Bioequivalence and Bioavailability Forum

Hi A-M-R,

❝ I will carrying out a comparative study between metformin on one side and

❝ on the other side (metformin and amoxycillin), the study will be carried

❝ out on rats to study the effect of amoxycillin on the pharmacokinetics of

❝ metformin. can you help me in finding the most suitable design for this

❝ study and if it is possible to take whole blood of the rat at each time

❝ interval (9 time points) especially the time interval will be for 8 hours

❝ and the project for statistical calculations in this case.

Difficult but doable.
A rat will not contain a lot of blood. Your samples need to be very moderate in size. Bleeding a rat may give as little as 8-10 ml depending on the size. What formulation is this? Gavage feeding is very reproducible in rats, but requires a liquid of some sorts.
You can do crossover but in that case should perhaps consider the blood volume loss needs time to recover. A parallel study might be a better way - I think I would do that. Rats come in many colors and flavours; from several sources the most common ones ("wt" is a misnomer) are for instance Wistar and Sprague-Dawley. These are typically sold as outbreds, but I would certainly pick inbreds to control genetic variation - this would be particularly important in a parallel design. Inbreds, though, are sometimes quite hysterical.

You can insert a hep-flushed catheter in v. jugularis under full anaesthesia and have the other end protruding from the neck. The rats will recover in a matter of a few days and be ready for the dosing. With a restrainer this setup can work well and is pretty easy to get approval for.

Dear Amr !

In this kind of studies, the sampling scheme is frequently a problem (i.e. it is not possible to get samples from all animals at all time points planned due to restrictions in blood volume). You may consider using a batch design to overcome these limitations which is for example described in the following papers:

Jaki T and Wolfsegger MJ (2009).
A theoretical framework for estimation of AUCs in complete and incomplete sampling designs.
Statistics in Biopharmaceutical Research, 1(2):176-184.
Abstract

Jaki T, Wolfsegger MJ, Lawo J-P (2010).
Establishing bioequivalence in complete and incomplete data designs using AUCs.
Journal of Biopharmaceutical Statistics, 20(4):803-820.
Abstract

The approaches described in these two papers are implemented in the R package PK:

Jaki T and Wolfsegger MJ.
Estimation of pharmacokinetic parameters with the R package PK.
Pharmaceutical Statistics, published online ahead of print
Abstract

hope this helps

martin

---

Edit: Abstracts and package PK linked. [Helmut]

Dear Amr !

You may find also the following paper regarding recommended blood sampling volumes of interest:

Diehl KH, Hull R, Morton D, Pfister R, Rabemampianina Y, Smith D, Vidal JM, van de Vorstenbosch C (2001).
A good practice guide to the administration of substances and removal of blood, including routes and volumes.
Journal of Applied Toxicology 21(1):15-23.
Paper

hope this helps

martin

Dear Amr

It is possbile to carry out kinetic study on rats. However, taking blood samples of same volume may be a concern here. You can go ahead with retro-orbital puncture which can give you up to 4 ml blood. But I am not sure you can get same 4 ml from same animal at all the time points. I prefer it is better to start with a pilot study with 2 or 3 rats and to to see how they respond.

A parallel design is better. Why dont you try the study in rabbits?? Rabbits can give you sufficient amount of blood through marginal ear vein.