How to Revamp Your Approach to Conventional IVIVC Models

How to Revamp Your Approach to Conventional IVIVC Models

An IVIVC (in vitro-in vivo correlation) is a predictive mathematical model describing the relationship between the in vitro properties of a dosage form and the in vivo responses. Drug developers frequently find IVIVCs useful for a number of reasons:

  • They can use dissolution tests as a surrogate for human bioequivalence (BE) studies.
  • They can support and/or validate the use of dissolution methods and specifications.
  • They can assist in quality control during manufacturing and selecting appropriate formulations.

Conventional IVIVC uses deconvolution methods such as Wagner Nelson (WN) and numerical deconvolution to estimate the rate of input of drug into the systemic circulation from observed plasma drug concentrations of the oral formulation with the use of IV bolus data as the unit impulse response (UIR). To help our customers generate these models, we offer the IVIVC Toolkit for Phoenix WinNonlin.

Clearly, there have been many instances of successful use of conventional IVIVC models, and numerical deconvolution remains the gold standard. However, there are also cases where conventional models have performed very badly. One reason for this may be that conventional methods do not separately consider the multiple mechanisms that determine a drug’s in vivo input rate— transit time, gut wall permeability, gut wall metabolism, and hepatic first-pass metabolism— from dissolution rate. In this blog post, I’ll discuss our work to develop physiologically-based pharmacokinetic (PBPK), mechanistic IVIVC and why you should consider using this approach when designing and evaluating the performance and safety of new drug formulations.

What is mechanistic, physiologically-based IVIVC?

Mechanistic, physiologically-based (PB) deconvolution models estimate in vivo drug dissolution profiles while separately accounting for permeation, GI transit and first pass elimination. This can allow more robust and transparent IVIVCs to be established against in vitro dissolution, and not the rate of systemic input. I’ve found the Simcyp Simulator‘s Advanced Dissolution Absorption and Metabolism (ADAM) model to a very useful tool for this approach. If you wanted to establish a PB-IVIVC in the Simcyp Simulator for a target formulation of a drug, you would follow these steps:

  1. Get the observed plasma concentrations (Cp), and in vitro dissolution profiles of the target formulation and higher and lower release formulations, if available, and reference Cp data (UIR) for the drug.
  2. Use reference oral formulation data for the drug to estimate disposition parameters and gut permeability.
  3. For each controlled release formulation, deconvolute the in vivo dissolution profiles from the corresponding Cp profile using the IVIVC module of the Simcyp Simulator.
  4. Establish and validate (internally and externally) a level A IVIVC between deconvoluted in vivo and in vitro dissolution profiles. Both two stage (deconvolution of in vivo dissolution profile followed by establishment of IVIVC) and single stage (a convolution based approach with a predefined IVIVC function) approaches are available.

If you wanted to study population variability in the PK of the designed formulation after a single dose and at steady state, you could use the Simcyp Simulator to perform clinical trial simulations in virtual populations. The Simcyp modeling platform allows users to define study populations (Caucasian, Chinese, Japanese, obese, diseased, etc.).

Some drugs are metabolized by CYPs that exhibit wide phenotypic variations. PB-IVIVCs allow the study of the PK of a formulation in subjects with various phenotypes of a CYP (extensive metabolizer (EM); poor metabolizer (PM), and ultra-rapid metabolizer (UM)). The retrograde calculator within Simcyp allows estimation from overall plasma clearance (CL) to intrinsic clearance (CLint) by particular enzymes if the fraction metabolized by that enzyme (fm) is known. This feature enables the study of the PK of the formulation in subjects with various drug metabolizing phenotypes.

What are the benefits of this approach?

Mechanistic deconvolution approaches, exemplified by the ADAM model, are potentially useful tools in designing and evaluating the performance and safety of new drug formulations. In some cases, the PB-IVIVC approach can improve the predictive performance of the IVIVC model and result in a simpler IVIVC model compared to conventional IVIVC models. Further validation using a range of drugs with different biopharmaceutical properties is needed to improve confidence in and increase the awareness and acceptance of mechanistic IVIVC approaches in formulation design and optimization.

Interested in learning more about physiologically based IVIVC?

Do you think that PB-IVIVC might be a useful approach for assessing different drug formulations? If so, I’ve provided some resources that provide more information about this method:

Learn about how the Simcyp Simulator is reshaping the approach the toxicological risk assessment

Clearly, PBPK models are widely used in drug development. Our colleagues performing toxicological risk assessment are also finding this approach useful for estimating chemical-specific PK variability across multiple life-stages and sub-populations. Dr. Barbara Wetmore of the Hamner Institute gave a webinar where she discussed how she led a team of Hamner, EPA, and Certara researchers to develop a new toxicity paradigm that considers the variability in responses to chemicals due to pharmacokinetic (PK) or pharmacodynamic (PD) differences among life-stages and sub-populations. I hope that you’ll watch it and let me know what you think in the comments below.

* This post received editorial support from Suzanne Minton, manager of scientific communications at Certara.

Nikunjkumar Patel

About the Author

Nikunjkumar Patel

More Posts by This Author

Nikunj Patel is a senior research scientist in Certara’s modelling and simulations group where he is leading oral and dermal absorption projects and is a member of the Cardiac Safety Simulator development team. He joined Certara in August 2011 and led the development of the physiologically based IVIVC (PB-IVIVC) module of the Simcyp Simulator and the Pharmaceutics module of SIVA (Simcyp In Vitro (data) Analysis) platform. Before joining Certara, he spent three years at the life science innovation labs of Tata Consultancy Services as a research scientist mainly working on pharmacokinetic/pharmacodynamic modelling and QSAR development for various ADMET properties. During his graduate studies, he used computer aided drug design (CADD) and molecular modelling to identify safe and potent novel anti-diabetic ligands.