Pharmos (PARS) Message Board

STOCKTALK

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor. We ask that you disable ad blocking while on Silicon Investor in the best interests of our community. If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.

Biotech / Medical : Pharmos (PARS)

PARS 2.700

+13.6%

Jan 21 4:00 PM EST

Public Reply Prvt Reply Mark as Last Read File

Previous 10 Next 10 Previous Next

To: Clarksterh who wrote (1277)	10/19/2004 6:17:30 PM
From: marin	Read Replies (1) of 1386

" I'll say that I hope (and have reason to believe) that we won't need the new covariate analysis techniques. I would never want to be the first one through a massive bureaucracy with a new technique. I am also a little worried that the new technique could cut both ways - that the label may be more limited because of this technique even when the overall results are very good. But OTOH labelling restrictions don't mean too much." The new covariate analysis will hopefully show a stronger statistical difference between dex and placebo. As far as subgroups showing efficacy, if the primary endpoint isn't reached imo the share price will drop regardless of what subgroups may be stat sig. The problem with subgroups is that some may be appear significant by chance. PARS did not have any interim analysis performed due to the statistical penalty involved. (no they didn't say what the penalty would be) Also there are some scientists that get pretty bent out of shape when any interim peeks at data are performed (except for safety issues). The problem with interim analyses is that you may get a difference that looks significant, but when the trial runs to completion it turns out to be a random fluctuation. Much like tossing 6 heads in a row. Considering that Pharmos is trying to do get approval on a single trial, that alone raises the statistical stakes. However, Pharmos has not really told us what would be considered a successful trial in the eyes of the FDA. They mentioned awhile ago that a 10% difference between groups would be sufficient, but they haven't mentioned it since then. One general paper on single study trials mentioned that the p value should be .00125 or less. to equal two stat sig trials with a p value of .05. On the other hand, PARS does have Orphan Drug Status, which means that the population is rather small. (Ignoring the off-label indications) So, what will the FDA accept? I think a result of p <.05 or should be sufficient given the unmet medical need. But it really comes down to the results, which of course we don't know, and the FDA and the CNS division (rather picky) which will do what they want to anyway. If you want a history of PARS it might be useful to read through all of the commentary on PARS on the neuroinvestment website. It's a collection of some of his earlier message board postings along with some newsletter selections. btw you can drop the mr. jonmarin

Report TOU Violation

Share This Post

Public Reply Prvt Reply Mark as Last Read File

Previous 10 Next 10 Previous Next