Abstract
AIM
The aim of this study was to evaluate the acceptability of 124 bioequivalence (BE) studies with 80 active substances categorized according to the Biopharmaceutics Classification System (BCS) in order to establish if there were different probabilities of proving BE between the different BCS classes.
METHODS
We evaluated the differences between pharmaceutical products with active substances from different BCS classes in terms of acceptability, number of subjects in the study (n), the point estimates, and intra- and inter-subject coefficients of variation data from BE studies with generic products.
RESULTS
Out of 124 BE studies 89 (71.77%) were performed with pharmaceutical products containing active substances classified by the BCS. In all BCS classes there were non-bioequivalent pharmaceutical products: 4 out of 26 (15.38%) in class 1, 14 out of 28 (50%) in class 2, 3 out of 22 (13.63%) in class 3 and 1 out of 13 (7.69%) in class 4. When we removed those pharmaceutical products in which intra-subject variability was higher than predicted (2 in class 1 active substances, 9 in class 2 and 2 in class 3) there were still non-BE pharmaceutical products in classes 1, 2 and 3.
CONCLUSIONS
Comparisons between pharmaceutical products with active substances from the four BCS classes have not allowed us to define differential characteristics of each class in terms of n, inter and intra-subject variability for Cmax or AUC. Despite the usually employed test dissolution methodology proposed as quality control, pharmaceutical products with active substances from the four classes of BCS showed non-BE studies.
Keywords: bioequivalence studies, biopharmaceutics classification system
WHAT IS ALREADY KNOWN ABOUT THIS SUBJECT
In USA and Europe the classical way to guarantee equivalence between different formulations is the bioequivalence (BE) study based on in vivo bioavailability.
The Biopharmaceutic Classification System (BCS) classifies active substances into four different groups according to their aqueous solubility and intestinal permeability.
Recently the European Medicines Agency (EMA) has released a new bioequivalence guideline for immediate release solid oral dosage forms that includes recommendations on BCS-based biowaivers.
HAT THIS STUDY ADDS
As far as we know this is the first time that the probability of proving BE of more than one hundred BE studies with 80 active substances categorized according to BCS was evaluated.
Despite the usually employed test dissolution methodology proposed as quality control, pharmaceutical products with active substances from the four classes of BCS showed non-BE studies.
Introduction
In many developed countries, the main purpose of regulation related to the manufacture and distribution of medicinal products for human use is the safeguard of public health. In many countries the marketing authorization process for generic drugs is clearly established and relies on the demonstration of bioequivalence (BE). BE is defined as the absence of a significant difference in the availability of the active ingredient at the site of drug action [1]. In Europe, the European Medicines Agency (EMA) has stated that an ‘essentially similar medicinal product’ shall refer to a medicinal product which has the same qualitative and quantitative composition in terms of active substances and the same pharmaceutical form as the reference medicinal product, and whose bioequivalence with the reference medicinal product has been demonstrated by appropriate bioavailability studies [2].
In the USA [3] and Europe [4] the classical way to guarantee equivalence between different formulations is the so-called in vivo equivalence (bioequivalence) based on comparative in vivo bioavailability using pharmacokinetic endpoints.
Other countries have incorporated equivalence regulation, with various levels of requirements according to their specific national context and depending on many factors, such as the characteristics of the active drug substance, the health risk and the availability of resources to carry out a specific type of study [5]. In many places there is an interest in developing definitions and guidelines on when in vivo equivalence studies are necessary and when in vitro studies are acceptable. In this context, some countries are trying to develop regulatory systems to establish equivalence between different formulations supported by in vitro testing and Biopharmaceutics Classification System (BCS) considerations.
The BCS, introduced by Amidon et al. in 1995 [6], classifies active substances into four different groups according to their aqueous solubility at the highest dose in a volume of 250 ml and intestinal permeability. The idea is that when a drug formulated as an immediate release solid oral dosage form has an in vivo high solubility into the gastrointestinal tract and has a high permeability (class 1 drugs) the rate and extent of drug absorption is expected to be equivalent to the in vitro dissolution test, being 85% dissolved in less than 30 min with three different buffers, making in vivo bioequivalence demonstration unnecessary. This approach has been accepted by the FDA [7] and EMA [4] but is still rarely used, at least in Europe.
Nevertheless the EMA has gone one step further, and has released new bioequivalence guidelines that include recommendations on BCS-based [8] biowaivers. Biowaivers would be applicable to products containing BCS class 1 drugs which exhibit high solubility, complete absorption and rapid in vitro dissolution and also to pharmaceutical products containing substances exhibiting high solubility, limited absorption (BCS class 3) and very rapid in vitro dissolution, once the product has been previously justified for the biowaiver and has demonstrated other specific requirements concerning active substances and excipients [9–12]. Also the new EMA guideline said that ‘excipients that might affect bioavailability are qualitatively and quantitatively the same. In general, the use of the same excipients in similar amounts is preferred’. More than that, the WHO [13] had relaxed the solubility ratio and permeability criterion for class 1 and class 3 allowing too some bioexceptions in pharmaceutical products containing BCS class 2 drugs that are weak acids with dose solubility ratio of 250 ml or lower at 37°C over a pH range of 1.2–6.8. Such waivers would increase the speed and decrease the cost of bringing orally administered multisource therapeutics to market.
The aim of this study was to evaluate the characteristics in terms of acceptability, sample size, the point estimates, and intra- and inter-subject coefficients of variation of 80 active substances categorized according to the BCS tested in 124 BE studies with generic pharmaceutical products in order to establish if there were different probabilities of proving BE between the different BCS classes.
Methods
Studies
One hundred and twenty-four BE randomized crossover studies with 80 active substances were conducted on approximately 4500 healthy, young volunteers. The studies were all designed according to the requirements of EMA for BE studies for immediate release dosage forms [4] and developed in the Department of Pharmacology of the School of Medicine. The design and the reference medicinal product was always the same for each active substance but the number of subjects was different depending on previous data. All subjects gave written informed consent for participation in the BE studies, which were conducted according to the Helsinki declaration and the Spanish laws and were approved by the Ethics Committee of La Paz University Hospital. In every case, the BE study was developed after Spanish Medicines Agency approval of the protocol and the dissolution in vitro tests. In compliance with BE European requisites, in each study, the in vitro test dissolution methodology was in accordance with European pharmacopoeial requirements and EMA regulations. Some data from 108 of 124 studies presented here have previously been published [14].
Drug classification
The BCS classification of drugs was obtained from the Therapeutic System Research Laboratories website (http://www.tsrlinc.com/services/bcs/search.cfm) directed by Amidon et al. [15] (Table 1). Active substances in bold (Cmax) or underlined (AUC) were non-BE formulations). In order to classify drugs in BCS categories, we selected the ClogP-based permeability list because it classifies more drugs. A sensitivity analysis was performed excluding all drugs with discordant classification in ClogP vs. log P permeability measurements of Amidon vs. Spanish data from Takagi et al. [16]. The drugs excluded are listed in Table 2.
Table 1.
Biopharmaceutics Classification System (BCS) of active substances (number of studies)
| High solubility | Low solubility | Non-classified | ||
|---|---|---|---|---|
| High permeability | ||||
| Class 1 | Class 2 | |||
| Amlodipine (2) Bisoprolol (1) Cetirizine (2) * Citalopram/escitalopram (2) Donepezil (1) Doxazosin (1) Enalapril (2) Loratadine (1) Mirtazapine (1) Ondansetron (1) | Pravastatin (1)Quinapril (2) Ramipril (1) Sertraline (1) Sildenafil (1)Sildenafil (1) Terbinafin (2) Tramadol (1) Venlafaxine (1) Zolpidem (1) | Aceclofenac (1) Aceclofenac (1) Carbamazepine (1) Carvedilol (1) Clopidogrel (3) Ebastine (1) Ibuprofen (1+3) Ibuprofen (4) Irbesartan (1) | Irbesartan (1) Lamotrigine (1) Lorazepam (1) Lovastatin (1+2)Mycophenolate mofetil (1) Quetiapine (1) Risperidone (1)Risperidone (1) Simvastatin (1) | Amisulpride (1) Azyithromycin (2)Azythromycin (2) Bromazepam (1)Bromazepam (1) Clarithromycin (1)Clarithromycin (1) Clavulanate (2)Clavulanate (1) Deflazacort (1) |
| Low permeability | Deflazacort (1) | |||
| Class 3 Alendronic acid (3)† Anastrazole (1) Cefaclor (2) Codeine (1) Codeine (1) Fluconazole (1) Gabapentin (1) Isoniazid (1) Lamivudine (1) Letrozole (1) | Levetiracetam (1)*Levofloxacin (1) Lisinopril (1)Losartan (1) Pyrazinamide (1) Ranitidine (1)Risedronic acid (1) Terazosin (1) Topiramate (1) | Class 4 Acetaminophen (1) Amoxicillin (5) Cefixime (1) Cefuroxime axetil (1) Famotidine (2) Hydrochlorothiazide (2) Oxcarbazepine (1) | Dypirone (4) Flutamide (1)Leflunomide (1) Lercanidipine (1)Loperamide (1) Lormetazepam (2)Naproxen (1) Norfloxacine (1) Pentoxifylline (1)Piroxicam (1) Pramipexole (1) Prednisone (1) Repaglinide (2) Rifampicin (1)Ropirinol (1) Torasemide (1) |
Classification of drugs according to ClogP permeability of Biopharmaceutics Classification System (BCS) of Therapeutic System Research Laboratories web site (http://www.tsrlinc.com/services/bcs/search.cfm). Drugs in bold for Cmax and underlined for AUC correspond to non-bioequivalent parameters.
Therapeutic Equivalence Evaluation AA Code of FDA (Rest of codes were AB or none classified).
The primary results were maximum observed excretion rate (Uratemax) and accumulated elimination (Ae(0,∞)).
Table 2.
Drugs excluded in the sensitivity analysis
Pharmacokinetics
The pharmacokinetic (PK) analysis was performed using the WinNonlin 2.0 program (Pharsight Corporation, Cary, USA) from individual plasma or urine (three studies) concentration values on a non-compartmental extravascular input model. PK was expressed in terms of the rate and extent of absorption as characterized by the maximum observed plasma concentration (Cmax) and area under the concentration–time curve (AUC(0,∞)), or the truncated AUC as appropriate. In three studies (three alendronic acid formulations) the primary results were the maximum observed excretion rate (Uratemax) and the accumulated elimination (Ae(0,∞)). Only one analyte per study was included in the analysis, usually the parent drug. In seven BE studies the pharmaceutical products had two active ingredients and in one study three (isoniazid, pyrazinamide and rifampicin). In these cases all active substance were analyzed independently for BE purposes.
Statistical analysis
Standard analysis of bioequivalence
Methodology and statistical notations of BE studies are well described in the EMA guidelines [4]. These calculations were made with the WinNonlin 2.0 program. From the WinNonlin statistical analysis, we calculated the power a posteriori, T : R mean ratios, the 90% CI and the mean square error (MSE) of anova on the log scale,
The products were considered bioequivalent if the 90% confidence intervals (CI) of the log means of ratios of Cmax and AUC (Uratemax and Ae) fell within 80% and 125%.
Calculating the point estimate, inter and intra-subject variability
We showed the T : R mean ratios as a subtraction from 100, resulting in a point estimate. Inter-subject variability (CVInter-subject) was obtained exclusively from the reference formulation. Intrasubject variability (CVIntrasubject) was calculated from the MSE of the corresponding analysis of variance (anova) using the following formula (CVIntra-subject= SQRT(EXP(σi2) − 1), where σi2 denotes the residual (intra-subject) variance of the logarithmically transformed characteristics data [17].
Recalculation of the sample size
In the case of data from non-BE studies we recalculated the number of subjects (nr) using the following formula (nr= ((Z1–β+ Z1–α)2 2 σi2)/(lnθ− ln(1.25)2[18].
where σi2 is the within-subject variability on the log scale and nr is the retrospective number of subjects calculation. Let θ=µT/µR where µT and µR denote the mean parameters for test and reference, and let ln denote the natural logarithm. This equation provides reasonable approximations of the sample size for µT/µR ≠ 1, as the power of a study is very sensitive to the assumption about the point estimate. nr was calculated for a type I error of 5% and θ= 0.05.
Analysis of BCS groups
To analyze possible differences between formulations containing active substances from different BCS classes we evaluated the study number of subjects (n), point estimate and CVIntrasubject and the CVInter-subject with respect to Cmax and AUC in BE studies by means of a Student's t-test for two samples with unequal variances or a Mann-Whitney U-test for non-Gaussian distribution, respectively, using SPSS 15.0 Statistical Analysis Software (SPSS, Chicago, IL, USA).
Results
Out of 124 BE studies 89 (71.77%) were performed with active substances classified by the BCS (Table 1) and 77 (62.1%) were classified in the sensitivity analysis described in the methodology. Table 3 summarizes the number of products that fulfilled BE (for Cmax and AUC) vs. those not fulfilling BE, and it also shows the distribution (mean ± SD) of the number of subjects in the study (n), the point estimate, the CVIntra-subject and the CVInter-subject for Cmax and AUC of studies with active substances grouped by BCS.
Table 3.
BCS classification (ClogP-based) of active substances tested in 124 BE studies with generic pharmaceutical products for maximum observed plasma concentration (Cmax) and area under the concentration–time curve (AUC) or maximum observed excretion rate (Uratemax) and accumulated elimination (Ae(0,∞)) as appropriate
 |
In all BCS classes there were non-bioequivalent pharmaceutical products, reaching 50% in class 2. Thirty-five out of 124 pharmaceutical products did not fulfil BE criteria because Cmax exceeded limits, and 15 did not comply with the AUC. Comparisons between all classes have showed some statistical differences (Table 3).
Table 4 shows the results from non-BE pharmaceutical products for Cmax and AUC classified by active substance according to the four categories of BCS. We also have grouped these non-BE products according to differences <20% of point estimate (Group A) or >20% (Group B) (meaning differences between formulations), and by the power a posteriori >0.8 (Group 1) or <0.8 (Group 2) (assuming the validity of the results).
Table 4.
Results of non-BE formulations (study number of subjects, retrospective sample size calculation) classified by BCS active substances for maximum observed plasma concentration (Cmax) and area under the concentration-time curve (AUC)
| Point estimate | |||
|---|---|---|---|
| Group A (<20%) | Group B (>20%) | ||
| Power a posteriori | |||
| Group 1 (>0.8) | |||
| Class 1Cmax: Pravastatin (36,34), Zolpidem (36, 27) | Class 2Cmax: Clopidogrel-2 (82,84), Ibuprofen-2 (36, 9), Irbesartan (35, 23) AUC: Clopidogrel-1 (36, 54), Clopidogrel-2 (82, 51), Risperidone (34, 24) | Class 1 | Class 2Cmax: Ibuprofen-1 (24, 15), Ibuprofen-3 (35, 10), Ibuprofen-4 (24, 18), Lovastatin-1 (36,37), Risperidone (34, 27), Simvastatin (36, 43) AUC: Ibuprofen-3 (35, 5), Simvastatin (36, 30) |
| Class 3 | Class 4 | Class 3Cmax: Codeine (35, 34), Isoniazid (24, 25) | Class 4 |
| NCCmax: Bromazepam (33, 43), Deflazacort (35, 34), Flutamide (24, 35), Lormetazepam-2 (24, 25), Rifampicin (24, 15) AUC: Bromazepam (33, 26), Clavulanate-2 (24, 18) | NCCmax: Azythromycin-2 (35, 33), Clarithromycin (36, 32) AUC: Pentoxifiline (12, 12) | ||
| Group 2 (< 0.8) | |||
| Class 1Cmax: Sildenafil (35, 66), Loratadine (35, 69) | Class 2Cmax: Aceclofenac (23, 42), Clopidogrel-1 (36, 137), Clopidogrel-3 (54, 235), Lovastatin-2 (36, 65), Lovastatin-3 (36, 53) AUC: Clopidogrel-3 (54, 88), Lovastatin-1 (36, 63) | Class 1 | Class 2 |
| Class 3Cmax: Ranitidine (23, 50) AUC: Lisinopril (24, 52) | Class 4Cmax: Oxcarbazepine (24, 69) | Class 3AUC: Codeine (35, 47) | Class 4 |
| NCCmax: Azythromycin-1 (35, 70), Lercanidipine (35, 97) AUC: Flutamide (24, 35) | NCCmax: Clavulanate-1 (24, 233), Clavulanate-2 (24, 235), Lormetazepam-1 (24, 41), Pentoxifiline (12, 30) AUC: Clarithromycin (24, 32), Clavulanate-1 (24, 170), | ||
Formulation-Number of study with the same drug. Behind formulation appears between parenthesis, two numbers (n, nr), showing the study number of subjects (n) and retrospective sample size calculation (nr).
In a sensitivity analysis (Table 5) the removal of five studies because of discordances in the classification of active substances of the BCS of three BE and two non-BE active substances of class 1 changed the ratio BE : Total number of studies of products from 15.38% to 9.52% and the removal of another five BE active substances from class 4 improved the ratio from 12.5% to 5.69%. For class 2 and 3 only one active substance were excluded, respectively. Nevertheless sample size, point estimates, CVIntra-subject and CVInter-subject between classes continued to show the same significant differences, reinforcing our results. Excluding the three formulations of class 3 active substances where the primary results were urine analysis, because of extra biological variability, differences remained for the point estimate with respect to class 4 drugs for Cmax and we found no other significant differences in both analyses.
Table 5.
Sensitivity analysis, BCS classification (ClogP-based) of active substances tested in 124 BE studies for maximum observed plasma concentration (Cmax) and area under the concentration–time curve (AUC) or or maximum observed excretion rate (Urate max) and accumulated elimination (Ae(0,∞)), as appropriate, excluding discordances
 |
Moreover, when we removed those studies in which the intra-subject variability of formulations was higher than predicted (Table 4, Group A2, defined as those active substances from studies with insufficient power a posteriori and a point estimate <20%) (two with class 1, nine with class 2 and two with class 3 active substances), there were non-BE formulations with active substances from classes 1, 2 and 3 (Table 4).
Discussion
Cmax is the cornerstone for the demonstration of bioequivalence because almost always it is more variable than AUC. In fact, in our data from 124 formulations 35 did not comply with BE acceptance criteria because of Cmax and 15 because AUC exceeded the limits, and between these 14 were also out of limits for Cmax. For class 1, non-BE was always due to Cmax, not AUC.
Comparison between all classes showed some statistical differences in terms of study n, the point estimate, CVIntra-subject and the CVInter-subject for Cmax and AUC, but in our opinion these differences did not define differential characteristics between classes.
Although it is believed that class 1 active substances may not require in vivo BEQ demonstration, 4 of 26 pharmaceutical products containing class 1 active substances (two in the sensitivity analysis) did not comply with EMA requirements, particularly because Cmax exceeded the limits. In two cases (sildenafil and loratadine) the CVIntra-subject was higher than predicted (Group A2 Table 4). In another two cases (pravastatin and zolpidem) the differences in point estimates of Cmax were higher than expected, falling out of the 90% CI limits of 80–125%, with an intra-subject variability as predicted, meaning that the disposition between formulations was truly different (Group A1 Table 4). With our data, the positive predictive value (PPV) of the usually employed in vitro dissolution test for quality control purposes was 92% (95% CI 73, 99%) (after excluding two un-powered studies (group A2)) for formulations containing class 1 BCS drugs.
In our study, more than 50% of pharmaceutical products containing class 2 active substances failed the BE demonstration of Cmax (50% or 51.85% in the sensitivity analysis) or AUC (25% or 25.93% in the sensitivity analysis). Ibuprofen is a class 2 weak acid proposed for biowaiver by WHO [13]. We conducted BE studies with eight formulations of ibuprofen of which four resulted in non-BE, three of them because of dramatic differences between formulations. These formulations may have presented more problems in determining bioequivalence due to the influence of bioavailable excipients and the highly variable behaviour of these compounds or maybe because the inability of usually employed in vitro dissolution tests to predict in vivo behaviour. The PPV of conventional usually employed in vitro dissolution tests for formulations containing class 2 BCS drugs (excluding two un-powered studies (group A2)) was 61% (95% CI 39, 80%).
Characteristics of class 3 active substances are high solubility and low permeability. Permeation through the intestinal membrane is the rate limiting process for drug absorption. In our results, 13.63% of studies were non-BE, which is similar to class 1 (15.38%). This type of drug seems to be, in terms of differences between point estimates and intra-subject variability, between class 1 and 2. In one case (ranitidine) the intra-subject variability was higher than predicted (Group A2). In another two cases (codeine and isoniazid) the differences in point estimates of Cmax exceeded 20% (<80% or >125%). In these cases, the probability of acceptance is zero because of dramatic differences between formulations (Group B1 Table 4). The PPV of usually employed in vitro dissolution tests is 86% (95% CI 70, 99%) (excluding two unpowered studies (Group A2)) for formulations containing class 3 BCS drugs. Wu & Benet [19] developed a Biopharmaceutics Drug Disposition Classification System (BDDCS) based on the extent of metabolism, instead of permeability or extent of absorption. They think it inappropriate to have a biowaiver for this class until more is known about the effect of formulation components on intestinal transport and until there is a validated methodology to predict these effects except, perhaps, in the case of identical excipients [9].
Active substances from class 4 are of low solubility and low permeability and show significant problems for effective oral delivery. The number of pharmaceutical products that fall into this class depends on the precise limit used for the definition of permeability and solubility. Surprisingly, class 4 has the lowest percentage of non-bioequivalent formulations (7.69%, 0% in the sensitivity analysis). In terms of point estimates, class 4 showed the lowest value, a value significantly lower than class 2 and 3, although this fact should be considered cautiously because of the limited number of studies in our series. We note that these compounds showed the lowest inter-subject variability in AUC of all four BCS classes, significantly lower than class 1 (Table 3), which may be due to the absence of saturation of intestinal transport and enzymes affected by the inter-subject differences in their function. The PPV of conventional in vitro dissolution tests is 100% for formulations containing class 4 BCS drugs.
The BCS developed by Amidon et al. [6] with the objective to predict from a more specific in vitro dissolution test the bioavailability of a drug has produced an important impact based on the possibility to waive in vivo bioavailability and bioequivalence testing for immediate release solid dosage forms for class 1 (FDA, EMA and WHO) [7, 8, 13], for class 3 (EMA and WHO) [8, 13] and for a specific group of class 2 (WHO) [13]. This means there is the possibility of decreasing the cost of bringing multisource orally administered drugs to the market. Nevertheless, more than a decade after the pioneering work of Amidon et al., 35 (28%) of the formulations of our series are still un-classified and in another 10% more there are classification discordances. To this point we must add that FDA guidelines and EMA and WHO proposals to biowaive BE requirements also show discordances in definitions of high solubility (the decrease in pH from 7.5 in the FDA guidance to 6.8 in WHO and EMA proposals) and high permeability (the permeability criterion was relaxed from 90% in the FDA guidance to 85% in the WHO and EMA proposals).
The primary concern from the regulatory point of view is the protection of the patient against approval of products that are not bioequivalent. The current practice in the in vivo studies of carrying out two one-sided tests at the 0.05 level of significance ensures that there is no more than a 5% chance that a generic product that is not truly equivalent to the reference will be approved [20]. However, with our data, with the usually employed in vitro dissolution tests, except for class 4, there were more than 5% of false positives.
In fact, our data show that some products containing class 1, class 2 and class 3 active substances do not comply with the EMA regulation requisites for bioequivalence although these formulations satisfied the test dissolution methodology proposed as quality control. Therefore we suggest the possibility that drug manufacturers may not develop these in vitro dissolution tests properly, or perhaps these usually employed in vitro tests may be insufficient to allow a waiver of in vivo bioequivalence.
However we must also point out some limitations in the study design that limit the comparison between BCS classes such as the limited number of studies with class 4 drugs and the absence of a calculation of sample size to ensure the power of comparisons between classes.
In our opinion, the demonstration of bioequivalence is a major concern for the approval and the use of generic products, and the possibility of a biowaiver for in vivo BE studies needs to be approached with greater caution and careful supervision in order to guarantee the efficacy and safety of these medicines.
In conclusion, comparisons between pharmaceutical products with active substances from the four BCS classes have not allowed us to define differential characteristics of each class in terms of n, CVIntra-subject and CVInter-subject for Cmax or AUC. Despite the usually employed test dissolution methodology proposed as quality control, pharmaceutical products with active substances from the four classes of BCS showed non-BE studies. Pharmaceutical products in classes 1 and 3 were similar but failed to comply with BE requirements in nearly 15% of the cases. Products containing class 2 drugs were the less compliant with BE requirements, failing in at least 50% of cases. Moreover, once those formulations with a CVIntrasubject higher than predicted were excluded, there were non-BE products containing drugs from classes 1, 2 and 3.
Acknowledgments
We would like to acknowledge Dr Alfredo Garcia's invaluable advice and help with this study.
Author's contributions: Elena Ramirez conceived the study, participated in the design of the study, conducted the review of Phase I Clinical Trial Unit studies and the extraction of data, the performance of the statistical analysis, the interpretation of the results and the writing of the paper. Olga Laosa participated in the design of the study, in the review and interpretation of the results and the writing of the paper. Pedro Guerra participated in the review and interpretation of the results and the writing of the paper. Blanca Duque participated in the review of Phase I Clinical Trial Unit studies, the extraction of data, in the review of the results and the writing of the paper. Beatriz Mosquera participated in the review of the Phase I Clinical Trial Unit studies, the extraction of data, in the review of the results and the writing of the paper. Alberto M. Borobia participated in the performance of the statistical analysis and in the review of the results. Suhua H. Lei participated in the in the performance of the statistical analysis and in the review of the results. Antonio J. Carcas participated in the interpretation of the results and the writing of the paper. Jesús Frias participated in the design of the study, in the review and interpretation of the results and the writing of the paper.
Competing interests
All authors have been working in the Phase I Clinical Trial Unit of the Clinical Pharmacology Service at La Paz University Hospital and the School of Medicine, Autonomous University of Madrid, Spain.
REFERENCES
- 1.FDA. 21CFR320.1 PART 320 – Bioavailability and Bioequivalence requirements. 2000. Revised as of April 1, 2003.
- 2.Directive 2004/27/EC of the European Parliament and the Council of 31 March 2004 amending Directive 2001/83/EC on the Community code relating to medicinal products for human use Official Journal L – 136, 30/04/2004. pp. 34–57.
- 3.U.S. Food and Drug Administration, Center for Drug Evaluation and Research. Guidance for Industry: Bioavailability and Bioequivalence Studies for Orally Administered Drug Products– General Considerations, Office of Training and Communications, Division of Communications Management, Drug Information Branch, HFD-210. 2003. Rockville MD 20857, March.
- 4.EMEA/CPMP/EWP/1401/98. Note for guidance on the investigation of bioavailability and bioequivalence. 2001. EMEA, London, 26 July.
- 5.Pan American Health Organization. Regional Office of the World Health Organization. Science based criteria for bioequivalence in vivo and in vitro, bio-waivers, and strategic framework for implementation. Draft. Available at: http://www.paho.org/english/ad/ths/ev/be-doct-draft-eng.pdf (last accessed 7 March 2010.
- 6.Amidon GL, Lennernäs H, Shah VP, Crison JR. A theorical basis for a biopharmaceutic drug classification: the correlation of in vitro drug product dissolution and in vivo bioavailability. Pharm Res. 1995;12:413–20. doi: 10.1023/a:1016212804288. [DOI] [PubMed] [Google Scholar]
- 7.Rockville MD. U.S. Food and Drug Administration, Guidance for Industry. Food and Drug Administration. Waiver of in vivo bioavailability and bioequivalence studies for immediate release solid oral dosage forms based on a biopharmaceutics classification system. 2000. Available at: http://www.fda.gov/RegulatoryInformation/Guidances/default.htm (last accessed 7 March 2010)
- 8.EMEA/CPMP/EWP/QWP/1401/98 Rev.1/Corr*. Guideline on the investigation of bioequivalence. London. 2010. 20 January. Available at: http://www.ema.europa.eu/docs/en_GB/document_library/Scientific_guideline/2010/01/WC500070039.pdf (last accessed 7 March 2010.
- 9.Yu LX, Amidon GL, Polli GL, Zhao H, Mehta MU, Conner DL, Shah VP, Lesko LJ, Chen ML, Lee VHL, Hussain AS. Biopharmaceutics classification system: the scientific basis for biowaiver extensions. Pharm Res. 2002;19:921–5. doi: 10.1023/a:1016473601633. [DOI] [PubMed] [Google Scholar]
- 10.Blume HH, Schug BS. The biopharmaceutics classification system (BCS): class III drugs-better candidate for BA/BE waiver? Eur J Pharm Sci. 1999;9:117–21. doi: 10.1016/s0928-0987(99)00076-7. [DOI] [PubMed] [Google Scholar]
- 11.Polli JE, Yu LX, Cook JA, Amidon GL, Borchardt RT, Burnside BA, Burton PS, Chen ML, Conner DP, Faustino PJ, Hawi AA, Hussain AS, Joshi HN, Kwei G, Lee VHL, Lesko LJ, Lipper RA, Loper AE, Nerurkar SG, Polli JW, Sanvordeker DR, Taneja R, Uppoor RS, Vattikonda CS, Wilding I, Zhang G. Summary workshop report: biopharmaceutics classification system-implementation challenges and extension opportunities. J Pharm Sci. 2004;93:1375–81. doi: 10.1002/jps.20064. [DOI] [PubMed] [Google Scholar]
- 12.Cheng CL, Yu LX, Lee HL, Yang CY, Lue CS, Chou CH. Biowaiver extension potential to BCS Class III high solubility-low permeability drugs: bridging evidence for metformin immediate-release tablet. Eur J Pharm Sci. 2004;22:297–304. doi: 10.1016/j.ejps.2004.03.016. [DOI] [PubMed] [Google Scholar]
- 13.Proposal to waive in vivo bioequivalence requirements for WHO Model List of Essential Medicines immediate-release, solid oral dosage forms. Technical Report Series, No 937, 40th Report, Annex 8 of WHO Expert committee on specifications for pharmaceutical preparations. Available at: http://whqlibdoc.who.int/trs/WHO_TRS_937_eng.pdf (last accessed 7 March 2010.
- 14.Ramirez E, Guerra P, Laosa O, Duque B, Tabares B, Lei AS, Carcas AJ, Frias J. The importance of sample size, log-mean ratios, and intra-subject variability in the acceptance criteria of bioequivalence studies. Eur J Clin Pharmacol. 2008;64:783–93. doi: 10.1007/s00228-008-0476-8. [DOI] [PubMed] [Google Scholar]
- 15.Biopharmaeutics Classification System (BCS) Therapeutic System Research Laboratories (TSRL inc.) Available at: http://www.tsrlinc.com/services/bcs/search.cfm.
- 16.Takagi T, Ramadrasekharan C, Bermejo M, Yamashita S, Yu LX, Amidon LG. A provisional Biopharmaceutical Classification of the Top 200 oral drugs products in the Unites States, Great Britain, Spain and Japan. Mol Pharm. 2006;3:631–43. doi: 10.1021/mp0600182. [DOI] [PubMed] [Google Scholar]
- 17.Hauschke D, Steinijans VW, Diletti E, Burke M. Sample size determination for bioequivalence assessment using a multiplicative model. J Pharmacokinet Biopharm. 1992;20:557–61. doi: 10.1007/BF01061471. [DOI] [PubMed] [Google Scholar]
- 18.Julious SA. Tutorial in biostatistics. Statistic Med. 2004;23:1921–86. doi: 10.1002/sim.1783. [DOI] [PubMed] [Google Scholar]
- 19.Wu CY, Benet LZ. Predicting drug disposition via application of BCS: Transport /absorption /elimination interplay and development of a biopharmaceutics drug disposition classification system. Pharm Res. 2005;22:11–23. doi: 10.1007/s11095-004-9004-4. [DOI] [PubMed] [Google Scholar]
- 20.Orange Book Preface, Statistical Criteria for Bioequivalence. Approved Drug Products with Therapeutic Equivalente Evaluations. 2009. 29th Edition. U.S. Food and Drug Administration. Center for Drug Evaluation and Research. 2009-06–18. Available at: http://www.fda.gov/drugs/developmentapprovalprocess/ucm079068.htm.