1 ivivc - A Tool for in vitro- in vivo Correlation Exploration with R Speaker: Hsin-ya Lee Advisors:...

1

ivivc ivivc - A Tool for - A Tool for in vitroin vitro--in vivo in vivo Correlation Correlation Exploration with Exploration with RR

Speaker: Hsin-ya LeeSpeaker: Hsin-ya LeeAdvisors: Pao-chu Wu,Advisors: Pao-chu Wu, Yung-jin Lee

College of Pharmacy, Kaohsiung Medical University, Kaohsiung, Taiwan (R.O.C)

2008/08/14

2

BackgroundBackground In vitro-in vivo correlation (IVIVC)

the correlation between in vitro drug dissolution and in vivo drug absorption

3

Purpose of IVIVC The optimization of formulations

may require changes in the composition, manufacturing process, equipment, and batch sizes.

In order to prove the validity of a new formulation, which is bioequivalent with a target formulation, a considerable amount of efforts is required to study bioequivalence (BE)/bioavailability(BA).

The main purpose of an IVIVC model to utilize in vitro dissolution profiles as a surrogate for

in vivo bioequivalence and to support biowaivers Data analysis of IVIVC attracts attention from the

pharmaceutical industry.

4

Purpose of our study

The purpose of this study is to develop an IVIVC tool (ivivc) in R.

ivivc in R is menu-driven package. The development of level A IVIVC model

5

Frameworks of IVIVC in RFrameworks of IVIVC in R

Input/Edit In Vivo Absorption Data: IV, Oral solution or IR drug

Input/Edit In Vitro Dissolution Data and In Vivo absorption Data: ER drug with Different Release Rates

Develop an IVIVC Model: Fitting IV, Oral solution or IR drug

6

Fitting IV, Oral solution or IR drug

PK parameters (kel and Vd) using PKfit Started with genetic algorithm (genoud is from

“rgenoud” package) fitting Nelder-Mead Simplex algorithm (optim) end with nls

7

Frameworks of IVIVC in RFrameworks of IVIVC in R

Input/Edit In Vivo Absorption Data: IV, Oral solution or IR drug

Develop an IVIVC Model: Fitting IV, Oral solution or IR drug

Input/Edit In Vitro Dissolution Data and In Vivo absorption Data: ER drug with Different Release Rates

Input/Edit In Vivo Absorption Data: IV, Oral solution or IR drug

Input/Edit In Vitro Dissolution Data and In Vivo absorption Data: ER drug with Different Release Rates

Develop an IVIVC Model: Model Dependent Method

ER drug with Different Release Rates

Model Dependent Method: deconvolution The observed fraction of the drug absorbed is based

on the Wagner-Nelson method

observed drug

plasma concentration

(conc.obs)

estimated fraction of the drug absorbed

(Fab)

9

Wagner-Nelson method

10

IVIVC model IVIVC model

fraction of the drug absorbed vs. the drug dissolved the predicted fraction of the drug absorbed is

calculated from the observed fraction of the drug dissolved.

α and β are the intercept and slope of the regression line, α and β are the intercept and slope of the regression line, respectively.respectively.

11

IVIVC modelIVIVC model

12

the predicted fraction of the drug absorbed is then convolved to the predicted drug plasma concentrations

Convolution

Gohel M. and et al. http://www.pharmainfo.net/reviews/simplified-mathematical-approach-back-calculation-wagner-nelson-method

predicted fraction of the drug absorbed

(PredFab)

predicted drug plasma concentration

(conc.pred)

13

Predicted drug plasma conc.Predicted drug plasma conc.

sciplot package

14

Frameworks of IVIVC in RFrameworks of IVIVC in R

Input/Edit In Vivo Absorption Data: IV, Oral solution or IR drug

Develop an IVIVC Model: Fitting IV, Oral solution or IR drug

Input/Edit In Vitro Dissolution Data and In Vivo absorption Data: ER drug with Different Release Rates

Develop an IVIVC Model: Model Dependent Method

Input/Edit In Vivo Absorption Data: IV, Oral solution or IR drug

Input/Edit In Vitro Dissolution Data and In Vivo absorption Data: ER drug with Different Release Rates

Develop an IVIVC Model: Model Dependent Method

Evaluate an IVIVC model: Prediction Error

15

Internal Validation of level A correlation

Predictability of a level A correlation estimating the percent prediction error (%PE)

between the observed and predicted drug plasma concentration profiles

pharmacokinetic parameters (Cmax, and the area under the curve from time zero to infinity, AUC∞).

16

Limitation and Future worksLimitation and Future works

LimitationLimitation Model dependent method

One-compartment model: One-compartment model: Wagner-Nelson method

Future worksFuture works Model dependent method

Two-compartment model: Loo-Riegelman method Model independent method

Numerical deconvolution Differential-equation based IVIVC model

17

AcknowledgmentAcknowledgment

Stephen D. Weigand (Departments of Biostatistics , Mayo Clinic Rochester, MN, USA): coding (by e-mail)

Henrique Dallazuanna (Curitiba-Paraná-Brasil): coding (by e-mail)

18

More informationMore information Reference

1997. Guidance for industry, extended release oral dosage forms: Development, evaluation, and application of in vitro/ in vivo correlations.

Dutta S, Qiu Y, Samara E, Cao G, Granneman GR. 2005. J Pharm Sci 94(9):1949-1956.

Gohel M. , Delvadia RR, Parikh DC, Zinzuwadia MM, Soni CD, Sarvaiya KG, Joshi R and Dabhi AS. Simplified Mathematical Approach for Back Calculation in Wagner-Nelson Method. http://www.pharmainfo.net/reviews/simplified-mathematical-approach-back-calculation-wagner-nelson-method

Email Yung-Jin Lee : : pkpd.taiwan@gmail.compkpd.taiwan@gmail.com Hsin-Ya Lee: hsinyalee@gmail.com

Web http://pkpd.kmu.edu.tw/ivivc/

19

Thanks for your attention!Thanks for your attention!