I gather you organized a great deal of the thoughtful, measured ass-kicking that got the APA to move on this, Dr. Mickey? Congratulations! We’ll all have to help keep the pressure on.
Along those lines, the folks at Mad in America suggested I get this article over to you, which is cross-posted on MIA and RxISK:
Otsuka is running a big trial of Abilify Maintena monthly injections for “early-stage schizophrenia.” They plan to take a few hundred patients ages 18-35 who are a year or less from their first psychotic episode, randomize them to depot injections, and keep them there for two whole years. John Kane and his new for-profit CRO are the captains of this nasty ship, but eight medical schools are joining in.
Don’t read it too soon after eating … but any ideas you have will be much welcomed. Thanks!
The obvious question is why did they just use different SNRI/SSRIs instead of treating melancholic with TCA and atypical with MAOI? Which could have made the study more worthwhile. I could have predicted that the type of SSRI/SNRI wouldn’t matter much (unless the groups were demographically separated, ie. male v. female, pre vs. postmenopausal). I’m wondering if there is simply a mind set that older drugs are obsolete. Maybe, I’m reading too much into it and the focus of the study is narrower that it could have been. Or maybe they are not all that interested in the older off-patent (and honorarium free) drugs.
To James O’Brien’s query: A quick look suggests this is a secondary report from the iSPOT-D project. They carved the original sample into 3 subtypes – melancholic, atypical, and anxious. It is a purely descriptive study, without any particular hypothesis being tested. Had it been based on past work, then the hypotheses you outlined would have been reasonable, but of course those aren’t present – further suggesting that this is a post hoc exercise. When they compared 3 fairly similar antidepressant drugs in the 3 subtypes there was no reason in advance to think that response rates would be different, and indeed they weren’t. Anyway, the experimental design did not allow the drug-attributable benefit to be identified, because there was no placebo-treated group. We always need to recall that observed response rates are a summation of nonspecific improvement and drug-attributable improvement. A placebo-treated group allows those two elements to be distinguished, and from that the specific drug effect as well as the Number Needed to Treat can readily be calculated. From this report we have no earthly idea what those key metrics were, for any of the drugs or for any of the subgroups. They can only report the pragmatic conclusion that nothing made any difference to outcome, but they cannot claim to have moved the ball down the field.
You pulled more than your weight for this, Dr. Mickey… kudos!
Better late than never. Another added benefit: heat is now off researchers who find negative results. Which should really be most studies.
I gather you organized a great deal of the thoughtful, measured ass-kicking that got the APA to move on this, Dr. Mickey? Congratulations! We’ll all have to help keep the pressure on.
Along those lines, the folks at Mad in America suggested I get this article over to you, which is cross-posted on MIA and RxISK:
http://www.madinamerica.com/2015/05/the-once-and-future-abilify-depot-injections-for-everyone/
Otsuka is running a big trial of Abilify Maintena monthly injections for “early-stage schizophrenia.” They plan to take a few hundred patients ages 18-35 who are a year or less from their first psychotic episode, randomize them to depot injections, and keep them there for two whole years. John Kane and his new for-profit CRO are the captains of this nasty ship, but eight medical schools are joining in.
Don’t read it too soon after eating … but any ideas you have will be much welcomed. Thanks!
Does this mean The American Journal of Psychiatry will be more careful about the articles it accepts for publication?
Speaking of AJP, I would like Dr. Carroll to weigh in on this recent article. Notice that one of the authors is a name often discussed here:
http://ajp.psychiatryonline.org/doi/10.1176/appi.ajp.2015.14020181
The obvious question is why did they just use different SNRI/SSRIs instead of treating melancholic with TCA and atypical with MAOI? Which could have made the study more worthwhile. I could have predicted that the type of SSRI/SNRI wouldn’t matter much (unless the groups were demographically separated, ie. male v. female, pre vs. postmenopausal). I’m wondering if there is simply a mind set that older drugs are obsolete. Maybe, I’m reading too much into it and the focus of the study is narrower that it could have been. Or maybe they are not all that interested in the older off-patent (and honorarium free) drugs.
To James O’Brien’s query: A quick look suggests this is a secondary report from the iSPOT-D project. They carved the original sample into 3 subtypes – melancholic, atypical, and anxious. It is a purely descriptive study, without any particular hypothesis being tested. Had it been based on past work, then the hypotheses you outlined would have been reasonable, but of course those aren’t present – further suggesting that this is a post hoc exercise. When they compared 3 fairly similar antidepressant drugs in the 3 subtypes there was no reason in advance to think that response rates would be different, and indeed they weren’t. Anyway, the experimental design did not allow the drug-attributable benefit to be identified, because there was no placebo-treated group. We always need to recall that observed response rates are a summation of nonspecific improvement and drug-attributable improvement. A placebo-treated group allows those two elements to be distinguished, and from that the specific drug effect as well as the Number Needed to Treat can readily be calculated. From this report we have no earthly idea what those key metrics were, for any of the drugs or for any of the subgroups. They can only report the pragmatic conclusion that nothing made any difference to outcome, but they cannot claim to have moved the ball down the field.
Thanks. I was missing the context entirely and couldn’t figure out why anyone would bother doing it as a de novo study.
I guess it’s nice that negative studies are published, it would be even better if they failed to confirm an actual hypothesis.