Shuanghe
★★  

Spain,
2022-07-10 02:36
(627 d 17:09 ago)

Posting: # 23124
Views: 2,976
 

 schedule time vs. actual time for AN­VI­SA BE [Regulatives / Guidelines]

Hi all,

I just had a wierd request from one of our clients regarding PK calculation in BE study intended to be submitted to ANVISA, and I'd like to know your experience with ANVISA in this regard. Detailed background below.

It's a pretty normal BE study of a solid oral drug product with systemic absorption, with a sentence in the protocol saying that PK parameters will be calculated with actual sampling time instead of scheduled time, which is pretty much universal standard these days.

Our client requested to change the protocol to say that PK will be calculated with scheduled time, not the actual time. Now that is an unusual request since I couldn't find such requirement in ANVISA's old (RDC 896, 2003) and new (RDC 1170, 2006) BE guideline, statistical guideline (RDC 898, 2003), and a few other guidelines that I thought might be remotely relevant.

When I requested the justification, I was told that this was their experince with ANVISA from deficient letters. It seems that if the protocol says using actual time for PK calculation the ANVISA will issue deficient letter requesting the scheduled time... I was told that our client had also asked some consultants and other companies in Brazil and they all had the similar experience...

For our study it really doesn't matter since we house the subject until the last PK sample is taken so the main expected time deviation would usually be a couple of minutes instead of possibly a few hours if ambulatory samles are involved and some subjects arrive late... But I really don't like the idea of writing such sentence in the protocol without justification from at least some regulatory guidelines...

What is your experience in this regard? My thought is that if this is a common request, someone would have had mentioned in the forum, but I couldn't find any post here ...

All the best,
Shuanghe
Helmut
★★★
avatar
Homepage
Vienna, Austria,
2022-07-10 17:37
(627 d 02:07 ago)

@ Shuanghe
Posting: # 23125
Views: 2,738
 

 schedule time for ANVISA?

Hi Shuanghe,

no personal experience with the ANVISA in this respect. IMHO, this is just bizarre. :thumb down:
I haven’t seen a single case where NCA was based on the scheduled sampling times in decades.

Dif-tor heh smusma 🖖🏼 Довге життя Україна! [image]
Helmut Schütz
[image]

The quality of responses received is directly proportional to the quality of the question asked. 🚮
Science Quotes
Relaxation
★    

Germany,
2022-07-12 15:50
(625 d 03:55 ago)

@ Shuanghe
Posting: # 23127
Views: 2,655
 

 schedule time vs. actual time for ANVISA BE

Hello Shuanghe.

I can only add a personal observation from reviews and discussion about a few reports that were going to be submitted to ANVISA.
Basically, the CRO provided both evaluations, one based on actual times and one on scheduled times with an emphasis on that with scheduled times.

When asked why this was done I received a template for the report to be submitted named "Relatorio Estatistico" I was told is a template from ANVISA.
My version here is from 2015 and outdated and I failed to find a newer template on the net (a quick search showed, that this is not a very unique name), but this may still be true.

Therein is a sample table named:
Tabela 4 – Parâmetros primários e secundários ... calculados com base no tempo nominal, referente ao tratamento A.
=> Table 4 - Primary and secondary parameters ... calculated based on the nominal time, referring to treatment A.

I was then told, that this was the root cause for the additional analysis, although we all agreed in the end, that the purpose of such an approach is not fully clear :confused:.

Well, I think that explains nothing, but maybe its a pointer to the origin of this unique requirement for ANVISA.

Best regards!
Helmut
★★★
avatar
Homepage
Vienna, Austria,
2022-07-12 18:15
(625 d 01:30 ago)

@ Relaxation
Posting: # 23128
Views: 2,701
 

 Not even wrong

[image]Hi Relaxation,

❝ Well, I think that explains nothing, but maybe its a pointer to the origin of this unique requirement for ANVISA.


Sad, so sad! To quote Wolfgang Pauli Not even wrong!”

Perhaps the ANVISA is recalculating the AUC in a spreadsheet. Given, with actual sampling times the number of columns would double (see there).

Dif-tor heh smusma 🖖🏼 Довге життя Україна! [image]
Helmut Schütz
[image]

The quality of responses received is directly proportional to the quality of the question asked. 🚮
Science Quotes
Shuanghe
★★  

Spain,
2022-07-13 17:23
(624 d 02:21 ago)

@ Helmut
Posting: # 23141
Views: 2,596
 

 Not even wrong

Hi Helmut

❝ Sad, so sad! To quote Wolfgang Pauli Not even wrong!”


I remember that you had given some talks about BE to ANVISA staff before. Maybe next time you can suggest why they should not request schedule time point :lol3:

All the best,
Shuanghe
Helmut
★★★
avatar
Homepage
Vienna, Austria,
2022-07-13 19:03
(624 d 00:41 ago)

@ Shuanghe
Posting: # 23144
Views: 2,617
 

 ANVISA: MD studies no more required

Hi Shuanghe,

❝ I remember that you had given some talks about BE to ANVISA staff before.


Twice – the last one dealing with the necessity of multiple dose studies.
Haha, now I can write it esp. for you: See this article. If you are short in time, go straight to its Postscript.
Even shorter executive summary (for the guy in the Armani suit): Multiple dose studies are no more required for the ANVISA.
In April 2021 a revision of the guideline was promised for the next summer. If there is one, I failed to find it at the ANVISA’s website, as usual.*

❝ Maybe next time you can suggest why they should not request schedule time point :lol3:


Sure. :-D


  • Edit: RDC Nº 742 of 10 August 2022 was published on 18 August 2022 and will be effective with 3 July 2023.
    PK metrics: \(\small{C_\text{max}}\), \(\small{AUC_{0-\text{t}}}\), \(\small{\textrm{p}AUC_{0-\tau/2}}\), \(\small{\textrm{p}AUC_{\tau/2-\tau}}\).

Dif-tor heh smusma 🖖🏼 Довге життя Україна! [image]
Helmut Schütz
[image]

The quality of responses received is directly proportional to the quality of the question asked. 🚮
Science Quotes
Ohlbe
★★★

France,
2022-07-13 12:23
(624 d 07:21 ago)

@ Relaxation
Posting: # 23135
Views: 2,571
 

 Relatorio estatistico

Dear Relaxation,

❝ My version here is from 2015 and outdated and I failed to find a newer template on the net (a quick search showed, that this is not a very unique name), but this may still be true.


I gave up trying to follow the ANVISA website years ago, but I managed to find that document. Latest version is 2.0, dated 09/09/2015. Contents is as you describe.

Regards
Ohlbe
Helmut
★★★
avatar
Homepage
Vienna, Austria,
2022-07-13 13:02
(624 d 06:43 ago)

@ Ohlbe
Posting: # 23136
Views: 2,631
 

 Relatorio estatistico

Dear Ohlbe,

❝ […] I managed to find …


Kudos! May I call you Sherlock in the future?

❝ … that document. Latest version is 2.0, dated 09/09/2015.


On top of the page: “Atualizado em 11/11/2020 14h43”

Interesting that no formula is given for the extrapolation.
In previous guidelines it was \(\small{AUC_{0-\infty}=AUC_{0-\text{last}}+C_{t_\text{last}}/\widehat{\lambda}_\text{z}}\). Since Hauschke et al.1 is in the references, may we use the recommended \(\small{AUC_{0-\infty}=AUC_{0-\text{last}}+\widehat{C}_{t_\text{last}}/\widehat{\lambda}_\text{z}}\) (in Phoenix parlance AUC_INFpred instead of AUCINF_obs)?
Quoting Gabrielsson and Weiner:2

In general, we recommend use of the predicted rather than the observed
last concentration when computing the extrapolated area.

See also this article.

In this goodie only nominal time as well. In Pharmacokinetic Parameters AUCINF_obs:-(


  1. Hauschke D, Steinjans V, Pigeot I. Bioequivalence Studies in Drug Development. Methods and Appli­ca­tions. Chichester: Wiley; 2007. p. 131.
  2. Gabrielsson J, Weiner D. Pharmacokinetic and Pharmacodynamic Data Analysis. Concepts and Appli­ca­tions. Stock­holm: Apote­kar­­so­ci­eteten; 5th ed. 2016. p. 147.

Dif-tor heh smusma 🖖🏼 Довге життя Україна! [image]
Helmut Schütz
[image]

The quality of responses received is directly proportional to the quality of the question asked. 🚮
Science Quotes
Ohlbe
★★★

France,
2022-07-13 14:37
(624 d 05:07 ago)

@ Helmut
Posting: # 23137
Views: 2,601
 

 ANVISA

Dear Helmut,

❝ ❝ […] I managed to find …


❝ Kudos! May I call you Sherlock in the future?


I have bookmarked this page from which you can find quite a few documents. Maybe you should add it to the guidance page. Not sure it would be worth the effort of updating the link to each document, considering ANVISA's practice of modifying them periodically.

❝ On top of the page: “Atualizado em 11/11/2020 14h43”


That's the date the web page was last updated, not the date the document was last updated ;-)
The document history is given in the Word document itself.

Regards
Ohlbe
Helmut
★★★
avatar
Homepage
Vienna, Austria,
2022-07-13 19:17
(624 d 00:27 ago)

@ Ohlbe
Posting: # 23145
Views: 2,607
 

 ANVISA

Dear Ohlbe,

❝ I have bookmarked this page… Maybe you should add it to the guidance page.


Time allowing.

❝ Not sure it would be worth the effort of updating the link to each document, considering ANVISA's practice of modifying them periodically.


Yep. Essentially I’ve given up.

❝ ❝ On top of the page: “Atualizado em 11/11/2020 14h43”


❝ That's the date the web page was last updated, not the date the document was last updated ;-)

❝ The document history is given in the Word document itself.


I know. I assumed that it was reviewed (without changes) in 2020 cause this date is hard-coded in the HTML.

Dif-tor heh smusma 🖖🏼 Довге життя Україна! [image]
Helmut Schütz
[image]

The quality of responses received is directly proportional to the quality of the question asked. 🚮
Science Quotes
pjs
★    

India,
2022-07-25 12:42
(612 d 07:03 ago)

@ Helmut
Posting: # 23168
Views: 2,289
 

 Relatorio estatistico

Hi all,

Can anyone share below sited reference full book.

❝ 2. Gabrielsson J, Weiner D. Pharmacokinetic and Pharmacodynamic Data Analysis. Concepts and Appli­ca­tions. Stock­holm: Apote­kar­­so­ci­eteten; 5th ed. 2016. p. 147.


Thanks
Helmut
★★★
avatar
Homepage
Vienna, Austria,
2022-07-25 14:32
(612 d 05:13 ago)

@ pjs
Posting: # 23172
Views: 2,254
 

 Full book?

Hi pjs,

❝ Can anyone share below sited reference full book.


What do you mean by ‘full book’? Scan its 1,040 pages and ‘share’ it with you violating the copyright? Since I know both authors personally, I guess that they won’t like that.
Invest 155 bucks and buy it.

If you don’t trust me in quoting correctly:

[image]


Dif-tor heh smusma 🖖🏼 Довге життя Україна! [image]
Helmut Schütz
[image]

The quality of responses received is directly proportional to the quality of the question asked. 🚮
Science Quotes
Shuanghe
★★  

Spain,
2022-07-13 17:21
(624 d 02:24 ago)

@ Relaxation
Posting: # 23140
Views: 2,505
 

 schedule time vs. actual time for ANVISA BE

Hi Relaxation,

❝ Well, I think that explains nothing, but maybe its a pointer to the origin of this unique requirement for ANVISA.


Thanks. At least now I know where this comes from...

All the best,
Shuanghe
ElMaestro
★★★

Denmark,
2022-07-13 01:10
(624 d 18:35 ago)

@ Shuanghe
Posting: # 23131
Views: 2,743
 

 schedule time vs. actual time for ANVISA BE

Hi Shuanghe,

like others I like the actual timings best.
But there are reasons and even if they are not entirely in the public domain I am happy to shed a little light over it here. To catch the squirrel....

You are testing A versus B on behalf of a sponsor.
For Cmax you get a 90% CI of 86.44-118.20. Pass.
For AUCt you get a 90% CI of 113.87-125.04. Bugger. Angry sponsor.

You're royally screwed. But wait... no-one signed off on the pk+stats yet....

You go back into the source. You find the sheets with TPDs. You "update" them a little. It is safe because our clinical updates to the Sponsor only mentions how many were dosed, how many dropped out, how many had AEs, did they resolve. But the clinical updates to do not contain TPD details.
And now a second round of pk+stats.
For Cmax you get a 90% CI of 86.44-118.20. Pass.
For AUCt you get a 90% CI of 113.87-124.96. Pass. Happy sponsor.

And one day the agency gets wind of it. One case is enough. Agency wants to see analysis with scheduled timings. It is a sensitivity analysis now. If the analysis with actual timings gives another result than with scheduled timings then it can be argued the study was not well planned or well conducted (much like if AUCinf is wildly different from AUCt).

Bear in mind that AUCt(A) or AUCt(B) is not of interest in BE, but the ratio is. And the expected ratio is (can be said to be) independent of TPDs as long as TPD occur at random, i.e. with no treatment preference.

Pass or fail!
ElMaestro
Helmut
★★★
avatar
Homepage
Vienna, Austria,
2022-07-13 01:29
(624 d 18:16 ago)

@ ElMaestro
Posting: # 23132
Views: 2,684
 

 TPD?

Hi ElMaestro,

excuse my ignorance: TPD?

Dif-tor heh smusma 🖖🏼 Довге життя Україна! [image]
Helmut Schütz
[image]

The quality of responses received is directly proportional to the quality of the question asked. 🚮
Science Quotes
ElMaestro
★★★

Denmark,
2022-07-13 01:32
(624 d 18:13 ago)

@ Helmut
Posting: # 23133
Views: 2,585
 

 TPD?

Sorry,

❝ excuse my ignorance: TPD?


Time point deviations.

Pass or fail!
ElMaestro
Helmut
★★★
avatar
Homepage
Vienna, Austria,
2022-07-13 01:55
(624 d 17:49 ago)

@ ElMaestro
Posting: # 23134
Views: 2,606
 

 TPD!

[image]

Dif-tor heh smusma 🖖🏼 Довге життя Україна! [image]
Helmut Schütz
[image]

The quality of responses received is directly proportional to the quality of the question asked. 🚮
Science Quotes
Shuanghe
★★  

Spain,
2022-07-13 17:15
(624 d 02:30 ago)

@ ElMaestro
Posting: # 23139
Views: 2,559
 

 schedule time vs. actual time for ANVISA BE

Hi ElMaestro,

Thanks. That's a bit dark (maybe because too much scandals such as GVK/Semler/Synchron… :ponder:?)… but at least it makes some sense as to why some agencies may ask for it…

All the best,
Shuanghe
ElMaestro
★★★

Denmark,
2022-07-14 16:57
(623 d 02:47 ago)

@ Shuanghe
Posting: # 23147
Views: 2,542
 

 schedule time vs. actual time for ANVISA BE

Hi Shuanghe,

❝ Thanks. That's a bit dark (maybe because too much scandals such as GVK/Semler/Synchron… :ponder:?)… but at least it makes some sense as to why some agencies may ask for it…


"Some agencies" is indeed the correct way to say it, it is not only ANVISA now.

Take a step back and look at it. CROs started out by testing R versus R to pass. Then regulators found out and started inspecting the sites to verify DA. How annoying. Then we had the switcharoo (credit to a well known WHO inspector for coining this term) which involved switching T and R for select subjects and reinjecting under false ID. It was extremely efficient in the sense it can make any study pass how ridiculously badly T matches R. Then someone published a paper about it. how ANNOYING. Initial disbelief, the author got angry emails saying that although the paper presented some interesting ideas, it was generally disconnected from real life. Only took a few months until the matter was re-confirmed by FDA's letters to two CROs. Basically, regulators pulled the carpet under CROs when they fiddled with the clinics (pharmacy) or bioanalysis to make a study pass.
So out of desperation, the bad CROs have turned their focus to data management (or would that be data mismanagement?) and pk/stats. This approach with TPDs works well, but it is difficult to plausibly rig a study that is wildly failing. So, it will be mainly applicable to studies failing "a little bit". Regulators are now starting to ask for analysis with scheduled time points. How annoying.

Dark? No.
Creative? Yes.

There is in fact one other reason which I can also mention: the author of the paper mentioned has worked with agencies to improve SaToWIB. One of the potential improvements was to make an objective function which is based on AUC match between any two PK profiles. So if you have two identical profiles (within the meaning of random analytical variability) they will have better matching AUCs if there is no difference in recorded time points. So if you use this kind of objective function you increase the chance of detecting reinjected profiles if you ignore the TPD and just go with the scheduled timings.
But linear regression and method 32 still works much better if you ask me.:-)

Pass or fail!
ElMaestro
UA Flag
Activity
 Admin contact
22,957 posts in 4,819 threads, 1,636 registered users;
65 visitors (0 registered, 65 guests [including 3 identified bots]).
Forum time: 18:45 CET (Europe/Vienna)

Nothing shows a lack of mathematical education more
than an overly precise calculation.    Carl Friedrich Gauß

The Bioequivalence and Bioavailability Forum is hosted by
BEBAC Ing. Helmut Schütz
HTML5