Log in
Register|
pash413 ★ India, 2014-04-05 16:56 (4057 d 03:42 ago) Posting: # 12775 Views: 15,056 |
|
|
Hello, We conducted a full replicated BE study (2T and 2R) with scaling approach for DR formulation of a highly variable class 4 molecule for US Interestingly we observed zero concentrations throughout all the sampling points for test formulation in at-least one period in ~ 45 % subjects and zero concentrations throughout the all the sampling points for reference formulation in at-least one period in ~ 50 % subjects [Note: LLOQ was sufficiently low to capture the PK profile]. Similar results were observed for this molecule in other literature too. When statistical analysis was conducted as per Progesterone guidance of USFDA, subjects having zero concentration in at-least one period were automatically deleted in the BE evaluation as log of zero value will lead to not estimable values for Cmax and AUC. This leads to a significant data loss and detrimental effect on sample size and study outcome. We are concerned about the data loss of subjects having non-zero PK parameters (Cmax and AUC) in at-least 2 or 3 periods, just because mathematically log (0) is not estimable.
|
|
ElMaestro ★★★ Denmark, 2014-04-05 22:17 (4056 d 22:21 ago) @ pash413 Posting: # 12776 Views: 13,700 |
|
|
Hi Pash413, ❝ We are concerned about the data loss of subjects having non-zero PK parameters (Cmax and AUC) in at-least 2 or 3 periods, just because mathematically log (0) is not estimable. ❝ ❝ 1. Is there any statistical methodology to include data of subjects who have non-zero PK parameters (Cmax and AUC) in at least 3 periods to prevent such huge data loss Come on, with these 45%+ figures something has gone seriously wrong in the planning or conduct of your study; your data do not shed light onto the comparability of the two products. This isn't the time to fiddle and tweak with BLQ data or do imputation or clever transforms - rather it is the time to think hard and find out what went wrong and how to prevent recurrence. ❝ [Note: LLOQ was sufficiently low to capture the PK profile] Errr..... so then the subjects didn't take the meds? Did the staff/PI check with a tongue spatula? — Pass or fail! ElMaestro |
|
luvblooms ★★ India, 2014-04-07 09:47 (4055 d 10:51 ago) @ pash413 Posting: # 12777 Views: 13,726 |
|
|
Hi Pash413 ❝ Interestingly we observed zero concentrations throughout all the sampling points for test formulation in at-least one period in ~ 45 % subjects and zero concentrations throughout the all the sampling points for reference formulation in at-least one period in ~ 50 % subjects [Note: LLOQ was sufficiently low to capture the PK profile]. Similar results were observed for this molecule in other literature too. We have observed this phenomenon in few of our studies including DR tablets even after staff/PI checking with a tongue spatula and keeping close eyes on volunteers that they do not vomit/cough the tablets out. Just two point I would like to make a) What were the time points used? In our studies we have observed that few volunteers showed one point Cmax between 24-48 hrs time points and after that the concentration fall drastically with no measurable concentrations at later points. b) DR tablets are quite notorious and highly variable when it comes to gastric emptying. Would it be possible to check that whether tablets came out in feces as such? This might give you an idea. c) Would it be possible for further lower the LLOQ to get better idea? ❝ 3. We have got a literature in which Cmax and AUC with zero value were assigned/imputed with a fixed value = (LLOQ/2) and data of all such subjects were included in statistical analysis. This methodology was accepted by FDA for one of the NDA. Can we adopt the same for our product for ANDA submission This approach looks reasonable and IMHO can be used. I think Stat Gurus will let us know the pros and cons of the method soon.  But again, first you should look for the reason of getting zero concentrations and on possibility of decreasing such occurrences. — ~A happy Soul~ |
|
Dr_Dan ★★ Germany, 2014-04-07 10:35 (4055 d 10:03 ago) @ luvblooms Posting: # 12779 Views: 13,572 |
|
|
Dear Pash413 In the concept of bioequivalence bioavailability is taken as a surrogate parameter for safety and efficacy. If you observe zero concentrations throughout all the sampling points for test formulation in at-least one period in ~ 45 % subjects this would mean that in consequence this formulation would not be effective in a great portion of patients. The originator has demonstrated safety and efficacy in clinical trials. So your argumentation that zero concentrations throughout the all the sampling points for reference formulation in at-least one period in ~ 50 % subjects was shown and similar results were observed for this molecule in other literature too does not help. If you claim that your formulation is as bad as the originator you have to proof that your observations are no artefact. If this is not possible you have to provide clinical data. Kind regards Dr_Dan — Kind regards and have a nice day Dr_Dan |
|
pash413 ★ India, 2014-04-07 10:37 (4055 d 10:01 ago) @ luvblooms Posting: # 12780 Views: 13,594 |
|
|
We have thought about all possibilities as mentioned above, but there is no evidence for noncompliance in the study. However the formulation (Mesalamine DR tablet) itself has such property. please refer Literature Reference. Our main concern is how to handle such incomplete data statistically in scaled average (full/partial replicate) design study  |
|
Helmut ★★★   Vienna, Austria, 2014-04-07 11:01 (4055 d 09:37 ago) @ pash413 Posting: # 12781 Views: 13,659 |
|
|
Hi Pash, ❝ However the formulation (Mesalamine DR tablet) itself has such property. please refer Literature Reference. Since you had evidence of such a behaviour beforehand, why didn’t you invest your intellectual horsepower in designing the study? Why didn’t you contact FDA’s review staff in a controlled correspondence? ❝ Our main concern is how to handle such incomplete data statistically in scaled average (full/partial replicate) design study I don’t think that you can derive a meaningful comparison from your data. — 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 |
|
pash413 ★ India, 2014-04-07 11:30 (4055 d 09:08 ago) @ Helmut Posting: # 12782 Views: 13,543 |
|
|
Hello HS ❝ I don’t think that you can derive a meaningful comparison from your data. Agreed to your point. But the reason behind us being curious to use partial data (subjects with non-zero Cmax/AUC in at-least 2/3 periods) is because one of the firm has used the method of Imputation for same molecule (MESALAMINE) which was well accepted by FDA. (Delzicol Clinical Pharmacology and Bio-pharmaceutics review link provided in my above posts). We would like to understand the applicability and scope of such statistical methodology. |
|
ElMaestro ★★★ Denmark, 2014-04-07 14:14 (4055 d 06:24 ago) @ pash413 Posting: # 12785 Views: 13,579 |
|
|
Hi Pash, ❝ (...) the firm has used the method of Imputation for same molecule (MESALAMINE) which was well accepted by FDA. ❝ ❝ We would like to understand the applicability and scope of such statistical methodology. As I see it, imputation has never really been widely used when it comes to proofs of bioequivalence (or ther. eq. / EMA). The only exception I can think of from the top of my head is the quite naïve LLOQ/2 for BLQ PK data (and I am not a fan, by the way). Imputation is sometimes used by originator firms when they do their proofs of efficacy or safety. EMA was toying with guidance in the area some years ago. Quite controversial, really, and my impression was that it was mainly intended for longitudinal data. I could be completely wrong but the idea of imputation as a way to save your BE study does not strike me as particularly useful; if you treat all your 'cases' in a similar fashion then you might simply be biasing the dataset towards bioequivalence. — Pass or fail! ElMaestro |
|
nobody nothing 2014-04-07 11:31 (4055 d 09:07 ago) @ Helmut Posting: # 12783 Views: 13,614 |
|
|
For mesalamine there is an FDA Draft Guidance (dated SEP-2012), requiring truncated AUC evaluation with "at least four non-zero measurements of concentrations"... Locally applied, locally acting might be tricky, some times... — Kindest regards, nobody |
|
luvblooms ★★ India, 2014-04-07 13:50 (4055 d 06:48 ago) @ pash413 Posting: # 12784 Views: 13,593 |
|
|
Dear Pash ❝ We have thought about all possibilities as mentioned above, but there is no evidence for noncompliance in the study. ❝ ❝ However the formulation (Mesalamine DR tablet) itself has such property. please refer Literature Reference. I too am talking about the Meslamine DR tablet formulations and my experience with it. And what I told you were from those learning only. You need to control your clinical part and design your study appropriately and belive me this value will decrease to a level of 10-15% from 45-50%. ❝ Our main concern is how to handle such incomplete data statistically in scaled average (full/partial replicate) design study IMHO, if your only concern is the data handling and stats, best option is to go for Controlled corresponence with FDA. — ~A happy Soul~ |
|
Mahesh M ★ India, 2014-04-07 14:23 (4055 d 06:14 ago) @ luvblooms Posting: # 12786 Views: 13,582 |
|
|
How many subjects were enrolled for mesalamine DR capsule replicated study ? Becose the variability is more than 100 % can you elaborate how we control our clinical phase for complete and accurate data. Regards M |
|
jag009 ★★★ NJ, 2014-04-07 18:18 (4055 d 02:19 ago) @ luvblooms Posting: # 12788 Views: 13,432 |
|
|
Hi Luv and Pash, ❝ ❝ Our main concern is how to handle such incomplete data statistically in scaled average (full/partial replicate) design study ❝ ❝ IMHO, if your only concern is the data handling and stats, best option is to go for Controlled corresponence with FDA. The implication / sensitivity analysis might have worked with NDA group. With ANDA group it might not (never hurt to ask). ANDA reviewers are usually more strict... John |
|
jag009 ★★★ NJ, 2014-04-07 17:29 (4055 d 03:09 ago) @ pash413 Posting: # 12787 Views: 13,539 |
|
|
Hi, ❝ However the formulation (Mesalamine DR tablet) itself has such property. please refer Literature Reference. ❝ ❝ Our main concern is how to handle such incomplete data statistically in scaled average (full/partial replicate) design study I just flipped through the reference you linked. The concentration table listed <10 ng/mL as the LoQ(?)... Why didn't you go lower in the LLQ? John |
|
Ohlbe ★★★ France, 2014-04-07 20:23 (4055 d 00:14 ago) @ pash413 Posting: # 12789 Views: 13,565 |
|
|
Dear Pash, Could this be due to a stability issue ? How long did you store your samples before analysis, at which temperature, using which anticoagulant ? I just came across an old paper describing stability issues with 5-ASA at -20°C, at acidic pH values. They had no problem at -80°C. Did you check the stability in whole blood too, before centrifugation ? — Regards Ohlbe |
|
Helmut ★★★ Vienna, Austria, 2014-04-07 21:39 (4054 d 22:59 ago) @ Ohlbe Posting: # 12791 Views: 13,431 |
|
|
Dear Ohlbe & Pash, ❝ Could this be due to a stability issue ? Very good point! I’m traveling, so I don’t have access to all my data. Both 5-ASA and N-acetyl-5-ASA are stable at <-70 ℃ for five months and 24 hours at room temperature (anticoagulant heparin). Since we never stored samples at <-70 ℃ for fun, for sure there were problems at -20 ℃. — 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 |
|
pash413 ★ India, 2014-04-09 15:00 (4053 d 05:38 ago) @ Helmut Posting: # 12798 Views: 13,340 |
|
|
Our method was well validated and points mentioned above were well taken care of. No issues with stability |
|
Mahesh M ★ India, 2014-04-09 16:51 (4053 d 03:47 ago) (edited on 2014-04-10 05:51) @ pash413 Posting: # 12800 Views: 13,342 |
|
|
Dear Pash, In healthy volunteers as well as in patients, the need for a high pH of 7.0 for drug release to occur as well as the gastrointestinal transit time variabilities across individuals may influence the amount of drug absorbed from the intestine. The reason for the absence of drug concentrations was not clear (i.e. whether it is related to lack of drug release or lack of drug absorption). Regards Mahesh |
|
Ohlbe ★★★ France, 2014-04-09 18:31 (4053 d 02:07 ago) @ pash413 Posting: # 12801 Views: 13,339 |
|
|
Dear Pash, ❝ Our method was well validated and points mentioned above were well taken care of. No issues with stability Including in whole blood ? If yes: how long after blood collection ? Did you incubate it at 37°C before spiking ? — Regards Ohlbe |
Ing. Helmut Schütz