Back to the Top
Dear Friends,
We ran a BE study on NSAID recently. The generic product is significantly different statistically in AUC and Cmax. Nevertheless, the 90%CI falls within 0.8-1.25. To establish bioequivalence, is it a must to fulfill both no statistically significant difference for log transformed AUC (or Cmax) values together with 90%CI within 0.8-1.25. I have come across some CROs take into account only 90%CI.
I would be very grateful if you could kindly advise.
Regards,
Ken
Back to the Top
The following message was posted to: PharmPK
Dear Ken,
The BE criteria to pass bioequivalence is to have the 90% CI within 0.8-1.25. The idea behind this is that clinically it is considered bioequivalent if the test formulations has 20% or less difference than the reference product in the ratios of AUC and Cmax (Test over reference). So the parameters could be statistically different and still bioequivalent if the 90% CI is within 0.8-1.25.
I hope this help,
Jean
Back to the Top
The following message was posted to: PharmPK
Dear Peh Kok-Khiang!
Agree with Ken; statistically significant differences are irrelevant in bioequivalence. Since in most regulations the minimum sample size is 12, there is a high chance to get a statistical significant difference in "small" studies if the intra-subject CV is low (i.e <~10%). Note that a statistical significant difference means that 100% is not included in the 90% confidence interval.
Only Denmark has a strange point of view (despite being a member of the European Union and should follow the respective guideline). See
http://laegemiddelstyrelsen.dk/en/topics/authorisation-and-supervision/lic ensing-of-medicines/marketing-authorisation/application-for-marketing-auth orisation/bioequivalence-and-labelling-of-medicina--bstitution
"In the opinion of the Danish Medicines Agency, the 90 % confidence intervals for the ratio of the test and reference products should include 100%, regardless that the acceptance limits are within 80.00-125.00 %. Deviations are usually accepted if it can be adequately proved that the deviation has no clinically relevant impact on the efficacy and safety of the medicinal product."
Best regards,
Helmut
-
Ing. Helmut Schuetz
BEBAC - Consultancy Services for
Bioequivalence and Bioavailability Studies
Neubaugasse 36/11
1070 Vienna, Austria
e-mail helmut.schuetz.-at-.bebac.at
web http://bebac.at/
Back to the Top
The following message was posted to: PharmPK
Dear Ken,
Jean Lavigne and Helmut Schuetz answered your question, and I agree. The Danish view is very strange and not logical, since a significant difference between the products is irrelevant, as follows from the following: It is very unlikely that the 'true' AUC and/or Cmax of both products would be exactly the same; there will be always a difference. The larger the sample size, the larger the chance to detect a significant difference, and the larger the risk that the products are considered inequivalent; so there is a penalty, rather than a bonus, for large sample size; for very large sample size, the chance to detect a very small difference is very large. In addition, if the 'true' difference is rather large, say 15%, the chance to meet the bioequivalence criteria is very small, so there is no need for this penalty.
best regards,
Hans Proost
Johannes H. Proost
Dept. of Pharmacokinetics, Toxicology and Targeting
University Centre for Pharmacy
Antonius Deusinglaan 1
9713 AV Groningen, The Netherlands
|
Want to post a follow-up message on this topic?
If this link does not work with your browser send a follow-up message to PharmPK@boomer.org with "Statistical significant vs 90%CI" as the subject | Support PharmPK by using the |
Copyright 1995-2011 David W. A. Bourne (david@boomer.org)