Hi everyone,
I am developing a POP PK model for an anti-infective drug, I am trying to
determine if dosing should be weight based or not. The range of weight in the
study was 40-100 kg.
Weight was statistically significant for Cl/F but only explained 9% of the
variability observed for Cl.
I used allometric scaling to describe weights effect on Cl/F and slope effect
of weight was 0.58, and scaled to 60 kg (the median).
Based on the slope effect estimated, AUC is predicted to decrease by 15% for an
80 kg individual, and increase by 25% for an individual that weights 40 kg
compared to a 60 kg individual.
How much should I trust the slope effect determined by my study? and should I
rely on it to develop the dosing regimen?
if weight only explained 9% of variability observed with Cl/F, could that
indicate that it is not clinically significant and weight based dosing is not
required?
Thanks,
Abdullah Sultan, PhD candidate
University of Florida
Weight based dosing
Hi Abdullah Sultan,
Since your estimate is not too far from 0.75 exponent in Clearances, did you
try using the theoretical allometric scaling (0.75 in all clearances and 1.00
in volumes)? With these, it would be easier to justify. Once you include the
body size, I would suggest you check the matrix plots of ETA in CL or V versus
other covariates (demo, labs, etc) to see if there is any other info would help
explain the variability. Without knowing more of the nature of the drug, I
think these would help build the model.
Hope this helps,
Rudy
Quoted reply history
From: [email protected] [mailto:[email protected]] On
Behalf Of Sultan,Abdullah S
Sent: Tuesday, December 08, 2015 4:40 PM
To: [email protected]
Subject: [NMusers] Weight based dosing
Hi everyone,
I am developing a POP PK model for an anti-infective drug, I am trying to
determine if dosing should be weight based or not. The range of weight in the
study was 40-100 kg.
Weight was statistically significant for Cl/F but only explained 9% of the
variability observed for Cl.
I used allometric scaling to describe weights effect on Cl/F and slope effect
of weight was 0.58, and scaled to 60 kg (the median).
Based on the slope effect estimated, AUC is predicted to decrease by 15% for an
80 kg individual, and increase by 25% for an individual that weights 40 kg
compared to a 60 kg individual.
How much should I trust the slope effect determined by my study? and should I
rely on it to develop the dosing regimen?
if weight only explained 9% of variability observed with Cl/F, could that
indicate that it is not clinically significant and weight based dosing is not
required?
Thanks,
Abdullah Sultan, PhD candidate
University of Florida
Hi Abdullah:
This depends on how you set your criteria for including covariates into the
model (the p-value, reduction in OFV, se of the estimated covariate parameter
etc).
Before starting running covariate models, it is always a good practice to plot
BSV (i.e. ETAs) against the available covariates your have (in your case: ETAs
versus weight) then if you see biologically plausible covariate relationships
then you start testing them. I usually use forward addition (p-value 0.05,
delta reduction 3.84 or p-value 0.01, delta OFV 6.63 units) and backward
elimination (p-value 0.001, delta increase OFV 10.8 units) and se should be <
51.2%. You don't need to stick with these criteria but you can develop your
own. If including a covariate passes all the criteria and is estimated
precisely, then I think it should stay in the model. Reduction in BSV > 5% may
be significant.
There are various methods for including covariates in the literature, you may
need to choose yours!
Sincerely,
Ahmad Abuhelwa
University of South Australia
Adelaide, South Australia
Australia
Quoted reply history
From: [email protected] [mailto:[email protected]] On
Behalf Of Sultan,Abdullah S
Sent: Wednesday, 9 December 2015 11:10 AM
To: [email protected]
Subject: [NMusers] Weight based dosing
Hi everyone,
I am developing a POP PK model for an anti-infective drug, I am trying to
determine if dosing should be weight based or not. The range of weight in the
study was 40-100 kg.
Weight was statistically significant for Cl/F but only explained 9% of the
variability observed for Cl.
I used allometric scaling to describe weights effect on Cl/F and slope effect
of weight was 0.58, and scaled to 60 kg (the median).
Based on the slope effect estimated, AUC is predicted to decrease by 15% for an
80 kg individual, and increase by 25% for an individual that weights 40 kg
compared to a 60 kg individual.
How much should I trust the slope effect determined by my study? and should I
rely on it to develop the dosing regimen?
if weight only explained 9% of variability observed with Cl/F, could that
indicate that it is not clinically significant and weight based dosing is not
required?
Thanks,
Abdullah Sultan, PhD candidate
University of Florida
Dear Abdullah,
I assume that you have only looked at the decrease in the variance around CL
comparing the base model and the covariate model, when you state : "only
explained 9% of the variability".
We have shown that total parameter variability is changing throughout model
building and a decrease in unexplained parameter variability is not equal to an
increase in explained parameter variability when adding covariates in your
model. So the explained parameter variability might have increased by more
than 9%.
Please see the reference below. This methodology is now also implemented in
PsN and you can perform it alongside your covariate model building. You might
want to try this. The manuscript below also discusses the difference between
improved model fit and clinical significance of a covariate.
(Hennig S, Karlsson MO. Concordance between criteria for covariate model
building. J. Pharmacokinet. Pharmacodyn. 2014;41:109-125.)
Further, I would like to highlight to you that others have previously discussed
on NMusers and in the literature that for an easier comparison of study results
it is preferred to use a standard weight of 70kg.
Also, if you did use an allometric scaling model on CL/F, you would have used
an power exponent of ¾ or estimated this exponent, but not a slope. So I am
unsure about the slope effect that you are talking about and cannot comment
further on this.
Best wishes and a Happy holiday season
Stefanie
_____________________________________________________________
Dr Stefanie Hennig
Lecturer | Pharmacometrics
School of Pharmacy| Pharmacy Australia Centre of Excellence (PACE) |The
University of Queensland, QLD 4072, Australia
Phone: +61 7 334 61970, Fax: +61 7 334 61999, Email:
[email protected]<mailto:[email protected]>
Please note my working days are Monday to Thursday.
"You can't fix by analysis what you bungled by design." Light, Singer and
Willett
The World Conference of Pharmacometrics in Brisbane 2016
http://www.wcop2016.com/
[WCoP 2016 website banner 3]
Quoted reply history
From: [email protected] [mailto:[email protected]] On
Behalf Of Sultan,Abdullah S
Sent: Wednesday, 9 December 2015 10:40 AM
To: [email protected]
Subject: [NMusers] Weight based dosing
Hi everyone,
I am developing a POP PK model for an anti-infective drug, I am trying to
determine if dosing should be weight based or not. The range of weight in the
study was 40-100 kg.
Weight was statistically significant for Cl/F but only explained 9% of the
variability observed for Cl.
I used allometric scaling to describe weights effect on Cl/F and slope effect
of weight was 0.58, and scaled to 60 kg (the median).
Based on the slope effect estimated, AUC is predicted to decrease by 15% for an
80 kg individual, and increase by 25% for an individual that weights 40 kg
compared to a 60 kg individual.
How much should I trust the slope effect determined by my study? and should I
rely on it to develop the dosing regimen?
if weight only explained 9% of variability observed with Cl/F, could that
indicate that it is not clinically significant and weight based dosing is not
required?
Thanks,
Abdullah Sultan, PhD candidate
University of Florida
While it is important to look at the variability explained by the covariate,
your question is not really a statistical one. You need to evaluate the
clinical significance of the change in exposure by weight. For a given dose,
how does exposure change by weight and how will this affect the efficacy of the
antibiotic under various assumptions of bacterial sensitivity? Think about how
you can use your model to answer the clinical question - you won't find the
answer in p-values or objective function changes.
Mike Fossler
Quoted reply history
On Dec 8, 2015, at 9:28 PM, Sultan,Abdullah S
<[email protected]<mailto:[email protected]>> wrote:
Hi Rudy Gunawan,
Thanks for the helpful information,
I tried both methods, fixing the exponents at 0.75 and 1 and estimating them,
results for both models are similar, No other covariate (demographic variables
or lab test) showed any correlation with Cl/F.
my main question is how important is it to look at the variability explained by
a covariate? And can it be used to determine if a covrariate is not clinically
significant
Thanks,
Abdullah Sultan
________________________________
From: Rudy Gunawan <[email protected]<mailto:[email protected]>>
Sent: Tuesday, December 8, 2015 8:10 PM
To: Sultan,Abdullah S; [email protected]<mailto:[email protected]>
Subject: RE: Weight based dosing
Hi Abdullah Sultan,
Since your estimate is not too far from 0.75 exponent in Clearances, did you
try using the theoretical allometric scaling (0.75 in all clearances and 1.00
in volumes)? With these, it would be easier to justify. Once you include the
body size, I would suggest you check the matrix plots of ETA in CL or V versus
other covariates (demo, labs, etc) to see if there is any other info would help
explain the variability. Without knowing more of the nature of the drug, I
think these would help build the model.
Hope this helps,
Rudy
From: [email protected]<mailto:[email protected]>
[mailto:[email protected]] On Behalf Of Sultan,Abdullah S
Sent: Tuesday, December 08, 2015 4:40 PM
To: [email protected]<mailto:[email protected]>
Subject: [NMusers] Weight based dosing
Hi everyone,
I am developing a POP PK model for an anti-infective drug, I am trying to
determine if dosing should be weight based or not. The range of weight in the
study was 40-100 kg.
Weight was statistically significant for Cl/F but only explained 9% of the
variability observed for Cl.
I used allometric scaling to describe weights effect on Cl/F and slope effect
of weight was 0.58, and scaled to 60 kg (the median).
Based on the slope effect estimated, AUC is predicted to decrease by 15% for an
80 kg individual, and increase by 25% for an individual that weights 40 kg
compared to a 60 kg individual.
How much should I trust the slope effect determined by my study? and should I
rely on it to develop the dosing regimen?
if weight only explained 9% of variability observed with Cl/F, could that
indicate that it is not clinically significant and weight based dosing is not
required?
Thanks,
Abdullah Sultan, PhD candidate
University of Florida
________________________________
Notice: This e-mail message, together with any attachments, contains
information of Trevena, Inc., 1018 West 8th Avenue, King of Prussia, PA 19406,
USA. This information may be confidential, proprietary, copyrighted and/or
legally privileged.
It is intended solely for use by the individual or entity named on this
message. If you are not the intended recipient, and have received this message
in error, please notify us immediately and delete it and any attachments from
your system.
Abdullah,
I agree with Stefanie's advice.
Here are some references for you to read to help you understand about using biology not statistics for weight based dosing.
The use of a standard (70 kg) and theory based allometry (exponent 3/4 for clearance and 1 for volume) is important to allow comparison of different studies and for integration of pharmacokinetic knowledge (Holford, Heo & Anderson 2013).
It is difficult to obtain reliable estimates of allometric exponents using empirical estimation (Anderson & Holford 2008) because you need a broad size distribution and also to account for all other factors correlated with weight (e.g. age) in order to obtain true allometric exponents. The theory of allometry is well explained by West & Brown (2005).
Best wishes,
Nick
Holford N, Heo YA, Anderson B. A pharmacokinetic standard for babies and adults. J Pharm Sci. 2013;102(9):2941-52. Anderson BJ, Holford NH. Mechanism-based concepts of size and maturity in pharmacokinetics. Annu Rev Pharmacol Toxicol. 2008;48:303-32. West GB, Brown JH. The origin of allometric scaling laws in biology from genomes to ecosystems: towards a quantitative unifying theory of biological structure and organization. J Exp Biol. 2005;208(9):1575-92.
Quoted reply history
On 09-Dec-15 14:59, Stefanie Hennig wrote:
> Dear Abdullah,
>
> I assume that you have only looked at the decrease in the variance around CL comparing the base model and the covariate model, when you state :“only explained 9% of the variability”.
>
> We have shown that total parameter variability is changing throughout model building and a decrease in unexplained parameter variability is not equal to an increase in explained parameter variability when adding covariates in your model. So the explained parameter variability might have increased by more than 9%.
>
> Please see the reference below. This methodology is now also implemented in PsN and you can perform it alongside your covariate model building. You might want to try this. The manuscript below also discusses the difference between improved model fit and clinical significance of a covariate.
>
> (Hennig S, Karlsson MO. Concordance between criteria for covariate model building. J. Pharmacokinet. Pharmacodyn. 2014;41:109-125.)
>
> Further, I would like to highlight to you that others have previously discussed on NMusers and in the literature that for an easier comparison of study results it is preferred to use a standard weight of 70kg.
>
> Also, if you did use an allometric scaling model on CL/F, you would have used an power exponent of ¾ or estimated this exponent, but not a slope. So I am unsure about the slope effect that you are talking about and cannot comment further on this.
>
> Best wishes and a Happy holiday season
>
> Stefanie
>
> _______________________________________________________________
>
> *Dr Stefanie Hennig*
>
> Lecturer | Pharmacometrics
>
> School of Pharmacy| Pharmacy Australia Centre of Excellence (PACE) |The University of Queensland, QLD 4072, Australia
>
> Phone: +61 7 334 61970, Fax: +61 7 334 61999, Email: [email protected] < mailto: [email protected] >___
>
> *Please note my working days are Monday to Thursday.*
>
> **
>
> /"You can't fix by analysis what you bungled by design." Light, Singer and Willett/
>
> //
>
> *The World Conference of Pharmacometrics in Brisbane 2016 www.wcop2016.com < http://www.wcop2016.com/ >*//
>
> //
>
> WCoP 2016 website banner 3//
>
> *From:* [email protected] [ mailto: [email protected] ] *On Behalf Of *Sultan,Abdullah S
>
> *Sent:* Wednesday, 9 December 2015 10:40 AM
> *To:* [email protected]
> *Subject:* [NMusers] Weight based dosing
>
> Hi everyone,
>
> I am developing a POP PK model for an anti-infective drug, I am trying to determine if dosing should be weight based or not. The range of weight in the study was 40-100 kg.
>
> Weight was statistically significant for Cl/F but only explained 9% of the variability observed for Cl.
>
> I used allometric scaling to describe weights effect on Cl/F and slope effect of weight was 0.58, and scaled to 60 kg (the median).
>
> Based on the slope effect estimated, AUC is predicted to decrease by 15% for an 80 kg individual, and increase by 25% for an individual that weights 40 kg compared to a 60 kg individual.
>
> How much should I trust the slope effect determined by my study? and should I rely on it to develop the dosing regimen?
>
> if weight only explained 9% of variability observed with Cl/F, could that indicate that it is not clinically significant and weight based dosing is not required?
>
> Thanks,
>
> Abdullah Sultan, PhD candidate
>
> University of Florida
--
Nick Holford, Professor Clinical Pharmacology
Dept Pharmacology & Clinical Pharmacology, Bldg 503 Room 302A
University of Auckland,85 Park Rd,Private Bag 92019,Auckland,New Zealand
office:+64(9)923-6730 mobile:NZ+64(21)46 23 53
email: [email protected]
http://holford.fmhs.auckland.ac.nz/
"Declarative languages are a form of dementia -- they have no memory of events"
Holford SD, Allegaert K, Anderson BJ, Kukanich B, Sousa AB, Steinman A, Pypendop,
B., Mehvar, R., Giorgi, M., Holford,N.H.G. Parent-metabolite pharmacokinetic models
- tests of assumptions and predictions. Journal of Pharmacology & Clinical
Toxicology. 2014;2(2):1023-34.
Holford N. Clinical pharmacology = disease progression + drug action. Br J Clin
Pharmacol. 2015;79(1):18-27.