Clinical significance of Tmax [Regulatives / Guidelines]

posted by Helmut Homepage – Vienna, Austria, 2017-10-12 15:22 (2359 d 19:26 ago) – Posting: # 17891
Views: 4,751

Hi M.tareq,

❝ "A statistical evaluation of tmax is not required …


Unfortunately the EMA is so allergic to a nonparametric test that it was removed from the GL.

❝ … if rapid release is claimed to be clinically relevant and of importance for onset of action or is related to adverse events, …


We had to accept that BE is seen a “quality measure”. Why the heck does clinical relevance come back here?

❝ … there should be no apparent difference in median tmax and its variability between test and reference product."


Oh dear! Apparent? Variability of the median? WTF? The median is an estimate. The data is variable.

❝ If a fast disintegrating tab has a tmax earlier than the conventional release, does it prove that it has an earlier onset of action?


Not the slightest idea. First of all, tmax is a very poor predictor of differences in the rate of absorption.
Example: One compartment model; FT=FR=1, kel 0.1733 h–1 (t½ 4 h), kabs,R 1.3863 h–1 (t½ 30 min), kabs,T 2.0794 h–1 (t½ 20 min), sigmoidal effect model identical for T and R; Emax 125, EC50 50, γ 0.25.

  t       C (R)   C (T)        E (R)   E (T)
 0.0000    0.00    0.00         0.00    0.00
 0.0833    9.48   14.48        49.69   52.89
 0.1667   17.78   26.44        54.47   57.53
 0.2500   25.05   36.30        57.11   60.00
 0.3333   31.39   44.39        58.87   61.57
 0.5000   41.70   56.35        61.08   63.43
 0.7500   52.46   66.79        62.87   64.76
 1.0000   59.09   71.59        63.80   65.30
 1.2500   62.85   73.09        64.29   65.46
 1.5000   64.61   72.69        64.50   65.42
 2.0000   64.46   69.15        64.48   65.03
 2.7500   59.88   61.76        63.91   64.15
 4.0000   49.61   49.98        62.44   62.50
 6.0000   35.33   35.35        59.79   59.79
 8.0000   25.00   25.00        57.10   57.10
12.0000   12.50   12.50        51.78   51.78
16.0000    6.25    6.25        46.61   46.61
24.0000    1.56    1.56        37.00   37.00
tmax       1.50    1.25  tmax   1.50    1.25
 Δ                –0.25   Δ            –0.25
Cmax      64.61   73.09  Emax  64.50   65.46
 T/R             113.13%  T/R          101.49%
λz       0.1733  0.1733  λz   0.0281   0.0281
AUCt     494.87  519.25  AUECt 1236.8  1240.5
AUC∞     503.89  528.26  AUEC∞ 2552.3  2556.0
 T/R             104.84%  T/R          100.15%


[image]


Although the test has a 50% (!) faster absorption than the reference, tmax of T is only 15 min earlier than the one of R (reflected both in PK and PD). The T/R-ratio of Cmax is 113% but the maximum effect increases by only 1.5%! Is that relevant?
This is a very simple PK/PD-link model. With others (e.g., incorporating a delay in onset) the observed difference in PK may disappear completely in PD. If we have no data about the PK/PD-relationship we are left out in the dark. I’m aware of only a few studies (rapid onset for multiphasic release products of zolpidem and methylphenidate).

Dif-tor heh smusma 🖖🏼 Довге життя Україна! [image]
Helmut Schütz
[image]

The quality of responses received is directly proportional to the quality of the question asked. 🚮
Science Quotes

Complete thread:

UA Flag
Activity
 Admin contact
22,957 posts in 4,819 threads, 1,636 registered users;
113 visitors (0 registered, 113 guests [including 5 identified bots]).
Forum time: 09:49 CET (Europe/Vienna)

With four parameters I can fit an elephant,
and with five I can make him wiggle his trunk.    John von Neumann

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