## OT: Bias of AUCs; example [Regulatives / Guidelines]

Massa,

I dunno understand wat ya sayin’.

» there is no doubt that AUC calculation can be biased.

OK.

» Whether the ln difference or ratio is biased, generally, is another matter and this is where I meant to draw the attention - this is the BE outcome. If one could make a case for saying that this bias affects T more often or to a larger extent than it affects R,

If we can measure all concentrations, any of the methods is unbiased as is the T/R estimate (yep, even by the linear trapezoidal and the bloody AUCall). I’m only talking about cases where we deal with BQLs.

If T = R (and for nitpickers: we have balanced sequences / equal group sizes) both treatments will be affected to the same extent, the bias of the AUC will mean out and the estimate will tell used T ≈ R.

If T < R, T will be more often affected than R. If T > R, R will be more often affected than T. In the first case the T/R estimate will have a negative bias (since we have more cases where T is underestimated) and in the second a positive one (since we have more cases where R is underestimated).

Example: One-compartment, D 100, V 1, t

As long as we can measure all concentrations of both T and R, the PE of AUC

The PEs of AUC

I dunno understand wat ya sayin’.

» there is no doubt that AUC calculation can be biased.

OK.

» Whether the ln difference or ratio is biased, generally, is another matter and this is where I meant to draw the attention - this is the BE outcome. If one could make a case for saying that this bias affects T more often or to a larger extent than it affects R,

*generally*, then that is where the use of this approach would be not just ugly but even worse also clearly biased.If we can measure all concentrations, any of the methods is unbiased as is the T/R estimate (yep, even by the linear trapezoidal and the bloody AUCall). I’m only talking about cases where we deal with BQLs.

If T = R (and for nitpickers: we have balanced sequences / equal group sizes) both treatments will be affected to the same extent, the bias of the AUC will mean out and the estimate will tell used T ≈ R.

If T < R, T will be more often affected than R. If T > R, R will be more often affected than T. In the first case the T/R estimate will have a negative bias (since we have more cases where T is underestimated) and in the second a positive one (since we have more cases where R is underestimated).

Example: One-compartment, D 100, V 1, t

_{½a}1 h, t_{½e}4 h, F_{R}1, F_{T}0.8–1, t 0, 1, 2, 2.5, 3, 3.5, 4, 4.5, 6, 9, 12, 16, 24; λ_{z}estimated from the last two measurable concentrations. LLOQ 3% of theoretical C_{ref,max}(62.996). Lin-up / log-down trapezoidal method for AUC_{0–t}, AUC_{0–∞}, and AUC_{0–t(common)}.As long as we can measure all concentrations of both T and R, the PE of AUC

_{0–t}is unbiased. Once we drop below the LLOQ, the PE will be negatively biased because we are comparing AUC_{0–16}of T with AUC_{0–24}of R. Here it happens below a true f of 0.9071 where C_{last}of T with 1.8898 will be below the LLOQ of 1.8899.The PEs of AUC

_{0–∞}and AUC_{0–t(common)}are always unbiased.*- Fisher D, Kramer W, Burmeister Getz E.
*Evaluation of a Scenario in Which Estimates of Bioequivalence Are Biased and a Proposed Solution: t*J Clin Pharm. 2016; 56(7): 794–800. doi:10.1002/jcph.663. Open access._{last}(Common).

—

Helmut Schütz

The quality of responses received is directly proportional to the quality of the question asked. 🚮

Science Quotes

*Dif-tor heh smusma*🖖Helmut Schütz

The quality of responses received is directly proportional to the quality of the question asked. 🚮

Science Quotes

### Complete thread:

- AUCt not covering at least 80% of AUCinf ratnakar1811 2013-03-05 08:30 [Regulatives / Guidelines]
- AUCt not covering at least 80% of AUCinf d_labes 2013-03-05 11:12
- AUCt not covering at least 80% of AUCinf ratnakar1811 2013-03-06 08:50
- AUCt not covering at least 80% of AUCinf ElMaestro 2013-03-06 11:10
- Science vs. regulations Helmut 2013-03-06 14:30
- Science vs. regulations cakhatri 2013-03-10 08:59
- ratio = difference of logs! Helmut 2013-03-11 01:17

- Science vs. (EMA) GL aka PK primer Helmut 2013-03-11 03:17
- Science vs. regulations qualityassurance 2020-04-23 14:11
- Science vs. regulations Helmut 2020-04-23 17:32
- Science vs. regulations Achievwin 2020-05-07 22:59
- Science vs. regulations Helmut 2020-05-07 23:54

- Science vs. regulations Achievwin 2020-05-07 22:59

- Science vs. regulations Helmut 2020-04-23 17:32

- Science vs. regulations cakhatri 2013-03-10 08:59
- AUCt not covering at least 80% of AUCinf Brus 2018-11-20 13:01
- AUCt not covering at least 80% of AUCinf Astea 2020-06-05 19:05
- So what? Helmut 2020-06-06 12:02
- Carrot and whip Astea 2020-06-06 21:11
- Gedankenexperiment Helmut 2020-06-07 12:41
- Dead dogs mittyri 2020-06-07 20:32
- Aber meine Herren das ist keine physik Astea 2020-06-08 00:08
- Aber meine Dame, das ist alles Unsinn! Helmut 2020-06-08 01:04
- Bizarre paper Helmut 2020-06-08 12:43
- Bizarre paper ElMaestro 2020-06-08 15:33
- OT: Bias of AUCt, AUCall, pAUC Helmut 2020-06-08 15:53
- OT: Bias of AUCt, AUCall, pAUC ElMaestro 2020-06-09 08:45
- OT: Bias of AUCs; exampleHelmut 2020-06-09 14:07

- OT: Bias of AUCt, AUCall, pAUC ElMaestro 2020-06-09 08:45

- OT: Bias of AUCt, AUCall, pAUC Helmut 2020-06-08 15:53
- Maxwell's demon Astea 2020-06-12 14:15

- Bizarre paper ElMaestro 2020-06-08 15:33

- Dead dogs Helmut 2020-06-08 10:40

- Aber meine Herren das ist keine physik Astea 2020-06-08 00:08

- Dead dogs mittyri 2020-06-07 20:32

- Gedankenexperiment Helmut 2020-06-07 12:41

- Carrot and whip Astea 2020-06-06 21:11

- So what? Helmut 2020-06-06 12:02

- AUCt not covering at least 80% of AUCinf Astea 2020-06-05 19:05

- AUCt not covering at least 80% of AUCinf ratnakar1811 2013-03-06 08:50

- AUCt not covering at least 80% of AUCinf d_labes 2013-03-05 11:12