Bioequivalence and Bioavailability Forum • Compartmental modelling

jag009
★★★

NJ,
2012-05-01 21:55
(4814 d 21:58 ago)

Posting: # 8484
Views: 4,712

Compartmental modelling [PK / PD]

Hi everyone,

Could you help me on the following:

Assuming the plasma conc-time profile of a drug follows a extravascular (oral) 1st order 2 compartment model represented by the exponential equation:

Equation A: C(t)=A*exp(-alpha*t)+B*exp(-beta*t)+C*exp(-k01*t), where C = -(A+B).

How do I go about to generate a theoretical bolus IV profile

I assumed the equation to be (by remove the last term from equation A):

Equation B: C(t)=A*exp(-alpha*t)+B*exp(-beta*t)?

I tried that but the AUC from a equation A did not equal to equation B.

Am I doing something wrong?

Thanks

John

Helmut
★★★

Vienna, Austria,
2012-05-02 05:24
(4814 d 14:29 ago)

@ jag009
Posting: # 8485
Views: 3,805

Hybrid constants in iv and ev models

Post reply

Hi John!

❝ Equation A: C(t)=A*exp(-alpha*t)+B*exp(-beta*t)+C*exp(-k01*t), where C = -(A+B).

❝ I assumed the equation to be (by remove the last term from equation A):

❝ Equation B: C(t)=A*exp(-alpha*t)+B*exp(-beta*t)?

Since you are using Phoenix/WinNonlin look up in the User’s Guide how hybrid constants (A/α, B/β) are calculated from rate constants (iv bolus: Model 7, ev no lag time: Model 11). See the difference?
You cannot simply drop the C, since A and B are differently calculated in these models. k₀₁ in the ev model is part of A and B – therefore if you simply drop C you cannot use the ev model’s A and B in the iv model. Assuming F=1 calculate A and B for the iv model from D, V, α, β, and k₂₁ according to the formula of Model 7. Then your “Equation B” should work (untested!).

—
Dif-tor heh smusma 🖖🏼 Довге життя Україна!
Helmut Schütz

The quality of responses received is directly proportional to the quality of the question asked. 🚮
Science Quotes

jag009 ★★★ NJ, 2012-05-04 00:54 (4812 d 18:59 ago) @ Helmut Posting: # 8509 Views: 3,716	Hybrid constants in iv and ev models Post reply
	Hi Helmut, Thanks! I went ahead and refitted with a microconstant model to get V, Cl (etc) and then use the parameters with an IV bolus model to generate the IV profile. John