Bioequivalence and Bioavailability Forum

Hi

I would like some advice on what type of bioequivalence study to do for a highly varible oncology drug. The drug in question, has variable F (6-30%) and in previous literature reports the intra-subject %CV has been quoted as almost 60%. It has to be said though, that these are old reports (>15 years old), in a patient (rather than healthy volunteer) population.

I have found that other generic companies have shown bioequivalence of two tablet formulations with a 2 x 2 cross over study and enrolling 60 subjects. Our formulation is a liquid and we would like to show bioequivalence to the tablets. We do not have any pilot data to estimate intra-subject CV. However, we are aimng to argue widening of the limits for Cmax (0.75-1.33) with the EMEA, since efficacy/safety is unlikely to be compromised. Since it is an oncology drug, we would like to limit the sample size to a minimum requirement.

What are our study design options?
1. 2 x 2 cross over with 60 subjects?
2. Or should we do a replicate design, 3 x 2 or 4 x 2?

Can we suggest an interim analysis?
1. For the 2 x 2 cross over design e.g. we will recruit 30-40 subjects and at interim analysis if bioequivalence is shown we will stop study, if not we will recuit another 30.
2. For the replicate design e.g. do an interim analysis after 24 subjects (determine intra subject CV) and then power to see how many more subjects are required.

Look froward to hearing your opinions.

regards

Meena Sarasia