PK Parameter Calculation Bug—AUC_tEnd

[SCZ1174]

Hello everyone,

I’ve noticed that PKsim may have some issues when calculating AUC_tEnd.
In PKsim, AUC_tEnd is defined as: the area under the concentration-versus-time curve from the "first" to the last data point.
However, based on my actual testing, I’ve found that in practice:

• Simulated curve: The starting point is “0” rather than “first,” even if the simulation setting "Start Time" does not begin at 0.
• Observed curve: It is “first”.
  This can lead to significant discrepancies between simulated and observed values when AUC_tEnd is used directly as an output in many situations (e.g., insufficient blood sampling during the absorption phase of oral formulations, or long intravenous infusion times with no blood sampling during the infusion).

I hope someone can verify my concerns.

[github-actions[bot]]

🤖 Discussion Assistant: I found some similar discussions that might be helpful:

• PK parameters during multiple dosing (78% similar)
• PK-Sim AUC_tEnd calculating method (75% similar)
• Add User Defined PK Parameters for Population Simulation Result in OSPsuite R (72% similar) ✅ (Answered)
• How is the PK parameter AUC_inf_tD1 calculated? (69% similar)
• Why the PK-analysis results of some population simulations in PK-sim are missing some parameters? (66% similar)

[tobiasK2001]

Hi,

Yes, I can confirm your concerns. This is exactly why we only use the internal PK-Sim PK-Analysis for informal or internal guidance. For official, external reports, we calculate both the simulated and observed metrics using R NCA packages to guarantee identical methodology and strict comparability.

Best, Tobias

[Yuri05]

• Simulated curve: The starting point is “0” rather than “first,” even if the simulation setting "Start Time" does not begin at 0.

This will be corrected in the upcoming Version 13 release of the OSP Suite.

[tobiasK2001]

Excellent! Great to hear ;-)