- Email: [email protected]
Conference Report: Bio-International 2005
COMMENTARY Conference Report: Bio-International 2005 KAMAL K. MIDHA,1,2 VINOD P. SHAH,3 GUR JAI PAL SINGH,4 RABI PATNAIK5 1
College of Pharmacy and Medicine, University of Saskatchewan, Saskatoon, SK, Canada
2
Pharmalytics Inc., Saskatoon, SK, Canada
3
International Pharmaceutical Federation (FIP) Scientific Secretary, North Potomac, Maryland
4
Division of Bioequivalence, United States Food and Drug Administration, Rockville, Maryland
5
Watson Laboratories, Corona, California
Received 14 June 2006; revised 21 August 2006; accepted 28 August 2006 Published online in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/jps.20786
ABSTRACT: This is a summary report of the International Pharmaceutical Federation/ Board of Pharmaceutical Sciences (FIP/BPS) international conference, Bio-International 2005, which was held October 24–26, 2005 at the Royal Pharmaceutical Society, in London, UK. Bioequivalence (BE) issues related to multisource locally delivered topical dosage forms, oral inhalation drug products, highly variable drug products (HVDP), and endogenously occurring drugs were discussed. The conference also focused on alternate approaches to assess BE for some of these drug products. Pharmacokinetic (PK) approaches like, dermatopharmacokinetics (DPK) for dermatological topical dosage forms, scaled average BE (s-ABE) where within-subject variability is considered for estimation of 90% confidence intervals to document BE for highly variable drugs (HVD) were recommended. In addition, issues and difficulties related to the BE assessment of oral inhalation products, role, and appropriateness of metabolites in BE assessment, importance of base line correction in BE assessment of endogenously occurring drugs, and waiver of BE study requirements for certain drugs based on a Biopharmaceutics Classification System (BCS), were also discussed. ß 2006 Wiley-Liss, Inc. and the American Pharmacists Association J Pharm Sci 96:747–754, 2007
Keywords: bioequivalence; bioavailability; Biopharmaceutics Classification System; regulatory science; transdermal
INTRODUCTION The availability of multisource (generic) drug products has made a notable impact on economics of the pharmaceutical market. More than 50% of the prescriptions filled in the United States are for multisource drug products. This represents Gur Jai Pal Singh’s present address is Watson Laboratories, Corona, CA. Correspondence to: Kamal K. Midha (Telephone: 306-6688585; Fax: 306-384-1726.; E-mail: [email protected]) Journal of Pharmaceutical Sciences, Vol. 96, 747–754 (2007) ß 2006 Wiley-Liss, Inc. and the American Pharmacists Association
remarkable success in providing medicines which are less expensive, yet provide safety and effectiveness comparable to that of the corresponding brand name drugs. Safety and effectiveness of multisource drug products is based on documentation of their bioequivalence (BE) to the corresponding reference listed drugs (RLDs).1 A large majority of the approved multisource drug products are intended for delivery to the systemic circulation, which represents the vehicle for drug delivery to the site(s) of action. For these drug products, BE based on drug concentrations in blood following
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 96, NO. 4, APRIL 2007
747
748
MIDHA ET AL.
administration of multisource and RLD products assures therapeutic equivalence.1 Compared with the drug products intended for systemic delivery, development of multisource drug products for local delivery has had much less success, partly due to the lack of suitable methods for the determination of equivalence of the locally delivered drug products. In this conference the locally delivered drug products included topical dosage forms and oral inhalation drug products. The conference also dealt with three other important BE topics which included highly variable drug products (HVDP), metabolites issues in BE assessment, and products of drugs with endogenous levels.
BE OF TOPICAL DOSAGE FORMS Assessment of BE of topical dosage forms is a challenge for pharmaceutical and regulatory scientists. Topical dosage forms such as creams, ointments, transdermal patches, etc. differ in their intended therapeutic effects. Generally the therapeutic actions are divided into two broad categories such as products for local action and products for systemic action. For the approval of a multisource drug product in USA, an Abbreviated New Drug Application (ANDA) is required with documentation of BE along with appropriate in vivo and in vitro tests. Possible methods for determination of BE of multisource topical drug products include clinical trials, pharmacodynamics (PD), dermatopharmacokinetics (DPK), dermalmicrodialysis (DMD), in vitro methods and other techniques.2 Comparative clinical trials require a large patient population, are cumbersome, expensive, time consuming, and lack sensitivity. PD method is limited to corticosteroids. DPK methodology measures the drug concentration in the stratum corneum (SC), and is applicable for all topical dermatological drug products. However, the relevance of DPK to clinical efficacy has not been established. DMD is a moderately invasive technique compared to DPK, but can provide valuable supportive data for DPK.3,4 Combination of in vitro release technique with other BE measures such as DPK can serve as an excellent method to document BE of dematological dosage forms. DPK approach to assess the topical bioavailability of drugs proposes to use tape-strip sampling of skin’s outermost layer, the SC which then gets qualified as a surrogate for the determination of JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 96, NO. 4, APRIL 2007
drug levels at the target site in the skin. This is similar to the way in which blood levels are used for systemically active drugs. The technique has the advantage of simplicity in its performance in vivo in man.5 The concept has a merit, but the implementation of the approach demands that (1) DPK be validated as a tool with which to determine the availability of a percutaneously absorbed drug to its site of action within the skin and (2) the DPK protocol be sufficiently robust for the reliable evaluation and comparison of topical bioavailability of drugs from different formulations. To achieve these broad objectives, tape stripping experiments were integrated with mathematical models of dermal absorption (1) to quantify uncertainty in DPK method, (2) to evaluate how modifications of DPK method improve the quality of information derived, and (3) to assess the suitability and limitations of DPK method for evaluating the rate and extent of drug delivery following diverse application regimens.6,7 Tape striping removes SC and the drug embedded in it. The amount of SC (corneocyte) removed depends on the type of formulation applied to the skin, type of tape used, and the skin itself. Furthermore, SC thickness varies between the subjects. Therefore, for the DPK methods to be quantitative, it is essential that the amount of tissue removed by tape stripping is known. This can be achieved gravimetrically, in combination with transepidermal water loss measurements or spectrophotometrically.8 Lateral spreading of topically applied formulations post application needs to be considered when using the DPK methodology. This phenomenon has been identified as the cause of artifactual differences in topical BE assessed by DPK.9 Nevertheless, DPK is a potentially powerful technique, allowing the determination of pharmacokinetic (PK) parameters for objective BE assessment. The feasibility of employing DMD for topical BE assessment has been evaluated. The approach showed a rank order correlation with DPK measurements for cream and ointment products. DMD has the unique advantage of being potentially useful in vivo in diseased skin; clearly this is less practical with DPK. The possibility of continuous sampling is also attractive. DMD, however, is technically demanding, and requires highly sensitive analytical methodology, particularly for poorly penetrating and/or extremely lipophilic drugs. DOI 10.1002/jps
CONFERENCE REPORT: BIO-INTERNATIONAL 2005
ORAL INHALATION DRUG PRODUCTS Many respiratory drugs are locally acting and they are manufactured as nebulizer solutions, nasal sprays, pressurized metered dose inhalers (MDIs), and dry powder inhalers (DPIs). Among these, nebulizer/inhalation solutions may be eligible for waiver of in vivo BE testing.10 The remaining inhalation aerosols require extensive testing of comparative in vivo and in vitro performance to establish BE. The conventional (PK) methods of documentation of BE are not applicable to establishing equivalence with respect to local delivery to the lung.11,12 Consequently, documentation of equivalence of local delivery is based on either PD or clinical endpoint studies. The FDA has previously issued guidances for PD studies for determination of in vivo BE of inhalation aerosols.13,14 The agency has previously approved multisource albuterol MDIs, based on PD studies that employed either bronchodilatation model based on comparison of AUEC (Area under the FEV1* vs. time curve) and FEV1max, or bronchoprovocation model based on comparison of PD20 or PC20{ of the challenge agent following administration of the multisource and RLD MDIs.12 For the PD studies, demonstration of adequate dose response is necessary to provide the study with the ability to distinguish between single and multiple actuations/doses of the marketed products. A number of factors including study drug, study population, and measures of drug effects may determine the quality of dose response. Selection of patient population which can provide a wider window for improvement of asthma,12 and assessments based on measures of greater severity of asthma, for example, rate of exacerbations15 may facilitate demonstration of dose response. Alterations in the design of the drug delivery systems can have profound effect on in vivo performance of the inhalation aerosols. Therefore, documentation of BE of such products also includes determination of equivalent performance of metering devices, based on a variety of in vitro tests that characterize the drug delivery and spray plumes.12,16 The FDA guidance for determination of BE of inhalation aerosols issued up to this time of this conference pertain to the chlorofluorocar*FEV1 ¼ Forced expiratory volume in one second. { PD20 or PC20 ¼ Provocative dose or concentration of the challenge agent required to reduce the bronchodilator-induced FEV1 to a level 20% below the baseline (control) FEV1. DOI 10.1002/jps
749
bon (CFC)-based shortacting beta-agonists. BE studies for hydrofluoroalkane (HFA)-based MDIs containing same drugs are expected to be similar, due to same mode of drug delivery. The bronchoprovocation or bronchodilation models used for the shortacting beta-agonists may also be extended to the longacting beta-agonists by modification of study designs. However, these models as such are not applicable to determination of BE of inhaled corticosteroid (ICS) products, principally due to the lack of measurable dose response relationships of marketed ICS products. It was indicated in this session at Bio-International 2005 that demonstration of the ICS dose response may be possible if the studies are based on measures of greater severity of asthma instead of the conventional measures-like FEV1. It was noted, however, that such measures have not been used for crossover evaluations which are preferred in the conduct of BE studies, because each subject provides its own control. A recently described ‘‘Asthma Stability’’ model17,18 appears promising in obtaining measurable dose response for ICS with possible evaluation of treatments in a crossover manner. Inhalation aerosols also include DPIs which provide an environmentally friendly alternative to the use of MDIs and help overcome limitations of the amount of drug that can be administered19 and problems of co-ordination20,21 associated with the use of MDIs. Development of recommendations to establish BE is complicated as the available DPIs differ with regard to the powder formulation and design of the delivery devices.19,22 Furthermore, due to the dependence of many DPIs for drug delivery upon the patient’s inspiratory effort, the inspiratory flow rate is an important determinant of in vitro drug delivery from DPIs, their in vivo bioavailability and clinical effectiveness.11 The available FDA draft guidance23 stipulates consideration for flow rate in the determination of drug delivery from DPIs. Up to the time of this conference, the FDA has made no recommendations for documentation of BE of DPIs. Presumably, the recommendations should include both in vitro and in vivo studies. Clinical models used for documentation of BE of pressurized MDIs may be applicable to the determination of comparative in vivo performance of the single entity DPIs which contain only one active drug. However, development of in vivo models for the determination of equivalent local delivery from combination products (e.g., DPIs containing a beta agonist and a corticosteroid) is complicated due to lack of JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 96, NO. 4, APRIL 2007
750
MIDHA ET AL.
methods that can distinguish between the pulmonary effects of the individual drugs in the combination product. At this meeting, a couple of proposals were made to address this complexity. One suggestion was to evaluate the bronchodilator component as done for conventional bronchodilators, in addition to demonstration of clinical equivalence based on rate of exacerbations or levels of a noninvasive marker (e.g., exhaled nitric oxide) unaffected by bronchodilatation. An alternative proposal included assessment of BE based on the bronchodilator component with supportive evidence for the comparable effectiveness of the corticosteroid component consisting of equivalent systemic exposure and pharmaceutical equivalence. It was noted that these proposals were presented for discussion purposes, and did not represent recommendations of any regulatory agency.
BE OF HIGHLY VARIABLE DRUGS Drugs with high within-subject variability (ANOVA %CV) of 30% or above in the PK measures are generally defined as HVDs.24 Highly variable drug products (HVDP) are products containing drugs that may not necessarily be highly variable, but because of poor pharmaceutical quality formulation which results in high variability. High variability may arise due to unpredictable response of either a significant number or a few subjects in a conventional BE study on this class of drug products. BE study of these drugs has shown that when the same drug is dosed on two occasions, often BE is not demonstrated between the two dosing regimens. Thus it would be difficult to establish BE between two different products containing HVD with conventional number of subjects (e.g., 24) using a 90% confidence interval with preset acceptance limits. Consequently, it becomes imperative to enroll unreasonably higher number of test subjects so that the 90% confidence interval for the PK measures would remain within the acceptable range to document BE. This results in unnecessary additional human testing without any tangible benefit to consumers from the product quality standpoint.25 HVDs usually have wide therapeutic ranges, having little safety concern. For wide therapeutic range HVDs, one should not have to study an excessive number of test subjects to prove that two equivalent products meet preset (one size fits all) JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 96, NO. 4, APRIL 2007
statistical criteria. Highly variable approved drugs must have a wide therapeutic range; otherwise there would have been significant safety issues and lack of efficacy during phase III testing.25 Various approaches including alternate study designs, statistical methods, and other considerations have been suggested to document BE of this class of drug products. Study designs include, replicate designs, group sequential designs, stable isotope techniques, and statistical approaches such as scaled average BE (s-ABE), individual BE (IBE).25–27 Some other approaches include multiple dose studies, BE on the basis of metabolite(s), and correction methods to reduce within-subject variability. Among all reported approaches, significant interest has been recently expressed regarding s-ABE.25–28 The concept of s-ABE has evolved from the original approach of IBE. The IBE concept addresses the correct question, that is, the switchability of drug products. It also considers subject-by-formulation interaction as well as offers incentive for less variable test products. In this approach, scaling is based on the within-subject variability of the reference product for highly variable as well as narrow therapeutic range drugs. In addition, it encourages the use of test subjects more representative of the general population.25,26 On subsequent extensive examination of this novel approach, the problem of assuring switchability by the current ABE approach and the occurrence of subject-by-formulation interaction could not be substantiated. However, it was shown that the incentive for less variable test products could be dealt with the ABE with scaling approach. The scaling is based on the pooled within-subject variabilities of the test and reference products in a two period design or based on within-subject variability of the reference product in a replicate design.25–28 Some expert scientists recommended that ABE with scaling may be considered. A point estimate (PE) criteria on mean AUC and mean CMAX should be required for both HVDs and NTI drugs. The limit of the PE criteria would depend on the magnitude of the within-subject variability. For example, for NTI drugs, where the within-subject variability of the reference product is <20%, PE criteria for mean AUC of 10%, and for mean CMAX of 15% should be required. For all other drugs including NTI drugs showing >20% within-subject variability of reference drugs, the PE criteria for mean AUC and mean CMAX should be 15% and DOI 10.1002/jps
CONFERENCE REPORT: BIO-INTERNATIONAL 2005
20%, respectively. Methodology should be developed to allow approvals based on weighting of ABE analyses for HVDs. Regulatory agencies should define the methodology to determine the weighting criterion for the first generics. Innovators should also be encouraged to provide measure of within-subject variability upon approval, which would be used in the weighting scheme for the first approval. However, multisource product manufacturers should have the option of determining within-subject variability as part of their study. Interestingly, it was noted at this meeting that a large majority of HVDs are BCS Class I compounds and therefore the BE issue of this class of drug products may become unpronounced in future.
ROLE OF METABOLITES IN BE DETERMINATION The use of metabolite data in BE studies has been the subject of many discussions and no clear consensus seems to emerge from the review of the scientific literature. These reports are derived from actual BE studies, simulated data involving parent drug, and one or more metabolite using often linear kinetics. In one such report involving parent drug and one metabolite with linear kinetics and no first-pass effect, it was observed that considering the ANOVA-CV (pooled withinsubject variability) and the width of the 90% confidence interval for CMAX, the parent drug exhibited the greater sensitivity than that of the metabolite based on the width of the 90% confidence intervals, and thus, for these types of drugs, it was proposed that the parent drug may be preferred over metabolite for BE determination as the data on parent drug will be more discriminatory.29 Using a more sophisticated PK modeling and applying intrinsic and renal clearances, it was reported that when intrinsic clearance was less than the liver blood flow, the BE based on the parent drug was also preferred over that of the metabolite. On the other hand, when renal clearance was low, the metabolite should be preferred in case of AUC.30 In another report on HVDs with linear kinetics and high first-pass effect, metabolite was preferred for measurement of AUC and CMAX, when intrinsic clearance was higher than the hepatic blood flow.31 In other analyses,32 data from actual BE studies were used to estimate the fraction of the dose of the parent DOI 10.1002/jps
751
drug reaching the systemic circulation intact33 and to calculate the hepatic extraction. Using the estimates of intrinsic clearance (Clint) and unbound fraction, it was observed that metabolite data could only be justified in simple uncomplicated situations such as in case of inactive prodrug which is converted to active metabolite in vivo. In cases were the parent drug is subjected to extensive phase I and phase II metabolism to multiple active metabolites and when the CLint is more that the hepatic blood flow, it is preferable to rely on the parent drug for BE determination.32 It is important to separate bioavailability from BE. Bioavailability is principally a drug rather than a formulation characteristic. In bioavailability, metabolism of the drug is characterized as an essential part of a PK profile. In BE, a major objective is to characterize the release of drug from the drug product. Thus it is more appropriate to measure the drug itself and parent drug should be measured where ever possible. Where it is not possible, a rationale should be provided to measure the metabolite(s). However there are certain exceptions. It is not acceptable to indicate that variability is driving the choice of analyte to be measured. Rationale for not measuring the parent drug could be based on the transient nature of the parent drug such as in the case of a pro-drug or if it is practically impossible to characterize the profiles of the parent drug. It is paramount that in BE studies, a proper characterization of concentration versus time profiles is prerequisite to establishing equivalence of the two products. Thus, where ever the parent drug concentration versus time profile can be adequately characterized BE decision should be based on the parent drug. It is recognized that as technology advances, it is expected that there will be significant improvement in the measures to establish BE and new approaches should be developed to establish BE.
BE OF DRUGS WITH ENDOGENOUS LEVEL BA and BE of drug products with endogenous levels of the drug need special considerations. There is lack of information and guidance regarding the appropriate way of assessing BE of this special type of drug products. Several drugs are available that are endogenous to the human and/ or present to a varying degree in the systemic circulation. Some of these include hormones, for example, estrogens, testosterone, progesterone, calcitriol, ursodiol, levothyroxine, insulin, human JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 96, NO. 4, APRIL 2007
752
MIDHA ET AL.
growth hormone, and electrolyte, for example, potassium. Several issues are associated with the assessment of BA and BE of drug products with endogenous levels. These include assay sensitivity, endogenous baseline, feedback inhibition of endogenous production, circadian rhythm, linear/ nonlinear PK. Thus, careful attention has to be paid to the selection of study design, biological matrix for drug analysis, and the need for baseline correction. For example, in order to assess BE of potassium drug product, careful attention is paid to the sodium and potassium intake in the diet, fluid intake so that the baseline of potassium levels in the biological matrix is stabilized before administration of the drug product. Usually, the matrix of choice for measuring drug levels is the blood. However in the case of potassium products, it is desirable to measure urinary potassium levels for several reasons. Serum potassium measurements are insensitive and homeostatic mechanisms maintain serum potassium level within a narrow concentration range. Moreover, after drug administration there is very limited rise in the serum potassium levels. Documentation of BE for potassium requires baseline correction. The baseline reflects excretion of potassium ingested in the diet, the amount of potassium excreted following dosing arises from the drug, and food intake. This correction is necessary to compare amounts of drug delivered from two formulations and must correct for the amount due to diet. BE is based on the urinary excretion PK, that is, cumulative excretion and maximum excretion rate. As usual, BE is assessed by the 90% confidence interval approach. By examining the results from the baseline-uncorrected and -corrected data, it was established that baseline correction is appropriate for evaluation of BE of potassium drug products as baselinecorrected data were more sensitive to changes in the formulations. Baseline for each subject in each period should be corrected and be reproducible during the two equilibration periods. Baseline correction should be evaluated for drug products which have endogenous levels of drug. Special attention should be paid to defining the baseline and then measuring the baseline in each period prior to dosing. Other considerations for these products would require establishing the method of correction of baseline, and the magnitude of baseline in relationship to the drug concentration after treatment. Endogenous baselines that change due to exogenous drug JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 96, NO. 4, APRIL 2007
administration, present a technical challenge to examine BA, and to demonstrate BE.
BIOPHARMACEUTICS CLASSIFICATION SYSTEM (BCS) Biopharmaceutics Classification System (BCS) is a scientific framework for classifying drug substance based on aqueous solubility and permeability. Based on these two parameters, drug substances are classified into one of the four classes—Class 1: high solubility/high permeability; Class 2: low solubility/high permeability; Class 3: high solubility/low permeability; and Class 4: low solubility/low permeability. The drug substance is highly soluble when the highest marketed dose dissolves in 250 mL of aqueous buffers in pH 1.2–6.8. The drug substance is highly permeable when its absorption is 85% or greater.34 When combined with the dissolution of the drug product, BCS takes into account three major factors that govern the rate and extent of absorption from immediate release solid oral dosage forms: dissolution, solubility, and intestinal permeability. According to FDA guidance of biowaiver based on BCS, Class 1 drugs can be waived from BE requirements if the drug substance belongs to Class 1, that is, highly soluble and highly permeable, and the drug product is either very rapidly dissolving or rapidly dissolving (85% dissolution in 15 or 30 min when it is carried out using basket method at 100 rpm or paddle method at 50 rpm in 900 mL of pH 1.2, 4.5, and 6.8). For biowaiver, the test product should be compared to the reference product, and its dissolution profile in all three media should meet dissolution profile comparison criteria, based on a similarity factor (f2) 50.35 Application and extension of BCS principles to predict drug disposition classification system (BDDCS) was also discussed. It was suggested that (1) all Class 1 and 2 drugs are generally extensively metabolized in the body whereas Class 3 and Class 4 drugs are eliminated unchanged, (2) For Class 1 drugs, GI transporter effects are minimum whereas for Class 2 drugs efflux transporter effects predominate, and for Class 3 drugs absorptive transporter effects predominate, and (3) Class 1 drugs had no food effect whereas Class 2 drugs showed increase in drug absorption. Thus the value of BCS in predicting other in vivo characteristics of drug absorption was expanded.36 DOI 10.1002/jps
CONFERENCE REPORT: BIO-INTERNATIONAL 2005
WHO working guidance on drug interchangeability, especially with respect to BCS-based biowaiver was discussed. Based on scientific principle and with the aim of reducing unnecessary human testing and reducing regulatory requirements, relaxation of criteria for biowaiver based on BCS was discussed. It was considered that (1) the dissolution testing requirements for weak acids for Class 2 drugs can be restricted to 85% dissolution in pH 6.8 (site for drug absorption) and maintaining profile similarity criteria under all three pHs, and (2) for Class 3 drugs exhibiting very rapid dissolution (85% in 15 min) in all three pH conditions as the reference product and containing known excipients with no GI effects and no PK effect, a biowaiver can be granted.37 Several aspects of this BCS-based biowaiver criteria were discussed including the following: – Underlying principle and mechanism of BCS-based biowiaver. – Comparison of BCS-based ‘essential WHO drug list’ based on published literature (articles by Amidon, Dressman, and Benet groups). – Criteria for BCS-based biowaiver (too stringent and should be relaxed). – Reasons why BCS has not been picked-up by the generic industry. – Extension of BCS-based biowaivers based on WHO document.
2.
3.
4.
5.
6.
7.
8.
In spite of good scientific basis for biowaiver, some countries were skeptical in adopting BCS criteria for biowaivers. However, it was concluded that there was no rationale for holding back the proposal for recommending BCS-based biowaivers, in accord with the FDA and WHO guidances.
10. 11.
ACKNOWLEDGMENTS
12.
9.
Contributors: Richard H Guy, Eva Benfeldt, Nina Otberg, Neil Barnes, Ian Smith, Richard Ahrens, Henning Blume, Gordon Amidon, Leslie Benet, and Iain McGilveray.
REFERENCES 1. Conner DP, Davit B. 2005. Bioequivalence and drug product assessment, in vivo. In: Shargel L,
DOI 10.1002/jps
13.
753
Kanfer I, editors. Generic drug product development: Solid oral dosage forms. New York: Marcel Dekker, pp 227–255. Shah VP, Flynn GL, Yacobi A, Maibach HI, Bon C, Fleischer NM, Franz TJ, Kaplan SA, Kawamoto J, Lesko LJ, Marty JP, Pershing LK, Schaefer H, Sequeira JA, Shrivastava SP, Wilkins J, Williams RL. 1998. Bioequivalence of topical dermatological dosage forms. Methods of evaluation of bioequivalence. Pharm Res 15:167–171. Benfeldt B, Growth G. 1998. Feasibility of measuring lipophilic or protein-bound drugs in the dermis by in vivo microdialysis after topical or systemic drug administration. Acta Derm Venereol 78:274– 278. Shah VP. 2001. Progress in methodologies for evaluating bioequivalence of topical formulations. Am J Clin Dermatol 2:275–280. Pershing LK, Nelson JL, Corlett JL, Shrivastava SP, Hare DB, Shah VP. 2003. Assessment of dermatopharmacokinetic approach in bioequivalence determination of topical tretinoin gel products. J Am Acad Dermatol 48:740–751. Alberti I, Kalia YN, Naik A, Bonny JD, Guy RH. 2001. Assessment and prediction of the cutaneous bioavailability of topical terbinafine in vivo. Pharm Res 18:1472–1475. Alberti I, Kalia YN, Naik A, Bonny JD, Guy RH. 2001. Effect of ethanol and isopropyl myristate on the availability of topical terbinafine in human stratum corneum, in vivo. Int J Pharm 219:11–19. Weigmann HJ, Lademann J, Schaefer H, Sterry W. 1999. Determination of the horny layer profile by spectroscopy in the visible range as a prerequisite to quantify percutaneous absorption. Skin Pharmacol Appl Skin Physiol 12:34–45. Shah VP. 2005. IV-IVC for topically applied preparations-a critical evaluation. Eur J Pharm Biopharm 60:309–314. [21Code of Federal Regulation. 320.22(b) (3)] Singh GJP. 2006. Evaluation of comparative performance of multisource locally acting inhalation aerosols and nasal sprays—An FDA Perspective. Am Pharm Rev (in press). Singh GJP, Adams WP. 2005. US regulatory and scientific considerations for approval of generic locally acting orally inhaled and nasal drug products. In: Dalby RN, Byron P, Peart J, Suman JD, editors. Respiratory drug delivery. A regulatory and analytical symposium. Richmond, VA: Virginia Commonwealth University, pp 115– 125. Guidance for the in-vivo bioequivalence studies of metaproterenol sulfate and albuterol inhalation aerosols (metered dose inhalers). 1989. Division of Bioequivalence, Food and Drug Administration, Rockville, Maryland.
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 96, NO. 4, APRIL 2007
754
MIDHA ET AL.
14. Interim guidance for documentation of in vivo bioequivalence of albuterol inhalation aerosols (metered dose inhalers). 1994. Office of generic drugs, US food and drug administration, Rockville, Maryland. 15. Pauwels RA, Lofdahl CG, Postma DS, Tatterfield AE, O’Byrne P, Barnes PJ, Ullman A. 1997. Effect of inhaled formoterol and budesonide on exacerbation of asthma. Fermoterol and Corticosteroid Establishing Therapy (FACET) International Study Group. New Engl J Med 337:1405– 1411. 16. Guidance for Industry (Draft): Bioavailability and bioequivalence studies for nasal aerosols and nasal sprays for local action. 2003. Center for drug evaluation and research, US food and drug administration, Rockville, Maryland. www.fda.gov/cder/ guidance/5383DFT. 17. Ahrens RC, Teresi ME, Han S-H, Donne D, Vanden Burgt JA, Lux CR. 2001. Asthma stability after oral prednisone—A clinical model for comparing inhaled corticosteroid potency. Am J Respir Crit Care Med 164:1138–1145. 18. Ahrens RC, Hendeles L, Teresi ME, Lux CR, Vanden Burgt JA, Hart KA, Ekholm BP. 2003. Relative potency of beclomethasone dipropionate (BDP) delivered by HFA-MDI, and fluticasone propionate (FP) delivered by Diskus. Eur Respir J 22:1576. 19. Smith IJ, Parry-Billing M. 2003. The inhalers of the future? A review of dry powder devices on the market. Pulmon Pharmacol Therap 16:79–95. 20. Crompton G. 1982. Problems patients have using pressurized aerosol inhalers. Eur J Respir Dis 119:101–104. 21. Larsen J, Hahn M, Elkholm B, Wick K. 1994. Evaluation of conventional press-and-breath metered-dose inhaler techniques in 501 patients. J Asthma 31:193–199. 22. Dunbar CA, Hickey AJ, Holzner P. 1998. Dispersion and characterization of pharmaceutical dry powder aerosols. KONA 16:7–14. 23. Guidance for Industry (Draft): Metered dose inhaler (MDI) and dry powder inhaler (DPI) drug products: CMC, manufacturing and control documentation. 1998. Center for drug evaluation and research, US Food and Drug Administration, Rockville, Maryland. 24. Blume HH, Midha KK. 1993. Bio-International 1992, Conference on bioavailability, bioequivalence, and pharmacokinetics studies. J Pharm Sci 82:1186–1189.
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 96, NO. 4, APRIL 2007
25. Midha KK, Rawson MJ, Hubbard JW. 2005. The bioequivalence of highly variable drugs and drug products. Int J Clin Pharmcol Ther 43:485–598. 26. Tothfalusi L, Endrenyi L, Midha KK, Rawson MJ, Hubbard JW. 2001. Evaluation of bioequivalence of highly variable drugs. Pharm Res 18:728–733. 27. Tothfalusi L, Endrenyi L, Midha KK. 2003. Scaling or wider bioequivalence limits for highly variable drugs and for the special case of Cmax. Int J Clin Pharmacol Therap 41:217–225. 28. Tothfalusi L, Endrenyi L. 2003. Limits for scaled average bioequivalence of highly variable drugs and drug products. Pharm Res 20:382–389. 29. Chen ML, Jackson AJ. 1991. The role of metabolite in bioequivalency assessment. I. Linear pharmacokinetics without first-pass effect. Pharm Res 8:25– 32. 30. Tucker GT, Rostami A, Jackson PR. 1993. Metabolite measurement in bioequivalence studies: Theoretical considerations. In: Midha KK, Blume H, editors. Bio-International: Bioavailability, Bioequivalence and Pharmacokinetics. International Conference of FIP. ‘‘Bio-International 0 92’’ held in Bad Homburg, Germany, May 20–23, 1992. Stuttgart: Medpharm Scientific Publishers, pp 163–170. 31. Jackson AJ. 2000. The role of metabolites in bioequivalency assessment. III. Highly variable drugs with linear kinetics and first-pass effect. Pharm Res 17:1432–1436. 32. Midha KK, Rawson MJ, Hubbard JW. 2004. Commentary: The role of metabolites in bioequivalence. Pharm Res 21:1331–1334. 33. Gibaldi M, Boyes RN, Feldman S. 1971. Influence of first-pass effect on availability of drugs on oral administration. J Pharm Sci 60:1338–1340. 34. Amidon GL, Lennernas H, Shah VP, Crison JR. 1995. 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. 35. Guidance for Industry: Waiver of in vivo bioavailability and bioequivalence studies for immediaterelease dosage forms based on biopharmaceutics classification system. 2000. Center for drug evaluation and research, US Food and Drug Administration, Rockville, Maryland. 36. Wu C-Y, Benet LZ. 2005. Predicting drug disposition via application of BCS. Transport/absorption/ elimination interplay and development of a biopharmaceutics drug disposition classification system. Pharm Res 22:11–23. 37. WHO Working Document QAS/04.093. 2005.
DOI 10.1002/jps
College of Pharmacy and Medicine, University of Saskatchewan, Saskatoon, SK, Canada
2
Pharmalytics Inc., Saskatoon, SK, Canada
3
International Pharmaceutical Federation (FIP) Scientific Secretary, North Potomac, Maryland
4
Division of Bioequivalence, United States Food and Drug Administration, Rockville, Maryland
5
Watson Laboratories, Corona, California
Received 14 June 2006; revised 21 August 2006; accepted 28 August 2006 Published online in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/jps.20786
ABSTRACT: This is a summary report of the International Pharmaceutical Federation/ Board of Pharmaceutical Sciences (FIP/BPS) international conference, Bio-International 2005, which was held October 24–26, 2005 at the Royal Pharmaceutical Society, in London, UK. Bioequivalence (BE) issues related to multisource locally delivered topical dosage forms, oral inhalation drug products, highly variable drug products (HVDP), and endogenously occurring drugs were discussed. The conference also focused on alternate approaches to assess BE for some of these drug products. Pharmacokinetic (PK) approaches like, dermatopharmacokinetics (DPK) for dermatological topical dosage forms, scaled average BE (s-ABE) where within-subject variability is considered for estimation of 90% confidence intervals to document BE for highly variable drugs (HVD) were recommended. In addition, issues and difficulties related to the BE assessment of oral inhalation products, role, and appropriateness of metabolites in BE assessment, importance of base line correction in BE assessment of endogenously occurring drugs, and waiver of BE study requirements for certain drugs based on a Biopharmaceutics Classification System (BCS), were also discussed. ß 2006 Wiley-Liss, Inc. and the American Pharmacists Association J Pharm Sci 96:747–754, 2007
Keywords: bioequivalence; bioavailability; Biopharmaceutics Classification System; regulatory science; transdermal
INTRODUCTION The availability of multisource (generic) drug products has made a notable impact on economics of the pharmaceutical market. More than 50% of the prescriptions filled in the United States are for multisource drug products. This represents Gur Jai Pal Singh’s present address is Watson Laboratories, Corona, CA. Correspondence to: Kamal K. Midha (Telephone: 306-6688585; Fax: 306-384-1726.; E-mail: [email protected]) Journal of Pharmaceutical Sciences, Vol. 96, 747–754 (2007) ß 2006 Wiley-Liss, Inc. and the American Pharmacists Association
remarkable success in providing medicines which are less expensive, yet provide safety and effectiveness comparable to that of the corresponding brand name drugs. Safety and effectiveness of multisource drug products is based on documentation of their bioequivalence (BE) to the corresponding reference listed drugs (RLDs).1 A large majority of the approved multisource drug products are intended for delivery to the systemic circulation, which represents the vehicle for drug delivery to the site(s) of action. For these drug products, BE based on drug concentrations in blood following
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 96, NO. 4, APRIL 2007
747
748
MIDHA ET AL.
administration of multisource and RLD products assures therapeutic equivalence.1 Compared with the drug products intended for systemic delivery, development of multisource drug products for local delivery has had much less success, partly due to the lack of suitable methods for the determination of equivalence of the locally delivered drug products. In this conference the locally delivered drug products included topical dosage forms and oral inhalation drug products. The conference also dealt with three other important BE topics which included highly variable drug products (HVDP), metabolites issues in BE assessment, and products of drugs with endogenous levels.
BE OF TOPICAL DOSAGE FORMS Assessment of BE of topical dosage forms is a challenge for pharmaceutical and regulatory scientists. Topical dosage forms such as creams, ointments, transdermal patches, etc. differ in their intended therapeutic effects. Generally the therapeutic actions are divided into two broad categories such as products for local action and products for systemic action. For the approval of a multisource drug product in USA, an Abbreviated New Drug Application (ANDA) is required with documentation of BE along with appropriate in vivo and in vitro tests. Possible methods for determination of BE of multisource topical drug products include clinical trials, pharmacodynamics (PD), dermatopharmacokinetics (DPK), dermalmicrodialysis (DMD), in vitro methods and other techniques.2 Comparative clinical trials require a large patient population, are cumbersome, expensive, time consuming, and lack sensitivity. PD method is limited to corticosteroids. DPK methodology measures the drug concentration in the stratum corneum (SC), and is applicable for all topical dermatological drug products. However, the relevance of DPK to clinical efficacy has not been established. DMD is a moderately invasive technique compared to DPK, but can provide valuable supportive data for DPK.3,4 Combination of in vitro release technique with other BE measures such as DPK can serve as an excellent method to document BE of dematological dosage forms. DPK approach to assess the topical bioavailability of drugs proposes to use tape-strip sampling of skin’s outermost layer, the SC which then gets qualified as a surrogate for the determination of JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 96, NO. 4, APRIL 2007
drug levels at the target site in the skin. This is similar to the way in which blood levels are used for systemically active drugs. The technique has the advantage of simplicity in its performance in vivo in man.5 The concept has a merit, but the implementation of the approach demands that (1) DPK be validated as a tool with which to determine the availability of a percutaneously absorbed drug to its site of action within the skin and (2) the DPK protocol be sufficiently robust for the reliable evaluation and comparison of topical bioavailability of drugs from different formulations. To achieve these broad objectives, tape stripping experiments were integrated with mathematical models of dermal absorption (1) to quantify uncertainty in DPK method, (2) to evaluate how modifications of DPK method improve the quality of information derived, and (3) to assess the suitability and limitations of DPK method for evaluating the rate and extent of drug delivery following diverse application regimens.6,7 Tape striping removes SC and the drug embedded in it. The amount of SC (corneocyte) removed depends on the type of formulation applied to the skin, type of tape used, and the skin itself. Furthermore, SC thickness varies between the subjects. Therefore, for the DPK methods to be quantitative, it is essential that the amount of tissue removed by tape stripping is known. This can be achieved gravimetrically, in combination with transepidermal water loss measurements or spectrophotometrically.8 Lateral spreading of topically applied formulations post application needs to be considered when using the DPK methodology. This phenomenon has been identified as the cause of artifactual differences in topical BE assessed by DPK.9 Nevertheless, DPK is a potentially powerful technique, allowing the determination of pharmacokinetic (PK) parameters for objective BE assessment. The feasibility of employing DMD for topical BE assessment has been evaluated. The approach showed a rank order correlation with DPK measurements for cream and ointment products. DMD has the unique advantage of being potentially useful in vivo in diseased skin; clearly this is less practical with DPK. The possibility of continuous sampling is also attractive. DMD, however, is technically demanding, and requires highly sensitive analytical methodology, particularly for poorly penetrating and/or extremely lipophilic drugs. DOI 10.1002/jps
CONFERENCE REPORT: BIO-INTERNATIONAL 2005
ORAL INHALATION DRUG PRODUCTS Many respiratory drugs are locally acting and they are manufactured as nebulizer solutions, nasal sprays, pressurized metered dose inhalers (MDIs), and dry powder inhalers (DPIs). Among these, nebulizer/inhalation solutions may be eligible for waiver of in vivo BE testing.10 The remaining inhalation aerosols require extensive testing of comparative in vivo and in vitro performance to establish BE. The conventional (PK) methods of documentation of BE are not applicable to establishing equivalence with respect to local delivery to the lung.11,12 Consequently, documentation of equivalence of local delivery is based on either PD or clinical endpoint studies. The FDA has previously issued guidances for PD studies for determination of in vivo BE of inhalation aerosols.13,14 The agency has previously approved multisource albuterol MDIs, based on PD studies that employed either bronchodilatation model based on comparison of AUEC (Area under the FEV1* vs. time curve) and FEV1max, or bronchoprovocation model based on comparison of PD20 or PC20{ of the challenge agent following administration of the multisource and RLD MDIs.12 For the PD studies, demonstration of adequate dose response is necessary to provide the study with the ability to distinguish between single and multiple actuations/doses of the marketed products. A number of factors including study drug, study population, and measures of drug effects may determine the quality of dose response. Selection of patient population which can provide a wider window for improvement of asthma,12 and assessments based on measures of greater severity of asthma, for example, rate of exacerbations15 may facilitate demonstration of dose response. Alterations in the design of the drug delivery systems can have profound effect on in vivo performance of the inhalation aerosols. Therefore, documentation of BE of such products also includes determination of equivalent performance of metering devices, based on a variety of in vitro tests that characterize the drug delivery and spray plumes.12,16 The FDA guidance for determination of BE of inhalation aerosols issued up to this time of this conference pertain to the chlorofluorocar*FEV1 ¼ Forced expiratory volume in one second. { PD20 or PC20 ¼ Provocative dose or concentration of the challenge agent required to reduce the bronchodilator-induced FEV1 to a level 20% below the baseline (control) FEV1. DOI 10.1002/jps
749
bon (CFC)-based shortacting beta-agonists. BE studies for hydrofluoroalkane (HFA)-based MDIs containing same drugs are expected to be similar, due to same mode of drug delivery. The bronchoprovocation or bronchodilation models used for the shortacting beta-agonists may also be extended to the longacting beta-agonists by modification of study designs. However, these models as such are not applicable to determination of BE of inhaled corticosteroid (ICS) products, principally due to the lack of measurable dose response relationships of marketed ICS products. It was indicated in this session at Bio-International 2005 that demonstration of the ICS dose response may be possible if the studies are based on measures of greater severity of asthma instead of the conventional measures-like FEV1. It was noted, however, that such measures have not been used for crossover evaluations which are preferred in the conduct of BE studies, because each subject provides its own control. A recently described ‘‘Asthma Stability’’ model17,18 appears promising in obtaining measurable dose response for ICS with possible evaluation of treatments in a crossover manner. Inhalation aerosols also include DPIs which provide an environmentally friendly alternative to the use of MDIs and help overcome limitations of the amount of drug that can be administered19 and problems of co-ordination20,21 associated with the use of MDIs. Development of recommendations to establish BE is complicated as the available DPIs differ with regard to the powder formulation and design of the delivery devices.19,22 Furthermore, due to the dependence of many DPIs for drug delivery upon the patient’s inspiratory effort, the inspiratory flow rate is an important determinant of in vitro drug delivery from DPIs, their in vivo bioavailability and clinical effectiveness.11 The available FDA draft guidance23 stipulates consideration for flow rate in the determination of drug delivery from DPIs. Up to the time of this conference, the FDA has made no recommendations for documentation of BE of DPIs. Presumably, the recommendations should include both in vitro and in vivo studies. Clinical models used for documentation of BE of pressurized MDIs may be applicable to the determination of comparative in vivo performance of the single entity DPIs which contain only one active drug. However, development of in vivo models for the determination of equivalent local delivery from combination products (e.g., DPIs containing a beta agonist and a corticosteroid) is complicated due to lack of JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 96, NO. 4, APRIL 2007
750
MIDHA ET AL.
methods that can distinguish between the pulmonary effects of the individual drugs in the combination product. At this meeting, a couple of proposals were made to address this complexity. One suggestion was to evaluate the bronchodilator component as done for conventional bronchodilators, in addition to demonstration of clinical equivalence based on rate of exacerbations or levels of a noninvasive marker (e.g., exhaled nitric oxide) unaffected by bronchodilatation. An alternative proposal included assessment of BE based on the bronchodilator component with supportive evidence for the comparable effectiveness of the corticosteroid component consisting of equivalent systemic exposure and pharmaceutical equivalence. It was noted that these proposals were presented for discussion purposes, and did not represent recommendations of any regulatory agency.
BE OF HIGHLY VARIABLE DRUGS Drugs with high within-subject variability (ANOVA %CV) of 30% or above in the PK measures are generally defined as HVDs.24 Highly variable drug products (HVDP) are products containing drugs that may not necessarily be highly variable, but because of poor pharmaceutical quality formulation which results in high variability. High variability may arise due to unpredictable response of either a significant number or a few subjects in a conventional BE study on this class of drug products. BE study of these drugs has shown that when the same drug is dosed on two occasions, often BE is not demonstrated between the two dosing regimens. Thus it would be difficult to establish BE between two different products containing HVD with conventional number of subjects (e.g., 24) using a 90% confidence interval with preset acceptance limits. Consequently, it becomes imperative to enroll unreasonably higher number of test subjects so that the 90% confidence interval for the PK measures would remain within the acceptable range to document BE. This results in unnecessary additional human testing without any tangible benefit to consumers from the product quality standpoint.25 HVDs usually have wide therapeutic ranges, having little safety concern. For wide therapeutic range HVDs, one should not have to study an excessive number of test subjects to prove that two equivalent products meet preset (one size fits all) JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 96, NO. 4, APRIL 2007
statistical criteria. Highly variable approved drugs must have a wide therapeutic range; otherwise there would have been significant safety issues and lack of efficacy during phase III testing.25 Various approaches including alternate study designs, statistical methods, and other considerations have been suggested to document BE of this class of drug products. Study designs include, replicate designs, group sequential designs, stable isotope techniques, and statistical approaches such as scaled average BE (s-ABE), individual BE (IBE).25–27 Some other approaches include multiple dose studies, BE on the basis of metabolite(s), and correction methods to reduce within-subject variability. Among all reported approaches, significant interest has been recently expressed regarding s-ABE.25–28 The concept of s-ABE has evolved from the original approach of IBE. The IBE concept addresses the correct question, that is, the switchability of drug products. It also considers subject-by-formulation interaction as well as offers incentive for less variable test products. In this approach, scaling is based on the within-subject variability of the reference product for highly variable as well as narrow therapeutic range drugs. In addition, it encourages the use of test subjects more representative of the general population.25,26 On subsequent extensive examination of this novel approach, the problem of assuring switchability by the current ABE approach and the occurrence of subject-by-formulation interaction could not be substantiated. However, it was shown that the incentive for less variable test products could be dealt with the ABE with scaling approach. The scaling is based on the pooled within-subject variabilities of the test and reference products in a two period design or based on within-subject variability of the reference product in a replicate design.25–28 Some expert scientists recommended that ABE with scaling may be considered. A point estimate (PE) criteria on mean AUC and mean CMAX should be required for both HVDs and NTI drugs. The limit of the PE criteria would depend on the magnitude of the within-subject variability. For example, for NTI drugs, where the within-subject variability of the reference product is <20%, PE criteria for mean AUC of 10%, and for mean CMAX of 15% should be required. For all other drugs including NTI drugs showing >20% within-subject variability of reference drugs, the PE criteria for mean AUC and mean CMAX should be 15% and DOI 10.1002/jps
CONFERENCE REPORT: BIO-INTERNATIONAL 2005
20%, respectively. Methodology should be developed to allow approvals based on weighting of ABE analyses for HVDs. Regulatory agencies should define the methodology to determine the weighting criterion for the first generics. Innovators should also be encouraged to provide measure of within-subject variability upon approval, which would be used in the weighting scheme for the first approval. However, multisource product manufacturers should have the option of determining within-subject variability as part of their study. Interestingly, it was noted at this meeting that a large majority of HVDs are BCS Class I compounds and therefore the BE issue of this class of drug products may become unpronounced in future.
ROLE OF METABOLITES IN BE DETERMINATION The use of metabolite data in BE studies has been the subject of many discussions and no clear consensus seems to emerge from the review of the scientific literature. These reports are derived from actual BE studies, simulated data involving parent drug, and one or more metabolite using often linear kinetics. In one such report involving parent drug and one metabolite with linear kinetics and no first-pass effect, it was observed that considering the ANOVA-CV (pooled withinsubject variability) and the width of the 90% confidence interval for CMAX, the parent drug exhibited the greater sensitivity than that of the metabolite based on the width of the 90% confidence intervals, and thus, for these types of drugs, it was proposed that the parent drug may be preferred over metabolite for BE determination as the data on parent drug will be more discriminatory.29 Using a more sophisticated PK modeling and applying intrinsic and renal clearances, it was reported that when intrinsic clearance was less than the liver blood flow, the BE based on the parent drug was also preferred over that of the metabolite. On the other hand, when renal clearance was low, the metabolite should be preferred in case of AUC.30 In another report on HVDs with linear kinetics and high first-pass effect, metabolite was preferred for measurement of AUC and CMAX, when intrinsic clearance was higher than the hepatic blood flow.31 In other analyses,32 data from actual BE studies were used to estimate the fraction of the dose of the parent DOI 10.1002/jps
751
drug reaching the systemic circulation intact33 and to calculate the hepatic extraction. Using the estimates of intrinsic clearance (Clint) and unbound fraction, it was observed that metabolite data could only be justified in simple uncomplicated situations such as in case of inactive prodrug which is converted to active metabolite in vivo. In cases were the parent drug is subjected to extensive phase I and phase II metabolism to multiple active metabolites and when the CLint is more that the hepatic blood flow, it is preferable to rely on the parent drug for BE determination.32 It is important to separate bioavailability from BE. Bioavailability is principally a drug rather than a formulation characteristic. In bioavailability, metabolism of the drug is characterized as an essential part of a PK profile. In BE, a major objective is to characterize the release of drug from the drug product. Thus it is more appropriate to measure the drug itself and parent drug should be measured where ever possible. Where it is not possible, a rationale should be provided to measure the metabolite(s). However there are certain exceptions. It is not acceptable to indicate that variability is driving the choice of analyte to be measured. Rationale for not measuring the parent drug could be based on the transient nature of the parent drug such as in the case of a pro-drug or if it is practically impossible to characterize the profiles of the parent drug. It is paramount that in BE studies, a proper characterization of concentration versus time profiles is prerequisite to establishing equivalence of the two products. Thus, where ever the parent drug concentration versus time profile can be adequately characterized BE decision should be based on the parent drug. It is recognized that as technology advances, it is expected that there will be significant improvement in the measures to establish BE and new approaches should be developed to establish BE.
BE OF DRUGS WITH ENDOGENOUS LEVEL BA and BE of drug products with endogenous levels of the drug need special considerations. There is lack of information and guidance regarding the appropriate way of assessing BE of this special type of drug products. Several drugs are available that are endogenous to the human and/ or present to a varying degree in the systemic circulation. Some of these include hormones, for example, estrogens, testosterone, progesterone, calcitriol, ursodiol, levothyroxine, insulin, human JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 96, NO. 4, APRIL 2007
752
MIDHA ET AL.
growth hormone, and electrolyte, for example, potassium. Several issues are associated with the assessment of BA and BE of drug products with endogenous levels. These include assay sensitivity, endogenous baseline, feedback inhibition of endogenous production, circadian rhythm, linear/ nonlinear PK. Thus, careful attention has to be paid to the selection of study design, biological matrix for drug analysis, and the need for baseline correction. For example, in order to assess BE of potassium drug product, careful attention is paid to the sodium and potassium intake in the diet, fluid intake so that the baseline of potassium levels in the biological matrix is stabilized before administration of the drug product. Usually, the matrix of choice for measuring drug levels is the blood. However in the case of potassium products, it is desirable to measure urinary potassium levels for several reasons. Serum potassium measurements are insensitive and homeostatic mechanisms maintain serum potassium level within a narrow concentration range. Moreover, after drug administration there is very limited rise in the serum potassium levels. Documentation of BE for potassium requires baseline correction. The baseline reflects excretion of potassium ingested in the diet, the amount of potassium excreted following dosing arises from the drug, and food intake. This correction is necessary to compare amounts of drug delivered from two formulations and must correct for the amount due to diet. BE is based on the urinary excretion PK, that is, cumulative excretion and maximum excretion rate. As usual, BE is assessed by the 90% confidence interval approach. By examining the results from the baseline-uncorrected and -corrected data, it was established that baseline correction is appropriate for evaluation of BE of potassium drug products as baselinecorrected data were more sensitive to changes in the formulations. Baseline for each subject in each period should be corrected and be reproducible during the two equilibration periods. Baseline correction should be evaluated for drug products which have endogenous levels of drug. Special attention should be paid to defining the baseline and then measuring the baseline in each period prior to dosing. Other considerations for these products would require establishing the method of correction of baseline, and the magnitude of baseline in relationship to the drug concentration after treatment. Endogenous baselines that change due to exogenous drug JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 96, NO. 4, APRIL 2007
administration, present a technical challenge to examine BA, and to demonstrate BE.
BIOPHARMACEUTICS CLASSIFICATION SYSTEM (BCS) Biopharmaceutics Classification System (BCS) is a scientific framework for classifying drug substance based on aqueous solubility and permeability. Based on these two parameters, drug substances are classified into one of the four classes—Class 1: high solubility/high permeability; Class 2: low solubility/high permeability; Class 3: high solubility/low permeability; and Class 4: low solubility/low permeability. The drug substance is highly soluble when the highest marketed dose dissolves in 250 mL of aqueous buffers in pH 1.2–6.8. The drug substance is highly permeable when its absorption is 85% or greater.34 When combined with the dissolution of the drug product, BCS takes into account three major factors that govern the rate and extent of absorption from immediate release solid oral dosage forms: dissolution, solubility, and intestinal permeability. According to FDA guidance of biowaiver based on BCS, Class 1 drugs can be waived from BE requirements if the drug substance belongs to Class 1, that is, highly soluble and highly permeable, and the drug product is either very rapidly dissolving or rapidly dissolving (85% dissolution in 15 or 30 min when it is carried out using basket method at 100 rpm or paddle method at 50 rpm in 900 mL of pH 1.2, 4.5, and 6.8). For biowaiver, the test product should be compared to the reference product, and its dissolution profile in all three media should meet dissolution profile comparison criteria, based on a similarity factor (f2) 50.35 Application and extension of BCS principles to predict drug disposition classification system (BDDCS) was also discussed. It was suggested that (1) all Class 1 and 2 drugs are generally extensively metabolized in the body whereas Class 3 and Class 4 drugs are eliminated unchanged, (2) For Class 1 drugs, GI transporter effects are minimum whereas for Class 2 drugs efflux transporter effects predominate, and for Class 3 drugs absorptive transporter effects predominate, and (3) Class 1 drugs had no food effect whereas Class 2 drugs showed increase in drug absorption. Thus the value of BCS in predicting other in vivo characteristics of drug absorption was expanded.36 DOI 10.1002/jps
CONFERENCE REPORT: BIO-INTERNATIONAL 2005
WHO working guidance on drug interchangeability, especially with respect to BCS-based biowaiver was discussed. Based on scientific principle and with the aim of reducing unnecessary human testing and reducing regulatory requirements, relaxation of criteria for biowaiver based on BCS was discussed. It was considered that (1) the dissolution testing requirements for weak acids for Class 2 drugs can be restricted to 85% dissolution in pH 6.8 (site for drug absorption) and maintaining profile similarity criteria under all three pHs, and (2) for Class 3 drugs exhibiting very rapid dissolution (85% in 15 min) in all three pH conditions as the reference product and containing known excipients with no GI effects and no PK effect, a biowaiver can be granted.37 Several aspects of this BCS-based biowaiver criteria were discussed including the following: – Underlying principle and mechanism of BCS-based biowiaver. – Comparison of BCS-based ‘essential WHO drug list’ based on published literature (articles by Amidon, Dressman, and Benet groups). – Criteria for BCS-based biowaiver (too stringent and should be relaxed). – Reasons why BCS has not been picked-up by the generic industry. – Extension of BCS-based biowaivers based on WHO document.
2.
3.
4.
5.
6.
7.
8.
In spite of good scientific basis for biowaiver, some countries were skeptical in adopting BCS criteria for biowaivers. However, it was concluded that there was no rationale for holding back the proposal for recommending BCS-based biowaivers, in accord with the FDA and WHO guidances.
10. 11.
ACKNOWLEDGMENTS
12.
9.
Contributors: Richard H Guy, Eva Benfeldt, Nina Otberg, Neil Barnes, Ian Smith, Richard Ahrens, Henning Blume, Gordon Amidon, Leslie Benet, and Iain McGilveray.
REFERENCES 1. Conner DP, Davit B. 2005. Bioequivalence and drug product assessment, in vivo. In: Shargel L,
DOI 10.1002/jps
13.
753
Kanfer I, editors. Generic drug product development: Solid oral dosage forms. New York: Marcel Dekker, pp 227–255. Shah VP, Flynn GL, Yacobi A, Maibach HI, Bon C, Fleischer NM, Franz TJ, Kaplan SA, Kawamoto J, Lesko LJ, Marty JP, Pershing LK, Schaefer H, Sequeira JA, Shrivastava SP, Wilkins J, Williams RL. 1998. Bioequivalence of topical dermatological dosage forms. Methods of evaluation of bioequivalence. Pharm Res 15:167–171. Benfeldt B, Growth G. 1998. Feasibility of measuring lipophilic or protein-bound drugs in the dermis by in vivo microdialysis after topical or systemic drug administration. Acta Derm Venereol 78:274– 278. Shah VP. 2001. Progress in methodologies for evaluating bioequivalence of topical formulations. Am J Clin Dermatol 2:275–280. Pershing LK, Nelson JL, Corlett JL, Shrivastava SP, Hare DB, Shah VP. 2003. Assessment of dermatopharmacokinetic approach in bioequivalence determination of topical tretinoin gel products. J Am Acad Dermatol 48:740–751. Alberti I, Kalia YN, Naik A, Bonny JD, Guy RH. 2001. Assessment and prediction of the cutaneous bioavailability of topical terbinafine in vivo. Pharm Res 18:1472–1475. Alberti I, Kalia YN, Naik A, Bonny JD, Guy RH. 2001. Effect of ethanol and isopropyl myristate on the availability of topical terbinafine in human stratum corneum, in vivo. Int J Pharm 219:11–19. Weigmann HJ, Lademann J, Schaefer H, Sterry W. 1999. Determination of the horny layer profile by spectroscopy in the visible range as a prerequisite to quantify percutaneous absorption. Skin Pharmacol Appl Skin Physiol 12:34–45. Shah VP. 2005. IV-IVC for topically applied preparations-a critical evaluation. Eur J Pharm Biopharm 60:309–314. [21Code of Federal Regulation. 320.22(b) (3)] Singh GJP. 2006. Evaluation of comparative performance of multisource locally acting inhalation aerosols and nasal sprays—An FDA Perspective. Am Pharm Rev (in press). Singh GJP, Adams WP. 2005. US regulatory and scientific considerations for approval of generic locally acting orally inhaled and nasal drug products. In: Dalby RN, Byron P, Peart J, Suman JD, editors. Respiratory drug delivery. A regulatory and analytical symposium. Richmond, VA: Virginia Commonwealth University, pp 115– 125. Guidance for the in-vivo bioequivalence studies of metaproterenol sulfate and albuterol inhalation aerosols (metered dose inhalers). 1989. Division of Bioequivalence, Food and Drug Administration, Rockville, Maryland.
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 96, NO. 4, APRIL 2007
754
MIDHA ET AL.
14. Interim guidance for documentation of in vivo bioequivalence of albuterol inhalation aerosols (metered dose inhalers). 1994. Office of generic drugs, US food and drug administration, Rockville, Maryland. 15. Pauwels RA, Lofdahl CG, Postma DS, Tatterfield AE, O’Byrne P, Barnes PJ, Ullman A. 1997. Effect of inhaled formoterol and budesonide on exacerbation of asthma. Fermoterol and Corticosteroid Establishing Therapy (FACET) International Study Group. New Engl J Med 337:1405– 1411. 16. Guidance for Industry (Draft): Bioavailability and bioequivalence studies for nasal aerosols and nasal sprays for local action. 2003. Center for drug evaluation and research, US food and drug administration, Rockville, Maryland. www.fda.gov/cder/ guidance/5383DFT. 17. Ahrens RC, Teresi ME, Han S-H, Donne D, Vanden Burgt JA, Lux CR. 2001. Asthma stability after oral prednisone—A clinical model for comparing inhaled corticosteroid potency. Am J Respir Crit Care Med 164:1138–1145. 18. Ahrens RC, Hendeles L, Teresi ME, Lux CR, Vanden Burgt JA, Hart KA, Ekholm BP. 2003. Relative potency of beclomethasone dipropionate (BDP) delivered by HFA-MDI, and fluticasone propionate (FP) delivered by Diskus. Eur Respir J 22:1576. 19. Smith IJ, Parry-Billing M. 2003. The inhalers of the future? A review of dry powder devices on the market. Pulmon Pharmacol Therap 16:79–95. 20. Crompton G. 1982. Problems patients have using pressurized aerosol inhalers. Eur J Respir Dis 119:101–104. 21. Larsen J, Hahn M, Elkholm B, Wick K. 1994. Evaluation of conventional press-and-breath metered-dose inhaler techniques in 501 patients. J Asthma 31:193–199. 22. Dunbar CA, Hickey AJ, Holzner P. 1998. Dispersion and characterization of pharmaceutical dry powder aerosols. KONA 16:7–14. 23. Guidance for Industry (Draft): Metered dose inhaler (MDI) and dry powder inhaler (DPI) drug products: CMC, manufacturing and control documentation. 1998. Center for drug evaluation and research, US Food and Drug Administration, Rockville, Maryland. 24. Blume HH, Midha KK. 1993. Bio-International 1992, Conference on bioavailability, bioequivalence, and pharmacokinetics studies. J Pharm Sci 82:1186–1189.
JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 96, NO. 4, APRIL 2007
25. Midha KK, Rawson MJ, Hubbard JW. 2005. The bioequivalence of highly variable drugs and drug products. Int J Clin Pharmcol Ther 43:485–598. 26. Tothfalusi L, Endrenyi L, Midha KK, Rawson MJ, Hubbard JW. 2001. Evaluation of bioequivalence of highly variable drugs. Pharm Res 18:728–733. 27. Tothfalusi L, Endrenyi L, Midha KK. 2003. Scaling or wider bioequivalence limits for highly variable drugs and for the special case of Cmax. Int J Clin Pharmacol Therap 41:217–225. 28. Tothfalusi L, Endrenyi L. 2003. Limits for scaled average bioequivalence of highly variable drugs and drug products. Pharm Res 20:382–389. 29. Chen ML, Jackson AJ. 1991. The role of metabolite in bioequivalency assessment. I. Linear pharmacokinetics without first-pass effect. Pharm Res 8:25– 32. 30. Tucker GT, Rostami A, Jackson PR. 1993. Metabolite measurement in bioequivalence studies: Theoretical considerations. In: Midha KK, Blume H, editors. Bio-International: Bioavailability, Bioequivalence and Pharmacokinetics. International Conference of FIP. ‘‘Bio-International 0 92’’ held in Bad Homburg, Germany, May 20–23, 1992. Stuttgart: Medpharm Scientific Publishers, pp 163–170. 31. Jackson AJ. 2000. The role of metabolites in bioequivalency assessment. III. Highly variable drugs with linear kinetics and first-pass effect. Pharm Res 17:1432–1436. 32. Midha KK, Rawson MJ, Hubbard JW. 2004. Commentary: The role of metabolites in bioequivalence. Pharm Res 21:1331–1334. 33. Gibaldi M, Boyes RN, Feldman S. 1971. Influence of first-pass effect on availability of drugs on oral administration. J Pharm Sci 60:1338–1340. 34. Amidon GL, Lennernas H, Shah VP, Crison JR. 1995. 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. 35. Guidance for Industry: Waiver of in vivo bioavailability and bioequivalence studies for immediaterelease dosage forms based on biopharmaceutics classification system. 2000. Center for drug evaluation and research, US Food and Drug Administration, Rockville, Maryland. 36. Wu C-Y, Benet LZ. 2005. Predicting drug disposition via application of BCS. Transport/absorption/ elimination interplay and development of a biopharmaceutics drug disposition classification system. Pharm Res 22:11–23. 37. WHO Working Document QAS/04.093. 2005.
DOI 10.1002/jps