In the late 1980s and early 1990s, a small group of influential and forward-thinking pharmaceutical scientists posited a somewhat radical hypothesis: the extent of absorption of an orally administered drug can be predicted by knowing 1) the intrinsic permeability of the active pharmaceutical ingredient (API) across the intestinal mucosa and 2) the concentration of the drug at the site of absorption. For well-behaved (highly permeable and highly soluble) drugs, assuming that the dose form dissolves rapidly (i.e., the drug product is an immediate-release formulation), the fraction absorbed (the amount of drug that actually makes it across the gut wall) should be consistent regardless of the composition of the formulated product. Thus —and this is the radical part —one ought to be able to accurately determine, from two intrinsic properties of the API (permeability and solubility) and one property of the drug product (dissolution rate), a human outcome.1 The beauty of the system is that it is all performed using in vitro test systems. In layman’s terms, three benchtop experiments using petri dishes, dissolution baths, test tubes, and beakers can eliminate the need for human testing. This is the basic premise underlying the Biopharmaceutics Classification System (BCS).
In 2000, the U.S. Food and Drug Administration (FDA) embarked on a bold initiative and went out on a limb by issuing a regulatory guidance on the BCS,2 thereby making it part of the drug approval process. Testing specifications were established for “high” permeability and “high” solubility characteristics of the raw API, namely:
- In a validated in vitro permeability model, a test API must be more permeable than a co-dosed drug whose in vivo human fraction absorbed is known to be at least 90%. The permeability model must be able to clearly distinguish high-permeability and low-permeability compounds. Note that this is a relative, not absolute, definition of high permeability. Alternatively, one can demonstrate in humans, using a suitably powered statistical design, that at least 90% of a drug is absorbed after oral administration, which is not a simple matter, due to intrinsic variability between subjects and the need to quantify the area under the concentration vs. time curve in plasma, urine, and/or feces for the parent compound and metabolites in order to account for everything that was absorbed and subsequently retained, cleared, and/or metabolized.
- The highest dose strength of the drug must be soluble in 250 mL or less of aqueous medium at all pH values between 1 and 7.5, inclusive. That volume (250 mL) is approximately equal to 8 oz., based on the premise that a patient would take a pill with a glass of water. The threshold for high solubility depends on the dose. Thus, low dose strengths of a drug could meet the criterion for a biowaiver (see below), whereas higher dose strengths might still require human testing.
The BCS pathway applies only to immediate-release, solid, orally administered drugs without a narrow therapeutic index; the threshold for rapid dissolution (immediate release) is at least 85% dissolution of the API from the formulated drug product in 30 minutes or less. A highly soluble and highly permeable API in an immediate-release formulation is eligible for a “biowaiver,” meaning that clinical bioequivalence (BE) studies are unnecessary during the drug’s development.
The assignment of criteria for “high” solubility and permeability creates a four-quadrant classification grid (Figure 1):
- BCS Class 1 compounds are highly soluble at all GI pH values and highly permeable in a validated in vitro model, and thus are eligible for a biowaiver.
- BCS Class 2 compounds are highly permeable but not highly soluble at all GI pH values.
- BCS Class 3 compounds are highly soluble at all GI pH values but not highly permeable.
- BCS Class 4 compounds are neither highly soluble nor highly permeable.
Note that I did not say, for example, that BCS Class 3 compounds have “low permeability,” only that they do not meet the BCS criterion for “high” permeability. A compound that is less permeable in vitro than a high-permeability internal reference standard cannot be classified as highly permeable, although its in vivo fraction absorbed might be as high as 85% (which is not exactly “low”). Likewise, weakly basic compounds are generally more soluble at acidic pH than at neutral or basic pH and cannot be classified as highly soluble even if they meet the BCS criterion at every pH tested except 7.5. Thus, the BCS criteria are extremely conservative.
A BCS biowaiver enables a drug developer to avoid a clinical BE study following changes in formulation composition, manufacturing method, manufacturing site, etc. In the course of the development of a new chemical entity (NCE), as many as eight clinical BE studies are typically done to prove, for example, that a new formulation is bioequivalent to the previous one. A BCS biowaiver cuts that down to one: the original first-in-man bioavailability study. The benefits are even more obvious for generics, theoretically eliminating the need for clinical BE studies altogether. For this reason, many of the progressive generic drug companies have been among the first to take full advantage of biowaivers. The cost savings can be substantial: costs are on the order of $60K (one time) for a definitive in vitro BCS classification vs. upwards of $100K to $400K per clinical BE study. The time savings may be even more significant (five weeks for an in vitro study vs. possibly several months for each clinical study): consider that for a drug with $1.2 billion in annual sales, every month saved during development equates to an additional $100 million in sales prior to loss of patent exclusivity.
The FDA’s decision in 2000 to embrace the BCS concept by creating a corresponding regulatory pathway was bold because, at that time, the scientific underpinnings were still considered controversial in some quarters. In the decade since, the approach has been vindicated.3 Suffice to say that, in our experience, there are no examples of compounds classified as BCS Class 1 in vitro that did not behave as such in vivo. This is partly a function of the conservative BCS criteria and partly a function of the test system most commonly used for in vitro permeability measurements, transport across Caco-2 cell monolayers. In 1989, Hidalgo et al. published the seminal paper demonstrating the utility of monolayers of human Caco-2 cell as an in vitro intestinal permeability model.4 Due to factors such as over-expression of apical efflux transporters relative to human enterocytes in vivo, the model is inherently conservative, sometimes leading to under-prediction of permeability but never over-prediction.
Jim Polli of the University of Maryland argues that in vitro BCS testing is actually better than a clinical BE study, for several reasons.5 First, in vitro studies reduce drug development costs. Second, they assess product performance (drug absorption) more directly than human BE studies, for which the approach is indirect and the interpretation often complicated. Third, it is unethical to expose healthy volunteers unnecessarily to NCEs, about which very little is known, considering that a validated in vitro alternative is available. Furthermore, in vitro permeability data is inherently more precise than clinical data due to intra- and inter-individual variability in the latter. This is especially true for highly variable drugs, which are generally subject to rapid first-pass metabolism, and for which a large number of subjects may be required to adequately power a clinical BE study.
Many large pharma companies take advantage of the BCS rationale without necessarily submitting biowaiver applications: the BCS classification of an API determines what actions are required to demonstrate BE of a new formulation (e.g., in vitro dissolution only for Class I or rapidly dissolving Class III compounds, clinical BE study for Class IV or slowly dissolving Class III compounds).6,7 Why aren’t more drug developers taking full advantage of the biowaiver regulatory pathway? Good question, and the answer, like drug development in general, is complicated. First of all, following the step to create the pathway in 2000, there was little or no promotion to raise awareness of it. To this day, a surprisingly large number of people in the pharmaceutical industry are unaware of the BCS. Second, for several years there was a perception that FDA reviews of biowaiver applications took longer than reviews of clinical BE studies, thereby negating any potential time savings. The agency tried to address that by designating a group of reviewers for all biowaiver applications. As further evidence that the agency is adapting to the new realities of the industry, the FDA Office of Generic Drugs announced on August 3, 2011 a reorganization designed to further “advance the public health as the use of generic drugs continues to increase.”8 Without question (and published data is available9), the volume of biowaiver applications has increased in the past couple of years, particularly for ANDA (generic) applications, where the benefits of the BCS are obvious and immediate. You know about the patent cliff and all of the generic versions of blockbuster drugs racing to market, so you can appreciate the reason for the dramatic increase in BCS biowaiver activity in recent years.
Of course, it’s not only generic drug companies that are pushing the generic envelope; governments worldwide are encouraging the trend as well. Every country from Greece to the U.S. is desperate to cut costs, and one way to do so is to accelerate the approval of generics, thereby reducing the potential cost burden for publicly funded prescription drug programs. By reducing the time and cost of BE testing for many generics (i.e., Class I compounds), the BCS pathway is in line with government cost-saving initiatives. The World Health Organization, while neither a drug regulatory agency nor a government, is nevertheless supportive of the BCS for its ability to make more-affordable generic drugs available sooner in developing nations.10,11
For some classes of drugs, the BCS makes sense for other reasons, in addition to the potential cost and time savings that apply to any drug. For example, imagine having to perform a clinical BE study to get regulatory approval of an oncology drug that is known to be cytotoxic. It would obviously be unethical to perform such a study in healthy volunteers. When the drug was first developed, its clinical pharmacokinetics would have been evaluated in cancer patients, in a combined Phase I/Phase II trial where subjects would be willing to trade the risk of side effects for the chance of therapeutic benefit, either for themselves or for future cancer patients. When it comes time to demonstrate BE of a generic product, the developer could either endure the time and cost of recruiting cancer patients or go for an in vitro BCS biowaiver and avoid the clinical BE route altogether. A few years ago, the FDA posted web-pages providing drug-specific guidance on the development of generic versions of many prescription drugs, including temozolomide, a marketed cancer drug.12
The European Medicines Agency (EMA) takes a somewhat different view of the BCS.13 In a couple of ways, the EMA’s position is less conservative than that of the FDA:
- The EMA’s threshold for a well-absorbed compound is fractional absorption in vivo of at least 85%, compared with 90% in the FDA guidance, holding that, given the variability of human clinical data, there is really little difference between 85% and 90% fractional absorption. (From my humble position, I would like to suggest that they should agree on a number, regardless of which they choose.)
- The EMA seems to be more willing to consider biowaivers for Class III and some Class II compounds by considering the ambient pH at the likely site of absorption (jejunum or ileum, typically) rather than throughout the entire GI tract.
On the other hand, the EMA is unwilling to grant bio-waivers based solely on in vitro data, insisting on clinical BE data in all cases. With more than 10 years of data supporting the validity of using in vitro data for BCS permeability classification, the failure of the EMA to accept this approach seems to be largely an issue of collation of data, followed by dissemination of these scientific findings throughout its rank and file. In light of the number of current initiatives encouraging cooperation and collaboration between the EMA and FDA and other drug regulatory agencies,14 harmonizing the BCS approach is an obvious and easy target for consideration.
In this time of unprecedented pharmaceutical industry crisis, the rise of generics, and governmental cost-sensitivity, the BCS is clearly a regulatory pathway in the right place at the right time. It’s hard to imagine why a drug developer would not take advantage of it.
- GL Amidon et al., Pharmac Res 1995 Mar;12(3):413-20
- U.S. Dept. of Health and Human Services, FDA, CDER, Guidance for Industry: Waiver of In Vivo Bioavailability and Bioequivalence Studies for Immediate-release Solid Oral Dosage Forms Based on a Biopharmaceutics Classification System, 2000: http://www.fda.gov/downloads/Drugs/GuidanceCompliance-RegulatoryInformation/Guidances/UCM070246.pdf
- M Mehta, Application of Biopharmaceutics Classification System (BCS) in Regulatory Submissions, AAPS webinar, September 2010
- IJ Hidalgo et al., Gastroenterology 1989 Mar;96(3):736-49
- JE Polli, AAPS J 2008 Jun;10(2):289-99
- MS Ku, AAPS J 2008 Mar;10(1):208-12
- J Cook et al., AAPS J 2008 Jun ;10(2):306-10
- AK Nair et al., Abbreviated New Drug Application Approvals by US-FDA: Biopharmaceutics Classification System Based Waivers vs. In vivo Bioequivalence Studies, poster presented at PSWC/AAPS Annual Meeting, New Orleans, LA, USA, November 2010
- WHO Technical Report Series, No. 937, Annex 7, 2006, 347-90: http://apps.who.int/prequal/info_general/documents/TRS937/WHO_TRS_937_eng.pdf#page=359
- WHO Technical Report Series, No. 937, Annex 8, 2006, 391-437: http://apps.who.int/prequal/info_general/documents/TRS937/WHO_TRS_937_eng.pdf#page=403
- U.S. Dept. of Health and Human Services, FDA, CDER, Draft Guidance on Temozolomide: http://www.fda.gov/downloads/Drugs/GuidanceComplianceRegulatoryInformation/Guidances/UCM082273.pdf
- EMA, Guideline on the Investigation of Bioequivalence, CPMP/QWP/EWP/1401/98 Rev.1, 2010: http://www.ema.europa.eu/docs/en_GB/document_library/Scientific_guideline/2010/01/WC500070039.pdf
- EMA, Final Report on the International API Inspection Pilot Programme, EMA/89978/2011: http://www.fda.gov/downloads/InternationalPrograms/FDABeyondOurBordersForeignOffices/EuropeanUnion/EuropeanUnion/EuropeanCommission/UCM266257.pdf
Patrick Dentinger is president and chief executive officer of Absorption Systems. He can be reached at email@example.com.