Back to the Top
The following message was posted to: PharmPK
Dear all,
In my work with different clients, I frequently hear about different plasma protein binding of a compound in animals and humans, which is then used to interpret a higher or lower efficacy in humans compared to animals. I'd like to know the general opinion about this reasoning and if it has any merit (published data).
Toufigh
Toufigh Gordi, PhD
Clinical Pharmacology, PK/PD analysis consultant
www.tgordi.com
Back to the Top
The following message was posted to: PharmPK
Toufigh,
I have difficulty understanding your question.
Toufigh Gordi wrote:
> In my work with different clients, I frequently hear about different plasma protein binding of a compound in animals and humans, which is then used to interpret a higher or lower efficacy in humans compared to animals. I'd like to know the general opinion about this reasoning and if it has any merit (published data). In the strict pharmacological sense the word 'efficacy' is equivalent to Emax i.e. the maximum effect predicted at infinite concentration. Emax will not be affected whether one uses unbound or total (bound+unbound) concentrations. This is a pharmacological principle that is impossible to test with data but it is the basis of using the Emax model.
Your subject line refers to 'free and total fractions'. A fraction in this context is a dimensionless quantity. The 'free' or unbound fraction is the ratio of unbound/(bound+unbound). The 'total' fraction must always be 1 i.e. the ratio of total/(bound+unbound).
I realize that people refer to 'free fraction' when the mean 'unbound concentration' and use 'efficacy' to mean the effect or effectiveness of drug. This usually because they were not properly trained in pharmacology. If you want an answer based on reasoning then it might be more helpful if you ask the question using pharmacologically appropriate terms.
Best wishes,
Nick
-- Nick Holford, Professor Clinical Pharmacology
Dept Pharmacology & Clinical Pharmacology
University of Auckland,85 Park Rd,Private Bag 92019,Auckland,New Zealand
Back to the Top
Dear Toufiugh,
There are some reports indicate that a given drug may demonstrate different plasma binding properties between various species. For example, warfarin shows very high (>99%) binding in human plasma and less binding (90-95%) in rat plasma. Furthermore, a-acid glycoprotein (AGP) levels and binding vary between humans and other species, so drugs that demonstrate significant AGP protein binding would be expected to show variable binding. Here are some references:
Collins JM, Klecker RW, Jr. Evaluation of highly bound drugs: interspecies, intersubject, and related comparisons. J Clin Pharmacol 2002;42: 971-975.
Lin JH. Species similarities and differences in pharmacokinetics. Drug Metab Dispos 1995;23: 1008-1021.
Hedaya MA, Daoud SS. Tissue distribution and plasma pharmacokinetics of UCN-01 at steady-state and following bolus administration in rats: influence of human alpha1-acid
glycoprotein binding. Anticancer Res 2001;21: 4005-4010.
Alkharfy KM, et.al. Disposition of abouthiouzine: a novel antihyperthyroid drug. Biopharm Drug Dispos. 2007;28: 105-11.
Best wishes,
Khalid M Alkharfy
College of Pharmacy
King Saud University
Riyadh, Saudi Arabia
Back to the Top
Dear Toufigh,
I agree with what Nick has said about the use of the terms "free fraction" and "free concentration". In many cases, these terms are used interchangeably but have a different meaning. While free concentrations are usually correlated to the pharmacodynamic effect, free/unbound fractions (fu) are the ratio of free, unbound (Cu) and total (Ctot) concentrations. If we now look at the equation for free concentrations Cu=Ctot*fu, then it becomes apparent that if only Ctot is experimentally determined, we have 1 equation with 2 unknowns. This can result in multiple scenarios, e.g. fu increases and Ctot stays constant -> Cu increases or fu increases and Ctot decreases (to the same extent) -> Cu stays constant, etc.. Again, when condering Cu as pharmacologically active, then these scenarios may result in different PD effects.
The situation becomes more complex as the body is not a "closed system" and input (bioavailability), elimination (clearance) and distribution have to be considered as well to sufficiently characterize free and total concentrations (which changes in protein binding may or may not have an effect on). Also, I think it is important to consider the dynamics of the system. If, for example, total average steady-state concentrations stay constant constant when fu chages, does that mean that total peak and through concentrations at steady-state have to be constant? And what about the respective free concentrations - do they change and what impact on the PD does that have?
We tried to summarize some of our thoughts on these questions/principles in a recent paper showing that a "one size fits all" approach may not always be suitable to answer these questions (Schmidt S, Gonzalez D, Derendorf H, Significance of protein binding in pharmacokinetics and pharmacodynamics, J Pharm Sci. 2010 Mar;99(3):1107-22).
To come back to your original question, yes, there can be differences in plasma protein binding between species. If these species differences in protein binding result in a different PD effect greatly depends on how they impact the PK of this very compound (see above).
Also, it has been shown that differences in protein binding may also occur depending on what type of binding agent is used. For example Beer et al. summarized some of these aspects for antimicrobials in Beer J, Wagner CC, Zeitlinger M, Protein binding of antimicrobials: methods for quatification and for investigation of its impact on bacterial killing. AAPS J. 2009 Mar; 11(1): 1-12).
Best Regards,
Stephan Schmidt, Ph.D.
Leiden/Amsterdam Center for Drug Research
Back to the Top
The following message was posted to: PharmPK
It seems that I have not been very clear in my previous posting (specially with regard to the title of the question) so I'll try to remedy it here. I will give an example and then re-phrase my question. Hope it works this time.
Let's assume that a drug candidate has been shown to reach its maximum effect (E-max) at a concentration level of 100 ng/mL in dogs (the animal model in the preclinical studies). In other words, when infusing dogs to different steady state levels, there is no increase in the observed effect at concentrations above 100 ng/mL. The concentration-response relationship in humans for this drug is unknown. Assuming the same concentration-response relationship in humans (same Emax and EC50), it is decided to target the same concentration level in humans to reach the maximum effect. The issue in hand is that the free fraction of the drug in plasma in dogs is 1%, while in humans it is 10%. So the simple question is: do we need to target a total concentration of 100 ng/mL in humans or should this be adjusted so the target concentration will be 10 ng/mL in humans?
So, re-phrasing my original question, when trying to use animal data to guide you in dose-selection in humans, is it enough to use the total concentrations or should one adjust for the different degrees in protein bindings in different species and adjust the dose (and hence concentrations) accordingly.
If your suggestion is that the dose selection should be based on Cu, please provide an example, where use of Cu has been shown to be superior to using total concentrations.
Toufigh
Toufigh Gordi, PhD
Clinical Pharmacology, PK/PD analysis consultant
www.tgordi.com
E-mail: tg.-a-.tgordi.com
Back to the Top
Hi Toufigh,
Here is one reference that describes differences in concentration-response relationships between animals and humans that are explained by differences in protein binding:
Modelling of biomarker response of naproxen in rats and humans. Dymphy Huntjens, Magret Blom-Roosemalen, David Spalding, Meindert Danhof and Oscar Della Pasqua. PAGE 12 (2003) Abstr 449 http://www.page-meeting.org/?abstract=449
Regards,
Katya
--
Ekaterina Gibiansky, Ph.D.
CEO&CSO, QuantPharm LLC
Web: www.quantpharm.com
Email: EGibiansky at quantpharm.com
Back to the Top
The following message was posted to: PharmPK
Toufigh,
You make your question over complicated by saying the drug reaches Emax at 100 ng/mL.
> Let's assume that a drug candidate has been shown to reach its maximum effect (E-max) at a concentration level of 100 ng/mL in dogs (the animal model in the preclinical studies). In other words, when infusing dogs to different steady state levels, there is no increase in the observed effect at concentrations above 100 ng/mL.
What you mean is that the drug effect approaches Emax at a conc of 100 ng/mL. Emax is only reached at infinite concentrations.
> Assuming the same concentration-response relationship in humans (same Emax and EC50), it is decided to target the same concentration level in humans to reach the maximum effect. The issue in hand is that the free fraction of the drug in plasma in dogs is 1%, while in humans it is 10%. So the simple question is: do we need to target a total concentration of 100 ng/mL in humans or should this be adjusted so the target concentration will be 10 ng/mL in humans?
If you asked this question:
I have a drug which produces an effect in dogs at 100 ng/mL. The free fraction in dogs is 0.01 and in humans is 0.1. What conc in humans would produce the same effect if the Emax and unbound EC50 were the same in dogs and humans?
Then the answer is 10 ng/mL.
Nick
-- Nick Holford, Professor Clinical Pharmacology
Dept Pharmacology & Clinical Pharmacology
University of Auckland,85 Park Rd,Private Bag 92019,Auckland,New Zealand
Back to the Top
The following message was posted to: PharmPK
Dear Toufigh,
Given the big difference in protein binding between species, with humans having an unbound fraction 10 times higher than dogs, caution should also be exerted in terms of safety margins. The same plasmatic exposure in humans as in dogs will correspond to a 10 fold higher unbound drug exposure, toxicity of the drug will probably appear in humans at much lower plasmatic exposure than in dogs.
Do you have plasma protein binding and in-vivo data for other species? If the compound is planned to go to the clinic, I suppose rat toxicity / toxicokinetic data is available. How does that compare to dog?
Note however that the difference in unbound fraction could also influence the dose / exposure relationship. As more drug is unbound the volume of distribution (calculated from total plasma concentrations) may not be as expected based on animal data. More drug will also be available to elimination organs possibly influencing clearance, this would have a major effect if passive renal filtration was a major route of elimination for your drug (not often the case with NCEs nowadays though).
With a higher unbound fraction in plasma in human compared to dog, active and toxic plasmatic exposures can be expected to be lower in humans than in dogs; scaling the dose / exposure relationship from dog to human will also be influenced by the difference in binding.
Patrice
Back to the Top
The following message was posted to: PharmPK
Dear Patrice,
You wrote:
> With a higher unbound fraction in plasma in human compared to dog,
>active and toxic plasmatic exposures can be expected to be lower in
>humans than in dogs;
This is true if you express exposure as AUC, but (usually) not if you
express exposure in terms of AUC of the unbound concentration. The
latter (usually) makes more sense, because it is the unbound
concentration that is (usually) responsible for drug effect and
toxicity. AUC of the unbound concentration is (usually) independent on
the degree of plasma protein binding.
best regards,
Hans Proost
Johannes H. Proost
Dept. of Pharmacokinetics, Toxicology and Targeting
University Centre for Pharmacy
Antonius Deusinglaan 1
9713 AV Groningen, The Netherlands
Want to post a follow-up message on this topic?
If this link does not work with your browser send a follow-up message to PharmPK@boomer.org with "Free and total fractions" as the subject | Support PharmPK by using the |
Copyright 1995-2011 David W. A. Bourne (david@boomer.org)