ivivc for R - A Tool for “in vitro-in vivo Correlation" Exploration with  (demo)

Created by Hsin-ya Lee (hsinyalee@gmail.com), and
Yung-jin Lee (mobilePK@gmail.com)
College of Pharmacy, Kaohsiung Medical University
Kaohsiung, Taiwan 807

Introduction
In vitro-in vivo correlation (IVIVC) is defined as the correlation between in-vitro drug dissolution profile and in vivo drug absorption profile.  The main purpose of conducting an IVIVC study is to utilize in vitro dissolution profiles as a surrogate for in vivo bioequivalence and to support biowaivers.  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/bioavailability.  Thus, data analysis of IVIVC attracts great attention from the pharmaceutical industry.  The purpose of this study is to develop an IVIVC tool (ivivc) running in R.

Methods
Development and validation are 2 critical steps in the evaluation of an IVIVC model.  In the first, the development of level A IVIVC model is usually estimated by a two-stage process.  (1) Deconvolution:  the observed fraction of the drug absorbed is estimated based on the Wagner-Nelson method.  IV, IR or oral solution was attempted as the reference.  Then, the pharmacokinetic parameters will be estimated using a nonlinear regression tool or obtained from literatures reported previously.  The IVIVC model is developed using the observed fraction of the drug absorbed and that of the drug dissolved.  Based on the IVIVC model, the predicted fraction of the drug absorbed is calculated from the observed fraction of the drug dissolved.  (2) Convolution: the predicted fraction of the drug absorbed is then convolved to the predicted plasma concentrations by using the convolution method.  In the second stage, evaluating the predictability of a level A correlation focuses on estimating the percent prediction error (%PE) between the observed and predicted plasma concentration profiles, such as the difference in pharmacokinetic parameters (C_max, and the area under the curve from time zero to infinity, AUC_∞).
    This package was performed together with several R packages.  "PKfit" package is used to fit the reference data such as IV, IR or oral solution by fitting an one-compartment model or a two-compartment model.  Moreover, we use "reshape" package to rearrange some data format in "ivivc" package.  Then, "sciplot" package assists us to plot the mean values and standard error (or other summary statistics) of observed or predicted concentrations.