New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Scaling partition coefficients in Population simulation #234

Open
karthikl6 opened this Issue Oct 8, 2018 · 5 comments

Comments

Projects
None yet
3 participants
@karthikl6

karthikl6 commented Oct 8, 2018

Hi,

I have PK sim model, in which I had to scale partition coefficients (Interstitial : plasma & Intracellular : plasma) during model development. Now, If I want an individual PK simulation, then I'm scaling partition coefficients in Parameters --> Distribution --> Partition Coefficients. But, if I'm doing population PK simulation, I couldn't find a way to scale partition coefficients under the same path. Is there any other way that I could scale partition coefficients for population simulations?

@msevestre msevestre added the question label Oct 8, 2018

@msevestre

This comment has been minimized.

Member

msevestre commented Oct 11, 2018

@karthikl6 You would typically NOT scaled partition coefficients. Instead, you could optimize lipophilicity for your individual PK simulation. The partition coefficients values are highly dependent on lipophilicity.

The optimized lipophilicity would by construction be used in your population simulation as well so you would not have to worry about that anymore.

Does it make sense?
Cheers,
Michael

@msevestre msevestre added the answer label Oct 11, 2018

@karthikl6

This comment has been minimized.

karthikl6 commented Oct 14, 2018

@msevestre I'm working on Ketoconazole and its log P reported in literature is in the range of 3.85 - 4.40. But if I optimize this it is going below 2 and I thought it might be hard to justify. Instead I found couple of full of PBPK models of ketoconazole developed in Simcyp (Pathak et al 2017 and Rodrigo et al 2016) where they have scaled partition coefficients tp 2% of what Rodgers-rowland menthod predicts. So, I've done the same in PKSim which gave better fits. What do you suggest in this case?

@Aedginto

This comment has been minimized.

Member

Aedginto commented Oct 15, 2018

By reducing the coefficients to 2 percent of the R&R values, they are assuming that the reduction in coefficients is irrespective of tissue composition. I personally would optimize lipophilicity instead which maintains the relative effect of tissue composition. There is probably no right answer here though.

@msevestre

This comment has been minimized.

Member

msevestre commented Oct 15, 2018

@karthikl6 As far as HOW To scale partition coefficients in population is concerned, which was your original question, you have two choices:

1/ Partition coefficient are defined as "Variable in population". Which means that you can add a distribution for those parameters (or even set a fixed value)
image

The downsize of that approach is that all distributions will be independent (unless you use a constant distribution)

2/ The other approach is to CLONE the individual simulation with the scaled partition coefficients and swap out the individual for the population. After cloning, the partition coefficient will have the value of the original simulation. You can verify that by adding a variability as described in 1/ and see that the default value is the one from the individual simulation
image

Hope that makes sense,

Cheers,
Michael

@karthikl6

This comment has been minimized.

karthikl6 commented Oct 16, 2018

Dear Michael,
Thank you, I hope this helps. @msevestre Also is there any way in PKSim, where I can use liver microsomal data (Vmax and Km) not specific to any enzyme but overall clearance in the liver, if there are multiple enzymes involved in metabolism?
@Aedginto : Dear Andrea,
Thank you for your inputs. The other reason for not optimizing lipophilicity, in this case, is because I found in the literature that ketoconazole has concentration-dependent protein binding in plasma. Percent unbound in plasma is 1% [@1 ug/mL], 3% [@ 10 ug/mL]. So, I thought this also might be affecting the KP's to be completely off as there is almost 100% increase in fraction unbound at different concentrations. Please correct if I'm wrong.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment