Replies to post #133413 on Biotech Values

[the_busy_ant]

but I don't see any p values or hard numbers (things are very vaguely worded) which would make me cautious if I were long.

"Most authors refer to statistically significant as P < 0.05 and statistically highly significant as P < 0.001 (less than one in a thousand chance of being wrong)."
- ref http://www.statsdirect.com/help/basics/pval.htm

The PR is vaguely worded, however.

"demonstrated highly statistically significant results for the primary endpoint in two of the three treatment arms[OK p < 0.001, but which doses?] when compared to the placebo treatment arm"


"Both the 300 and 450 mg treatment arms[Here, they are very specific about which doses were p < 0.001] demonstrated highly statistically significant improvements in rescue-free bowel movement (RFBM) within 4 hours of administration over 28 days of dosing when compared to placebo treatment."


"Statistically significant efficacy was also seen in both the 300 and 450 mg treatment groups for the two key secondary efficacy endpoints including one assessing response (responder/non-responder) to study drug during Weeks 1 to 4 where “responder” is defined as having 3 or more RFBMs per week, with an increase of at least one RFBM per week over baseline, for at least 3 out of the first 4 weeks.". Were there any secondary endpoints where stat sig wasn't met?

I don't know much about Relistor or PGNX, but I'm reasonably adept at reading a PR in a skeptical fashion.

[vinmantoo]

ghmm,

I do agree it would have been nice to see some P values, but I am sure the final data will be presented soon. I think you are missing the point a bit in a comparison to NKTR. Relistor has already completed the required one-year safety study, and the PR today is for an oral formulation as opposed to the already approved sub-Q formulation. NKTR is a long way behind as they will need to conduct a one-year safety study. The edge NKTR did have was their drug was an oral formulation, and today's news appears to have removed any NKTR edge.

[mcbio]

SLXP/PGNX: I would like to see the data to better understand it. I am not as familiar with their sub-q product to easily compare with NKTR. The secondary end point is what would be a meaningful comparison. Not saying the data are bad but I don't see any p values or hard numbers (things are very vaguely worded) which would make me cautious if I were long.

I would imagine, as is sometimes the case, that they are saving the full detailed data for a 2012 medical conference (as alluded to in the PR). The PR makes it very clear that they will be submitting an NDA based on this data; doesn't really seem to be much wiggle room in that. So, I imagine there is confidence that this data supports approval of oral Relistor in this patient population (chronic, non-cancer). I think it's clearly this indication, not subQ, that NKTR needs to be more worried about from a competitive standpoint.

Keep in mind that PGNX is due 15-19% royalty rate on U.S. sales of Relistor from SLXP ( #msg-62151367 ).