- On 28 Aug 2002 at 10:28:05, "SK. Abdul Mohammed Jafar Sadik Basha" (h2002032.-a-.bits-pilani.ac.in) sent the message

Back to the Top

How to calculate Ka value for a drug that is showing first order

absorption process, one compartment kinetics when administered

orally, but showing multicompartment kinetics when administered as

i.v. bolus at same dose?

SK.ABDUL MOHAMMED JAFAR SADIK BASHA - On 28 Aug 2002 at 13:08:17, "Gobburu, Jogarao V" (GOBBURUJ.-at-.cder.fda.gov) sent the message

Back to the Top

Hello,

Ideally you could model PO and IV data simultaneously using a 2-comp

model. Usually you see this type of behaviour when absorption process is

slower than the distribution process. On the other hand, you could get your

disposition model from the IV data and deconvolve the PO data to obtain ka

(you do not need to assume first-order absorption a priori).

Regards,

Joga Gobburu - On 29 Aug 2002 at 23:48:28, Walt Woltosz (walt.-at-.simulations-plus.com) sent the message

Back to the Top

The following message was posted to: PharmPK At 08:28 AM 8/28/02,

Abdul Mohammed Jafar Sadik Basha wrote:

>How to calculate Ka value for a drug that is showing first order

>absorption process, one compartment kinetics when administered

>orally, but showing multicompartment kinetics when administered as

>i.v. bolus at same dose?

>

This implies that Ka has a single value for a drug, which is simply

not true. See other posts here for a more complete discussion.

The use of a single value for Ka ignores so many factors that cause

the absorption rate to change with time and location. It is an

approximation that you can sometimes get away with, but it is as

limited as assuming that all drugs always have one-compartment

pharmacokinetics and are always cleared at a constant rate. Many, of

course, have multi-compartment pharmacokinetics and their clearance

is saturable.

We do not recommend using constant Ka absorption models except under

the rare conditions that the drug is highly soluble, very rapidly

absorbed in the upper small intestine, and has no saturable effects

of any kind.

Walt Woltosz

Chairman & CEO

Simulations Plus, Inc. (SIMU)

1220 W. Avenue J

Lancaster, CA 93534-2902

U.S.A.

http://www.simulations-plus.com

E-mail: walt.at.simulations-plus.com - On 29 Aug 2002 at 23:56:27, Roger Jelliffe (jelliffe.-at-.usc.edu) sent the message

Back to the Top

Dear Joga and All:

About bioavailability, and simultaneous IV and PO regimens. You

can also take subjects that have received both IV and PO dosage regimens

and estimate, for example, Ka, Fa, V, and Kel, and also Kcp and Kpc for a

peripheral compartment if desired. The USC*PACK software does that with our

population modeling software, both with the parametric iterative 2 stage

Bayesian IT2B (FOCE) and the nonparametric NPAG software. What is needed is

that each subject or patient receive the dosage by both routes in a single

overall dosage history. For example, it can be intermixed IV and PO, or

first IV then PO, or vice versa, ad lib.

If you would like to compare the behavior of the nonparametric

NPAG population modeling approach with that of the parametric IT2B

approach, which uses the FOCE approximation to the likelihood, the

performance of the IT2B with the FOCE and the NPAG is examined, for a

carefully simulated population of 800 subjects, where the parameter

distributions were in fact very Gaussian. Both approaches were compared by

Bob Leary at the San Diego Supercomputer Center with respect to

mathematical consistency (the results should get closer to the true results

as the number of subjects in the population increases), efficiency, and

convergence (how many times the original number of subjects are needed to

achieve twice the precision of parameter estimates (half the SD, for

example). Theoretically, 4 times the number are required for half the SD.

This is true for NPAG, but with the FOCE approximation, 16 times the number

are required. There is general agreement that NP methods are better when

the parameter distributions are not Gaussian. Here though, even when the

parameter distributions are specifically defined to be Gaussian, the

performance of NPAG was consistent, was more efficient, and had better

convergence than the parametric method with the FOCE approximation. Look on

our web site (www.lapk.org) under New Advances in Population Modeling.

At any rate, both methods will let you estimate Fa when the

subjects receive an intermixed regimen. For this purpose, I would be

inclined to use a consistent and efficient method with good convergence

properties.

Very best regards,

Roger Jelliffe

Roger W. Jelliffe, M.D. Professor of Medicine,

Division of Geriatric Medicine,

Laboratory of Applied Pharmacokinetics,

USC Keck School of Medicine

2250 Alcazar St, Los Angeles CA 90033, USA

email= jelliffe.at.usc.edu

Our web site= http://www.lapk.org - On 30 Aug 2002 at 00:01:22, "Willi Cawello" (Willi.Cawello.aaa.schwarzpharma.com) sent the message

Back to the Top

The first procedure to do this could be graphically (as we did 20 years

ago):

- draw the concentrations using a semilogaritmic scale (conc is

log-scaled)

- do an regression of the elimination phase (by a ruler and a pencil)

- 'extrapolate' the elimination phase to time zero (draw the regression

line to the conc scale)

- draw the differences (d) between line and measured concentrations for

all the samples from zero to the maximum concentration

- the new points (d) show the rate of the absorption; use a regression

of this points, the slope is ka

Best regards,

Willi

Willi Cawello, PhD

Senoir Scientist Pharmacokinetics

SCHWARZ BIOSCIENCES GmbH

Alfred Nobel Str. 10

D40789 Monheim am Rhein, Germany

willi.cawello.-at-.schwarzbiosciences.com

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 "Calculating ka" as the subject

PharmPK Discussion List Archive Index page

Copyright 1995-2010 David W. A. Bourne (david@boomer.org)