Tuesday, May 26, 2009

Latest Asenapine Study Primary Endpoint -- Clank!

Like one of Shaq's free throws (Clank!) -- and more, soon -- this from MedPage Today:

. . . .Long-term findings for the novel antipsychotic agent asenapine did not confirm any advantage over second-generation antipsychotics for negative symptoms of schizophrenia. . . .

Earlier in the clinical trials program for asenapine, an unhypothesized benefit turned up -- it appeared to improve negative symptoms of schizophrenia significantly better than risperidone (Risperdal).

These negative symptoms, such as lack of motivation and social withdrawal, are far more likely to persist over time than are the core, positive symptoms of the disorder, "which have a much more waxing and waning form," Dr. Schooler said.

To examine this potential advantage prospectively, the researchers conducted the Aphrodite 25543 study, which included 26 weeks of double-dummy treatment with asenapine (5 to 10 mg bid) or olanzapine (5 to 20 mg qd) in 481 schizophrenia patients who had higher negative than positive symptom scores.

Dr. Schooler's presentation focused on the 26-week extension phase that followed for the 134 asenapine-treated patients and 172 olanzapine-treated patients that remained in the trial.

For the primary outcome of Negative Symptom Assessment Scale 16 scores, there was no significant difference between the drugs at week 52 (P=0.2344).

Likewise, they found no differences -- and little change over time -- for the positive symptom portion of the Positive and Negative Syndrome Scale (PANSS) or for the anxiety and depression portion. . . .



Anonymous said...

When I saw my pseudonym in the labels I knew I had to read this post.

As for the proposed mechanism in the article. HA HA HA HA! That's a good one.

I don't know where they initially saw the improvement in negative symptoms but I can say that I have heard from people at FDA that companies (SP?) have been interested in this. (It seems to me that I hear these sort of comments repeatedly even before the studies are done leaving me to wonder if there isn't some communication going on in addition to the submission of studies and meetings etc.)

For those who don't know schizophrenia is an illness where psychosis and other symptoms last for more than 6 months. A common saying to remember psychosis is hallucination, delusion, and confusion. The hallucinations and delusions (the explanation you come up with to explain the hallucination) are positive symptoms. Even if positive symptoms are controlled by drugs problems like thinking, and depressive like symptoms may still remain and are called the negative symptoms. Current antipsychotics have little to no effect on the 'negative' symptoms.

Baseline PANSS (Postive and Negative Symptom Scores) rating scales in antipsychotic trials are typically around 90 and the average change from baseline with drugs over 6 weeks may be in the range of 10 - 15 points with most of it being due to a change in hallucinations and delusions, and the response to placebo may only be 2 points less. Since the difference from placebo that results in a statistically significt difference in one of these trials is so small and the change in negative score is only a couple of points even a difference of 1 or 2 points in the negative score could make the negative score difference statistically significant or if the positive score diffeerence was either not or only borderline different than a slight difference in negative score could make the total PANSS statistically differenet. Thus it's possible to see something by chance that really doens't pan out. Since the major differences in the 'atypical antipsychotics though tot be associated with efficacy is which serotonin receptor subtypes they hit and their affinities there may or may not be a pharmacologic basis for differences in effect but more likely than not there probably isn't much difference.

Now I don't know which study or studies the original signal was in but they claim it was in comparison to risperidone and as I recall from clintrials.gov that there were only 2 such trails. In addition as reported in meeting abstracts one of these trials had a difference in baseline values in the different arms with the baseline asenapine arm being around 94 and the risperidone and placebo arms being around 90. Consequently since they all came down to the same scores at the end of 6 weeks, it erroneously looks like asenapine works when in fact it's simply not an adequately controlled study. (Thus it amazes me how the FDA can claim this study supports approval when it clearly doesn't.) However it's also possible that the difference in negative score could be due to the same reason.

So to me failure of this study should not be a major surprise. But neither do I want to criticize them doing the study without more information on their rational. So as to the asenapine program it sure would have been nice if it had been true, but it isn't.


Anonymous said...


Now people could say it wasn't less effective than olanzapine at 1 year, but there was no placebo arm and as I read more literature I'm beginning to think that over time that symptoms might actually be worse in people who take drugs.

Consequently, I wouldn't say that this study supports long term efficacy. Especially as we don't have long term toxicity and especially not at greater than 1 year which is when Vioxx's toxicities really began to take off.


Anonymous said...


delusions, and the (difference in)response -- (from) placebo may only be 2 points less. Since the difference from placebo that results in a...


Anonymous said...

So what, if it works as well as Zyprexa and no weight gain that's a huge plus.

Anonymous said...

Better side effects with efficacy will sell this drug. Mind numbing, appetite enhancing drugs that cause diabetes serve no one. If this drug has a good SE profile and helps patiets it's a godsend.......don't spin it just because it isn't any better without discussing SE and discontinuation rate.