# Bioequivalence and Bioavailability Forum 03:55 CET

CECIF
Junior

Colombia,
2017-11-02 22:45

Posting: # 17964
Views: 2,775

## Vd/F calculation in bear [R for BE/BA]

Hello everybody

I have been using bear for bioequivalence calculation since version 6.2. I consider it very useful and approachable and I am really grateful with yjlee168 for it.

For regulations in my country I have to make my own validation of bear's estimates. therefore I have been using a study of my own and I comapre my result with the ones I got from bear with the same study. Most of the values came out quite well, on exception of some of the AUC0inf and AUC0t mostly on the last sampling times which came out a little higher in my calculations (I can only suposse it is because of the decimal numbers).

However, the one parameter I have been struggling to get right in any way was Vd/F. Therefore I really need to ask what method bear uses to calculate it. I have been trying different equations to get it, but I failed in every one.
is it Vd/F= D*Ka/(I(Ka-Ke))? with ka from the residuals method or the wagner-nelson method?
or is it simply Vd/F=D/(AUC0inf*Ke)?
Vd/F= Apparent volume of distribution
D=dosage
I=intercept of the concentration vs time graph (x=0) when extrapolating the excretion line
ke=elimination constant (lambda z)
ka=absortion constant

Feel free to correct any missconception from my equations too.

I would really appreciate any light you could shed on this matter.
Helmut
Hero

Vienna, Austria,
2017-11-02 23:46

@ CECIF
Posting: # 17965
Views: 2,413

## Vd/F calculation in bear

Hi CECIF,

» […] the one parameter I have been struggling to get right in any way was Vd/F. Therefore I really need to ask what method bear uses to calculate it. […]
» or is it simply Vd/F=D/(AUC0inf*Ke)?

Exactly. The wonderful thing with open source is that you can perform a “whitebox” validation. With commercial software you can only perform a “blackbox” validation (rubbish in, rubbish out).

bear’s source code is available at

Inspecting the script `NCA.r` you find in line 284
`  VdFRef[j]<-Dose/(aucINF*ke)`
and in line 520
`  VdFTest[j]<-Dose/(aucINF*ke1)`

Hope that helps.

PS: The same formula is used in Phoenix/WinNonlin.
PPS: IMHO, the “volume of distribution” after an extravascular dose (if no i.v. dose was administered in the same study) bears [pun!] limited – if any – meaning. Since F is unknown, what’s the purpose of reporting V/F?

Cheers,
Helmut Schütz

The quality of responses received is directly proportional to the quality of the question asked. ☼
Science Quotes
yjlee168
Senior

Kaohsiung, Taiwan,
2017-11-02 23:59

@ CECIF
Posting: # 17966
Views: 2,400

## Vd/F calculation in bear

Hi CECIF,

» I have been using bear for bioequivalence calculation since version 6.2.

uhh... since version 6.2? bear? the current version is 2.8.3 (as shown in my signature panel).

» ... on exception of some of the AUC0inf and AUC0t mostly on the last sampling times which came out a little higher in my calculations (I can only suposse it is because of the decimal numbers).

What was your selection for trapezoidal method? `linear-up/log-down` (default method) or `all linear`? That could make a little differences.

» ... I have been trying different equations to get it, but I failed in every one.
» is it Vd/F= D*Ka/(I(Ka-Ke))? with ka from the residuals method or the wagner-nelson method?
» or is it simply Vd/F=D/(AUC0inf*Ke)?

Simply Vd/F=D/(AUC0inf*Ke). Again please check your selected trapezoidal method. It also affects Vd/F too as you can see. What about Cl/F?

Hope this message can help.

All the best,
---Yung-jin Lee
bear v2.8.3-2:- created by Hsin-ya Lee & Yung-jin Lee
Kaohsiung, Taiwan http://pkpd.kmu.edu.tw/bear
CECIF
Junior

Colombia,
2017-11-03 23:54

@ yjlee168
Posting: # 17968
Views: 2,305

## Vd/F calculation in bear

» uhh... since version 6.2? bear? the current version is 2.8.3 (as shown in my signature panel).

You are right, I meant 2.6.2. Sorry for the brain fart.

» What was your selection for trapezoidal method? `linear-up/log-down` (default method) or `all linear`? That could make a little differences.

My selection was default

Now That I looked for teh code as Helmut told me is all very clear. It seems however that both Vd/F and Cl/F are affected by the same misscalculation from AUC0inf. I will check with the other methods and tell you later.

Bioequivalence and Bioavailability Forum |  Admin contact
18,914 posts in 4,036 threads, 1,283 registered users;
online 16 (0 registered, 16 guests [including 14 identified bots]).

Statistics is, or should be, about scientific investigation
and how to do it better, but many statisticians believe
it is a branch of mathematics.    George E.P. Box

The BIOEQUIVALENCE / BIOAVAILABILITY FORUM is hosted by
Ing. Helmut Schütz