Regulatory requirements for BE and Existing Guidelines Alfredo García – Arieta, PhD WHO Workshop on Assessment of Bioequivalence Data, 31 August – 3 September, 2010, Addis Ababa

Subjects: Drop-outs and withdrawals Healthy volunteers: inclusion / exclusion criteria Sufficient number to compensate for drop-outs or withdrawals Replacement of subjects complicates the statistical model and analysis Reason for withdrawal (e.g. AE or personal reason) must be reported Protocol must state replacement policy or add-on designs

Subjects: Drop-outs and withdrawals Recruit a few subject more to compensate for drop-outs or withdrawals Protocol should state whether samples from this extra subjects will be assayed if not required for stat. analysis EU EMA, on the contrary: The data from all treated subjects should be treated equally It is not acceptable to have a protocol which specifies that ‘spare’ subjects will be included in the analysis only if needed as replacements for other subjects who have been excluded It should be planned that all treated subjects should be included in the analysis, even if there are no drop-outs.

EU Subject Accountability Ideally, all treated subjects should be included in the statistical analysis. However, subjects in a crossover trial who do not provide evaluable data for both of the test and reference products (or who fail to provide evaluable data for the single period in a parallel group trial) should not be included. The analysis for each comparison should be conducted excluding the data from the treatments that are not relevant for the comparison in question (e.g. a three period study including two references, US and EU).

Exclusion of subjects in EU Unbiased assessment of results from randomised studies requires that all subjects are observed and treated according to the same rules These rules should be independent from treatment or outcome In consequence, the decision to exclude a subject from the statistical analysis must be made before bioanalysis In principle any reason for exclusion is valid provided it is specified in the protocol and the decision to exclude is made before bioanalysis However the exclusion of data should be avoided, as the power of the study will be reduced and a minimum of 12 evaluable subjects is required

Exclusion of subjects in EU Examples Vomiting Diarrhoea the use of concomitant medication, in exceptional cases,. It should be noted in the CRF as the study is being conducted Clearly described and listed in the study report Exclusion of data cannot be accepted on the basis of statistical analysis or for pharmacokinetic reasons alone, because it is impossible to distinguish the formulation effects from other effects influencing the pharmacokinetics.

Exception where subject can be excluded A subject with lack of any measurable concentrations or only very low plasma concentrations for reference Very low plasma concentrations: if its AUC is less than 5% of reference medicinal product geometric mean AUC (which should be calculated without inclusion of data from the outlying subject) The exclusion of data due to this reason will only be accepted in exceptional cases and may question the validity of the trial.

Exceptions where subject must be excluded Subjects with non-zero baseline concentrations > 5% of Cmax Excluded from bioequivalence calculation to avoid carry- over effects In case of immediate release products it can be caused by: subject non-compliance (mouth check) insufficient wash-out period (sufficient wash-out) The samples from subjects excluded from the statistical analysis should still be assayed and the results listed

Not to exclude Subjects whose AUC(0-t) covers less than 80% of AUC(0- ∞) But if the percentage is less than 80% in more than 20% of the observations then the validity of the study may need to be discussed This does not apply if the sampling period is 72 h or more and AUC(0-72h) is used instead of AUC(0-t)

Add-on designs If bioequivalence cannot be demonstrated because: Larger variability than expected Larger difference than expected but still within 20% An add-on design can be performed using not less than half the number of subjects in the initial study Provided this was anticipated in the protocol Combining data only if: Same protocol Same batches

Add-on designs Add-on designs must be given appropriate statistical treatment Canadian add-on design requires: No interaction formulation effect x study phase No statistical difference in variability of both phases But the consumer’s risk is increased Statistically inadequate It must be a sequential design with control of the alpha expenditure (EMA)

Two-stage design An initial group of subjects can be treated and their data analysed. If bioequivalence has not been demonstrated an additional group can be recruited and the results from both groups combined in a final analysis. If this approach is adopted appropriate steps must be taken to preserve the overall type I error of the experiment and the stopping criteria should be clearly defined prior to the study. The analysis of the first stage data should be treated as an interim analysis and both analyses conducted at adjusted significance levels (with the confidence intervals accordingly using an adjusted coverage probability which will be higher than 90%)

Two-stage design For example, using 94.12% confidence intervals for both the analysis of stage 1 and the combined data from stage 1 and stage 2 would be acceptable, but there are many acceptable alternatives and the choice of how much alpha to spend at the interim analysis is at the company’s discretion. The plan to use a two-stage approach must be pre- specified in the protocol along with the adjusted significance levels to be used for each of the analyses. When analysing the combined data from the two stages, a term for stage should be included in the ANOVA model.

Genetic Phenotyping Phenotyping for metabolizing activity for high-clearance drugs that are metabolized by enzymes that are subject to genetic polymorphism, e.g. propranolol Phenotyping of subjects can be considered for studies of drugs that show phenotype-linked metabolism and for which a parallel group design is to be used, because it allows fast and slow metabolizers to be evenly distributed in the two groups of subjects Phenotyping could also be important for safety reasons, determination of sampling times and wash-out periods in cross-over design studies

Study standardization Standardization of study conditions is important to minimize the magnitude of variability other than in the pharmaceutical products Standardization should cover: Exercise Diet fluid intake Posture Restriction of the intake of alcohol, caffeine, certain fruit juices and concomitant medicines for a specified time period before and during the study Volunteers should not take any other medicine, alcoholic beverages or over the-counter (OTC) medicines and supplements for an appropriate interval either before or during the study

Study standardization In the event of emergency, the use of any non-study medicine must be reported (dose and time of administration). Physical activity and posture should be standardized as far as possible to limit their effects on gastrointestinal blood flow and motility The same pattern of posture and activity should be maintained for each day of the study The time of day at which the study drug is to be administered should be specified All meals should be standardized and the composition stated in the study protocol and report.

Selection of dose In bioequivalence studies the molar equivalent dose of multisource and comparator product must be used If possible If not, potency correction (see Canadian Guideline) Applicable if PK is dose-proportional Generally the marketed strength with the greatest sensitivity to bioequivalence assessment should be administered as a single unit

Selection of dose This will usually be the highest marketed strength Drug with low solubility and dose-proportional PK Drug with more than proportional increase in PK (AUC) R T

Selection of dose A lower strength in the linear part if PK shows a less than proportional increase with dose due to saturation of absorption process Both a lower strength in the linear part and the highest strength if PK shows a less than proportional increase with dose due to solubility limitations

Selection of dose A higher dose (i.e. more than one dosage unit) may be employed when analytical difficulties exist In this case the total single dose should not exceed the maximal daily dose of the dosage regimen Alternatively, the application of area under the curve (AUC) truncated to 3 × median tmax of the comparator formulation would avoid problems of lack of assay sensitivity in many cases In certain cases a study performed with a lower strength can be considered acceptable if this lower strength is chosen for reasons of safety.

Sample fluids and their collection Usually blood should be the biological fluid sampled In most cases the analyte is measured in serum or plasma Urine only if the analyte cannot be measured in plasma, etc. Not necessary that “the API is excreted predominantly unchanged in the urine” The volume of each sample must be measured at the study centre, where possible immediately after collection, and included in the report In most cases the exclusive use of urine excretion data should be avoided as this does not allow estimation of the tmax and Cmax Samples should be processed and stored under conditions that have been shown not to cause degradation of the analytes The sample collection methodology must be specified in the study protocol.

Parameters to be assessed The shape of the plasma concentration versus time curves is assessed by: Cmax is peak exposure AUC is extent of exposure Tmax is the time of peak exposure Focus now on exposure instead of absorption AUC is extent of absorption if PK is linear /not saturated Tmax and Cmax gives some information on rate of absorption

Primary parameters to be assessed For single-dose studies, the following parameters should be measured or calculated: AUC0–t, where t is the last sampling time point with a measurable concentration in the individual formulation tested. The method of calculating AUC-values should be specified. In general AUC should be calculated using the linear/log trapezoidal integration method. In practice linear/linear trapezoidal rule is more frequent The use of compartmental-based parameters is not recommended. Cmax is the maximum or peak concentration observed representing peak exposure of API (or metabolite) in plasma, serum or whole blood.

Secondary parameters to be assessed AUC0-inf representing total exposure, where AUC0- = AUC0–t + Clast/; Clast is the last measurable drug concentration  is the terminal or elimination rate constant tmax is the time after administration of the drug at which Cmax is observed. Additional information: T1/2 is the plasma (serum, whole blood) half-life

Application of truncated AUC in BE In US-FDA and EU: AUC0-72 h for drugs with long half-life There are two important advantages to the use of truncated areas: more blood samples can be clustered around tmax to give greater precision in the estimation of both tmax and Cmax; high assay sensitivity to define the disposition phase is not required. The applicability of the truncated AUC approach merits particular consideration in the following cases: where low concentrations occur in the terminal portion of the plasma concentration versus time curve, which may not be quantifiable by means of an adequately validated, sensitive analytical method; and for products of APIs with long half-lives.

Application of truncated AUC in BE When considering BE of immediate-release formulations for systemic delivery, the most important portion of the plasma concentration versus time curve is until the absorption phase is complete On the other hand, the disposition phase does not illustrate formulation differences between the multisource product and comparator product in the bioequivalence decision-making process Gaureault examined the use of partial (truncated) AUC using Monte Carlo simulations and found a high degree of concordance between the bioequivalence decision based on the partial area truncated to four times tmax and the area extrapolated to infinity The evidence suggests that for immediate-release formulations it is unnecessary to take blood samples beyond four times tmax.

Parameters to be assessed in SS and Urine For steady-state studies (prolonged release): AUC is AUC over one dosing interval () at steady-state; Cmax; Cmin is concentration at the end of a dosing interval; Peak trough fluctuation is percentage difference between Cmax and Cmin Immediate release: only AUC and Cmax. When urine samples are used Cumulative urinary recovery (Ae) instead of AUC Maximum urinary excretion rate instead of Cmax.

Parameters to be assessed in SS For steady-state studies (prolonged release): AUC Cmax Cmin PTF

Analyte: parent or metabolite Generally, evaluation of pharmacokinetic bioequivalence will be based upon the measured concentrations of the parent drug released from the dosage form rather than the metabolite The concentration–time profile of the parent drug is more sensitive to changes in formulation performance than a metabolite, which is more reflective of metabolite formation, distribution and elimination.

Analyte: parent or metabolite In some situations it may be necessary to measure metabolite concentrations rather than those of the parent drug: The measurement of concentrations of therapeutically active metabolite is acceptable if the substance studied is a pro-drug. Measurement of a metabolite may be preferred when concentrations of the parent drug are too low to allow reliable analytical measurement in blood, plasma or serum for an adequate length of time, or when the parent compound is unstable in the biological matrix.

Analyte: parent or metabolite It is important to state a priori in the study protocol which chemical entities (pro-drug, drug (API) or metabolite) will be analysed in the samples. It is important to note that measurement of one analyte, API or metabolite, carries the risk of making a type-I error (the consumer risk) to remain at the 5% level However, if more than one of several analytes is selected retrospectively as the bioequivalence determinant, then both the consumer and producer risks change. When measuring the active metabolites wash-out period and sampling times may need to be adjusted to enable adequate characterization of the pharmacokinetic profile of the metabolite.

Enantiomers A non-stereoselective assay is currently acceptable for most pharmacokinetic bioequivalence studies When the enantiomers have very different pharmacological or metabolic profiles, assays that distinguish between the enantiomers of a chiral API may be appropriate Stereoselective assay is also preferred when systemic availability of different enantiomers is demonstrated to be non-linear

Enantiomers in US FDA For BE studies, the racemate using an achiral assay. Measurement of individual enantiomers in BE studies is recommended only when all of the following conditions are met: the enantiomers exhibit different PD characteristics, the enantiomers exhibit different PK characteristics, primary efficacy and safety activity resides with the minor enantiomer, and Nonlinear absorption is present (as expressed by a change in the enantiomer concentration ratio with change in the input rate of the drug) for at least one of the enantiomers

Enantiomers in EU The use of achiral bioanalytical methods is generally acceptable. However, the individual enantiomers should be measured when all the following conditions are met: the enantiomers exhibit different pharmacokinetics the enantiomers exhibit pronounced difference in pharmacodynamics the exposure (AUC) ratio of enantiomers is modified by a difference in the rate of absorption.

Enantiomers in EU The individual enantiomers should also be measured if the above conditions are fulfilled or are unknown If one enantiomer is pharmacologically active and the other is inactive or has a low contribution to activity, it is sufficient to demonstrate bioequivalence for the active enantiomer

Statistical Analysis: Concern The primary concern in BE assessment is to limit the risk of a false declaration of equivalence Statistical analysis of the BE trial should demonstrate that a clinically significant difference in bioavailability between the multisource product and the comparator product is unlikely

Statistical Analysis: Method The statistical procedures should be specified in the protocol before the data collection starts The statistical method for testing PK BE is based upon the determination of the 90% confidence interval around the ratio of the log-transformed population means (multisource/comparator) for the pharmacokinetic parameters under consideration and by carrying out two one-sided tests at the 5% level of significance

Statistical Analysis: Conclusion To establish PK BE, the calculated confidence interval should fall within a preset bioequivalence limit 80-125% for the 90% CI is the conventional acceptance range 80-125% for the 90% CI for AUC and Cmax always in FDA Except scaling for HVD based on intra-subject variability 80-125% for 90 CI% of AUC and for PE of Cmax in Canada 90-111% for 90% CI of AUC and 80-125% for CI of Cmax in Canada 80-125% for 90% CI of AUC and Cmax normally in EU Except scaling for Cmax in HVD based on intra-subject variability Narrowing to 90-111 in AUC and/or Cmax in NTI drugs

Statistical Analysis: Why 0.80-1.25? The procedures should lead to a decision scheme which is symmetrical with respect to the two formulations (i.e. leading to the same decision whether the multisource formulation is compared to the comparator product or the comparator product to the multisource formulation)

Statistical Analysis: log-transformation All concentration-dependent pharmacokinetic parameters (e.g. AUC and Cmax) should be log-transformed using either common logarithms to the base 10 or natural logarithms The choice of common or natural logs should be consistent and should be stated in the study report.

Statistical Analysis: ANOVA Logarithmically transformed, concentration-dependent pharmacokinetic parameters should be analysed using analysis of variance (ANOVA) Usually the ANOVA model includes the formulation, period, sequence or carry-over and subject factors Parametric methods, i.e. those based on normal distribution theory, are recommended for the analysis of log-transformed bioequivalence measures

Statistical Analysis: in log scale The general approach is to construct a 90% confidence interval for the quantity μT−μR and to reach a conclusion of pharmacokinetic equivalence if this confidence interval is within the stated limits The nature of parametric confidence intervals means that this is equivalent to carrying out two one-sided tests of the hypothesis at the 5% level of significance The antilogs of the confidence limits obtained constitute the 90% confidence interval for the ratio of the geometric means between the multisource and comparator products

Statistical Analysis The same procedure should be used for analysing parameters from steady state trials or cumulative urinary recovery, if required: For steady-state studies: AUC is AUC over one dosing interval () at steady-state; Cmax; Cmin is concentration at the end of a dosing interval; Peak trough fluctuation is percentage difference between Cmax and Cmin. When urine samples are used Cumulative urinary recovery (Ae) instead of AUC Maximum urinary excretion rate instead of Cmax.

Statistical Analysis: Tmax For tmax descriptive statistics should be given If tmax is to be subjected to a statistical analysis this should be based on non-parametric methods and should be applied to untransformed data For drugs where onset of action is important EU before 90% CI, but now not required 90% but PE A sufficient number of samples around predicted maximal concentrations should have been taken to improve the accuracy of the tmax estimate

Statistical Analysis: other parameters For parameters describing the elimination phase (T1/2) only descriptive statistics should be given

Statistical Analysis: outliers Methods for identifying and handling of possible outlier data should be specified in the protocol. Medical or pharmacokinetic explanations for such observations should be sought and discussed As outliers may be indicative of product failure, post hoc deletion of outlier values is generally discouraged. An approach to dealing with data containing outliers is to apply distribution-free (non-parametric), statistical methods No longer valid or acceptable No parametric methods ignore outlier values and it is equivalent to remove them

Statistical Analysis: non-parametric If the distribution of log-transformed data is not normal, non-parametric statistical methods can be considered No longer valid Outliers make the distribution non-normal Only tmax is analyzed with non-parametric methods ANOVA is robust despite deviations from Normality The justification of the intent to use nonparametric statistical methods should be included a priori in the protocol.

Acceptance ranges: AUC-ratio The 90% confidence interval for this measure of relative bioavailability should lie within a bioequivalence range of 0.80–1.25 If the therapeutic range is particularly narrow, the acceptance range may need to be reduced based on clinical justification Canada: 90.00-111.11% EU: 90.00-111.11.% A larger acceptance range may be acceptable in exceptional cases if justified clinically HVD in US FDA (not in Guidance for Industry)

Acceptance ranges: Cmax-ratio In general the acceptance limit 0.80–1.25 should be applied However, this measure of relative bioavailability is inherently more variable, and in certain cases a wider acceptance range (e.g. 0.75– 1.33) may be acceptable EU for HVD only widening in Cmax based on intra-subject variability The range used must be defined prospectively and should be justified, taking into account safety and efficacy considerations. In exceptional cases, a simple requirement for the point estimate to fall within bioequivalence limits of 0.80–1.25 may be acceptable with appropriate justification in terms of safety and efficacy Canada Not statistically correct.

Acceptance ranges: tmax-difference Statistical evaluation of tmax makes sense only if there is a clinically relevant claim for rapid onset of action or concerns about adverse effects The nonparametric 90% confidence interval for this measure of relative bioavailability should lie within a clinically relevant range. EU presently only based on Point Estimate For other pharmacokinetic parameters the same considerations as outlined above apply

Fixed-dose combination products The study design should follow the same general principles as described in previous sections The multisource FDC product should be compared with the pharmaceutically equivalent comparator FDC product In certain cases (e.g. when no comparator FDC product is available on the market) separate products administered in free combination can be used as a comparator

Fixed-dose combination products Sampling times should be chosen to enable the pharmacokinetic parameters of all APIs to be adequately assessed. The bioanalytical method should be validated on respect to all compounds measured Statistical analyses should be performed with pharmacokinetic data collected on all active ingredients; the 90% confidence intervals of test/comparator ratio of all active ingredients should be within acceptance limits.

Highly variable drugs: The problem A “highly variable drug” has been defined as an API with a within- subject variability of > 30% in terms of the ANOVA-CV Moreover “highly variable drugs” are generally safe drugs with shallow dose–response curves Proving the bioequivalence of medicinal products containing “highly variable drugs” is problematic because the higher the ANOVA-CV, the wider the 90% confidence interval Thus large numbers of subjects must be enrolled in studies involving highly variable drugs to achieve adequate statistical power

Highly variable drugs: The solutions Some regulatory authorities permit the use of broadened bioequivalence limits provided there is adequate justification For example, the regulatory agency could broaden the bioequivalence limits from 0.8–1.25 to 0.75–1.33 taking into consideration the therapeutic category of the drug In EU previously. Not presently Now acceptance range is widen based on intra-subject variability if clinically justified for Cmax only

Highly variable drugs: The solutions Some regulatory authorities permit the use of scaling to broaden the bioequivalence limits South-Africa In a two-period design, the limits are scaled to the residual standard deviation, or in a replicate design, to the within-subject standard deviation of the comparator formulation In US-FDA and EU a replicate design in necessary to estimate the real intra-subject variability of the comparator

Highly variable drugs: Japan’s solution Some regulatory authorities allow the following acceptance criteria: “Products are considered to be bioequivalent, if the 90% confidence interval of average ratios of AUC and Cmax between test and reference products is within the acceptable range of 0.8–1.25 If the confidence interval is not in the above range, test products are accepted as bioequivalent, if the following three conditions are satisfied: the total sample size of the initial bioequivalence study is not less than 20 (n = 10/group) or pooled sample size of the initial and add-on subject studies is not less than 30 the ratio of geometric least squares means of AUC and Cmax between the multisource and comparator product are between 0.9 and 1.11; and dissolution rates of test and reference products are evaluated to be equivalent under all dissolution testing conditions This rule cannot be applied to slowly dissolving products from which less than 80% of a drug dissolves within the final testing time (2 hr in pH 1.2 medium and 6 hr in others) under any conditions of the dissolution tests described

Highly variable drugs: Canada’s solution Some regulatory authorities do not allow for any adjustments In Canada Cmax variability is not a problem because it is usually assessed with Point Estimate In US-FDA no widening if the drug is not HVD Now in EU no widening for AUC Now in EU no widening for Cmax if the drug is not HVD or if it is not clinically justified

Highly variable drugs: Proposal AUC is usually critical: no widening despite variability Like Canada and EU Cmax can be widening if the comparator Cmax is HV A clinical justification is necessary (EU) A replicate design is necessary (EU and US-FDA) Predefined in the protocol Proportionality constant of EU is statistically more adequate than the one of US-FDA with respect to consumer’s risk Endrenyi L, Tothfalusi L. Regulatory conditions for the determination of bioequivalence of highly variable drugs. J Pharm Pharm Sci 2009; 12(1):138-149

Limits at CV = 30 or 25%

Highly variable drugs: EU Highly variable drug products (HVDP) are those whose intra-subject variability for a parameter is larger than 30% If an applicant suspects that a drug product can be considered as highly variable in its rate and/or extent of absorption, a replicate cross-over design study can be carried out Any replicate design is acceptable The FDA recommends a specific one (RTR/TRR)

Highly variable drugs: EU Those HVDP for which a wider difference in Cmax is considered clinically irrelevant based on a sound clinical justification can be assessed with a widened acceptance range If this is the case the acceptance criteria for Cmax can be widened to a maximum of 69.84 – 143.19% For the acceptance interval to be widened the bioequivalence study must be of a replicate design where it has been demonstrated that the within-subject variability for Cmax of the reference compound in the study is >30% The applicant should justify that the calculated intra-subject variability is a reliable estimate and that it is not the result of outliers

Highly variable drugs: EU If this is the case the acceptance criteria for Cmax can be widened to a maximum of 69.84 – 143.19% The extent of the widening is defined based upon the within-subject variability seen in the bioequivalence study using scaled-average-bioequivalence according to [U, L] = exp [±k·sWR], Where U is the upper limit of the acceptance range, L is the lower limit of the acceptance range, k is the regulatory constant set to 0.760 and sWR is the within-subject standard deviation of the log-transformed values of Cmax of the reference product

Highly variable drugs: EU The table below gives examples of how different levels of variability lead to different acceptance limits using this methodology

Highly variable drugs: EU The geometric mean ratio (GMR) should lie within the conventional acceptance range 80.00-125.00%. The possibility to widen the acceptance criteria based on high intra-subject variability does not apply to AUC where the acceptance range should remain at 80.00 – 125.00% regardless of variability It is acceptable to apply either a 3-period or a 4-period crossover scheme in the replicate design study

Narrow Therapeutic Index Drugs It is difficult to define NTI drugs US-FDA list in SUPAC guideline JAPAN list is also available in BE guideline Canada Guideline on Critical Dose Drugs “Critical dose drugs” are defined as those drugs where comparatively small differences in dose or concentration lead to dose-and concentration-dependent, serious therapeutic failures and/or serious adverse drug reactions which may be persistent, irreversible, slowly reversible, or life threatening, which could result in inpatient hospitalization or prolongation of existing hospitalization, persistent or significant disability or incapacity, or death. Adverse reactions that require significant medical intervention to prevent one of these outcomes are also considered to be serious Cyclosporine , Digoxin, Flecainide, Lithium, Phenytoin, Sirolimus, Tacrolimus, Theophylline, Warfarin EU: case by case Ciclosporine: AUC and Cmax should be narrowed in both fed and fasting state studies Tacrolimus: AUC should be narrowed

Narrow Therapeutic Index Drugs FDA 80.00 – 125% Canada 90% CI AUC T/R should be within 90.0 to 112.0% 90% CI Cmax T/R should be between 80.0 and 125.0% Both the fasted and fed states EU In specific cases, the acceptance interval for AUC should be tightened to 90.00-111.11% Where Cmax is of particular importance for safety, efficacy or drug level monitoring the 90.00-111.11% acceptance interval should also be applied for this parameter

Thank you very much for your attention!