Log in
Register|
drsinghs ☆ India, 2011-11-09 13:45 (5344 d 11:55 ago) Posting: # 7643 Views: 5,046 |
|
|
Dear All, Does anybody have experience of filing Statistically Borderline failed studies to any of the regulatory agencies? Say for e.g. if the 90% CI upper limit is 125.51. Thanks & Regards, Sandeep Edit: Category changed. [Helmut] |
|
Helmut ★★★   Vienna, Austria, 2011-11-09 14:26 (5344 d 11:14 ago) @ drsinghs Posting: # 7644 Views: 4,362 |
|
|
Dear Sandeep, in the recent past regulators generally became more and more strict. One the other hand see this thread. ❝ Say for e.g. if the 90% CI upper limit is 125.51. You can try. If the metric was AUC and the acceptance range was fixed at 80.00–125 % in the protocol I’m afraid cards are stacked against you. If the PE was not too far away from 1 (=the study likely failed due to higher CV than anticipated) you may consider repeating it in a larger sample size. If you wait until getting a deficiency letter your timeframe will be rather close. If the metric was Cmax and the study is not intended for submission in the US, EU, Australia, or New Zealand, you have more headroom for starting an argument (some countries still accept a range wider than 80.00–125 %). Even if not stated in the protocol, try to get some references on safety (because upper CL > upper AL). Discuss the relevance. Good luck. — 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 |
|
drsinghs ☆ India, 2011-11-10 06:14 (5343 d 19:26 ago) (edited on 2011-11-10 10:04) @ Helmut Posting: # 7645 Views: 4,386 |
|
|
Thanks Helmut, The matrix is AUCt and the study is for Australian submission. Also the molecule is a prodrug and we have observed this result for the same. We have analysed the metabolite as well for supportive data. Active metabolite results are very convincing i.e. 90% CI from 95 - 105. Can we justify these results with following points: 1) The elimination half life of prodrug is short and of metabolite is long enough for once daily dosing. 2) Metabolite is 5-6 times potent than prodrug. 3) Major theraputic effect is due to metabolite and not prodrug. Can we conclude that there would not be any major safety issues and go ahead with the submission along with a justification for the same?  Thanks & regards, Sandeep Edit: Full quote removed. Please delete anything from the text of the original poster which is not necessary in understanding your answer; see also this post! [Ohlbe] |
|
Helmut ★★★ Vienna, Austria, 2011-11-10 14:51 (5343 d 10:48 ago) @ drsinghs Posting: # 7649 Views: 4,338 |
|
|
Dear Sandeep! ❝ The matrix is AUCt… Matrix ≠ Metric.  ❝ … the study is for Australian submission. That’s bad. Australia has adopted the current European guideline in Juni 2011 (get it here). Little space left for interpretation; especially for AUC. ❝ Also the molecule is a prodrug and we have observed this result for the same. We have analysed the metabolite as well for supportive data. Active metabolite results are very convincing i.e. 90% CI from 95 - 105. The GL is very clear: If your analytical method is sensitive enough to characterize the profile of the parent, assess BE based on the parent. No need to measure the metabolite – not even ‘supportive’ (a term which isn’t mentioned in the GL at all). If the method is not sensitive enough (e.g., very fast half-life…) measure the main metabolite (irrespective whether active or not!). Sometimes a pro-drug is metabolized to an inactive metabolite which is further metabolized to an active one. In such a case measure the first one, because considered to reflect (especially the rate of) absorption of the pro-drug better. According to current thinking BE is seen as a surrogate of pharmaceutical quality rather than of clinical equivalence (safety/efficacy). ❝ Can we justify these results with following points: ❝ ❝ 1) The elimination half life of prodrug is short and of metabolite is long enough for once daily dosing. ❝ 2) Metabolite is 5-6 times potent than prodrug. ❝ 3) Major theraputic effect is due to metabolite and not prodrug. ❝ ❝ Can we conclude that there would not be any major safety issues and go ahead with the submission along with a justification for the same? Your justification is a clinical one. This is not what regulators want (see above). BTW, regulators are allergic to post-hoc changes of a study’s primary endpoint. You can try, of course. — 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 |
|
drsinghs ☆ India, 2011-11-11 05:52 (5342 d 19:47 ago) @ Helmut Posting: # 7654 Views: 4,357 |
|
|
Thanks Helmut, We have an outlier in the same study and the same is confirmed by Lund's test. Can we exclude this subject and present the data to regulators. After excluding that subject 90% CI upper limit has come down to 119.37 from 125.51 I know this is again not as per GL. Thanks....... Sandeep |
|
Helmut ★★★ Vienna, Austria, 2011-11-11 06:54 (5342 d 18:46 ago) @ drsinghs Posting: # 7655 Views: 4,297 |
|
|
Dear Sandeep! ❝ We have an outlier in the same study […]. Can we exclude this subject and present the data to regulators. […] I know this is again not as per GL. Let’s review your posts:
You are free to do anything. Do you want to get an approval from me‽ — 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 |
|
richas ● 2011-11-11 13:00 (5342 d 12:39 ago) @ Helmut Posting: # 7657 Views: 4,263 |
|
|
❝ Even if all your assumptions in sample size estimation were correct and you powered the study to 80 % one out of five studies will fail due to pure chance. Hi Helmut, Is there a way to prove that the study failed due to pure chance?  Do you have any idea about the opinion of regulators on this fact? Richa Edit: standard quote restored [Ohlbe]. |
Ing. Helmut Schütz