Skip to main content
Home / Resources / Blog / Analyzing Complex In Vitro Experiments: It’s Not as Hard as You Think

Analyzing Complex In Vitro Experiments: It’s Not as Hard as You Think

Analyzing in vitro experiments can be challenging and time consuming. Yet, crucial decisions depend on accurate data analysis and interpretation early in development. Unfortunately, most lab-based scientists lack access to state-of-the-art models for analyzing in vitro data. A new tool enables the analysis of data generated from complexin vitro studies. These studies include assays using whole cells, tissue samples and solid dosage forms.

The need for a user-friendly in vitro data analysis tool

Many PK/PD software tools are currently available. But, most of them are general in application and not specifically designed for in vitro data analysis. Some of them are not able to fit complex models. Or they are mainly designed for pre-clinical/clinical PK/PD data analysis, not for in vitro experiments. Or they require users to have considerable scripting and statistical skills. None of the available tools meet the needs of in vitro pharmacological research.

A powerful, yet easy to use tool

The Simcyp In Vitro (Data) Analysis (SIVA) Toolkit provides model-based data analysis of drug metabolism, transport and dissolution/solubility. It uses powerful optimization algorithms in a sophisticated statistical environment. SIVA includes a pre-defined library of state of the art models for various in vitro assays and can handle complex protocols such as buffer changes within an experiment. The key components of SIVA include:

  • Permeability and Transport: Models for 2-compartment cellular uptake, 3&5- compartment permeability, and sandwich cultured hepatocyte assays
  • Enzyme inhibition: Mechanistic time-dependent inhibition models
  • Metabolic intrinsic clearance: Substrate depletion, metabolite formation
  • Solid dosage form: In vitro solubility and dissolution models e.g. USP II apparatus
  • Additional tools:
    • IVIVE clearance (rat, dog, mouse, man)
    • In silico prediction of ADME (absorption, distribution, metabolism, excretion) parameters

There are also model fitting and statistical analysis tools to assess quality of fit that support all the modules of SIVA in a user-friendly manner. The graphical user interface (GUI) has an intuitive layout. Users can cut and paste data from Microsoft® Excel® into SIVA for analysis. After analysis, users can export tables and graphs back into Excel spreadsheets.

Furthermore, SIVA provides greater statistical rigor. Fitted parameters can be assessed for “goodness of fit.” After parameter fitting, SIVA generates an array of diagnostic plots. This approach ensures that scientists can use SIVA without previous modeling or statistical experience.

Analysis of drug permeability and transport assays

SIVA provides pre-defined library of mechanistic models to incorporate in vitro data generated in various cellular transport/permeability assays to estimate specific transporter kinetics describing uptake, metabolism and efflux, including the interplay of these multiple dynamic processes. These studies include three different types of assays.

Cellular drug uptake studies

These assays characterize the uptake of drugs into hepatocytes or transfected cell lines. After adding the drug to the suspended or plated cells, the cells are lysed at one or more time points. Then, the lysates are analyzed for drug concentration using LC-MS/MS or scintillation counting.

Limitations of conventional modeling approaches

Conventional modeling approaches are commonly used to estimate transporter kinetics of drugs. The passive transport rate is estimated with the following methods:

  • Using cells that do not express any drug transporters
  • Using pharmacological inhibitors of transporters
  • Using the PAMPA assay

However, these approaches have several important limitations.

  • Conventional models lack time dependency. They use nominal concentrations when media and cell drug concentrations change with time.
  • Conventional modeling assumes that sink conditions are present. Thus, sink conditions for the drug must also be maintained experimentally. This can be difficult as small cell volumes can rapidly lead to high intracellular drug concentrations.
  • Specific transporter inhibitors are not available.

This non-integrated and non-physiological approach results in reduced precision of pharmacokinetic parameter estimates.The mechanistic models available in SIVA could help overcome the caveats of conventional modeling approaches.

Bidirectional transport studies

These assays characterize drug permeability and transport (efflux). They use either cells lines (e.g. Caco-2, MDCK II, LLC-PK1) or artificial membranes (PAMPA). The drug is added to the apical buffer. At various time points, the apical and basolateral buffer are sampled and analyzed for drug concentration. This assay can be modeled as 3- or 5 compartment models. The 3-compartment model allows assessment of transport rates and permeability of the drug between the apical and basolateral media and the cells. The 5-compartmental model allows for a time-lag as the drug crosses the cell’s apical and basolateral membranes.

Sandwich cultured human hepatocyte assays

These assays characterize a drug’s hepatic permeability and transport. Fresh or cryopreserved hepatocytes are seeded onto plates and covered with matrigel or collagen. Compared to immortalized cell lines, primary hepatocytes have a more physiological cell structure. They express transporters along apical/basolateral membranes and form bile networks. After drug addition, the cells are lysed and analyzed at various time points. The formation of tight junctions between hepatocytes is calcium-dependent. Thus, hepatobiliary drug transport can be examined by exchanging calcium-containing media with calcium-free media. SIVA allows simultaneous fitting of experiments with and without transport inhibitors. Likewise, SIVA can simultaneously fit experiments conducted with wash steps and buffer changes.

Analysis of drug solubility and dissolution assays

  • Solid dosage form dissolution studies: SIVA provides a mechanistic framework for modeling in vitrodissolution profiles permitting the estimation of the parameters required for in vivo simulation of the dissolution of solid dosage forms. These include a fluid dynamics model to account for stirring rate and a particle surface pH model to account for the impact of buffer species on dissolution of ionizable drugs. Library files defining the composition of widely used media such as FaSSIF, FeSSIF and simpler buffer systems are also provided.
  • Drug solubility studies: SIVA also provides a framework within which to model aqueous as well as bio-relevant solubility permitting estimation of pKa(s), intrinsic solubility, bile or surfactant micelle:water partition coefficients and related properties of active pharmaceutical ingredients.

Using in vitro data to inform later stages of drug development

The SIVA Toolkit platform is designed to analyze complex in vitro data. It calculates parameter values that can subsequently be used in in vitro-in vivo extrapolation (IVIVE) paradigms, that are necessary to successfully predictin vivo behavior of drugs in physiologically-based pharmacokinetic models.

The SIVA Toolkit can also be used to provide reliable in vitro data for decision trees in regulatory guidelines, mechanistic-static or static models. It is a standalone product and can be purchased independently of Certara’sSimcyp Simulator. It will be useful for researchers analyzing experimental data in the pharmaceutical industry, universities and contract research organizations.

Learn more about the SIVA Toolkit.

BLOG_Analyzing-Complex-In-Vitro-Experiments-It’s-Not-as-Hard-as-You-Think_2

Examples of in vitro cellular assays and dissolution tests for which mathematical models are available within the SIVA Toolkit


Learn more about how SIVA can help you!

My colleagues, Nikunjkumar Patel and Howard Burt, gave a webinar where they presented several case studies that illustrate how SIVA can help you get the most out of your in vitro data!

About the author

By: Krishna Machavaram