1 reply to this topic

[Bokhyeon]

Hi, I have 3 questions. Please see below questions and answer.

All of questions are related to contents of intermediate PK/PD modeling course.

 

1. For the solution file for PK 1 from Intermediate PK/PD modeling course,

    As far as I know, when we make a PK model, initial estimates should be set to make good correspondence between predicted values based on initial estimates and obs. data. The soultion file is set well. I arbitrarily changed the initial estimates to make bad correspondence to see how the theta and CV values and graphs(Ind DV vs IPRED, Ind IWRES vs IPRED and Ind IWRES vs IVAR etc) are affected. (Please see attached PK solution(for question 1),phxproj file) However, the theta and CV and graphs between the PK model in solution file and eddited PK model were not different. The results were not changed. why this happened?

 

2. For example of PK52(see attached PK52.png file), we calculated effective t1/2 from the formula, Effective t1/2 = In(2)*Vss/Cl.

    We calculated secondary parameter, eff_HL from the formula, eff_HL = Ln(2)*(tvV+tvV2)/tvCl

    How can we use Vss to get a eff_HL even though this drug doesn't appear to reach the steady state from the concentration vs time graph for PK52? Please explain how do we conclude whether this drug reached the steady state or not?

 

3. According to the concentration vs time graph for PK52(see attached PK52.png file), this drug was excreted very slowly in clamped rat compared to normal rat, so I judged the the majority of this drug was excreted through the kidney. Do we say that the main clearance site of this drug is kidney? 

 

Thank you

BH

Attached Thumbnails

• 

Attached Files

•  PK1 Solution(for question1).phxproj   665.69KB   378 downloads

[Simon Davis]

Hi BokHyeon, I'll try to clarify on your three questions.

 

1) in this example the data is quite rich and being fitted with a simple model, so *despite* the poor initial estimates the algorithm was still able to converge on what looks to be the optimal solution.

 

2)  You are correct this profile is from a single dose however since it was IV we can make an estimated of Vss = MRT * Cl,  since there is no absoprtion to complicate the question of how much drug is in the system at any one time.

 

there are a few blog posts that might be useful background reading

 

https://www.certara....en-vd-and-vss/?

 

https://www.certara....f-distribution/

 

3).Yes we saw that in rats with no kidney function the effective half-life increased nine-fold so we can say that in healthy animals the main clearance of this drug is via the kidney? 

 Simon.