NONMEM Users Network Archive

Hosted by Cognigen

RE: Using sparse data to develop Population PK model for pediatrics

From: Jacob Brogren <jacob.brogren>
Date: Mon, 8 Jun 2015 08:25:48 +0000

Dear Daping,

In order to provide more than just general comments it would be useful to get some more information.

- What is the purpose of developing the PopPK model?

- What is the medical condition/disease?

- What is the relation between plasma concentration and clinical effect/safety?

- What age groups are included in your data set?

You have a limited number of patients in your data set, which is often the case with pediatric studies. There are different opinions about this, but I think you are doing the right thing to compare your estimates with literature values as a sanity check. Maybe you should consider including previous data or use parameters found in the literature as priors.

As you probably know, the dependence of PK parameters on body size and organ maturation is frequently incorporated in pediatric PopPK models. You can find more information in the scientific literature (Anderson and Holford, 2013) about how to implement this.

You assumed that all concentrations in one individual was resulting from the same dose. Further, you assumed no drug accumulation (three times daily dosing seem to be in line with that). These assumptions can in theory be tested by using alternative models and comparing how well the different models fit your data. But again, your data is limited and some models may not run.

Please feel free to tell us more about your problem.

Best Regards

Jacob Brogren

Anderson, Brian J., and Nick HG Holford. "Understanding dosing: children are small adults, neonates are immature children." Archives of disease in childhood 98.9 (2013): 737-744.

Jacob Brogren, MSc
Senior Consultant
[Description: QPharmetra_noTag_FA little.png]
+46 72-350 88 69 (M)
jacob.brogren .brogren

This e-mail communication is confidential and is intended only for the individual(s) or entity named above and others who have been specifically authorized to receive it. If you are not the intended recipient, please do not read, copy, use or disclose the contents of this communication to others. Please notify the sender that you have received this e-mail in error by replying to the e-mail or by telephoning +46723508869. Please then delete the e-mail and any copies of it. Thank you.

From: owner-nmusers
Sent: den 3 juni 2015 03:29
To: nmusers
Subject: [NMusers] Using sparse data to develop Population PK model for pediatrics

Dear all,

I would like to develop a population PK model for a drug used in pediatrics. We got drug concentrations from 15 patients. The drug was given 3 times a day. Only one plasma sample were drawn every week for 4 weeks. They record the exact sampling time after drug was given. So for each patient, we had 4 concentrations from 4 weeks.

I assumed all the 4 samples were from one dose and developed a Pop PK model. This assumption did not consider disease progress and drug accumulation. The result was still comparable with literature. I will be very appreciated your comments and suggestions for this model.


Daping Zhang
Ph.D. Candidate
University of Houston

(image/png attachment: image001.png)

Received on Mon Jun 08 2015 - 04:25:48 EDT

The NONMEM Users Network is maintained by ICON plc. Requests to subscribe to the network should be sent to:

Once subscribed, you may contribute to the discussion by emailing: