Log in
Register|
ratnakar1811 ★ India, 2009-12-01 12:09 (5618 d 21:47 ago) Posting: # 4402 Views: 14,009 |
|
|
Dear Forum Members, We are planning a multiple dose crossover Bioequivalence study of Nasal Spray, wherein TID dose (08:00, 16:00 and 24:00 hrs) will be given for 4 days i.e. on 4th day after 08:00 hrs, profiling for PK will be done and pre-dose sample will be collected prior to the first dose of the study (1st dose of the study. Now my query is that do we need to take blood samples till 5 half life of the drug and pre-dose sample also require to be taken prior to 4th day dosing? If we following parameters to be considered for BE Parameters to be calculated: Cmax AUCtau Tmax Kel and t1/2 Bioequivalence to be concluded if the Cmax and AUCtau are within 80-125%. Your views/guidance will be highly appreciated in this regard. Best Regards, Ratnakar Edit: Category changed. [Helmut] |
|
Helmut ★★★   Vienna, Austria, 2009-12-02 01:20 (5618 d 08:35 ago) @ ratnakar1811 Posting: # 4406 Views: 12,020 |
|
|
Dear Ratnakar! ❝ [...] do we need to take blood samples till 5 half life of the drug and ❝ pre-dose sample also require to be taken prior to 4th day dosing? If we ❝ following parameters to be considered for BE Parameters to be calculated: ❝ ❝ ❝ ❝ ❝ Some points:
❝ Bioequivalence to be concluded if the Cmax and AUCtau are within 80-125%. Fine. Note, that some regulators are also interested in %PTF (Peak-Trough Fluctuation: 100[Cmax–Cmin]/Cav, where Cav=AUCτ/τ) and Cmin itself! — 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 |
![[image]](img/uploaded/image28.png)
Cmax![[image]](img/uploaded/image32.png)
Kel|
ratnakar1811 ★ India, 2009-12-03 08:36 (5617 d 01:19 ago) @ Helmut Posting: # 4408 Views: 11,594 |
|
|
Dear HS, Many thanks for your prompt reply! We are doing this study as a multiple dose just because single dose Nasal spray dose not give detectable concentrations. Now is it necessary that we have to prove Steady state in such situations also? If yes, then we will have to follow the all steady state requirements. Secondly for other parameters such as %fluctuations and Cmin criteria of 80-125% CI required to be applied or not? Cav should also be analysed with confidence interval or not? Thanking you in advance! Ratnakar |
|
Helmut ★★★ Vienna, Austria, 2009-12-03 15:53 (5616 d 18:03 ago) @ ratnakar1811 Posting: # 4416 Views: 11,753 |
|
|
Dear Ratnakar! ❝ We are doing this study as a multiple dose just because single dose Nasal spray dose not give detectable concentrations. Given the advancements in bioanalytics you should have very strong arguments before claiming that it's not possible to measure the single dose profile. Even if AUC cannot be characterized after SD, EMEA's BE-draft suggests to measure Cmax after SD and AUC in steady state! ❝ Now is it necessary that we have to prove Steady state in such situations also? IMHO yes, because the superposition principle in linear PK simply is AUC∞ (SD) = AUCτ (SS). ❝ If yes, then we will have to follow the all steady state requirements. I would say so. ❝ Secondly for other parameters such as %fluctuations and Cmin criteria of 80-125% CI required to be applied or not? Difficult question. Depends on both the regulatory requirements and the point of view of the assessor(s).  Generally %PTF as an aggregate metric is less variable than Cmax, so it should not be problematic (for a nasty example see D. Labes' experiences). Cmin is a pain in ... - especially if you have only little accumulation and values close to the LLOQ.In many cases the variability of Cmin prevents the demonstration of BE in a reasonbly powered study. You should study the literature whether this metric is important based on clinical grounds and state it in the protocol. On the other hand I have seen a lot of studies in the EU where assessors insisted on Cmin within conventional BE limits... ❝ Cav should also be analysed with confidence interval or not? No. Cav is only needed to calculate %PTF. It doesn't make sense: the PE and CI should be the same as the ones of AUCτ. — 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 |
|
ratnakar1811 ★ India, 2009-12-04 09:21 (5616 d 00:35 ago) @ Helmut Posting: # 4429 Views: 11,552 |
|
|
Dear HS, Thanks for your guidance, now I have a last question relating to taking of minimum three pre-dose sample prior to dosing of day4 as per your advice and suggested in guidline. Now should we take pre-dose samples prior to dosing of day-3 (1600 of day2, 0800, 1600 of day3 and 2400 of day3 which is same as that of pre-dose of day-4)? OR Pre-dose of all 0800hr dosing from day 1-4? Thanks Ratnakar |
|
d_labes ★★★ Berlin, Germany, 2009-12-03 09:58 (5616 d 23:58 ago) @ Helmut Posting: # 4409 Views: 11,767 |
|
|
Dear Helmut! ❝ Only if your drug shows no diurnal variability, the first 8 h profile ❝ on day 4 is sufficient. If your drug's PK is influenced by some ❝ circadian rhythms, you have to measure a full 24 hours cycle ❝ (i.e., three profiles). I have a question. How would you describe these three profiles?
Or can we restrict the bioequivalence decision to lets say profile 1? And see the other profiles only as secondary? I remember a study with Aescin (12 h dosing interval), we have done long ago, with a marked difference between the morning and evening profiles. Much lower concentrations in the second profile, but also much more variable in Cmax / Cmin there. Using the metrics of each profile we were able to show equivalence for the first profile but not for the second with respect to Cmax. PTF was beyond good and evil (CV very much higher than 30%). Bioequivalent? — Regards, Detlew |
|
Helmut ★★★ Vienna, Austria, 2009-12-03 16:50 (5616 d 17:05 ago) @ d_labes Posting: # 4419 Views: 11,639 |
|
|
Dear D. Labes! ❝ How would you describe these three profiles?
❝ If we take the first or last option and it comes to the statistical evaluation we have the burden to show bioequivalence for at least 3x more parameters than usual. ❝ Or can we restrict the bioequivalence decision to lets say profile 1? ❝ And see the other profiles only as secondary? I love your challenges! I would say it depends on the clinical impact (i.e., is drug dependent and not primarily a statistical question). If the important effect is within the day - and the evening dose is only administered to maintain steady state - I would see the first profile as the primary target. For bid antiasthmatics the evening dose should be more important. If nothing particular in terms of effect is documented for the drug, I use a pragmatic approach: 24 h values (primary, confirmatory) for BE + values of individual profiles (secondary, descriptive). But it's evident that the global Cmax (and even more tmax) is some kind of apples-and-oranges-comparison! Imagine a situation where tmax is two hours post dose. At bid we may end up with a median of 8 h... ❝ I remember a study with Aescin (12 h dosing interval), Ha, Aescin, nasty! You've been another victim.  ❝ ... with a marked difference between the morning and evening profiles. ❝ Much lower concentrations in the second profile, but also much more variable in Cmax / Cmin there. ❝ Using the metrics of each profile we were able to show equivalence for the first profile but not for the second with respect to Cmax. ❝ PTF was beyond good and evil (CV very much higher than 30%). So what was your primary target in that study? I would say that the high variability of %PTF in that study was bad luck. In my experience as an aggregate metric it 'behaves' quite nicely in most cases. ❝ Bioequivalent? Hhm. — 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 |
|
d_labes ★★★ Berlin, Germany, 2009-12-03 17:02 (5616 d 16:53 ago) @ Helmut Posting: # 4421 Views: 11,528 |
|
|
Dear Helmut! ❝ I love your challenges! I love your always Solomonic answers! Is it your experience with the regulatory acceptance in doing so as you described? — Regards, Detlew |
|
Helmut ★★★ Vienna, Austria, 2009-12-03 17:11 (5616 d 16:44 ago) @ d_labes Posting: # 4423 Views: 11,629 |
|
|
Dear D. Labes! ❝ I love your always Solomonic answers! Oh, not always! I can be quite strict - and even rude - sometimes... ❝ Is it your experience with the regulatory acceptance in doing so as you described? Yes. But not fail-safe.  Just this week I saw an example what can go wrong. Nonlinear PK, three dose strengths, the highest strength cannot be administered in SD due to AEs. In a scientific advisory meeting in the RMS is was agreed to go with a dose-escalating study to steady steady state; %PTF primary rate parameter, Cmax secondary. In the MRP one member state insisted in BE of Cmin (which failed due to the high CV). — 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 |
|
d_labes ★★★ Berlin, Germany, 2009-12-03 17:17 (5616 d 16:39 ago) @ Helmut Posting: # 4424 Views: 11,671 |
|
|
Dear Helmut! Answered to hasty, so I got not your second point. ❝ So what was your primary target in that study? I would say that the high variability of %PTF in that study was bad luck. In my experience as an aggregate metric it 'behaves' quite nicely in most cases. The study was long ago. So I don't remember exactly what was going on. But I have occasionally found that the variability of PTF was high if PTF was low. In a additive model this makes immediately sense: CV=sd/Mean is increasing if the mean is decreasing. BTW: For me it's questionable if a logarithmic transformation could and should be applied for this metric (also I do because of regulators). — Regards, Detlew |
|
Helmut ★★★ Vienna, Austria, 2009-12-03 17:42 (5616 d 16:14 ago) @ d_labes Posting: # 4425 Views: 11,800 |
|
|
Dear D. Labes! ❝ The study was long ago. So I don't remember exactly what was going on. But I have occasionally found that the variability of PTF was high if PTF was low. In a additive model this makes immediately sense: ❝ ❝ is increasing if the mean is decreasing. Sounds plausible. But only if the sd would stay constant. I would guess that the sd also decreases. Remember the good old days of 'the flatter is better'? Imagine two formulations (IR, MR) in steady state, AUC 2400. We get e.g.: Cmax Cmin %PTFI would guess that the variability of %PTF of the MR is lower, despite the lower value itself. But this assumption depends mainly on the variability of Cmin, which should be lower if moving upwards from the critical zone. The rest is silence (or the law of error propagation). ❝ BTW: For me it's questionable if a logarithmic transformation could and should be applied for this metric (also I do because of regulators). Right. I had a conversation with Regulator XY, * where he expressed also strong doubts about the distributional assumption. In our conversation we talked about Cmax/AUC in SD. We assume both AUC and Cmax follow lognormal - so what's the distribution of the ratio of two lognormal distributions? The same holds for %PTF (just see what happens if no accumulation occurs and Cmin=0). On the other hand Martin (who is on the contrary to myself a professional statistician) told me that this is no problem. He talked about monotonous transformations, but a little bit too much beer was involved that evening...  * Anonymized upon request; 2010-01-18 [Helmut]— 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 |
|
d_labes ★★★ Berlin, Germany, 2009-12-04 09:46 (5616 d 00:10 ago) @ Helmut Posting: # 4430 Views: 11,466 |
|
|
Dear Helmut! ❝ Remember the good old days of 'the flatter is better'?  Do me the favor and enlighten me. — Regards, Detlew |
|
Helmut ★★★ Vienna, Austria, 2009-12-04 13:35 (5615 d 20:21 ago) @ d_labes Posting: # 4431 Views: 11,520 |
|
|
Dear D. Labes! The phrase 'the flatter is better' was used in the mid-1980s of the last century. For some drugs AEs were found to be correlated with fast absorption rates / high Cmax-values (the classical example is nifedipine). Some clinicians went so far to demand modified release formulations ('flatter' profile) for all drugs intended for long term use. Questionable developments following this concept were modified release formulations of drugs with long half-lives (example: amitriptyline). In the meantime it became evident that a 'one-size-fits-all-concept' is not suitable. There are drugs where a fast initial 'ramp' is necessary for the effect. Therefore the concept was abandoned and filed in the history section of biopharmaceutics. — 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 |
Ing. Helmut Schütz