Log in
Register|
Dr_Dan ★★ Germany, 2013-08-14 14:27 (4696 d 16:41 ago) Posting: # 11282 Views: 2,344 |
|
|
Dear all, I just reviewed the FDA recommendations for a specific drug product with a long elimination half life and I am wondering why in the suggested parallel design study (or cross-over study) the administered single dose should be 3x1 capsule. There is only one strength available and the recommended dose in patients is 1x1 capsule daily as this represents the therapeutic plateau. The pharmacokinetics are doseproportional over a very wide range, the lowest dose investigated was 4 times lower than the actual dose. So the sensitivity of the bioanalytical method could not be the reason for dosing three capsules. The absorption is slow and extensive, bioavailability is nearly 100%. Do you have any idea? If I use the FDA recommended study design for European submission what could be a possible reason for rejection? Is a BE study with dosing 3x1 capsule less or more sensitiv to detect possible differences in formulation than 1x1 capsule? Looking forward to your reply. Thanks in advance Dan — Kind regards and have a nice day Dr_Dan |
|
ElMaestro ★★★ Denmark, 2013-08-14 15:09 (4696 d 15:58 ago) @ Dr_Dan Posting: # 11283 Views: 1,942 |
|
|
Hi Dan, ❝ Do you have any idea? Not really. Steady-state, comparison of Css? ❝ If I use the FDA recommended study design for European submission what could be a possible reason for rejection? Is a BE study with dosing 3x1 capsule less or more sensitiv to detect possible differences in formulation than 1x1 capsule? I don't think any general rule can apply here. If the Cmax is 20x LLOQ then all should be adequate, and this would certainly be the case here. The S:N's impact on accuracy of determinations is likely positive for all samples, but on the other hand depending on your calibration curve you may need dilutions which drag things in the other direction. Dosing at 3x recommended dose is not per se an issue for EU. Having said all that, due to intra-batch variation of the IMPs it is always in theory better with 1 unit than 2,3,4... units (example: if the content of a tablets is normal with some mean and some variance, then the sum of three tablets is also normal with mean 3x the single mean, and the variance of the sum of three will be 3x the variance of one). This will in theory translate into an increased residual in the linear model. I am not sure I would consider this to have too much practical importance though. — Pass or fail! ElMaestro |
Ing. Helmut Schütz