Log in
Register|
hiren379 ★ India, 2014-02-19 14:38 (4508 d 10:48 ago) Posting: # 12446 Views: 5,858 |
|
|
Hello friends, We have developed a combination product [Drug A + Drug B] and we are planning a drug interaction study. Design: 3 way cross over with following arms Arm 01: Drug A alone Arm 02: Drug B alone Arm 03: Drug A + Drug B (Newly developed combination product)
|
|
Helmut ★★★   Vienna, Austria, 2014-02-19 15:29 (4508 d 09:57 ago) @ hiren379 Posting: # 12448 Views: 5,209 |
|
|
Hi Hiren, are you aware of FDA’s guidance? ❝ 1. Is it mandatory for USFDA that Cmax, AUC meet the traditional BE limits of 80 - 125 for claiming no DDI?
❝ 2. If the study fails for meet 80-125 limits, is it possible to justify on clinical grounds that such an effect is of no clinical relevance? If you follow approach 2 above you should design the study with a sufficiently large samples size. ❝ 3. How to calculate sample size of DDI studies? Based on the drug with the larger CVintra and the “no effect boundaries”. ❝ 4. Since all the three products are developed by us on the same platform, we can assume zero / minimal variability associated with formulation difference.  ❝ Is there any method to calculate the sample size based on true intrinsic variability of molecule only? This can reduce our sample size to some extent. Doesn’t make sense. You administer formulations, not the APIs in solution. — 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 |
|
hiren379 ★ India, 2014-02-19 15:51 (4508 d 09:35 ago) @ Helmut Posting: # 12449 Views: 5,166 |
|
|
Thanks HS for reply, ❝ are you aware of FDA’s guidance? Yes I saw that but since it is still in draft phase (old-2006) wanted to confirm with practical experience ❝ ❝ 1. Is it mandatory for USFDA that Cmax, AUC meet the traditional BE limits of 80 - 125 for claiming no DDI? ❝
❝ ❝ 2. If the study fails for meet 80-125 limits, is it possible to justify on clinical grounds that such an effect is of no clinical relevance? ❝ ❝ If you follow approach 2 above you should design the study with a sufficiently large samples size. Ok. This is from draft guidance ❝ ❝ 3. How to calculate sample size of DDI studies? ❝ ❝ Based on the drug with the larger CVintra and the “no effect boundaries”. ❝ ❝ ❝ 4. Since all the three products are developed by us on the same platform, we can assume zero / minimal variability associated with formulation difference. ❝ ❝ We are sure that there shall be no/very less variability due to formulation as all the three products are manufactured by us on same platform with same excipients/process/manufacturing location, which is not the case in BE study as Test and reference product are based on different platforms/manufacturing process/location etc. Hence we want to remove this variability part from total variability while calculating sample size ❝ ❝ ❝ Is there any method to calculate the sample size based on true intrinsic variability of molecule only? This can reduce our sample size to some extent. ❝ ❝ Doesn’t make sense. You administer formulations, not the APIs in solution. Explained above |
|
Helmut ★★★ Vienna, Austria, 2014-02-19 16:36 (4508 d 08:49 ago) @ hiren379 Posting: # 12452 Views: 5,314 |
|
|
Hi Hiren, ❝ ❝ are you aware of FDA’s guidance? ❝ ❝ Yes I saw that but since it is still in draft phase (old-2006) wanted to confirm with practical experience Didn’t you bother following the link? The guidance is from February 2012. Practically the FDA does not distinguish between drafts and final versions. ❝ ❝ ❝ 4. Since all the three products are developed by us on the same platform, we can assume zero / minimal variability associated with formulation difference. ❝ ❝ ❝ We are sure that there shall be no/very less variability due to formulation as all the three products are manufactured by us on same platform with same excipients/process/manufacturing location, which is not the case in BE study as Test and reference product are based on different platforms/manufacturing process/location etc. Hence we want to remove this variability part from total variability while calculating sample size. Please don’t mix up variability with the expected deviations of test from references. F.i. in line extensions I generally assume a ratio of 1 – and not ±5%. The problem here is that DDI depend mainly on the properties of the active moieties, not the formulations. Do you have any hints (in vitro cocktail studies, PK/PD, …) suggesting that you should not expect a relevant interaction? ❝ ❝ ❝ Is there any method to calculate the sample size based on true intrinsic variability of molecule only? This can reduce our sample size to some extent. ❝ ❝ Doesn’t make sense. You administer formulations, not the APIs in solution. ❝ Explained above Nope. — 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 |
|
jag009 ★★★ NJ, 2014-02-20 22:41 (4507 d 02:45 ago) @ hiren379 Posting: # 12467 Views: 5,125 |
|
|
Hi, ❝ We are sure that there shall be no/very less variability due to formulation as all the three products are manufactured by us on same platform with same excipients/process/manufacturing location, which is not the case in BE study as Test and reference product are based on different platforms/manufacturing process/location etc. Hence we want to remove this variability part from total variability while calculating sample size How do you plan to remove this variability? John |
|
ElMaestro ★★★ Denmark, 2014-02-20 23:27 (4507 d 01:59 ago) @ hiren379 Posting: # 12468 Views: 5,151 |
|
|
Hi Hiren379, ❝ We are sure that there shall be no/very less variability due to formulation as all the three products are manufactured by us on same platform with same excipients/process/manufacturing location, which is not the case in BE study as Test and reference product are based on different platforms/manufacturing process/location etc. Hence we want to remove this variability part from total variability while calculating sample size This is not meaningful. By using treatment as a fixed factor (which is always done in BE regardless of design) you account for that. In other words, changes in excipients/process/manuf. location may be expected to affect the GMR but not the variability. — Pass or fail! ElMaestro |
Ing. Helmut Schütz