Plasma Protein Binding:
Friend or Foe in Drug Discovery?
Kevin Beaumont and Tristan Maurer
Pharmacokinetics, Dynamics and Metabolism
Pfizer Worldwide Research and Development
Memorial Drive, Cambridge
Massachusetts USA
Total and Free Plasma Concentration (ng/mL)
0.1
atorvastatin
digoxin
losartan
clonidine
amlodipine
warfarin
rosiglitazone
glburide
tolterodine
montelukast
sertraline
simvastin
alendronate
clonazepam
lorazepam
ceteirizine
amitryptilline
sildenafil
alprazolam
lansoprazole
diclofenac
zoldipem
pravastatin
pioglitazone
propranolol
celecoxib
verapimil
cephalexin
furosemide
diltiazem
fluoxetine
buproprion
lisinopril
hydrochlorth
rofecoxib
citalopram
venlaxfaxine
valsartan
trazodone
naproxen
metoprolol
ranitidine
ibuprofen
Plasma Protein Binding of Some
Top 100 Most Prescribed Drugs
1000000
100000
Total
Free
10000
1000
100
10
1
Drugs with 2 log units between total and free plasma
concentration, percent bound greater than 98-99.9 %.
If ppb drives dose, why are these compounds successful drugs?
Smith, D.A. et. Al. Nat. Rev. Drug Disc. 2010, 9, 929-939.
Individual Drugs
• Fluconazole
Naproxen
• fu = 0.88
fu = 0.003
• Identical free drug
Identical free drug
concentration (SD or MD)
concentration (SD or MD)
– Plasma
Plasma
– CSF, Saliva
Synovial fluid (deep tissue
– Vaginal secretions
fluid)
– Breast milk, sputum
– Prostatic seminal vesicle fluid
Suggests high plasma protein binding is not a hindrance to distribution or success
Importance
of selecting your battles
Title
We have many foes in
Drug Discovery
Why pick a fight with a neutral agent?
Attacking ppb is like invading Switzerland!
No benefit and a whole heap of trouble!
Toxicity
Potency
Selectivity
Intrinsic Clearance
Drug Drug Interactions
Reactive metabolites
Etc
Pfizer Internal Use Only
Plasma
Protein Binding is a Neutral Parameter
Title
Plasma protein binding is a key parameter to measure
Important to unmask free concentration
Important in PK and dose prediction
Plasma protein binding is not a key parameter to modulate
Neutral on free concentration
Neutral on half-life*
Neutral on dose
Exception is when Vd is fixed to blood volume*
Pfizer Internal Use Only
Free drug concentrations and
antimuscarinic activity of zamifenacin
Zamifenacin, potent and
selective M3 muscarinic antagonist.
10000
1000
Dose mg/kg
Cmax (total) ng/ml
Cmax (free) ng/ml
99.80%
99.98%
100
Translatable
Biomarker
Inhibition of
pupillary
response to
light (rat) or
transient
blurred vision
(man)
10
1
.1
Rat
Man
10-fold difference in free fraction,
0.002 and 0.0002 respectively
Highly bound in both species
Free concentration driving effect is identical
Beaumont et al, 1994, Brit J. Clin. Pharm. 38, 179P
Comparison of M3 Muscarinic Receptor
Antagonists
Zamifenacin
Compounds
In Vitro K (nM)
b
Intrinsic Clearance
Fu
Daily Clinical Dose
Darifenacin
Zamifenacin / Darifenacin
2 fold
Similar
300 fold (0.06 vs. 0.0002)
2 fold
Dose depends on potency and intrinsic clearance, but not PPB
Dose – Steady State Equation
Objective
Potency
Clearance
Dose Css, avg  CL

τ
F
Absorption + CL
Wagner, Nature(1965)
Determinants of Dose
Clearance
Potency
Objective
Dose

τ
Css, avg  CL
F
Substitute and simplify
Wagner, Nature(1965)
Absorption + CL
For hepatically cleared drugs dosed orally
CL 
Qh  fu  CLint
Qh  fu  CLint
Cancel fu
and Qh terms
Css, u   
Qh  CLint 
  τ Simplify
Qh
 (fu  CLint )

denominator
Dose 
  fu  CLint  
fa  1  
 
  Qh  (fu  CLint ) 
Css, u   
Qh  CLint 
τ
Qh
 (fu  CLint)

Dose 
Qh  (fu  CLint)- fu  CLint
fa 
Qh  (fu  CLint)
Pang, J Pharmacokin Biopharm (1973)
 CL 
F  fa  1 

 Qh 
Substitute terms to reveal protein binding influence
 Css, u   Qh  fu  CLint 

  
  τ
 fu   Qh  (fu  CLint)
Dose 
  Qh  fu  CLint  
 

Qh  (fu  CLint) 

fa  1  

Qh






Consolidate numerator and denominator
 Qh  CLint 
Qh  (fu  CLint)
  τ 
Dose  Css, u   
fa  Qh  (fu  CLint)- fu  CLint
 Qh  (fu  CLint)
simplify
 Css, u  CLint  τ 
Dose  

fa


“the chemists most important
pharmacokinetic equation”
Important:
No requirement for fu
CLint is the unbound intrinsic clearance
(the theoretical maximum rate of clearance in the absence of flow and binding limitations)
“The only determinants of unbound steady state exposure are intrinsic clearance and dose”
- Harold Boxenbaum
“Plasma protein binding alone cannot be regarded as either
a positive or negative aspect of a compound” - van de Waterbeemd, J Med Chem (2001)
Determinants of Dose –
Determinants
of
Dose
Oral drugs, hepatically cleared
Clearance
Potency
Objective
Dose

τ
Css, avg  CL
F
Substitute and simplify
Wagner, Nature(1965)
Absorption + CL
For hepatically cleared drugs dosed orally
CL 
Qh  fu  CLint
Qh  fu  CLint
Cancel fu
and Qh terms
Css, u   
Qh  CLint 
  τ Simplify
Qh
 (fu  CLint )

denominator
Dose 
  fu  CLint  
fa  1  
 
  Qh  (fu  CLint ) 
Css, u   
Qh  CLint 
τ
Qh
 (fu  CLint)

Dose 
Qh  (fu  CLint)- fu  CLint
fa 
Qh  (fu  CLint)
Pang, J Pharmacokin Biopharm (1973)
 CL 
F  fa  1 

 Qh 
Substitute terms to reveal protein binding influence
 Css, u   Qh  fu  CLint 

  
  τ
 fu   Qh  (fu  CLint)
Dose 
  Qh  fu  CLint  
 

Qh  (fu  CLint) 

fa  1  

Qh






Consolidate numerator and denominator
 Qh  CLint 
Qh  (fu  CLint)
  τ 
Dose  Css, u   
fa  Qh  (fu  CLint)- fu  CLint
 Qh  (fu  CLint)
simplify
 Css, u  CLint  τ 
Dose  

fa


“the chemists most important
pharmacokinetic equation”
No requirement for fu
“The only determinants of unbound steady state exposure are intrinsic clearance and dose”
- Harold Boxenbaum
“Plasma protein binding alone cannot be regarded as either
a positive or negative aspect of a compound” - van de Waterbeemd, J Med Chem (2001)
Determinants of Dose –
Determinants
of
Dose
Oral drugs, hepatically cleared
Clearance
Potency
Objective
Dose

τ
Css, avg  CL
F
Substitute and simplify
Wagner, Nature(1965)
Absorption + CL
For hepatically cleared drugs dosed orally
CL 
Qh  fu  CLint
Qh  fu  CLint
Cancel fu
and Qh terms
Css, u   
Qh  CLint 
  τ Simplify
Qh
 (fu  CLint )

denominator
Dose 
  fu  CLint  
fa  1  
 
  Qh  (fu  CLint ) 
Css, u   
Qh  CLint 
τ
Qh
 (fu  CLint)

Dose 
Qh  (fu  CLint)- fu  CLint
fa 
Qh  (fu  CLint)
Pang, J Pharmacokin Biopharm (1973)
 CL 
F  fa  1 

 Qh 
Substitute terms to reveal protein binding influence
 Css, u   Qh  fu  CLint 

  
  τ
 fu   Qh  (fu  CLint)
Dose 
  Qh  fu  CLint  
 

Qh  (fu  CLint) 

fa  1  

Qh






Consolidate numerator and denominator
 Qh  CLint 
Qh  (fu  CLint)
  τ 
Dose  Css, u   
fa  Qh  (fu  CLint)- fu  CLint
 Qh  (fu  CLint)
simplify
 Css, u  CLint  τ 
Dose  

fa


“the chemists most important
pharmacokinetic equation”
No requirement for fu
“The only determinants of unbound steady state exposure are intrinsic clearance and dose”
- Harold Boxenbaum
“Plasma protein binding alone cannot be regarded as either
a positive or negative aspect of a compound” - van de Waterbeemd, J Med Chem (2001)
Determinants of Dose –
Determinants
of
Dose
Oral drugs, hepatically cleared
Clearance
Potency
Objective
Dose

τ
Css, avg  CL
F
Substitute and simplify
Wagner, Nature(1965)
Absorption + CL
For hepatically cleared drugs dosed orally
CL 
Qh  fu  CLint
Qh  fu  CLint
Cancel fu
and Qh terms
Css, u   
Qh  CLint 
  τ Simplify
Qh
 (fu  CLint )

denominator
Dose 
  fu  CLint  
fa  1  
 
  Qh  (fu  CLint ) 
Css, u   
Qh  CLint 
τ
Qh
 (fu  CLint)

Dose 
Qh  (fu  CLint)- fu  CLint
fa 
Qh  (fu  CLint)
Pang, J Pharmacokin Biopharm (1973)
 CL 
F  fa  1 

 Qh 
Substitute terms to reveal protein binding influence
 Css, u   Qh  fu  CLint 

  
  τ
 fu   Qh  (fu  CLint)
Dose 
  Qh  fu  CLint  
 

Qh  (fu  CLint) 

fa  1  

Qh






Consolidate numerator and denominator
 Qh  CLint 
Qh  (fu  CLint)
  τ 
Dose  Css, u   
fa  Qh  (fu  CLint)- fu  CLint
 Qh  (fu  CLint)
simplify
 Css, u  CLint  τ 
Dose  

fa


“the chemists most important
pharmacokinetic equation”
No requirement for fu
“The only determinants of unbound steady state exposure are intrinsic clearance and dose”
- Harold Boxenbaum
“Plasma protein binding alone cannot be regarded as either
a positive or negative aspect of a compound” - van de Waterbeemd, J Med Chem (2001)
Thought Experiment 1 –Modulate Fraction Unbound,
Maintain Dose and Intrinsic Clearance (unbound)
Cliu = 1000 mL/min/kg, Q = 50 mL/min/kg, Complete absorption, Fixed dose
u
u
ClH = Q(fClearance
u.Cli )/(Q+fu.Cli )
AUCu
AUC u= dose/ Cl u
2
50
1.6
40
1.2
30
0.8
20
0.4
10
0
0
0
0.2
0.4
fu
0.6
0.8
1
0
0.2
0.4
fu
0.6
0.8
1
Bioavailability
AUC total = dose/Cl
F = 1 - (ClH/Q)
AUC
10000
100
1000
80
60
100
40
10
20
0
1
0
0.2
0.4
fu
0.6
0.8
1
0
Math courtesy of Don Walker
0.2
0.4
fu
0.6
0.8
1
It is tough to modulate fraction
unbound - IPRL
Isolated perfused rat liver model: modulate ppb with amount of BSA in perfusate
Total
Total
Free
Free
In vivo: Ctot decreases with increase in fu but Cfree does not change
Liu, X. et. Al. Curr. Top. Med Chem. 2011, 11, 450-466.
It is tough to modulate fraction
unbound – NAR Rats
Nagase analbuminemia rats (NAR): mutant strain from wild type Spraque-Dawley
rats (less albumin)
Total
Free
3X
Decrease in fu does not change the Concu ONLY the Conctot
Liu, X. et. Al. Curr. Top. Med Chem. 2011, 11, 450-466.
15
Dose is the
most important parameter to
Modulating
Dose
controlClearance and Potency (unbound concentration)
- Unbound
t is dose interval
F
Css is steady state Cav (unbound)
Related to potency
Cl is the clearance of the compound
(unbound)
Related to metabolism/excretion rate
Average plasma concentration
at a dose of 200mg
10000
Average unbound concentration
(nM)
Dose 
Css, avg  CL  τ
Clu 10ml/min/kg
Cav 520nM
1000
Clu 1000ml/min/kg
100
Cav 5nM
10
Lower dose
More potent
Lower Clun
1
0.1
1
10
100
1000
10000
Unbound clearance (ml/min/kg)
16
Modulating Dose
- Dose calculation for marketed drugs
1000000
Unbound intrinsic clearance (Clint), in vitro
potency and dose size
100000
Dose =
20mg
Acetaminophen (4 g)
Potency (nM)
10000
Dose =
200mg
Carbemazepine (1000mg)
1000
Quinine (2g)
Dose =
2000mg
Atenolol (75mg)
Artemisinin(250mg)
100
Dofetilide (10mg)
Sildenafil (100mg)
Indomethacin (100mg)
Propranolol (60mg)
Revatropate (1mg)
Fluoxetine (20mg)
10
1
0.1
0.1
1
10
100
CLint (ml/min/kg)
1000
10000
A Conundrum Wrapped up in a Riddle
Why is the literature replete with
approaches that target lower plasma
protein binding?
18
The Paradox:
The tactics to modulate fraction unbound are the same as
those to modulate unbound intrinsic clearance
A strategy to lower plasma protein binding by lowering log D(7.4)
may appear to improve dose………
……… but only by lowering unbound intrinsic clearance!!
van de Waterbeemd, J Comp Aid Mol Des (2001)
Unless we want to make life complicated!
Compartmental Model:
Physiological Parameter Models:
CL 
Qh  CLint fu plasma
Qh  CLint fu plasma
 fu plasma 

Vd  Vc  Vt 
 fu tissue 
F  fa  1  CLh 
Qh 

or GFR  fu plasma
Target Exposure:
5nM as Cu,ss,max, Cu,ss,min or Cu,ss,avg
Primary parameters:
fu: 0.01 – 1
CLint: 1 – 100,000 ml/min/kg
fa=1
Tau=24 hrs
ka=1.8 hr-1
futissue=0.01
Vc = 0.2 L/kg (extracellular fluid)
Vt = 0.4 L/kg (intracellular fluid)
Secondary parameters:
Vd = 0.6 – 40 L/kg
CL:0.01 – 19.996 ml/min/kg
t1/2: 1 – 800 hr
Dose to Achieve Target Cav
Css, avg  CL  τ
Dose 
F
Css, avg  CL  τ
Dose 
1
Hepatic / IV
Hepatic / PO
IV + Hepatic CL
PO + Hepatic CL
100000
CLint = 100000
1000
CLint = 100000
Bottom Line:
Easier to modulate dose with Clint
(and potency) rather than fu
Dose (mg)
Dose (mg)
10000
1000
100
100
10
10
CLint = 1
1
0.01
0.1
CLint = 1
1
fu
1
0.01
0.1
1
fu
As fu Css,av but
is offset by Cl and F
Overall, dose is independent of fu
As fu Css,av but
is only partially offset
by Cl
Effect is only seen at high
values of intrinsic clearance
Math courtesy of Tristan Maurer
Dose to Achieve Target Cmax
Dose 
Cmax
 Vd  (1 - e -k el t )  e k el t max
F
Hepatic / IV
Hepatic / PO
IV + Hepatic CL
PO + Hepatic CL
Dose (mg)
10000
1000
100
10
100
CLint = 100000
Bottom Line:
Easier to modulate dose with Clint
CLint = 1
(and potency) rather than
fu
CLint = 100000
Dose (mg)
100000
10
1
1
0.1
0.01
CLint = 1
0.1
1
fu
As fu Vd and Cl but Cmax and F
Increased dose requirements
only observed at high Clint
due to nonlinear effects on Cl and F
0.1
0.01
0.1
1
fu
As fu Vd and Cl but Cmax
Decreased dose requirements
with fu only observed at high Clint
Math courtesy of Tristan Maurer
Dose to Achieve Target Cmin
Cmin  Vd  k el  k a   (1- e-k el τ )
Dose 
F  k a  ekaτ  ekelτ


Hepatic / IV
Hepatic / PO
IV + Hepatic CL
PO + Hepatic CL
1E+17
Dose (mg)
1E+15
1E+13
1E+11
1E+09
1E+07
100000
1000
10
Bottom Line:
Easier to modulate dose with Clint
(and potency) rather than fu
0.1
0.01
100000
CLint = 100000
10000
Dose (mg)
1E+19
1000000
CLint = 100000
1E+21
1000
100
10
1
CLint = 1
CLint = 1
0.1
1
fu
Increasing fu, affects parameters the
same way leading to a decrease in dose,
but in a way that is nonlinearly related
to CLint with a impact only at higher
values of Clint and very high doses
0.1
0.01
0.1
1
fu
With loss of F, the effect is less
pronounced, but still evident at high
Clint values
Math courtesy of Tristan Maurer
Plasma
Protein Binding is a Neutral Parameter
Title
Plasma protein binding is a key parameter to measure
Important to unmask free concentration
Important in PK and dose prediction
Plasma protein binding is not a key parameter to modulate
Neutral on free concentration
Neutral on half-life*
Neutral on dose
Exception is when Vd is fixed to blood volume*
Unless you want to make it
complicated!
Bottom Line:
Easier to modulate dose with Clint
(and potency) rather than fu
Pfizer Internal Use Only
Summary and Recommendation
The next time your Discovery colleagues want to lower ppb…….
…….ask them to consider:
Dose route (iv or po)
In Vitro potency
Target (Cav, Cmax or Cmin)
Clearance route (hepatic metabolic or renal)
Clintu
Keep calm and carry on!
Acknowledgements
Dennis Smith
Li Di
Don Walker
Tristan Maurer
Back Ups
Dose to Achieve Target Cav
Dose Css,avg  CL

τ
F
Hepatic
PO
HepaticPO
IV
Renal
Math courtesy of Tristan Maurer
Dose to Achieve Target Cmax
Dose 
Cmax
 Vd  (1 - e -k el t )  e k el t max
F
Hepatic
PO
IV
Renal PO
Math courtesy of Tristan Maurer
Dose to Achieve Target Cmin
Cmin  Vd  k el  k a   (1- e-k el τ )
Dose 
F  k a  ekaτ  ekelτ


Hepatic
PO
IV
Renal PO
Math courtesy of Tristan Maurer