A. S. Hussain*, L.J. Lesko, K. Y. Lo, V.P. Shah, D. Volpe, and R. L. Williams
Office of Pharmaceutical Science
Center for Drug Evaluation and Research
Food and Drug Administration
Rockville, MD

*e-mail address for correspondence: hussaina@cder.fda.gov

Introduction
The Biopharmaceutics Classification System (BCS) is a drug development tool that allows estimation of the contributions of three major factors, dissolution, solubility, and intestinal permeability, that effect oral drug absorption from immediate release (IR) solid oral products (1). It was first introduced into regulatory decision-making process in the guidance document on Immediate Release Solid Oral Dosage Forms: Scale-Up and Post-Approval Changes (2).
A recent draft guidance document entitled "Waiver of In Vivo Bioavailability and Bioequivalence Studies for Immediate Release Solid Oral Dosage Forms Containing Certain Active Moieties/Active Ingredients Based on a Biopharmaceutics Classification System" proposes to further expand the regulatory applications of BCS and also recommends methods for classifying drugs and IR drug products (3). This article provides an overview of this draft guidance document.

In Vitro Dissolution Tests and Bioequivalence Assessment:
Need for BCS
The most widely applied dissolution test methods for IR products are based on the USP's apparatus I (basket) or II (paddle) at agitation rates of 100 and 50 rpm, respectively. Typically 900 ml of an aqueous dissolution media are used. Historical data suggests that in vitro dissolution methods are generally sensitive to formulation factors that affect drug dissolution process and often it is observed that two products that exhibit some dissolution differences, in vitro, may provide similar drug concentration time profiles in blood. These observations suggest that for many IR products, dissolution in vivo may not be the rate-limiting process. The observed variability in blood level profiles may be due to variability in the physiological processes and not due to minor dissolution differences in products being tested. In other instances significant differences have been observed in the blood level profiles of two products that meet the "one-point" acceptance criteria outlined in an application or compendial dissolution specification. On rare occasions an inverse in vitro - in vivo relationship, i.e., higher peak drug concentration in blood for a product that exhibits a relatively slow rate of dissolution in vitro, in a particular media, compared with another product, have also been observed. Such examples of "failure" of in vitro dissolution tests to signal bio-in-equivalence have hindered the use of dissolution tests for assessing bioequivalence between two pharmaceutically equivalent products. Comprehensive research studies designed to elucidate mechanistic reasons for such failures are generally not available in the public domain. Also, data from such failed studies are generally not submitted to the Agency.

Possible causes for such differences may include; 1) inappropriate specification (dissolution test conditions, primarily media composition, and acceptance criteria), (2) presence of an excipient that may alter drug absorption, and (3) other reasons (for example, statistical type II error).

Experience gained through development of traditional in vitro - in vivo correlations (e.g., Level A, B, or C correlations) for IR products containing poorly soluble drugs and for extended release products suggests a significant degree of formulation dependency or specificity associated with such correlations. Therefore, for products that are likely to exhibit slow in vivo dissolution, in vitro - in vivo correlations need to be established and their predictive performance verified through experimentation. Future research in this area should address how to a priori identify dissolution test conditions that yield robust in vitro - in vivo correlations that are applicable to a wide range of formulations.

If the regulatory utility of dissolution tests for IR products are to be expanded, their reliability must be improved. For IR products this may be achieved by considering the mechanistic relationships between drug dissolution, physico-chemical characteristics of drugs, gastrointestinal physiology and absorption or permeation processes. To this effect BCS provides, with minimal reliance on in vivo pharmacokinetic data, a rational mechanistic frame work for developing reliable dissolution tests for assessing bioequivalence. The BCS also provides a means for identifying when dissolution in vivo is likely or not likely to be rate- limiting and allows for managing risks associated with reliance on in vitro dissolution for bioequivalence assessment.

The Biopharmaceutics Classification System: Class Boundaries
In the BCS, a drug is classified as belonging to 1) high or low solubility class, 2) high or low permeability class, and 3) an IR dosage form is categorized as belonging to a rapid or slow dissolving class. The solubility class boundary is based on the highest dose strength of an IR product that is the subject of a bioequivalence assessment. A drug substance is considered highly soluble when the highest dose strength is soluble in 250 ml or less of aqueous media over the pH range of 1-8. The volume estimate of 250 ml is derived from typical bioequivalence study protocols that prescribe administration of a drug product to fasting human volunteers with a glass (about 8 ounces) of water. This boundary value is a reflection of the minimum fluid volume anticipated in the stomach at the time of drug administration during a typical fasting bioequivalence study.

The permeability class boundary is intended to identify drugs that exhibit complete absorption such as Class I drugs. When administered as a solution, highly permeable drugs would exhibit complete absorption. Rapidly dissolving IR products of a highly soluble and highly permeable drugs are expected to exhibit complete absorption. The high permeability class membership may be documented when: 1) the extent of absorption in humans is greater than 90% (of an administered dose or that of an intravenous reference dose) and the drug is determined to be stable in the gastrointestinal tract, or 2) intestinal permeability is determined to be high compared with selected reference compounds in well characterized experimental methods such as, in vivo (e.g., intestinal perfusion in humans), in situ (e.g., intestinal perfusion in animals), or in vitro (e.g., Caco-2 cell cultures) tests and the drug is determined to be stable in the gastrointestinal tract.

The four BCS drug classes with some examples are shown in Figure 1. The permeability values depicted in this figure were obtained from jejunal perfusion studies in humans performed at the Universities of Uppsala (by Professor H. Lennernas) and University of Michigan (by Professor Gordon Amidon). The solubility values were determined at the FDA's Division of Product Quality Research.

The dissolution class boundary is based on the in vitro dissolution rate of an IR dosage form under specified test conditions and is intended to demonstrate rapid in vivo dissolution in relation to the average rate of gastric emptying in humans under fasting conditions. Historical data (literature and in-house) suggests that rapid (gastric) dissolution will generally occur in normal subjects when products exhibit near complete (80% or higher) dissolution, in vitro, in 15 - 30 minutes, in the usual volume of 900 ml (or less) of suitable media (e.g., 0.1 N HCl) using the USP apparatus I or II at 100 and 50 rpm, respectively (for example, see reference 4). These observations are consistent with the average gastric emptying time (t50%) in human of about 15 minutes (5). It is important to note that, in the fasting state, pH of gastric juices and rate of gastric emptying is highly variable. This variability may result in a fraction of an administered dose dissolving in less acidic environment compared to 0.1 N HCl. Therefore, dissolution tests in only one acidic medium may not provide sufficient discriminatory power to distinguish between two bioinequivalent products. Also, excipients used in a product may be either acids, bases, or salts, and therefore be sensitive to media pH. Presence of such excipients could alter both dosage form and granule disintegration /dissolution characteristics. Therefore, an IR drug product may be considered rapidly dissolving when not less than 85% of the label amount of the drug substance dissolves within 30 minutes using the USP apparatus I at 100 rpm (or apparatus II at 50 rpm) in a volume of 900 ml, or less, in each of the following media: (1) acidic media, such as 0.1 N HCl or Simulated Gastric Fluid USP without enzymes; (2) a pH 4.5 buffer; and (3) a pH 6.8 buffer or Simulated Intestinal Fluid USP without enzymes.
Figure 2, below, summarizes the key objectives of the BCS class boundaries.

Biowaivers Based on BCS
Considering the uncertainties associated with in vitro dissolution tests, the proposed draft guidance recommends biowaivers only for rapidly dissolving products of highly soluble and highly permeable drugs, that are not considered, by the FDA, to be "Narrow Therapeutic Index Drugs." The criteria for defining the therapeutic index of a drug is currently under consideration at the FDA.
It is proposed that BCS based biowaivers apply for situations during both pre- (IND/NDA and ANDA) and post approval phases. Figure 3 depicts these applications in the drug development process.
In the current proposal the following criteria are recommended for justifying the request for a waiver of in vivo biostudies

1. The drug substance should be highly soluble and highly permeable, defined as a Class I drug above.
2. An IR drug product should be rapidly dissolving, as defined above.
3. For waivers of an in vivo relative bioavailability study, dissolution should be greater than 85% in 30 minutes in the three recommended dissolution media (acidic media, such as 0.1 N HCl or Simulated Gastric Fluid USP without enzymes, a pH 4.5 buffer; and a pH 6.8 buffer or Simulated Intestinal Fluid USP without enzymes). For waivers of in vivo bioequivalence, test and reference products should exhibit similar dissolution profiles under the dissolution test conditions defined for rapidly dissolving products.
Two dissolution profiles may be considered similar when compared using the f2 metric (f2 > 50) as described in the guidance for industry on dissolution testing (6). When both the test and the reference products dissolve 85% or more of the label amount in < 15 minutes, in all three dissolution media recommended above, a profile comparison is unnecessary.
4. The drug should not be a narrow therapeutic index drug. This limitation is expected to be applied primarily to NDA and ANDA bioequivalence studies after approval, as well as bioequivalence studies submitted in an ANDA, recognizing that during the IND period an investigational drug may not be clearly identified as a narrow therapeutic index drug.
5. Excipients used in the dosage form should have been used previously in FDA approved IR solid dosage forms. The quantity of excipients in the IR product should be consistent with their intended function. Large quantities of certain excipients, such as surfactants (e.g., sodium lauryl sulfate) or osmotic ingredients (e.g., sorbitol) may be problematic.
6. All other application commitments should be met.

Summary
For the last two decades bioequivalence assessments have predominantly been based on in vivo pharmacokinetic measures of rate and extent of absorption. This "one size fits all" approach may not fully consider the underlying mechanisms of drug release and absorption and could therefore impose unnecessary in vivo tests under certain situations. Regulatory application of BCS aids in the development of meaningful dissolution test specifications and for identifying immediate release solid oral products for which in vivo bioequivalence tests may not always be necessary.

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. Pharm. Res. 12: 413-420 (1995).
2. Guidance for industry, Immediate Release Solid Oral Dosage Forms: Scale-Up and Post-Approval Changes, November 1995, CDER/FDA.#
3. Draft Guidance for Industry, Waiver of In Vivo Bioavailability and Bioequivalence Studies for Immediate Release Solid Oral Dosage Forms Containing Certain Active Moieties/Active Ingredients Based on a Biopharmaceutics Classification System, February 1999, CDER/ FDA.#
4. Tammara, V. K., Hossain, M., Malinowski, H., and Hussain, A. S. Evaluation of the Proposed Biopharmaceutics Classification System (BCS) Class Boundaries: A Survey of Recent Neuropharmacological Drugs. AAPS Annual Meeting, Boston, MA, November 1997.
5. Kaus, L. C., Gillespie, W. R., Hussain, A. S., and Amidon, G. A. The effect of in vivo dissolution, gastric emptying rate, and intestinal transit time on the peak concentration and area-under-the-curve of drugs with different gastrointestinal permeabilities. Pharm. Res. 16: 272-280 (1999).
6. Guidance for Industry, Dissolution Testing of Immediate Release Solid Oral Dosage Forms, FDA, CDER, August 1997, CDER/FDA.#
# These documents may be obtained from the CDER's Internet website:
http://www.fda.gov/cder/regguide.htm

Acknowledgments
The information presented here reflects the views of the BCS Working Group and not FDA policy at this time. Contributions of former working group members (Lydia Kaus and Ram Mhatre), consultants (Gordon Amidon, Hans Lennernas, and Tom Tozer), and members of an FDA Expert Panel are also acknowledged.