics plainly included -- nice!).

I am hopeful that one particular Salmon will rise, here, in comments, over the long Labor Day weekend -- and offer his ever-welcome perspectives on it all. Of course, at the outset, due to Salmon's earlier efforts, we all know there are several studies on asenapine in humans, despite MacGuffin's concern in footnote 7, below -- but that is decidely not good news -- as we also know what those other human trials have shown -- see image at right; click to enlarge.

In any event, without additional ado, then:

. . . .Question: Should I be at all be concerned that there are more actual published peer-reviewed articles of this drug being tested in rats2, 3, 4, 5, 6 than in humans7 (seriously, this is the only published peer-reviewed article I can find)?. . . .

The manufacturer [Schering-Plough/Merck] is pushing the drug on the premise that it improves the negative and cognitive symptoms of schizophrenia better than other drugs, in addition to having a better safety profile.

The safety profile angle has been used before11. While the drug demonstrated less weight gain compared to risperidone or olanzapine, it has an elevated level (18%) of extrapyramidal symptoms (EPS) comparable to first generation antipsychotics12.

Don't worry folks, there's a way to obscure that little fact. All you have to do is generate a drug trial where you compare the drug to the very potent D2 blocker, haloperidol, thus making any incidence of EPS seem minuscule13. . . .

~~~~~~~~~~
UPDATED: 09.07.09 @ 9:16 AM EDT
~~~~~~~~~~


Indeed -- cue Salmon -- here is part one, of his two-part elaborations (also set as a new post, here):
. . . .I do believe MacGuffin has led me to strike gold once again and I will address this in a second post (in Part 2, coming soon), but first let me address MacGuffin's questions.

Yes, the lack of published human studies is a signal, but minor compared to what we’ve already discussed.

The rat and receptor studies are as MacGuffin suggests selective publication of data to highlight the marketing angles that SP wants to emphasize in fact the FDA clinical pharmacologist’s review points this out for cognitive effects and worries that they may be used for pushing asenapine off-label in the dementia in the elderly. In addition to this and the push for marketing for negative symptoms they also appear to be direct hits at the clinical pharmacologist’s concerns about the potential for 5HT2B inverse agonism by metabolites and his concerns about phen-fen like effects as SP has published showing antagonism of asenapine at the suspect receptor while providing no data on metabolites that are present in much much higher amounts that would likely be more than enough to counteract any effect of asenapine itself. (In fact someone even put the asenapine info sans comments on metabolite receptor binding in Wikipedia.)

The FDA reviews provide sufficient detail to show the problems in these human studies but by not publishing them they are much harder to find and thus likely won’t be critiqued in the medical literature. Plus as the FDA reviews won’t show up in pubmed no one will think to look for them other than in the labeling (which is misleading) as basically academics don’t know to look for FDA reviews at Drugs@FDA. However even if the human studies are published, many times the data, and the spin, and the take home message of published studies are much different than the FDA analysis of them.

As for the peer reviewed asenapine study. This is the phase II study (study 41004) and the 174 patients are spread over 3 treatments, so less than 60 subjects per arm with the drop out rate so high that only 21 – 28 subjects remained at the end so that the statisticians felt it wasn’t adequately controlled. Plus if you notice the abstract from MacGuffin’s footnote 7 it doesn’t report PANSS score change for risperidone because it wasn’t significant and thus as the clinical pharmacologist points out this was a failed study. [Yet Saphris won FDA approval(!) -- Ed.] As for significance for positive symptoms alone this is a post-hoc analysis and typically no one ever does this a priori. . . .

-- Salmon
September 6, 2009 @ 11:07 PM

2 comments:

Anonymous said...

Thank you Condor and MacGuffin.

Condor I gave you at shout out on the Carlat blog. It’s not your usual fair but I do believe it’s interesting so do read

https://www.blogger.com/comment.g?blogID=1562387750337311798&postID=1454317276140379660&page=1.

As for this post I do believe MacGuffin has led me to strike gold once again and I will address this in a second post (part 2), but first let me address your questions.

Yes, the lack of published human studies is a signal, but minor compared to what we’ve already discussed.

The rat and receptor studies are as MacGuffin suggests selective publication of data to highlight the marketing angles that SP wants to emphasize in fact the FDA clinical pharmacologist’s review points this out for cognitive effects and worries that they may be used for pushing asenapine off-label in the dementia in the elderly. In addition to this and the push for marketing for negative symptoms they also appear to be direct hits at the clinical pharmacologist’s concerns about the potential for 5HT2B inverse agonism by metabolites and his concerns about phen-fen like effects as SP has published showing antagonism of asenapine at the suspect receptor while providing no data on metabolites that are present in much much higher amounts that would likely be more than enough to counteract any effect of asenapine itself. (In fact someone even put the asenapine info sans comments on metabolite receptor binding in Wikipedia.)

The FDA reviews provide sufficient detail to show the problems in these human studies but by not publishing them they are much harder to find and thus likely won’t be critiqued in the medical literature. Plus as the FDA reviews won’t show up in pubmed no one will think to look for them other than in the labeling (which is misleading) as basically academics don’t know to look for FDA reviews at Drugs@FDA. However even if the human studies are published, many times the data, and the spin, and the take home message of published studies are much different than the FDA analysis of them.

As for the peer reviewed asenapine study. This is the phase II study (study 41004) and the 174 patients are spread over 3 treatments, so less than 60 subjects per arm with the drop out rate so high that only 21 – 28 subjects remained at the end so that the statisticians felt it wasn’t adequately controlled. Plus if you notice the abstract from MacGuffin’s footnote 7 it doesn’t report PANSS score change for risperidone because it wasn’t significant and thus as the clinical pharmacologist points out this was a failed study. As for significance for positive symptoms alone this is a post-hoc analysis and typically no one ever does this a priori.

Salmon

(to be continued in the morning in Part 2. I'm going to bed.)

condor said...

Hey… at 3:29 am… smiling. Almost Christmas!