The role of the Biopharmaceutics Classification System and In Vitro-In Vivo Correlation in the approval of drug products

March 23-25, 1998 Frankfurt am Main, Germany.

AIM OF THE MEETING

This 3-day conference aimed to unite regulators (including several delegates from the FDA), members of the pharmaceutical industry and academics in an appraisal of the impact of the Biopharmaceutics Classification System (BCS) on the drug approval process.

Biopharmaceutics Classification System (BCS)

This system1, pioneered by Professor Gordon Amidon of the University of Michigan, classifies all drug molecules into one of four classes based on their solubility and permeability through the intestinal cell layer. The classes are as follows:

• Class I-high solubility/high permeability,
• Class II-low solubility/high permeability,
• Class III-high solubility/low permeability,
• Class IV-low solubility/low permeability.

The interest in this classification system stems largely from its application in early drug development and then in the management of product change through its life-cycle. In early drug development knowledge of the class of a particular drug is an important factor influencing the decision to continue or stop its development. Obviously a low solubility/low permeability drug will never be presented as an orally administered product and will probably encounter serious formulation difficulties. A company wishing to produce an oral dosage form will wish to limit its development to those molecules that are high permeability. Increasingly, these considerations are incorporated from the very earliest phases, with the concept of property-based design being used in combinatorial chemistry to target production of compounds showing optimal properties. Subsequent in vitro models are then applied to evaluate the drug in early development, the accuracy and predictive capacity of these models being discussed during the conference. In the future, the experts predict an increasing shift from the in vitro cell-line screening models to absorption studies occurring predominately ‘in computro’, as computer models are developed to link molecular motives to absorption properties.

REGULATORY INTEREST IN THE BCS

The large regulatory interest in the BCS comes from its combination with in vitro – in vivo correlation (IVIVC) to manage the issues of product quality throughout its life-cycle. For example, orally administered class II drugs are expected to show a correlation between observed dissolution rate in vitro and rate and extent of absorption in vivo. The correlation is expected as it is the rate of release from the product that is the controlling factor in the absorptive process, the intestinal barrier not being rate-limiting. A reliable IVIVC can therefore be used to predict in vivo performance based on in vitro results. This tool would thus permit an assessment of the effect of change on in vivo product performance based on in vitro tests. This change is incurred for all products at several stages, for example: scale-up, change of manufacturing site, excipients, suppliers etc. At its most extreme, this could be a change of manufacturer, i.e. a generic product, and raises the ultimate possibility of approval of certain generic drugs using comparative in vitro results only. For class I drugs, which dissolve rapidly and are highly permeable, bioavailability problems are not to be expected and dissolution testing could suffice to assure bioequivalence in the circumstances discussed above.

‘BIO-WAIVERS’ BASED ON THE BCS

The opportunities for cost-reduction through the limitation of costly in vivo studies make this a current and widely-debated topic. The BCS Working Group of the FDA will propose in a subsequent draft guidance that a ‘bio-waiver’ (i.e. assessment of bioequivalence using in vitro testing only) could be considered for class I drugs, provided they meet a list of criteria (including not being narrow therapeutic index drugs). A bio-waiver would also be possible for certain class II drugs on a case-by-case basis. This represents a continuing shift in FDA policy away from the systematic requirement for bioequivalence testing. A similar proposal is under consideration by a European CPMP group of experts but consensus on this point has not been reached. Some EU members argue that bioequivalence studies will always be required, even for class I drugs. This topic and other bioavailability/bioequivalence issues have not been treated under the ICH process as this was considered not to be the easiest subject to discuss with the pharmaceutical industry given the close links with the hot topic of generic substitution. There have however been several conferences (including this one) which help to bring some consensus between the international regulatory communities.

CONCLUSIONS

One take-home message was that in drug development the move is away from the old paradigm of identifying development candidates on the basis of in vitro and in vivo pharmacological assessment and subsequently encountering the drug delivery and pharmacokinetic issues in later phases. This approach is being replaced by an increasingly accurate evaluation of drug delivery and pharmacokinetic properties in vitro and in computro before making the development recommendation.

On the regulatory side, the future goal continues to be harmonisation. Dr Roger Williams of the FDA expressed the wish for the mutual recognition of reviews between the FDA and the EMEA. The FDA will continue to progress their policy of issuing guidelines for industry, with future guidance on bioavailability and locally acting drug products expected. Within these guidances will be the concept of using the BCS to identify drugs for which in vivo bioequivalence studies are not required. This approach includes collaboration with academia and discussion with the pharmaceutical industry and bodes well for scientifically driven change to regulatory policy.

REFERENCES

1. Amidon G. L., Lennernas H, Shah V P, and Crison J R. ‘A theoretical basis for a biopharmaceutic drug classification: the correlation of in vitro drug product dissolution and in vivo bioavailability’. Pharmaceutical Research, 12, 413-420.