Obinoscopy
★    

USA,
2020-07-16 14:30
(95 d 19:52 ago)

Posting: # 21713
Views: 2,425
 

 Reanalysis for Pharmacokinetic Reason [Bioanalytics]

Dear All,

I am trying to get my head around why reanalyzing a blank sample because it had values above LLOQ in the initial analysis is not considered reanalysis for pharmacokinetic reason.

My thinking is...if a pre-dose sample gives a detectable peak area on analysis and on evaluation, there seem to be no explanation for this, why is it okay to reanalyze it again? Isn't it not considered reanalysis based on PK? Could it be because the concentration at time t = 0 is not used in calculating AUC? I don't know.

Also please what are the examples of reanalysis that are considered reanalysis due to PK reasons?

Thanks.

Scopy
Helmut
★★★
avatar
Homepage
Vienna, Austria,
2020-07-16 15:10
(95 d 19:12 ago)

@ Obinoscopy
Posting: # 21716
Views: 2,323
 

 Reanalysis for PK Reason: Gone with the Wind

Hi Scopy,

» I am trying to get my head around why reanalyzing a blank sample because it had values above LLOQ in the initial analysis is not considered reanalysis for pharmacokinetic reason.

Congratulations for discovering this inconsistency.

» My thinking is...if a pre-dose sample gives a detectable peak area on analysis and on evaluation, there seem to be no explanation for this, why is it okay to reanalyze it again?

That’s against scientific thinking. If regulators read this: Sorry to say, no offense intended.

» Isn't it not considered reanalysis based on PK?

Actually it is. At least in a drug-naïve subject in the first period it should be zero. See Harold Boxenbaum’s quote in this post. In any higher period we hope that the washout was sufficiently long enough…
If there would be a true unequal carry-over, we have no means to get an unbiased estimate of the treatment effect.

» Could it be because the concentration at time t = 0 is not used in calculating AUC?

No. A limited carry-over is acceptable. That means:
  • If C0 ≤5% Cmax of a subject in this particular period the concentration is used as it is in the calculation of AUC (the first partial AUC is indeed a trapezoid and not a triangle).
  • If C0 >5% Cmax the subject (or if in a replicate design, data of this period) can be excluded.

» Also please what are the examples of reanalysis that are considered reanalysis due to PK reasons?

According to the current guidelines (EMA 2011, FDA 2018, ICH draft 2019) it is no more acceptable. IMHO, bad science. See the two case studies in this presentation (slides 22–35). The first one was an obvious sample mix-up in the clincal phase and the second one likely an – undocumented – problem in sample handling. However, in both cases samples were reanalyzed (good scientific practice = ignoring the GL) and results confirmed. That reveals a common problem: In most cases errors occur in the clinical phase and not in bioanalytics. Now what?
In the past it was acceptable to perform a blinded review of data and have rules for reanalysis / exclusion in the protocol. Regrettably, those days are gone and – understandable – paranoia (driven by the many cases of fraud) prevails.
Nowadays reanalysis is only acceptable for the obvious reasons: Concentration >ULOQ, batch not valid, poor chromatography* (e.g., interference, degraded column), malfunctioning equipment, forgotten to add IS or derivatization reagent, etc.


  • Funny term. What is ‘rich chromatography’?

Dif-tor heh smusma 🖖
Helmut Schütz
[image]

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

France,
2020-07-16 17:16
(95 d 17:06 ago)

@ Helmut
Posting: # 21718
Views: 2,289
 

 Reanalysis for PK Reason: Gone with the Wind

Dear Scopy and Helmut,

» » I am trying to get my head around why reanalyzing a blank sample because it had values above LLOQ in the initial analysis is not considered reanalysis for pharmacokinetic reason.
»
» Congratulations for discovering this inconsistency.

Not necessarily an inconsistency: in the EMA guideline, PK re-analysis is not acceptable only for BE trials, but the guideline is applicable to all bioanalytical work. Confirming the presence of the analyte in a pre-dose sample could be relevant for other types of studies (and even worse, finding analyte in a placebo-treated subject, or animal in a TK study).

» » My thinking is...if a pre-dose sample gives a detectable peak area on analysis and on evaluation, there seem to be no explanation for this, why is it okay to reanalyze it again?
»
» That’s against scientific thinking.

I do see a scientific value: trying to understand whether this could be due to a contamination or analytical carry-over. But I agree this could be done under the reanalysis for "laboratory investigations" allowed by the guideline.

Other potential argument: the potential influence on AUC is very limited and there is no influence on Cmax, so there is no risk that such re-analysis may be done in order to make a failing study pass (which I did see once in the past).

» According to the current guidelines (EMA 2011, FDA 2018, ICH draft 2019) it is no more acceptable. IMHO, bad science.

Agreed.

» In the past it was acceptable to perform a blinded review of data and have rules for reanalysis / exclusion in the protocol. Regrettably, those days are gone and – understandable – paranoia (driven by the many cases of fraud) prevails.

Strangely enough, according to discussions I had with people involved in drafting the EMA guideline, it seems that it was the assessors who did not want to hear about PK repeats, not the inspectors – though the latter are well known for their paranoia, and for good reasons.

» * Funny term. What is ‘rich chromatography’?

:-D

Regards
Ohlbe
Helmut
★★★
avatar
Homepage
Vienna, Austria,
2020-07-16 18:39
(95 d 15:42 ago)

@ Ohlbe
Posting: # 21720
Views: 2,278
 

 Reanalysis for PK Reason: Gone with the Wind

Dear Ohlbe,

» » According to the current guidelines (EMA 2011, FDA 2018, ICH draft 2019) it is no more acceptable. IMHO, bad science.
»
» Agreed.

Agreed on what? That is not acceptable or that it is bad science? Don’t feel pushed to answer. :-D

» Strangely enough, according to discussions I had with people involved in drafting the EMA guideline, it seems that it was the assessors who did not want to hear about PK repeats, not the inspectors…

That’s very strange indeed.

» … – though the latter are well known for their paranoia, and for good reasons.

Of course.

Dif-tor heh smusma 🖖
Helmut Schütz
[image]

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

France,
2020-07-16 19:15
(95 d 15:07 ago)

@ Helmut
Posting: # 21721
Views: 2,274
 

 Reanalysis for PK Reason: Gone with the Wind

Dear Helmut,

» » » According to the current guidelines (EMA 2011, FDA 2018, ICH draft 2019) it is no more acceptable. IMHO, bad science.
» »
» » Agreed.
»
» Agreed on what? That is not acceptable or that it is bad science?

That it is bad science, of course ! To accept as a matter of principle data which are obviously incorrect, and to refuse as a matter of principle for these data to be corrected, is something I fail to understand.

I understand that assessors may be suspicious if there were some re-analyses for PK reasons. If the study is only borderline passing, they always have the possibility to perform (or ask for) a statistical analysis with the original data. Depending on the outcome, they may ask for an inspection.

Regards
Ohlbe
Obinoscopy
★    

USA,
2020-07-16 21:48
(95 d 12:34 ago)

@ Ohlbe
Posting: # 21723
Views: 2,267
 

 Reanalysis for PK Reason: Gone with the Wind

Hi Ohlbe

» Confirming the presence of the analyte in a pre-dose sample could be relevant for other types of studies (and even worse, finding analyte in a placebo-treated subject, or animal in a TK study).

Besides non-BE studies, I also noticed that reanalysis to confirm the presence of the analyte in a pre-dose sample is allowed by some BE assessors. Do you think those group of BE assessors are inconsistent? I ask this because, the same assessors will not accept reanalysis of post-dose samples to correct an irregular PK profile.

Scopy
Ohlbe
★★★

France,
2020-07-16 23:07
(95 d 11:14 ago)

@ Obinoscopy
Posting: # 21725
Views: 2,253
 

 PK assessors variability

Hi Scopy,

» Do you think those group of BE assessors are inconsistent?

There is indeed a high between-subject variability when it comes to PK assessors, sometimes even some within-subject variability, depending on their degrees of freedom :-D.

In Europe: add the between-country variability. In the US: possibly a between-review-division variability ?

Regards
Ohlbe
Helmut
★★★
avatar
Homepage
Vienna, Austria,
2020-07-17 16:11
(94 d 18:10 ago)

@ Ohlbe
Posting: # 21734
Views: 2,179
 

 Sometimes PK assessors ≈ scientists

Hi Ohlbe,

» There is indeed a high between-subject variability when it comes to PK assessors, sometimes even some within-subject variability, depending on their degrees of freedom :-D.
» In Europe: add the between-country variability.

Perfect description.
Last year I had a discussion with members of an European agency about irregular profiles and exclusion of data. They told me to regularly (‼) accept exclusion of “physiologically not plausible concentrations”. I was surprised and said that it is against the GL. Raised eyebrows…

Dif-tor heh smusma 🖖
Helmut Schütz
[image]

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

USA,
2020-07-17 23:48
(94 d 10:34 ago)

@ Ohlbe
Posting: # 21735
Views: 2,104
 

 PK assessors variability

Hi Ohlbe,

» In Europe: add the between-country variability. In the US: possibly a between-review-division variability ?

:-D

I wonder the study design we can use to measure these variabilities...especially the variability within the USFDA.

But this is sad...I had thought standardization and uniformity of regulatory review methods should be one of the foundation of regulatory science.

Scopy
ElMaestro
★★★

Belgium?,
2020-07-16 23:19
(95 d 11:03 ago)

@ Obinoscopy
Posting: # 21726
Views: 2,256
 

 Reanalysis for PK Reason: Gone with the Wind

Hi Obi,

» Besides non-BE studies, I also noticed that reanalysis to confirm the presence of the analyte in a pre-dose sample is allowed by some BE assessors. Do you think those group of BE assessors are inconsistent? I ask this because, the same assessors will not accept reanalysis of post-dose samples to correct an irregular PK profile.

What I was enforcing back in my day (roughly 200 million years ago), and what I think it still widely accepted, is not so complicated but perhaps convoluted: PK-repeat analysis may be used as part of a lab investigation, but it still requires a root cause to report a repeat value. A root cause is not a statistical figure but something tangible that proves that the original value is not indicative of the concentration in the sample. Whatever the result of a repeat sample is, that in itself is not a root cause.
Some CROs/labs have elaborate schemes with samples being flagged for repeats if they are out of expectation by some objective or subjective measure, and then three repeats are done, and we enter a decision scheme involving a comparison of the original value with the mean or median of the three repeats in the absence of a root cause. That is not at all universally agreed upon as ideal, I think.

I could be wrong, but...

Best regards,
ElMaestro

R's base package has 274 reserved words and operators, along with 1761 functions. I can use 18 of them (about 14 of them properly). I believe this makes me the Donald Trump of programming.
dshah
☆    

India,
2020-07-17 06:40
(95 d 03:42 ago)

@ ElMaestro
Posting: # 21729
Views: 2,217
 

 Reanalysis for PK Reason: Gone with the Wind

Greetings team!

I believe that Helmut is correct and without a predefined SOP for repeat analysis, repeat would not be acceptable!

If we are focusing on root cause, I believe that for pre-dose concentration even in period-I would not cause a significant issue to final BE out come as it may only impact AUC and it would not be significant. :-D If the value is MT 5% of Cmax, the subject would by default not considered for statistical evaluation :-). And for Root cause- most of CRO are not able to identify and thus there would not be a CAPA. :confused:


Thanks,
Dshah


Edit: Full quote removed. Please delete everything from the text of the original poster which is not necessary in understanding your answer; see also this post #5! [Ohlbe]
Obinoscopy
★    

USA,
2020-07-18 00:27
(94 d 09:54 ago)

@ dshah
Posting: # 21737
Views: 2,112
 

 Reanalysis for PK Reason: Gone with the Wind

Hi dshah

» If we are focusing on root cause, I believe that for pre-dose concentration even in period-I would not cause a significant issue to final BE out come as it may only impact AUC and it would not be significant. :-D If the value is MT 5% of Cmax, the subject would by default not considered for statistical evaluation :-).

How can you be so sure that its effect on AUC won't be significant? The removal/or inclusion of a subject from a statistical evaluation is enough is change the outcome especially if the result is borderline.

Scopy
dshah
☆    

India,
2020-07-20 07:22
(92 d 03:00 ago)

@ Obinoscopy
Posting: # 21755
Views: 1,886
 

 Reanalysis for PK Reason: Gone with the Wind

Dear Scopy!

Pls note that over here, we are only taking for pre-dose concentration. If the predose concentration>5%, the subject would be excluded from analysis as per most of the guideline. Pls find following example to illustrate that the impact on AUC would not be significant.

» » If we are focusing on root cause, I believe that for pre-dose concentration even in period-I would not cause a significant issue to final BE out come as it may only impact AUC and it would not be significant.:-D If the value is MT 5% of Cmax, the subject would by default not considered for statistical evaluation:-).
»
» How can you be so sure that its effect on AUC won't be significant? The removal/or inclusion of a subject from a statistical evaluation is enough is change the outcome especially if the result is borderline.

The concentration in below table is shown where only Pre-dose concentration is changed ;-) and then AUC is given using PKSolver.

time concentration 1 concentration 2 concentration 3 concentration 4
0          0               2               4               5
0.25       9.5             9.5             9.5             9.5
0.5       23              23              23              23
0.75      45              45              45              45
1         80              80              80              80
1.5      100             100             100             100
2         95              95              95              95
3         85              85              85              85
4         75              75              75              75
6         65              65              65              65
8         50              50              50              50
10        35              35              35              35
12        20              20              20              20
14        16              16              16              16
16        12              12              12              12
20        10              10              10              10
24         5               5               5               5
AUC      826.125         826.375         826.625         826.75


I believe that such small change in AUC can be considered non-significant. :-)

Regards,
Dshah
Obinoscopy
★    

USA,
2020-07-20 14:12
(91 d 20:10 ago)

@ dshah
Posting: # 21761
Views: 1,821
 

 Reanalysis for PK Reason: Gone with the Wind

Hi Dshah

» time concentration 1 concentration 2 concentration 3 concentration 4
» AUC      826.125         826.375         826.625         826.75

» I believe that such small change in AUC can be considered non-significant. :-)

I agree with you that these changes are very little. But you'd be amazed at how the difference between 826.125 and 826.375 is enough to make a study that failed (eg CI of T/R: 79.99 - 111.11) to pass (eg CI of T/R: 80.02 - 111.11).

Regards,

Scopy
Ohlbe
★★★

France,
2020-07-20 14:43
(91 d 19:39 ago)

@ dshah
Posting: # 21762
Views: 1,823
 

 Positive pre-dose

Dear Dshah,

» Pls note that over here, we are only taking for pre-dose concentration. If the predose concentration>5%, the subject would be excluded from analysis as per most of the guideline.

» I believe that such small change in AUC can be considered non-significant. :-)

The change in AUC is indeed extremely limited, and unless the study outcome is just at the boundary it will indeed not have any influence in itself. But excluding or including a subject, depending on whether his pre-dose concentration is > or < 5% of Cmax, can have dramatic consequences.

Regards
Ohlbe
Obinoscopy
★    

USA,
2020-07-18 00:00
(94 d 10:22 ago)

@ ElMaestro
Posting: # 21736
Views: 2,112
 

 Reanalysis for PK Reason: Gone with the Wind

Hi ElMaestro,

» PK-repeat analysis may be used as part of a lab investigation, but it still requires a root cause to report a repeat value.

From my understanding of the current thinking of assessors, PK-repeats can be done for lab investigation. However the values are not to be used for pharmacokinetic calculations and Bioequivalence determination.

But, during your enforcing days, PK-repeats can be used for BE determination provided a root cause of the initial result was established, right? If that's true, then I'd say those were the good old days for bioanalytical/pharmacokinetics scientists.

Scopy
ElMaestro
★★★

Belgium?,
2020-07-20 07:49
(92 d 02:32 ago)

@ Obinoscopy
Posting: # 21756
Views: 1,883
 

 Reanalysis for PK Reason: Gone with the Wind

Hi obi,

» From my understanding of the current thinking of assessors, PK-repeats can be done for lab investigation. However the values are not to be used for pharmacokinetic calculations and Bioequivalence determination.

If there is an established root cause which can be said to disqualify the initial value, then the (or a) repeat value is the natural value to report (provided of course that the repeat value is in itself considered valid and not suffering the same issue).
When I audit CROs I often encourage dialogue about the distinction between recording a value and reporting a value. It is a bit more than word play.

» But, during your enforcing days, PK-repeats can be used for BE determination provided a root cause of the initial result was established, right? If that's true, then I'd say those were the good old days for bioanalytical/pharmacokinetics scientists.

That's right. Even back then, when dinosaurs were roaming the jungle, Lucy was still not born, TV shows about the Kardasians were in black and white, flint axes were popular etc, I was enforcing it that way.
Actually, I can see I make it sound like I was influential and competent. Nothing could be farther from the truth :crying: (and that's still the lamentable state of things, by the way).
I rarely opened the analytical report in dossiers I assessed, but when I on occasion did then repeats (samples, runs or of individual validation tests), deviations, integrations, failed runs, exclusions/inclusion for A&P calculation, and stabilities tended to be the topics that had my attention.

I could be wrong, but...

Best regards,
ElMaestro

R's base package has 274 reserved words and operators, along with 1761 functions. I can use 18 of them (about 14 of them properly). I believe this makes me the Donald Trump of programming.
Helmut
★★★
avatar
Homepage
Vienna, Austria,
2020-07-20 10:52
(91 d 23:29 ago)

@ ElMaestro
Posting: # 21758
Views: 1,858
 

 Reanalysis for PK Reason: Gone with the Wind

Hi ElMaestro,

» Even back then, when dinosaurs were roaming the jungle, Lucy was still not born, TV shows about the Kardasians were in black and white, flint axes were popular etc,…

May I clarify the “etc.” partly? When I was barefooted dressed in furs and estimated rate constants from concentrations plotted on semi-log paper by means of a transparent ruler, a blinded review of data was common and excluding not plausible values was the rule rather than an exception.

If you have the stamina to watch the recordings mentioned there (navigate to 02:00:00): One participant asked “We all have seen concentrations which are ten times higher than adjacent ones. You can repeat the value a hundred times and will always get the same result. Is it possible to use M&S to justify exclusion?” Note that they were talking about a full-blown PopPK model and not just a simple lin-log regression and the likes. The answer of a guy of the FDA was in the spirit of [image] Little Britain’s Vicky:

[image]

See also my post above.

Dif-tor heh smusma 🖖
Helmut Schütz
[image]

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

Belgium?,
2020-07-20 12:12
(91 d 22:10 ago)

@ Helmut
Posting: # 21759
Views: 1,833
 

 Sightly OT (for once)

Hi Hötzi,

» May I clarify the “etc.” partly? When I was barefooted dressed in furs and estimated rate constants from concentrations plotted on semi-log paper by means of a transparent ruler, a blinded review of data was common and excluding not plausible values was the rule rather than an exception.

I still consider you kind of furry or, at the very least, hairy :-D:-D:-D

Thanks for linking to presentations. I'll listen over them.:ok:

I could be wrong, but...

Best regards,
ElMaestro

R's base package has 274 reserved words and operators, along with 1761 functions. I can use 18 of them (about 14 of them properly). I believe this makes me the Donald Trump of programming.
Achievwin
★    

US,
2020-07-27 23:47
(84 d 10:34 ago)

@ Obinoscopy
Posting: # 21793
Views: 917
 

 Reanalysis for Pharmacokinetic Reason

» I am trying to get my head around why reanalyzing a blank sample because it had values above LLOQ in the initial analysis is not considered reanalysis for pharmacokinetic reason.

IMHO most of your comments would have been anticipated and included in the Clinical and/or Bioanalytical study protocol on handling (Identifying, analyzing, and reporting) "PK repeats" there was a round table discussion in the AAPS annual meeting sometime around 2006, where three speakers discussed this issue "PK repeats- to do or not do"

Key take home message: "You have to follow the same criteria (whatever it is) which is defined before you start the sample analysis and follow tat criteria for the entire study" (NO dropping dead rat from the safety analysis)"

Hope you got my point.

AchievWin
Activity
 Admin contact
21,165 posts in 4,410 threads, 1,476 registered users;
online 7 (0 registered, 7 guests [including 3 identified bots]).
Forum time: Tuesday 10:22 CEST (Europe/Vienna)

In theory, there is no difference between theory and practice.
But, in practice, there is.    Jan L.A. van de Snepscheut

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