I am working with TDM data and am wondering when to use the M3 method for
dealing with BLQ data, or, when to just exclude data?
For instance: If I have <1% BLQ data in a large dataset (~1000 patients),
should I just exclude those patients/values? If I just have one patient with
one value BLQ, is it worth using the M3 method for that one patient? Is there
a breakpoint for which using these methods does/does not make a difference?
Thanks in advance...
______________________________________________________________________
CONFIDENTIALITY NOTICE:
The information in this e-mail may be confidential and/or
privileged. If you are not the intended recipient or an
authorized representative of the intended recipient, you
are hereby notified that any review, dissemination, or
copying of this e-mail and its attachments, if any, or
the information contained herein is prohibited. If you
have received this e-mail in error, please immediately
notify the sender by return e-mail and delete this e-mail
from your computer system. Thank you.
______________________________________________________________________
BLQ Data
Hi Brady,
Have a look at this recently published paper by Ron Keizer and colleagues;
will answer your question.
"Incorporation of concentration data below the limit of quantification in
population pharmacokinetic analyses". Keizer et al, Pharmacol Res Perspect,
2015.
Best Regards,
Quoted reply history
On Fri, Oct 2, 2015 at 4:22 PM, Moffett, Brady S. <[email protected]> wrote:
> I am working with TDM data and am wondering when to use the M3 method for
> dealing with BLQ data, or, when to just exclude data?
>
>
>
> For instance: If I have <1% BLQ data in a large dataset (~1000 patients),
> should I just exclude those patients/values? If I just have one patient
> with one value BLQ, is it worth using the M3 method for that one patient?
> Is there a breakpoint for which using these methods does/does not make a
> difference?
>
>
>
> Thanks in advance…
>
> ______________________________________________________________________
> CONFIDENTIALITY NOTICE:
> The information in this e-mail may be confidential and/or
> privileged. If you are not the intended recipient or an
> authorized representative of the intended recipient, you
> are hereby notified that any review, dissemination, or
> copying of this e-mail and its attachments, if any, or
> the information contained herein is prohibited. If you
> have received this e-mail in error, please immediately
> notify the sender by return e-mail and delete this e-mail
> from your computer system. Thank you.
> ______________________________________________________________________
>
--
Ahmed Abbas Suleiman, MSc
PhD Candidate at Bonn University
Institute of Pharmacology - Clinical Pharmacology Unit
Hospital of the University of Cologne
Gleueler Str. 24, 50931 Cologne,
Germany
Hi Brady,
I think there are three main articles you can refer to in terms of using M3
method for BLQ data:
Likelihood based approaches to handling data below the quantification limit
using NONMEM VI.
http://www.ncbi.nlm.nih.gov/pubmed/18686017
Impact of low percentage of data below the quantification limit on parameter
estimates of pharmacokinetic models.
http://www.ncbi.nlm.nih.gov/pubmed/21626437
Handling data below the limit of quantification in mixed effect models.
http://www.ncbi.nlm.nih.gov/pubmed/19452283
As a general rule of thumb, most people tend to ignore(=exclude) the datapoints
(not the whole ID) if BLQ data are less than 10%.
However, I think it is important to look at where these BLQ data occur mainly
to know if they can bring more info into your model and sometimes,
having these extra BLQ data left in the analysis and using the M3 method can be
better for parameter estimation. Nonetheless beware that M3 method does not
always help in terms of run convergence (covariance step).
Hope it helps!
Best regards,
Camille
Quoted reply history
From: [email protected] [mailto:[email protected]] On
Behalf Of Moffett, Brady S.
Sent: Friday, October 2, 2015 10:23 AM
To: [email protected]
Subject: [NMusers] BLQ Data
I am working with TDM data and am wondering when to use the M3 method for
dealing with BLQ data, or, when to just exclude data?
For instance: If I have <1% BLQ data in a large dataset (~1000 patients),
should I just exclude those patients/values? If I just have one patient with
one value BLQ, is it worth using the M3 method for that one patient? Is there
a breakpoint for which using these methods does/does not make a difference?
Thanks in advance...
______________________________________________________________________
CONFIDENTIALITY NOTICE:
The information in this e-mail may be confidential and/or
privileged. If you are not the intended recipient or an
authorized representative of the intended recipient, you
are hereby notified that any review, dissemination, or
copying of this e-mail and its attachments, if any, or
the information contained herein is prohibited. If you
have received this e-mail in error, please immediately
notify the sender by return e-mail and delete this e-mail
from your computer system. Thank you.
______________________________________________________________________
Hi Brady,
There are many papers discussing methods of handling BQL data. Below are
some examples:
[1] Ways to fit a PK model with some data below the quantification limit.
Beal SL. Journal of Pharmacokinetics and Pharmacodynamics 2001; 28:481-504
[3] Analysis of toxicokinetic data using NONMEM: Impact of quantification
limit and replacement strategies for censored data. Hing JP. et al. Journal
of Pharmacokinetics and Pharmacodynamics 2001; 28:465-479
[4] Impact of censoring data below an arbitrary quantification limit on
structural model misspecification. Byon W. et al. Journal of
Pharmacokinetics and Pharmacodynamics 2008; 35:101-116
[5] Likelihood based approaches to handling data below the quantification
limit using NONMEM VI. Ahn JE. et al. Journal of Pharmacokinetics and
Pharmacodynamics 2008; 35:401-424
[6] Impact of low percentage of data below the quantification limit on
parameter estimates of pharmacokinetic models. Xu XS. et al. Journal of
Pharmacokinetics and Pharmacodynamics 2011; 38:423-432
Also, I have conducted some analysis using real clinical PK models and will
present a poster in ACoP next Monday with a recommendation of decision tree
on handling BQL. My experience is that it really depends. If you have a
simple model with low BQL% (<10%), I think excluding BQL is fine. But if
you have a complex model with large BQL, M3 method should be considered.
Regards,
Li Li
Research Scientist, Global PK/PD and Pharmacometrics
Eli Lilly and Company
Quoted reply history
On Fri, Oct 2, 2015 at 10:22 AM, Moffett, Brady S. <[email protected]>
wrote:
> I am working with TDM data and am wondering when to use the M3 method for
> dealing with BLQ data, or, when to just exclude data?
>
>
>
> For instance: If I have <1% BLQ data in a large dataset (~1000 patients),
> should I just exclude those patients/values? If I just have one patient
> with one value BLQ, is it worth using the M3 method for that one patient?
> Is there a breakpoint for which using these methods does/does not make a
> difference?
>
>
>
> Thanks in advance…
>
> ______________________________________________________________________
> CONFIDENTIALITY NOTICE:
> The information in this e-mail may be confidential and/or
> privileged. If you are not the intended recipient or an
> authorized representative of the intended recipient, you
> are hereby notified that any review, dissemination, or
> copying of this e-mail and its attachments, if any, or
> the information contained herein is prohibited. If you
> have received this e-mail in error, please immediately
> notify the sender by return e-mail and delete this e-mail
> from your computer system. Thank you.
> ______________________________________________________________________
>
Sent from my iphone
Quoted reply history
On 3/10/2015, at 5:45 AM, Ahmed Suleiman
<[email protected]<mailto:[email protected]>> wrote:
Hi Brady,
Have a look at this recently published paper by Ron Keizer and colleagues; will
answer your question.
"Incorporation of concentration data below the limit of quantification in
population pharmacokinetic analyses". Keizer et al, Pharmacol Res Perspect,
2015.
Best Regards,
On Fri, Oct 2, 2015 at 4:22 PM, Moffett, Brady S.
<[email protected]<mailto:[email protected]>> wrote:
I am working with TDM data and am wondering when to use the M3 method for
dealing with BLQ data, or, when to just exclude data?
For instance: If I have <1% BLQ data in a large dataset (~1000 patients),
should I just exclude those patients/values? If I just have one patient with
one value BLQ, is it worth using the M3 method for that one patient? Is there
a breakpoint for which using these methods does/does not make a difference?
Thanks in advance...
______________________________________________________________________
CONFIDENTIALITY NOTICE:
The information in this e-mail may be confidential and/or
privileged. If you are not the intended recipient or an
authorized representative of the intended recipient, you
are hereby notified that any review, dissemination, or
copying of this e-mail and its attachments, if any, or
the information contained herein is prohibited. If you
have received this e-mail in error, please immediately
notify the sender by return e-mail and delete this e-mail
from your computer system. Thank you.
______________________________________________________________________
--
Ahmed Abbas Suleiman, MSc
PhD Candidate at Bonn University
Institute of Pharmacology - Clinical Pharmacology Unit
Hospital of the University of Cologne
Gleueler Str. 24, 50931 Cologne,
Germany