BE-proff ● 2016-06-22 13:26 (2836 d 20:05 ago) Posting: # 16437 Views: 3,278 |
|
Hi All, There is a guideline for BE-studies developed for EAEC which can be implemented somewhen in future. This gude contains interesting paragraph: "...AUC(0–t) must overlap at least 80 % of AUC(0–∞). If this rule is not fulfilled subjects should not be excluded from statistical analysis. But if AUC(0–t) doesn't overlap 80 % AUC(0–∞) in more then 20 % of cases you should doubt this study results. This requirement is not applicable to studies with blood sampling duration equal or over 72 h So what to do if AUC(0–t) doesn't overlap 80 % AUC(0–∞) following logic above? Nobody wants to bury studies with money... |
d_labes ★★★ Berlin, Germany, 2016-06-22 17:17 (2836 d 16:13 ago) @ BE-proff Posting: # 16442 Views: 2,526 |
|
Dear BE-proff ❝ So what to do if AUC(0–t) doesn't overlap 80 % AUC(0–∞) following logic above? ❝ Nobody wants to bury studies with money... If this happens, it's too late honey. It's only a post mortem finding. Plan your studies with sampling times long enough not to come in such trouble. — Regards, Detlew |
BE-proff ● 2016-06-22 23:30 (2836 d 10:01 ago) @ d_labes Posting: # 16446 Views: 2,531 |
|
Hi d_labes, I am scared with your verdict... Does it mean that all studies with immediate release forms should have points 72h and more to avoid that rule? |
Astea ★★ Russia, 2016-07-21 00:43 (2808 d 08:48 ago) @ BE-proff Posting: # 16507 Views: 2,042 |
|
Dear BE-proff! As usually one should be guided by the following rule: the sampling period should last not less than four times T1/2 (when it less than 72). Practically it depends on many different factors and it is very important to know the pharmacokinetics from the previous studies. It is often can be analysed from the previously obtained mean pharmacokinetics curves. There exists several approaches to estimate sufficient sampling period and adequate wash-out period. It can be represented by N times T1/2 or MRT, where N is a numerical factor (usually 4 for sampling time and 5-6 for wash-out) or it may be obtained from the previously obtained standard deviation of T1/2 or confidence interval (see this thread). As I understand numerical factors goes from the simplified model of logarithmical ellimination (like radioactive decay)... I will be grateful for some scientific articles dedicated to that topic. — "Being in minority, even a minority of one, did not make you mad" |