OT: BW in ANCOVA [Regulatives / Guidelines]

posted by ElMaestro  – Belgium?, 2019-06-18 15:51 (461 d 01:29 ago) – Posting: # 20336
Views: 1,788

Hi Hötzi,

» At the 2nd GBHI conference (Rockville 2016) body weight as a covariate was a hot topic. Body weight changes quite often in studies of anti-cancer drugs. The FDA was happy with that. No way for the EMA – as usual.


So odd.

"The statistical analysis should take into account sources of variation that can be reasonably assumed to have an effect on the response variable."

A larger person has larger plasma volume and will, all other factors equal, be expected to have lower AUC and Cmax. Thus, body weight is definitely a source of variation that can be assumed to have some effect. And this is how it is in practice; those metrics correlate significantly with body weight in many trials albeit not always in a p<0.0000001 fashion when doing e.g. Spearman's rank or Pearson (but hey, period for example is usually not significant but has to be included in the model).

The fact that one can do potency correction in the right circumstances is in fact an introduction of a kind of ANCOVA principle. So why they decide to shut the door on body weight is not clear to me, is it clear to anyone? Does this really have to do with concern for sources of variation, or are we in an altogether different territory here?

I could be wrong, but...

Best regards,
ElMaestro

R's base package has 274 reserved words and operators, along with 1761 functions. I can use 18 of them (about 14 of them properly). I believe this makes me the Donald Trump of programming.

Complete thread:

Activity
 Admin contact
21,068 posts in 4,392 threads, 1,466 registered users;
online 10 (1 registered, 9 guests [including 4 identified bots]).
Forum time: Monday 17:21 CEST (Europe/Vienna)

I have finally come to the konklusion
that a good reliable set ov bowels
iz worth more to a man
than enny quantity of brains.    Josh Billings

The Bioequivalence and Bioavailability Forum is hosted by
BEBAC Ing. Helmut Schütz
HTML5