Log in
Register|
yuvrajkatkar ★ Pune, Maharashtra (India), 2009-12-17 07:20 (6027 d 06:43 ago) Posting: # 4489 Views: 8,191 |
 Thread locked |
|
Dear Sir, Recently, I have analysed one Pilot study data. In this study, T/R ratios for Cmax, AUC0-t and AUC0-inf were 99.64%, 96.14% and 116.66%. 90% CI were not within the Bioequivalence Limit (80%-125%). Now, I want to calculate sample size for Pivotal Study, Is it right way to calculate sample size based on assumption (i.e maximum 10% T/R ratio difference? Please suggest me Edit: New thread created, category and subject line changed. [Helmut] — Best Regards, Yuvraj |
|
Helmut ★★★   Vienna, Austria, 2009-12-17 19:17 (6026 d 18:46 ago) @ yuvrajkatkar Posting: # 4497 Views: 6,962 |
|
|
Dear Yuvraj, just some desultory thoughts. ❝ Recently, I have analysed one Pilot study data. ❝ T/R ratios for Cmax, AUC0-t and AUC0-inf were 99.64%, 96.14% and 116.66%. I’m surprised by the large discrepancy in PEs between AUCt and AUCinf. It looks like the test has a longer apparent elimination than the reference. For immediate release products this should not be the case (actually we assume constant clearances). If you had modified release formulations, you may be trapped by flip-flop-pharmacokinetics (the slowest phase is absorption). In such a case AUCt as the primary target parameter is not trustworthy (remember: in BE we are interested in equivalence of absorption). Therefore we must concentrate on AUCinf. What’s the percentage of extrapolated AUC (mean, min, max) by formulations? If you really have to deal with a flip-flop, you must sample long enough (e.g., residual area 5% instead of 20%). ❝ 90% CI were not within the Bioequivalence Limit (80%-125%). Doesn't matter. That's not the target of a pilot. ❝ Now, I want to calculate sample size for Pivotal Study, ❝ Is it right way to calculate sample size based on assumption (i.e maximum 10% T/R ratio difference? No, don’t do that! Give us more information. First we have to find an explanation for the discrepancy between AUCt and AUCinf. I have my doubts whether it will be possible to demonstrate BE for AUCinf in a reasonable sample size (80% power, PE 116.66%): CV% n— Dif-tor heh smusma 🖖🏼 Довге життя Україна! ![[image]](https://static.bebac.at/pics/Blue_and_yellow_ribbon_UA.png) Helmut Schütz ![[image]](https://static.bebac.at/img/CC by.png) The quality of responses received is directly proportional to the quality of the question asked. 🚮 Science Quotes |
|
yuvrajkatkar ★ Pune, Maharashtra (India), 2009-12-19 09:05 (6025 d 04:57 ago) @ Helmut Posting: # 4507 Views: 6,852 |
|
|
Dear Sir, Actually it is a parallel study. maximum Pooled CV=53.0% Therefore, if we take actual maximum difference 116.66%. then 1286 completed subjects will be appropriate sample size to get 90% confidence interval of Log Transformed Cmax, AUC0-t and AUC0-inf data within 80 - 125 % to achieve 80% power at 5% level of significance. Is sample size based on actual maximum T/R ratio difference? Sir, I need your guidence. — Best Regards, Yuvraj |
|
Helmut ★★★ Vienna, Austria, 2009-12-19 13:40 (6025 d 00:23 ago) @ yuvrajkatkar Posting: # 4508 Views: 6,837 |
|
|
Dear Yuvraj, why don’t you answer our questions first? Your calculation is based on the flawed assumption that in the pivotal study the PE and CV will be exactly like in the pilot – do you really believe in that? Your sample size would be 1290 if the PE 116.66->116.67. Do you want to set a new world record in BE? Remember that based on your calculations human subjects will be treated – this is not a game to play with numbers. — Dif-tor heh smusma 🖖🏼 Довге життя Україна! Helmut Schütz The quality of responses received is directly proportional to the quality of the question asked. 🚮 Science Quotes |
|
yuvrajkatkar ★ Pune, Maharashtra (India), 2009-12-19 14:38 (6024 d 23:25 ago) @ Helmut Posting: # 4509 Views: 6,852 |
|
|
Dear Sir, I dont want to make a world record in BE. Sir, Please tell me, what is your opinion regarding sample size in this case? — Best Regards, Yuvraj |
|
Helmut ★★★ Vienna, Austria, 2009-12-19 19:41 (6024 d 18:21 ago) @ yuvrajkatkar Posting: # 4511 Views: 6,756 |
|
|
Dear Yuvraj, since you are not in the mood to give us more information – which we would need for a serious discussion, the answer is simple: Forget it. Reformulate! <°)))o><— Dif-tor heh smusma 🖖🏼 Довге життя Україна! Helmut Schütz The quality of responses received is directly proportional to the quality of the question asked. 🚮 Science Quotes |
|
ElMaestro ★★★ Denmark, 2009-12-19 00:32 (6025 d 13:30 ago) @ yuvrajkatkar Posting: # 4505 Views: 6,855 |
|
|
Hello yuvrajkatkar, ❝ T/R ratios for Cmax, AUC0-t and AUC0-inf were 99.64%, 96.14% and 116.66%. are you sure it's not the Cmax ratio that is 116.66% and the other two ratios being for the AUCs? Best regards EM. SCNR  [Helmut] |
Ing. Helmut Schütz