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

Issues: Obese population, GOF plot and Renal impairment population #71

Closed
vbalbas opened this Issue Sep 1, 2017 · 15 comments

Comments

Projects
None yet
5 participants
@vbalbas

vbalbas commented Sep 1, 2017

Dear all,

I am simulating PK data using PBPK model. I want to obtain the PK profile different individuals and compare the output using data coming from a clinical trial. I have a few questions that maybe you could help me with:

  1. I have an Obese patient in my clinical trial data set. PKSim creates properly an individual with its specific characteristics (gender, age, height and weight). But when I create 200 ID population (either from healthy subject or the previously generated ID) with the same specific ID characteristics, PKSim crashes with the following warning:

image

How may I create a 200 ID population with the obese patient characteristics?

  1. Is there a way in which I can plot model predicted concentration vs my data (GOF Plot) for all the simulated ID I have?

  2. Is there a rapid alternative in which I can simulate different IDs from the same data set? (by uploading a data set with NONMEM format, for example). Or should I create an administration protocol, a specific population and a simulation per each ID?

  3. Is there a possibility of simulating different degree of Renal impairment? (not only by updating GFR per ID).

Thank you in advance
Kind regards
Violeta
``

@msevestre msevestre added the question label Sep 1, 2017

@msevestre

This comment has been minimized.

Show comment
Hide comment
@msevestre

msevestre Sep 1, 2017

Member

Hi @vbalbas,

Regarding 1: This is a bug.
Could you please send a screenshot of the settings that you used to created your obese Individual as well as the settings you used to create the population

2/ Cannot you use observed data import for that? You can display these data in individual and population simulation concentration plots

3/ Can you explain what you mean with IDs from the same data set? Are you asking if there is a way to create a bunch of unrelated individuals and administration protocols automatically?

Thanks,
Michael

Member

msevestre commented Sep 1, 2017

Hi @vbalbas,

Regarding 1: This is a bug.
Could you please send a screenshot of the settings that you used to created your obese Individual as well as the settings you used to create the population

2/ Cannot you use observed data import for that? You can display these data in individual and population simulation concentration plots

3/ Can you explain what you mean with IDs from the same data set? Are you asking if there is a way to create a bunch of unrelated individuals and administration protocols automatically?

Thanks,
Michael

@JanSchlender

This comment has been minimized.

Show comment
Hide comment
@JanSchlender

JanSchlender Sep 1, 2017

Hi Violeta,

  1. Currently, there is no build-in template provided for different stages of renal impairment. Solely adapting GFR will by far not meet the complexity of this disease. Not only renal parameters need to be adjusted but also the crosstalk of the hepato-renal axis should be considered.

Best,
Jan

JanSchlender commented Sep 1, 2017

Hi Violeta,

  1. Currently, there is no build-in template provided for different stages of renal impairment. Solely adapting GFR will by far not meet the complexity of this disease. Not only renal parameters need to be adjusted but also the crosstalk of the hepato-renal axis should be considered.

Best,
Jan

@vbalbas

This comment has been minimized.

Show comment
Hide comment
@vbalbas

vbalbas Sep 4, 2017

Hi @msevestre,

  1. The settings I used to create both, my obese individual and the obese population are:

    image

  2. Yes, I can add observed data to plot of simulated concentration vs time. I´m referring to plot concentration vs concentration, which I know I can obtain after estimating a parameter. But in my case, I just would like to obtain this plot, collecting all my individual data (observed) and their simulated mean concentration (predicted) for model evaluation:

    image

  3. Yes, that is what I meant.

Hi @JanSchlender,

  1. I understand that only by altering GFR, renal impairment is not captured. For that reason I was asking if there would be a renal impairment build-in template provided in PKSim. As there is not, would you, by any chance, know any publication where renal impairment is simulated using PKSim?

Thank you @msevestre and @JanSchlender,

Best,
Violeta

vbalbas commented Sep 4, 2017

Hi @msevestre,

  1. The settings I used to create both, my obese individual and the obese population are:

    image

  2. Yes, I can add observed data to plot of simulated concentration vs time. I´m referring to plot concentration vs concentration, which I know I can obtain after estimating a parameter. But in my case, I just would like to obtain this plot, collecting all my individual data (observed) and their simulated mean concentration (predicted) for model evaluation:

    image

  3. Yes, that is what I meant.

Hi @JanSchlender,

  1. I understand that only by altering GFR, renal impairment is not captured. For that reason I was asking if there would be a renal impairment build-in template provided in PKSim. As there is not, would you, by any chance, know any publication where renal impairment is simulated using PKSim?

Thank you @msevestre and @JanSchlender,

Best,
Violeta

@msevestre

This comment has been minimized.

Show comment
Hide comment
@msevestre

msevestre Sep 4, 2017

Member

@vbalbas

  1. What about the population settings? What ranges did you use? Could you post a screenshot as well?

  2. I see. No those plots are not available at the moment outside of the parameter identification. Would you please add a feature request for PK-Sim explaining your use case?

  3. No this is also not available at the moment. This was in the pipeline at some points but the momentum was lost before of a lack of interest at the time. I would appreciate if you could also add a feature request for that.

Cheers,

Member

msevestre commented Sep 4, 2017

@vbalbas

  1. What about the population settings? What ranges did you use? Could you post a screenshot as well?

  2. I see. No those plots are not available at the moment outside of the parameter identification. Would you please add a feature request for PK-Sim explaining your use case?

  3. No this is also not available at the moment. This was in the pipeline at some points but the momentum was lost before of a lack of interest at the time. I would appreciate if you could also add a feature request for that.

Cheers,

@vbalbas

This comment has been minimized.

Show comment
Hide comment
@vbalbas

vbalbas Sep 4, 2017

Hi @msevestre,

  1. For the population are exactly the same values for the specific ID, so there are not ranges (I wanted to simulate 200 ID with similar characteristics than for my specific ID).

    image

2 -3. I will add a feature for each one.

Thank you very much for your early reply.

Best,
Violeta

vbalbas commented Sep 4, 2017

Hi @msevestre,

  1. For the population are exactly the same values for the specific ID, so there are not ranges (I wanted to simulate 200 ID with similar characteristics than for my specific ID).

    image

2 -3. I will add a feature for each one.

Thank you very much for your early reply.

Best,
Violeta

@msevestre

This comment has been minimized.

Show comment
Hide comment
@msevestre

msevestre Sep 4, 2017

Member

@vbalbas
I see thanks. PK-Sim should definitely not crash (I will create a bug report for that).
Would you mind explaining your use case ?e.g. Why do you need to create 200 (almost) identical individuals? Unless you have some user defined variability in your population, the only variability you will get out of this population is in SA and Q and some few other parameters).

Member

msevestre commented Sep 4, 2017

@vbalbas
I see thanks. PK-Sim should definitely not crash (I will create a bug report for that).
Would you mind explaining your use case ?e.g. Why do you need to create 200 (almost) identical individuals? Unless you have some user defined variability in your population, the only variability you will get out of this population is in SA and Q and some few other parameters).

@vbalbas

This comment has been minimized.

Show comment
Hide comment
@vbalbas

vbalbas Sep 4, 2017

@msevestre
Just to see how good is the model prediction for 200 ID (almost identical), comparing the prediction with my data for the specific ID, but taking into account the variability for those mentioned parameters.

As an example:

Simulation of 200 ID with the same age, weight and height:
image

vbalbas commented Sep 4, 2017

@msevestre
Just to see how good is the model prediction for 200 ID (almost identical), comparing the prediction with my data for the specific ID, but taking into account the variability for those mentioned parameters.

As an example:

Simulation of 200 ID with the same age, weight and height:
image

@StephanSchaller

This comment has been minimized.

Show comment
Hide comment
@StephanSchaller

StephanSchaller Oct 10, 2017

Member

@vbalbas, great thread!

  1. Take into consideration, that even though you can create a "heavy" population in PKSim, they may not reflect the pathophysiology of obese patients (blood flow, fat/muscle distribution). But it's a start. An Obese population should be another feature request.

  2. & 3. Great feature requests. Especially 3. has great potential aspecially if we want to use PKSim for analysis of Phase III Datasets.

  3. I believe there are a number of publications that have covered renal impaired:

Member

StephanSchaller commented Oct 10, 2017

@vbalbas, great thread!

  1. Take into consideration, that even though you can create a "heavy" population in PKSim, they may not reflect the pathophysiology of obese patients (blood flow, fat/muscle distribution). But it's a start. An Obese population should be another feature request.

  2. & 3. Great feature requests. Especially 3. has great potential aspecially if we want to use PKSim for analysis of Phase III Datasets.

  3. I believe there are a number of publications that have covered renal impaired:

@vbalbas

This comment has been minimized.

Show comment
Hide comment
@vbalbas

vbalbas Oct 13, 2017

Hi @StephanSchaller,

Thank you very much for your comments and for the suggested publications. They are very interesting indeed.

Best,
Violeta

vbalbas commented Oct 13, 2017

Hi @StephanSchaller,

Thank you very much for your comments and for the suggested publications. They are very interesting indeed.

Best,
Violeta

@msrinivasan1

This comment has been minimized.

Show comment
Hide comment
@msrinivasan1

msrinivasan1 Nov 2, 2017

Dear @vbalbas

I found this paper that might help for model building in obesity

https://link.springer.com/article/10.2165%2F11594420-000000000-00000

msrinivasan1 commented Nov 2, 2017

Dear @vbalbas

I found this paper that might help for model building in obesity

https://link.springer.com/article/10.2165%2F11594420-000000000-00000

@vbalbas

This comment has been minimized.

Show comment
Hide comment
@vbalbas

vbalbas commented Nov 6, 2017

Thank you @msrinivasan1!

@msrinivasan1

This comment has been minimized.

Show comment
Hide comment
@msrinivasan1

msrinivasan1 Nov 7, 2017

Dear @msevestre ,

I have published PK data from only ONE obese individual in a case report. I would like to build a model on PKSim to understand how to dose the drug in that population. What should my general strategy be to build this model?

msrinivasan1 commented Nov 7, 2017

Dear @msevestre ,

I have published PK data from only ONE obese individual in a case report. I would like to build a model on PKSim to understand how to dose the drug in that population. What should my general strategy be to build this model?

@msevestre msevestre added the answer label Nov 7, 2017

@msevestre

This comment has been minimized.

Show comment
Hide comment
@msevestre

msevestre Nov 7, 2017

Member

@msrinivasan1 Please try to create one entry for each question you have. This will help other users in the community to find answer to similar questions.
In the meantime, check the best practices to get started with PK-Sim

Member

msevestre commented Nov 7, 2017

@msrinivasan1 Please try to create one entry for each question you have. This will help other users in the community to find answer to similar questions.
In the meantime, check the best practices to get started with PK-Sim

@msrinivasan1

This comment has been minimized.

Show comment
Hide comment
@msrinivasan1

msrinivasan1 Nov 8, 2017

Dear @msevestre ,

  1. I created a virtual patient population 100 patients weighting between 90- 165 kgs in weight.
  2. I put in all the drug related information, administration information and simulated the conc-time profile
  3. However, I'm getting a Cmax of the drug at 0.02 umol/L. It is really small as compared to the data in the literature.

Any idea what I am missing here?

msrinivasan1 commented Nov 8, 2017

Dear @msevestre ,

  1. I created a virtual patient population 100 patients weighting between 90- 165 kgs in weight.
  2. I put in all the drug related information, administration information and simulated the conc-time profile
  3. However, I'm getting a Cmax of the drug at 0.02 umol/L. It is really small as compared to the data in the literature.

Any idea what I am missing here?

@msevestre

This comment has been minimized.

Show comment
Hide comment
@msevestre

msevestre Nov 8, 2017

Member

Please create a new entry explaining your issue as requested. Thanks for your efforts

Member

msevestre commented Nov 8, 2017

Please create a new entry explaining your issue as requested. Thanks for your efforts

@msevestre msevestre closed this Nov 8, 2017

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