Below is Merrill Lynch's commentary on the FDA Panel's 14 - 0 rejection of Zimulti. Merrill is essentially saying that the panel acted inappropriately. I'm struck in particular by the sentence "Essentially the drug was considered guilty until proven innocent in this case."
While I can understand that the company bears the burden of proof when it comes to safety, it appears that in this case the panel ignored previously-established standards of statistical significance. I invite anyone's commentary as to whether the panel indeed screwed up in making this decision or not.
"Industry concerns increasing
Importantly, we believe that the negative outcome for Zimulti raises our concerns over the fundamentals for the pharmaceutical industry even further.
From the outset of the panel, FDA focused almost entirely on the neurological and psychiatric safety risks based on preclinical data and a signal from a metanalysis of trials respectively. Importantly, although the absolute event rates were low (<1%) for individual safety risks, FDA was sufficiently concerned about a lack of understanding on their mechanism of action to openly express its concern about rimonabant’s use in the general population, particularly in the context of the limited efficacy demonstrated.
The key industry concern in our opinion is that the FDA seems to be increasingly focusing on pre-clinical mechanism of action data suggestive of safety risk and even inconclusive meta-analyses, with seemingly less emphasis on whether there are robust statistically significant safety concerns highlighted in large clinical trials. Essentially the drug was considered guilty until proven innocent in this case.
In our opinion this increases safety hurdles that new products must now overcome in the US and increases level of concern over other products where safety concerns have been a recent focus. We specifically highlight two products where the FDA focus could be similar to that seen in the Zimulti panel discussion:
GSK’s Avandia: the safety concerns raised for Avandia are the result of a meta-analysis conducted by pooling a number of clinical studies, similar to the analysis conducted for Zimulti. Although this is not as scientifically robust as data from prospective controlled clinical trials, in the absence of data from the latter for Avandia, FDA may be unwilling to accept the potential of long term cardiovascular risk, particularly given other viable diabetes drugs exist.
Novartis’s Galvus: Safety concerns raised for Galvus were related to toxicity seen in pre-clinical animal models and understanding mechansim of this toxicity for a new therapeutic class (DPP-IV inhibtors). This has similarities, in our opinion, to FDA's discussion on Zimulti as a first in class CB1 blocker which has unspecified neurotoxicity in pre-clinical data."
