Insights on Prospects of Prediction of Drug Bioavailability from in Vitro Models

Introduction

Following the oral administration of a solid dosage form, such as an immediate release tablet, the dosage form goes through the process of disintegration followed by dissolution/solubilization of the active pharmaceutical ingredient (API). The drug in solution within the gastrointestinal tract (GIT) is then absorbed into the systemic circulation. The rate and extent at which the drug molecules appear into the systemic circulation in an intact form is termed bioavailability. It is noteworthy that the bioavailability of drugs that fall into the Biopharmaceutical Classification System (BCS) Class II or IV are poorly water- soluble, can be limited by their dissolution rate and/or solubility in the GIT. Thus, studies from our laboratories and by several other authors have shown that the improvement of the solubility of poorly water-soluble drugs can be achieved through various approaches including crystal form modification, micronization, use of surfactants, salt formation, cyclodextrin complexation and complexation with other agents, solid dispersion, nanoparticles formulations and others [1, 2, 3]. Although dissolution testing devices and processes as described in the various pharmacopeia have been in use worldwide for decades to assess the performance of solid oral dosage forms, these devices do not reflect the variable GIT conditions and the dynamic processes associated with the drug passage through the GIT. Rather than going through the long process of determining the bioavailability of a new solid oral dosage formulation in humans, different biorelevant media or predictive dissolution models aimed at being able to predict the in vivo performance of the drug have been developed [4, 5, 6]. With availability of a variety of biorelevant dissolution media, an investigator must be able to select the most appropriate media that can predict the performance of the drug in human body. Replacing bioavailability studies carried out in healthy human volunteers with suitable in- vitro substitutes has the potential of saving enormous costs and time involved in the biostudies.

Biorelevant Media and its Selection

Over the years, different biorelevant dissolution media have been developed which differ in compositions, properties, buffer species and capacities, while some contain bile salts, lipids, surfactants and additional compounds which enhance drug solubility. Table 1 thus, fluid composition and hydrodynamics in the GIT are important considerations in the development of biorelevant dissolution media.

NB:FaSSIF = Fasted State Simulated Intestinal Fluid FeSSIF = Fed State Simulated Intestinal Fluid FaSSGF = Fasted State Simulated Gastric Fluid FeSSGF = Fed State Simulated Gastric Fluid Table 1: Sample Compositions of Some Biorelevant Dissolution Media [4, 5].

Appropriate selection of a biorelevant dissolution media requires • Knowledge of the physiochemical properties of the API so that the media should contain constituents that promote solubility and dissolution rate of the drug. Generally BCS class II (Low solubility and high permeability) and BCS Class IV (Low solubility and low permeability) drugs require special biorelevant media for improvement of their solubility while the dissolution of BCS class

1 drugs (High solubility and high permeability) do not require special biorelevant media. To aid in the selection of appropriate biorelevant dissolution media and parameters, these BCS II and IV drug classes have been further sub-classified into acidic, basic, and neutral compounds.

• An important consideration of the need to create sink conditions in various dissolution media so as to prevent drug precipitation. Various techniques including altering pH, use of surfactants and co-solvents like alcohols, are widely used to enhance drug solubility thereby ensuring sink condition.

• That all these media are made to simulate the gastric or intestinal fluid in the fed and fasted states [6]. These are Fasted State Simulated Intestinal Fluid (FaSSIF), Fed State Simulated Intestinal Fluid (FeSSIF), Fasted State Simulated Gastric Fluid (FaSSGF) and Fed State Simulated Gastric Fluid (FeSSGF). The intraluminal compositions in the postprandial stomach and upper small intestine can also be made differentially to simulate “early”, “middle,” and “late” phases of digestion.

These media can be streamlined by grouping them into categories:

Prediction of Bioavailability

In recent years, several biorelevant models have been developed aimed at predicting the bioavailability of oral drug products. Unlike the standardized dissolution methods used in quality control (QC), these biorelevant tools have often been produced by research groups and commercial companies and they are deployed for few specific formulations. The biggest challenge in predicting the in vivo performance of a drug from in vitro dissolution tests is the simulation of a dynamic in-vivo environment in an in-vitro test. Although these biorelevant media can reflect the biochemical aspect of the GIT conditions, the media alone can hardly simulate the mechanical aspects. Thus, bio-relevant dissolution methods have been combined with different devices to design several predictive models to replicate the physiological environment which more accurately predict in vivo drug availability [7, 8].

Prediction of bioavailability from dissolution data involves the establishment of an in vitro-in vivo correlation (IVIVC) between an in vitro measurement such as dissolution rate data and the in vivo performance of the drug formulation such as the bioavailability parameters (eg maximum concentration (Cmax), time for maximum concentration (Tmax) and the area under the plasma drug concentration-time curve (AUC) ). For example, from the in vitro data, parameters such as time to have 10, 50 or 90 % of drug dissolved, dissolution rate, or dissolution efficiency can be related to the in vivo data such as Cmax, Tmax, AUC, and time to have 10, 50, 90 % of drug absorbed [5].The FDA guidance provides that to develop and validate an IVIVC model, upto three different formulations should be studied both in vitro and in vivo [6]. The drug products should contain the same amount of API but the drug release-rates from the product may vary due to differences in the nature and amount of the excipients. If a good correlation is established between the in vitro data and in vivo measurements, the drug formulation can be exempted from a bioequivalence study [6].

Where the use of biorelevant dissolution media alone cannot produce an adequate in vitro-in vivo correlation, the use of an in vivo predictive dissolution model becomes necessary. These models Table 2 include dissolution- permeation simulating systems, biphasic dissolution system, dynamic gastric model and artificial stomach-duodenum models [8, 9]. In the recent years, the prediction of in vivo performance has also been achieved using physiologically based absorption modeling developed with dissolution data input from biorelevant media [8, 9]. Thus, mathematical and computational tools have been hugely exploited to correlate dissolution testing with in vivo data.

Conclusion

Most of the dissolution media described in international pharmacopoeia are meant for drug quality control purposes and are not adequate for predicting in vivo performance. To achieve this prediction, various Biorelevant dissolution media have been developed which simulate physiological processes of in vivo drug availability. With advancement in technology and application of computational tools, significant progress has been made in developing different in vivo predictive models that improve the possibility of prediction of drug bioavailability from in vitro data with resultant lowering of overall cost of drug product development.