`
`
`
`Jennifer Dressman,* James Butter, John Hempenstall, and Christos Reppas
`
`
`
`
`
`
`
`
`
`System (BCS)has proven tobea valuablejf
`tool for the regulation ofchangesiin oral’.
`drug products during scale-up and after
`productapproval.Thisarticle reviews the
`criteria for.classifying drugsaccording|to”
`the BCS anddiscusses further potential
`applicationsof theBCS,including the
`developmentof new drugs, the approval
`of generics, and the regulation of
`controlied-release products.
`
`
`
`Jennifer Dressman, PhD,currently is
`‘researching the developmentof a predictive
`-Modelfor the delivery of drugs via the
`gastrointestinal tract, based on
`physiological as weil as drug- and dosage-
`form-related considerations. She is a
`professerat the Institute of Pharmaceutical Technology,
`Johann Wolfgang Goethe University, Marie-Curie-str. 9,
`Biozentrum 60439 Frankfurt, Germany,tel. 49 69 7982 9680,
`fax 49 69 7982 9694, e-mail dressman@em.uni-frankfurt.de.
`Dr. Dressman is a memberof the Editorial Advisory Board of
`Pharmaceutical Technology. James ButlerIs principal
`scientist, strategic technologies, at Pharmaceutical Sciences,
`GlaxoSmithKline R&D, Ware, UK, John Hempenstall is
`director of product-line extensions at Pharmaceutical
`Development, GlaxoSmithKtine R&D, Harlow, UK. Christos
`Reppas js an assistant professor at Schoal of Pharmacy,
`The University of Athens, Athens, Greece.
`“fo whom all correspondence shoud be addressed,
`
`68
`
`Pharmaceutical Technology suty 2001
`
`
`nce the Biopharmaceutics Classification System
`
`_
`(BCS) was introducedseveral years ago,it has be-
`ome a benchmark in the regulation of bioequiva-
`
`lence of oral drug products both in the United
`States and abroad (1). The concept behind the BCSis that
`if two drug products yield the same concentration pro-
`file along the gastrointestinal (GI) tract, they will result
`in the same plasmaprofile after oral administration. This
`concept can be summarized by the following equation
`
`J = Py Cy
`
`in which Jis the flux across the gutwall, P,is the permea-
`bility of the gut wall to the drug, and C,, is the concen-
`tration profile at the gut wall. In terms of bioequivalence,
`it is assumed that highly permeable, highly soluble drugs
`housed in rapidly dissolving drug products will be bio-
`equivalent and that, unless major changes are made to the
`formulation, dissolution data can be used as a surrogate
`for pharmacokinetic data to demonstrate bioequivalence
`of two drug products. The BCS thus enables manufac-
`turers to reduce the costs of approving scale-up and post-
`approval changes (SUPAC) to certain oral drug products
`{rapidly dissolving products of Class I drugs: see TableI)
`without compromising public safety interests.
`After several years of experience with the BCS, several
`issues have arisen: First, is the BCS fail safe? Second, should
`biowaivers be limited to Class I drugs, or could we extend
`them to other classes? Third, what about controlled-
`release dosage forms? Fourth, how early in the develop-
`ment process can we apply the BCS principles, and should
`the sarne cutoff values be applied to developing both new
`drug products and SUPAC applications? Although these
`issues already have been addressed to some extent in the
`literature, we must continue to gather data and experi-
`ence in order to resolve them. In this article we havetried
`to summarize current thinking and to make some sug-
`gestions about where we should head with the BCS in the
`coming years.
`
`Is the BCSfail safe?
`FDAhasset quite restrictive limitations on which drugs
`and drug products would be candidates for biowalvers
`under the BCS, The permeability requirementstates that
`MYLAN EXHIBIT 1031
`MYLAN EXHIBIT 1031
`
`

`

`
`
`ample time to be dissolved. As longas
`these drugs meet the permeability crite-
`rion, biowaivers for products that dissolve
`rapidly at pH values typical of the small
`intestine could be considered.
`Anotherissue 1s that the requirement
`for “not less than 85% dissolution within
`36 min” may be too conservative in some
`dosing circumstances. Although in the
`fasted state it is quite possible that tran-
`sit time through the stomach is short
`(half-emptying times for water as short
`as 8-16 min have been reportedin thelit-
`erature}, if the dosage form is given with
`a meal, more than likely it will spend at
`least an hour or two in the stomach.
`Under these circumstances even slowly
`dissolving products still may show ab-
`sorption patterns that are controlled by
`gastric emptying. A case example is that
`of certain immediate-release (IR) parace-
`tamol tablets. Galia et al, showed that
`Panadoltablets release very stowly in sim-
`ulated fed-state conditions (milk) (6). It
`was subsequently shown by Reppas and
`Nicolaides that gastric emptying contin-
`ues to be rate limiting to absorption of
`paracetamol, even in the fed state (7),
`These results suggest that in cases in
`which the drug is routinely administered
`with meals, it may be possible to relax the
`criteria for dissolution.
`
`Can the BCSbe extended to
`rapidly dissolving products of
`Class III substances?
`It has been suggested by Blume and Schug
`that because the absorption of Class Ii]
`drugs is essentially controlled by the gut
`wall permeability to the drug and not by
`the drug's solubility, biowalvers for rapidly
`dissolving products of Class III drugs also
`could bejustified (8), Although in terms
`of the BCS theory this conceptts clearly
`valid, some physiological issues would
`have to be addressed on a case-by-case
`basis, First, one must establish why the
`
`© Paracetamol formulation, 630 mg sodium bicarbonate
`+ Paracetamol formutation, 400 mg sodium bicarbonate
`4 Paracetamol formufation, 375 mg sodium bicarbonate
`* Panadoltablets
`
`0 Panadol soluble, 1342 mg sodium bicarbonate
`
`
`Intrinsic solubility: 0.664 mg/ml. and pk: 4.39
`
`k :
`
`Figure 1: Mcan paracetamol serum concentrationsfollowing 500 mg oral paracetamol.
`{see Figure 1) even
`though the dissolution
`ofthe praducts in vitro
`was similar (4). This ex-
`ample shows that even
`though an excipient
`change may seem com-
`pletely innocuous,if the
`new excipient alters the
`GI physiology, then it
`mayvery well alter the
`plasma profile also,
`Regulatory authorities
`must be very careful
`about defining what
`constitutes a “major
`change”to the formu-
`lation to address the potential physiolog!-
`cal issues.
`
`Figure 2: The pH-solubility profile of ibuprofen at 37 °C.
`
`the permeability of the drug is commen-
`surate with 290% absorption from a so-
`lution, The solubility requirementis that
`the dose-to-solubility ratio (D:S) of the
`drug must be =250 mi. over a pH range
`of I to 7.5, and the dissolution require-
`ment for the drug product is that disso-
`lution must be >85% complete within
`30 min (3), For products meeting these
`criteria, gastric emptying, rather than the
`release performance of the drug product,
`will be the key factor in determining the
`plasmaprofile; therefore,variability in the
`plasmaprofile will be under physiologi-
`cal control and notdictated by the dosage
`form.
`Even for rapidly dissolving products of
`Class I drugs, however, it is possible to man-
`ufacture bioinequivalent products if ex-
`cipients that modify gastric emptying are
`added. For example, Grattan et al, showed
`that the addition of sodium bicarbonate to
`the paracetamol(acetaminophen) formu-
`lation produced a faster and higher peak
`concentration of paracetamol in plasma
`
`Are the BCScriteria too restrictive?
`On the other hand, some drugs that are
`currently classified as Class IT are consis-
`tently and completely absorbed after oral
`administration. These are typically poorly
`soluble weak acids with pK, values of <4.5
`and intrinsic solubilities (solubility of the
`un-ionized form) of =0.01 mg/mL. At pH
`values typical of the fasted state in the je-
`junum (about pH 6.5), these drugs will
`have solubilities of =1 mg/mL, resulting
`in fast and reliable dissolution of the drug.
`Currently, these drugs are classified as
`Class II drugs because they are poorly sol-
`uble at gastric pH, in which pli<< pX,,
`Figure 2 shows a typical solubility versus
`pHprofile for ibuprofen (5).
`Because the small-intestinal transit time
`is more reliable, and in the fasted state,
`longer than the gastric residence time
`(generally on the orderof 3 h), drugs with
`these physical characteristics will have
`
`www.pharmaportal.com
`70=Pharmaceutical Technolagy suty 2007
`
`

`

`
`
`
`a With 1132 mg SAPP
`4 Alone
`° All excipiants of effervescent tablet except SAPP
`« Effervescent tablet containing 1132 mg SAPP
`
`
`
`Figure 3: Mean serum ranitidine concentrations following 150 mg oral solution doses of
`ranitidine.
`
`permeability of the gut wall to the drugts
`low. If the permeability is low but uniform
`along the entire GItract (including the
`proximal colon), biowaivers might be con-
`sidered. However,if there is an absorption
`window ora gradientin the permeability
`of the gut wall to the drug (with decreas-
`ing permeability in distal regions}, excip-
`ients that accelerate gut motility could sig-
`nificantly reduce the contact time of the
`drug with the sites at which permeability
`is favorable and therefore lower the
`bioavallability of the drug.
`Several compounds belonging to the
`H2 receptor antagonist group are classi-
`cai examples of Class III drugs. It was
`shown in the literature some years ago that
`the shapeof the plasmaprofile of cimeti-
`dine is highly dependent uponthe gastric
`pH atthe time of administration, with the
`characteristic double peak eliminated if
`the drug is given underelevated gastric
`pH conditions (9), Further, excipients that
`accelerate transit in the upper GI tract
`such as sodium acid pyrophosphate (10)
`and mannitol {11} have been clearly
`shownto reduce the extent of absorption
`of ranitidine and cimetidine, respectively,
`The results from Koch etal. are shown in
`Figure 3 (10). The 50% reduction in C,...
`illustrates how importantthe influence of
`excipients that can alter the GI motility
`can be to the absorption of Ciass Il drugs.
`
`Can the BCS be applied to
`controlied-release drug products?
`Under the current definition, the BCS Is
`applicable only to immediate-release
`dosage forms because only the perme-
`72
`Pharmaceutical Technology duty 2007
`
`ability in the jejunum is considered, To
`extend the BCS to controlled-release (CR)
`dosage forms, one must assess the per-
`mieability at all points in the GI tract where
`release of the drug is foreseen (12). As
`pointed out by Corrigan, it is unlikely that
`drugs with low permeability in either the
`ileum or colon will prove to be suitable
`candidates for CR dosage forms, jet alone
`for biowaivers based on dissolution tests
`(5). He has proposed a useful subclassifi-
`cation scheme for CR products that is
`based on the site dependency of both the
`drug solubility and permeability,
`A further consideration is the selection
`of appropriate dissolution conditions to
`simulate the release profile of the dosage
`form as it moves through the GI tract.
`Conditions for dissolution in the stom-
`ach, the smail intestine, and the colon dif-
`fer greatly. Important parameters that
`vary with location in the GI tract include
`the volumeof fluid available for dissolu-
`tion, osmolarity of the contents, the hy-
`drodynamic (motility) conditions, and
`the secretion of various enzymes and
`other para-GIsecretions that could po-
`tentially affect the release rate, Similarity
`of the dissolution profiles under ail ap-
`propriate G] conditions would have to be
`shown for the two drug products, Al-
`though our understanding of the com-
`position of lumenal contents as they move
`along the GI tract is far better than it was
`a decade ago, a more complete charac-
`terization is still needed, Still almost to-
`tally lacking is an understanding ofthe
`relationship between the hydrodynamics
`in the gut and those in the currently avail-
`
`able dissolution testers. This throws a de-
`gree of uncertainty into the interpreta-
`tion of dissolution results in terms of in
`vivo performance, even when the com-~-
`position of the lumenal contents can be
`simulated well in the in vitro tests, Al-
`though a problem is posed by the limi-
`tation to establishing in vivo~in vitro cor-
`relations for IR products, the problem is
`compounded for CR dosage forms be-
`cause the hydrodynamics at severalsites
`within the GI tract must then be simu-
`lated. As a result, In vitro release profiles
`of CR dosage forms with different release
`mechanisms must be interpreted very
`cautiously,
`
`Application of the BES to the
`development of new
`drug substances
`Because the BCS was originally developed
`as a basis for determining bioequivalence
`of oral drug products, it assumes that the
`drugis suffictently well absorbed to make
`an oral dosage form feasible. When new
`drug substances are belng developed,
`however, this assumption is not appro-
`priate, and one must consider other fac-
`tors than just the solubility and permea-
`bility to determine whether an oral dosage
`form can be successfully developed. An
`overview of the events in the Gi tract fol-
`lowing oral drug administration is de-
`picted in Figure 4.
`First, it should be remembered that the
`drug substance does not have to meet the
`Class I criteria of high permeability and
`solubility for the drug to be successfully
`formulated in an oral solid dosage form.
`ManyClass II and Class IIT drugs are avail-
`able on the market, and several that meet
`Class IV criteria are available (see Table I).
`One problem with applying the BCScri-
`terla to new drug substances is that, early
`in preformutation/formulation, the dose
`is not yet accurately known, So at this
`point, the D:S can only be expressed as a
`likely range, A helpful rule of thumbis that
`compounds with aqueous solubilities
`> 100 pg/mL seldom exhibit dissolution
`rate-limited absorption. Alternatively, one
`can estimate the maximum absorbable
`dose on the basis of the usual volumes of
`GI fluids available underthe anticipated
`dosing conditions and the solubility of the
`drug. With regard to the solubility of the
`drug, it may be useful to consider the
`www. pharmaportal.com
`
`

`

`
`Drug in sofution
`at uptake sites
`Pt
`f
`
`Decomposition
`Adsorption
`Complexation
`
`---4-¢-<I
`j
`j
`I
`+
`4
`¥
`Decomposition
`Adsorption
`Drug in solution
`Complexation
`
`at uptake sites
`
`--Ne—o
`
`omeo
`Transitee ee
`
`
`
`
`small intestine, and up to 1 L for the post-
`prandial stomach and small intestine.
`A further consideration is the choice of
`model for assessing the permeability. Al-
`though perfusions in humans will pro-
`duce the mostreliable results (18) and are
`clearly the “gold standard,” these require
`too much time and money to make them
`practicable for screening new drug sub-
`stances, Many animal- and cell-culture
`models have been developed, each with its
`ownset of advantages and disadvantages.
`For example, the Caco 2 cells can be used
`with confidenceto assess transcellular dif-
`fusion and can be standardized to ensure
`reproducible results, but they tend to un-
`derestimate paracellular and active mecha-
`nisms, cannot be employed to determine
`regional permeability within the gut, and
`tend to overestimate efflux via the P-
`glycoproteins. In situ perfusions in rats,
`although they are muchbetter in termsof
`forecasting active transport and can be
`used to determine regional permeability,
`take more time and effort to produce a re-
`liable permeability estimate. In any case,
`it is a good idea to have more than one
`permeability screen at the disposal of the
`laboratory in order to build confidence
`and robustness into the screening system.
`If the drug is pooriy soluble but highly
`permeable, formulation efforts will con-
`centrate on improving the dissolution
`profile. For example, the combined effects
`of formulating the drug as amorphous
`solid dispersion and administering it in
`the fed state are shownfor troglitazone
`in Figure 5, Combined, these two ap-
`proachesshift the solubility-dissolution
`characteristics from those of a very poorly
`soluble drug (D:S >10,000 mL) to those
`of a drug product with a D:S within the
`range of values encountered in the gut
`after meals.
`Figure 6 summarizes some further pos-
`sibilities for improving the absorption of
`drugs with less than optimal permeabil-
`ity and solubility characteristics. If per-
`meability rather than solubility is the
`main problem, formulation approaches
`are less numerous andless reliable, In ex-
`treme cases, It may be appropriate to con-
`sider developing another analog with
`more appropriate biopharmaceutical
`characteristics.
`Even when allowance is made for the
`differences in solubility and permeability
`
`
`
`
`
`
`
`
`
`
`
`Figure 4: Steps in drug absorption and sourcesof incomplete bioavailability following or
`administration of a solid dosage form.
`
`would lead to a quite different interpreta-
`tion of the chances for absorption in vivo.
`For promising compounds that are both
`ionizable and Hpophilic, extensive solubil-
`ity experiments in biorelevant media will
`help characterize thelikely solubility be-
`haviorin vivo. Several publications address
`the composition and applications of these
`media (6,13-16). An alternative approach
`is to use aspirates from human volunteers,
`although volumes aspirated typically are
`small and the choice of experiments and
`apparatusthereforeis limited (17).
`Anotherissue is the use of 250 mL as
`the volume in which a dose must be dis-
`solved, This amount is a conservative es-
`timate of the volume offluid available in
`the gut under fasting-state conditions and
`is based on the volume usually ingested
`along with the dosage form in a pharma-
`cokinetic study (the so-called FDA glass of
`water). The actual volume available ts a
`composite of the ingested fluid and the se-
`cretions of the GI tract. Although these
`amounts tend to be modest in the fasted
`state, secretions in the fed state contribute
`substantially to the overall fluid volume,
`which may be as high as 1.5 L in both the
`stomach and upper smail intestine. De-
`pending on whether drug administration
`is to be on an empty stomach or with
`meals,it is reasonable to adjust the volume
`used to assess the capacity of the GIfluids
`to dissolve the dose, A useful starting point
`would be to use a volume of 300 mL for
`the fasted stomach, 5€0 mL. for the fasting
`
`physicochemical properties of the drug
`when deciding which media to use for the
`solubility determinations. For example,
`measuring solubility at all pH values rec-
`ommended by the BCSis unnecessary for
`neutral compoundsin early development,
`Later, when formulations are compared,
`dissolution data for the drug product over
`the entire GI pH range will be useful in es-
`tablishing the robustness of release from
`the formulation under GI conditions,
`Lipophilic drugs may be very poorly sol-
`uble in water and in simple buffers, but in
`the GI fluids they can often be solubilized
`by the bile to a significant extent, Increases
`in solubility of one to two orders of mag-
`nitude are passible for compounds with
`log P values of 24. In some cases this
`www, pharmaportai.com
`74~—Pharmaceutical Technology vury 2007
`
`

`

`ty
`Amorphous!
`{fed}
`
`fy
`
`Crystalline
`(fasted)
`
`proval, In the future,
`BCSconcepts probably
`will be used increasingly
`in the early develop-
`ment of new drugs, In-
`cluding for analog se-
`lection as well as for
`
`
`Figure 5: Troglitazone 200 mg: the effect of food and form on the
`potential for solubility limited bioavailability.
`
`Amorphous (fasted) scale-up and postap-
`
`
`. E, Gatia et al., “Evaluation of Various Disso-
`lution Media for Predicting In Vivo Perfor-
`mance of Class and H Compounds,” Pharm.
`Res. 15 (5), 698-705 (1998).
`. C, Reppas and E, Nicolaides, “Analysis of
`Drug Dissolution Data” in Methods for As-
`sessing Oral Drug Absorption, J, Dressman
`and H. Lennernas, Eds. (Marcel Dekker, New
`York, 2000}, pp. 228-254.
`. H. Blume andB. Schug, “The Biopharma-
`initial formulation ap-
`ceutles Classification System (BCS): ClassII
`Drugs —- Better Candidates for BA/BE
`proaches. As our knowl-
`Watver?” Bur. J. Pharm, Sci. 9 (2), 117-121
`edge of GI physiology
`(1999}.
`becomes more sophis-
`. ¥, Mummanert and J}. Dressman, “Gastric
`ticated, in vitro dissolu-
`pH Influences on the Appearance of Double
`tion tests will be able to
`Peaks in the Plasma Concentration-Time
`Profiles of Cimetidine after Oral Adminis-
`better simulate the con-
`tration in Dogs,” Pharm. Res, 12 (6), 780-786
`ditions in the GI tract.
`(995).
`This in turn will lead to
`10,
`K.M.Kochet al., “Effect of Sodium Acid Py-
`more powerful predic-
`<——_
`rophosphate on Ranitidine Bioavailability
`=
`k
`and Gastrointestinal Transit Time,” Pharm.
`tions of In vivo perfor-
`Nanoparticles
`FSelebt more soluble salt/polymorph
`Res, 10 (7), 1027-1030 (1993).
`mance and ultimately to
`. DA. Adkin, et al., “The Effect of Mannitol
`|
`“Si Liquid-filed capsutes (ofy
`a significant reduction
`on the Oral Bioavailability of Cimetidine,” j,
`i
`1
`& self-amulsifying vehicias)
`in the numberof ani-
`Solic dispersions|
`Pharm. Sel, 84(12),1405-1409 (1995).
`TAddltion of surfactaatto
`mal and human studies
`I. Wilding, “Evolution of the Biopharma-
`solld dosage form
`required to optimize the
`ceutics Classification System (BCS) to Mad-
`Liquid-filled
`ified Release (MR) Formulations: What Do
`formulation, Together
`capsules with
`absorption-
`We Need to Consider?” Eur. f. Pharm. Sci, 8
`with screens for other
`enhancing
`(3), 157-159 (1999).
`excipients?
`limitations to oral ab-
`J, Dressmanetal., “Dissolution Testing as a
`sorption,the BCS paves
`Prognostic Tool for Oral Drug Absorption:
`the way for (r)evolution
`Immediate Release Dosage Forms,’ Pharm.
`Res, 15 (1), 11-22 (1998}.
`in the drug develop-
`E, Nicolaideset al,, “Forecasting the In Vivo
`ment process,
`Performance of Four Low-Solubility Drugs
`from Their In Vitro Dissolution Behaviour,”
`Pharm, Res, 16 (12}, 1876-1882 (1999).
`R, Lébenberget al., “Dissolution Testing as
`a Prognostic Tool for Oral Drug Absorption:
`Dissolution Behaviour of Glibenclamide,”
`Pharm. Res, 17 {4}, 439-444 (2000),
`J, Dressman and C., Reppas, “In Vitro In Vivo
`Correlations for Lipophitic, Poorly Water-
`Soluble Deugs,” Bur. J. Pharm, Sei, 14 Suppl.
`2,573-S80 (2001).
`B, Pedersonetal., “A Comparisonofthe Sot-
`ubility of Danazol in Human and Simulated
`Gastrointestinal Fluids,” Pharm. Res. 17 (1),
`891-894 (2000).
`H. Lennernas, “Human Perfusion Studies,’
`in Methods in Oral Drug Absorption, J. Dress-
`man and H. Lennernas, Eds. (Marcel Dekker,
`New York, 2000}, pp. 99-17. PT
`
`
`
`12.
`
`13.
`
`14,
`
`£5,
`
`16.
`
`17.
`
`18,
`
`www. pharmaportal.cam
`
`3—
`
`Mucoadhesion,
`absorption-
`1 erchancing HE
`excipients,
`effiux
`inhibitors
`
`WV
`
`Figure 6: Possible effects of various formulations on developabilily.
`
`requirements for oral drug product de-
`velopmentvis-a-vis biowaivercriteria ac-
`cording to the BCS,further factorsstill
`must be considered for new drugs. These
`include the possibility of decomposition
`under GI conditions and the assessment
`of first-pass metabolism both in the gut
`wail and theliver. Appraising decompo-
`sition in the gutis relatively simple using
`biorelevant media and exposure times
`based on longest anticipated exposure
`times. For sensitive compounds, appro-
`priate enzymes(e.g., pepsin and gastric li-
`pases for the stornach, pancreatic enzymes
`for the jejunum, and bacterial enzymes
`for the colon) must be added to the
`medium in relevant concentrations. As far
`as first-pass metabolism in the gut wailis
`concerned, it may be possible to screen for
`metabolites in the permeability model de-
`pending on how the modelis set up.
`
`Summary
`In summary, the BCS has proven to be an
`extremely useful tool for the regulation of
`bioequivalence of drug products during
`76
`Pharmaceutical Technology suvy 2007
`
`References
`1. G. Amidon etal., “Theoretical Basls for a Bio-
`pharmaceutical Drug Classification: The Cor-
`relation of In Vitro Drug Product Dissclu-
`tion and In Vivo Bioavailability,” Pharm, Res.
`12 (3), 413-420 (1995),
`2, Center for Drug Evaluation, FDA,“Guidance
`for Industry on Dissolution Testing of Im-
`mediate Release Solid Oral Dosage Forms”
`{August 1997).
`3. Center for Drug Evaluation, FDA,“Guidance
`for Industry on the Waiver of In Vivo
`Bioavailability and Bioequivalence for Im-
`mediate Release Solid Oral Dosage Forms
`Based on a Biopharmaceuties Classification
`System” (August 2.000).
`4, T. Grattan et al., "A Flve-way Crossover
`Human Volunteer Study to Compare the
`Pharmacokinetics of Paracetamol Following
`Oral Administration of Two Commerciaily
`Available Paracetamol Tablets and Three De-
`velopment Tablets Containing Paracetamol
`in Combination with Sodium Bicarbonate
`or Calctum Carbonate,” Bur. J. Pharm. Bio-
`pharm. 49 (3), 225-229 (2000).
`5, OJ. Corrigan. “The Biopharmaceutics Drug
`Classification and Drugs Administered In Ex-
`tended Release (ER) Formulations,” Adv. Exp.
`Med. Biol. 423, 11-128 (1997).
`
`